An Essay Defending the View that Diagnosis of Disease is not Objective, and that Diagnosis of Mental Illness is Significantly Normative

Disease/Illness: Messy and Unavoidably Normative

The concept of “illness” is far more vexed and confusing than probably most people think. In the philosophy of medicine literature, there are two major, fully independent (or ‘pure’) naturalistic (as opposed to normativist) accounts of the nature of illness in biological organisms: the “biostatistical account” and the disease-as-dysfuntion account. Both are massively unsatisfactory in their own ways.

The biostatistical account says, as the name implies, that an illness or disease is any kind of isolable physical abnormality in an organism, where abnormality is defined as abnormality for the relevant age-group and sex of that species, and for the kind of environment which the organism inhabits (there is an unavoidable vagueness about all of the key terms in this definition (the term “abnormality” and the specification of the reference class), which is a problem we’ll explore in a minute). 

Meanwhile, the disease-as-dysfunction account says that we should understand illness as some kind of disruption of the selected function of the organ or organs in question. As “Life history” theory shows, many parts of the physiology of organisms have been selected to perform different roles at different stages of life, so we may also need to look at more specific reference classes than just the species according to this account, too.

The main problems with the biostatistical account are as follows:

1.)    The word “abnormality” is extremely vague (and making it more precise is doomed, for reasons we’ll see), and the specification of the reference class is also (unavoidably) extremely vague. As I make clear below, if we wanted a statistical account that fitted our concept of disease best, it couldn’t lay down a general reference class for ‘abnormalities’: it would sometimes have to take a very broad reference-class, and sometimes a very narrow one. (This is another way of saying that the statistical account is fucked.)

2.)    The account implies that very common diseases of aging like osteoarthiritis or dementia are not diseases for very old people but are for people who get them prematurely. It implies, even more strangely, that very unusually healthy old people (say, 80-year-olds without any debilitative osteoarthritis and minimal cognitive decline) are diseased! This evidently clashes violently with how the medical profession sees things (which ought to be fatal for an account of disease).

3.)    It simply cannot handle epidemics, or times and places where a large slice of the population has had a disease far less prevalent in other environments. It implies that the Black Death was not a disease between 1346 and 1353, and that malaria isn’t a disease in malaria-infested tropical regions. It seems to imply that the common lung diseases in Victorian England weren’t diseases then, that cholera and dysentery aren’t diseases in environments where they run rampant due to lack of clean water, that having asthma or several allergies doesn’t make you pathological if you are a contemporary First World child, and that being grossly overweight doesn’t make you diseased if you are an American. This failure to handle environmentally-widespread disease seems an immense flaw.

4.)    Freakish athleticism or freakish intelligence, or other beneficial mutations (in any organism), or unusual psychological traits we say are worthy of respect, would be designated as diseases on this account. More intriguingly, all forms of gender dysphoria would be labelled a mental illness according to this account, simply because gender dysphoria is still very rare; in recent times, such a view of ‘gender non-conformity’ has come to be seen, among a certain very vocal and growing group of young people, as a very reactionary position. The account may also designate pure lesbianism as a sexual pathology, just as it designates (say) paedophilia or extreme sexual sadism as sexual pathologies, seeing as pure lesbianism is very rare (though presumably (one hopes) not quite as rare as paedophilia)). Contra Steve Silberman and other advocates of the goodness of ‘neurodiversity’, it probably implies that even high-functioning Asperger’s Syndrome is a disease for women (and maybe men), and definitely implies that any more severe forms of autism are a disease for men and women.

5.)    It implies that wounds and severe injuries are diseases. It implies that a man missing a leg from a traumatic injury is diseased. Obviously, this is absurd.

6.)    Most generally, it’s not explanatory. In itself, the account doesn’t make sense of why disease is a bad thing.

Although the disease-as-dysfunction account solves many of the major problems that the biostatistical account – a bit of reflection will make clear that it overcomes the environmentally-widespread-disease problem, the severe-injuries-as-disease absurdity, and is explanatory (diseases are bad because they represent a form of dysfunction) – the main problems with it are as follows:

1.)    It seems to have morally difficult or even repugnant conclusions in the case of mental illness, and clashes violently with the DSM. Although it has been long been suggested that, given its prevalence in our species, homosexuality may somehow be an adaptation in our species (and one or two not-completely-absurd just-so stories have been proposed, the most popular of which is the “gay-uncle theory” that homosexuality (in men) may have been weakly selected for in ancestral populations because children with gay uncles or other close, older gay relatives did better because they had an extra carer in their early development, and those children also had genes more likely to create gay men, even if the gay men didn’t themselves reproduce)), no-one can claim with any confidence that homosexuality was selected for (in fact, I would suggest that a Bayesian aware of the evidence and arguments could rationally have negative-confidence that it was selected for). Thus, this account implies that homosexuality may well be a mental illness; fixed homosexual orientation may represent a strong disruption of the adapted sexual behaviour of humans. The account may also imply that men with ‘anger-management’ issues are not typically pathological, or that jealous rage in men is not typically pathological, or that spouses who fall into deep depression when they are widowed are not typically pathological. It may also imply, if Randy Thornhill and Craig Palmer are right about male-on-female rape being a selected behaviour for ‘low-status’ males (a classic example of an evo-psych hypothesis which it may be entirely pointless to debate, because it is not clear that we will ever be able to establish with any confidence that it is right or wrong, or how much selection pressure was applied to this very particular kind of behaviour over (recent? distant?) evolutionary history)), that young, ‘low-status’ men in the right kind of environment (an important clause because many adaptive behaviours in organisms do not get expressed in all environments (a proponent of the disease-as-dysfunction account wouldn’t have to say that pacifist young men raised in a pacifist environment are pathological, even if tendencies for aggression have been selected for in young men, for various reasons, because that presupposes a silly, genetic-determinist view of evolution which makes the account ridiculous)) who do not have any urges to rape women are pathological, and that young, ‘low-status’ men in the right environment who do have urges to rape women (and perhaps do rape women) are not pathological.

2.)    It’s very hard to say what is required for a trait to be an adaptation and, as I made clear in my parenthetical interjection about pacifist males above, some traits can be adaptations while not being very strongly developmentally/environmentally canalised (because of culture and language and shit, probably many human psychological traits and behavioural tendencies are not strongly developmentally/environmentally canalised). This means that, in practice, ‘applying’ this account would be very difficult. Also, as everyone learns in high school, humans actually have adapted responses to infections and viruses, produced by our immune system (symptoms of disease are often (usually?) symptoms of an adapted immune-system attack on the disease), which makes it extremely difficult (if not impossible) to state this account in a non-question-begging way. You can’t say diseases are disruptions of the optimal selected function of an organ/organs for an organism of that phase in the life cycle, or disruptions of the healthy selected function…. There may be a way around this but it hasn’t occurred to me straight away, and thinking about this stuff is very mentally taxing.

Basically, all this should make it clear that our concept of illness is not one which can be defined according to sufficient and necessary conditions; instead, it is a vague, ‘family-resemblances’-type notion which combines ideas of rareness and defection and deformity, and yet is also firmly conceptually distinct from ‘injury’ (diseases are forms of defect which aren’t brought about by physical force). That’s why neither pure account is fully satisfactory. The above discussion also made it very clear that there is a very significant normative element to diagnosis of mental illness. In 2015, I tentatively came to the conclusion that there was nothing objective about the move in modern times of the psychiatric community to declare gender dysphoria (any kind of gender dysphoria, whether the ‘homosexual-transsexual’-type or the ‘autogynephile’-type) a normal personality trait, rather than a mental illness, because: a) it is extremely rare to have an overpowering feeling that one has been ‘born in the wrong body’, or to be an autogynephile; and b) to be gender-dysphoric is dysfunctional both in the sense that being gender-dysphoric is likely to lead to social exclusion and censure (social dysfunctionality is very relevant to the intuitive concept of mental illness), and also seems to represent a big disruption of adapted sex-specific psychological and behavioural tendencies (as anyone who’s read any of my other posts on sex and gender would know, I don’t believe that none of the psychological/behavioural sex differences that appear across multiple cultures were selected for, and in fact I think that’s incredibly implausible). At the time, the following question struck me as quite important: if you are not thinking of mental illness in heavily normative terms, what makes gender dysphoria a form of psychopathology? What I have realised is that you are thinking of mental illness in heavily normative terms, and that this is unavoidable. Mental illness is not objective; an intelligent alien species couldn’t pick out which people our psychiatrists say are mentally ill and which aren’t without becoming just like us, and understanding all of the details of our psychological make-up and the culture in which we are embedded. There’s no way an alien could possibly work out the difference between a serially self-aggrandising person and someone with “narcissistic personality disorder”, or between a “psychopath” and someone with low-empathy, or an “eccentric genius” and a total weirdo, without seeing the world in almost exactly the way we do (‘becoming us’, as it were). In fact, as the main discussion of this essay made clear, an intelligent alien species couldn’t pick out physically “diseased” versus “non-diseased” organisms in the same way we do it, even if they had developed a strong understanding of earth biology. No doubt they would have a similar concept to our “disease” but, seeing as our concept of “disease” is so messy, it surely wouldn’t be the same.

To be clear, I'm not denying that there are unambiguous cases of (even) mental illness (let alone denying that there are unambiguous cases of non-mental illness): if a big movement started up declaring that it was morally admirable to be delusional, it still wouldn’t make sense to declassify schizophrenia. Obviously, if a specific gender-dysphoric person has a more generalised anxiety on top of their gender-related anxiety, or also has schizophrenia, we can definitely say they’re mentally ill (and maybe, as controversial as this may sound, speculate that their gender dysphoria is bound up with their general mental ill-health). But there’s a fucktonne of fuzziness in the way we conceptualise illness and disease, and that’s where our ethics comes in to try to make things a little more sharp. That's my point. 

It may have occurred to you that this conclusion seems to raise some interesting questions about the fashion among certain popular scientists and commentators (speaking on behalf of the psychiatric community) to 'internalise' common mental illnesses like depression and anxiety - a fashion symbolised by the refrain that such pathologies are really "chemical imbalances in the brain". This talk could be interpreted as simply a way of expressing the truism that the mind supervenes on the brain: that is, that mental states can't change without something changing in the brain. More often, though, it is a way of expressing commitment to a theory about the instantiated signature of such common mental illnesses which is, despite the rhetoric, not settled science. The following extract from Marcia Angell's June 2011 article in The New York Review of Books explains the origin of this theory, and the problems with it:

"When it was found that psychoactive drugs affect neurotransmitter levels in the brain, as evidenced mainly by the levels of their breakdown products in the spinal fluid, the theory arose that the cause of mental illness is an abnormality in the brain’s concentration of these chemicals that is specifically countered by the appropriate drug. For example, because Thorazine was found to lower dopamine levels in the brain, it was postulated that psychoses like schizophrenia are caused by too much dopamine. Or later, because certain antidepressants increase levels of the neurotransmitter serotonin in the brain, it was postulated that depression is caused by too little serotonin. (These antidepressants, like Prozac or Celexa, are called selective serotonin reuptake inhibitors (SSRIs) because they prevent the reabsorption of serotonin by the neurons that release it, so that more remains in the synapses to activate other neurons.) Thus, instead of developing a drug to treat an abnormality, an abnormality was postulated to fit a drug.

That was a great leap in logic, as all three authors point out. It was entirely possible that drugs that affected neurotransmitter levels could relieve symptoms even if neurotransmitters had nothing to do with the illness in the first place (and even possible that they relieved symptoms through some other mode of action entirely). As Carlat puts it, “By this same logic one could argue that the cause of all pain conditions is a deficiency of opiates, since narcotic pain medications activate opiate receptors in the brain.” Or similarly, one could argue that fevers are caused by too little aspirin.

But the main problem with the theory is that after decades of trying to prove it, researchers have still come up empty-handed. All three authors document the failure of scientists to find good evidence in its favor. Neurotransmitter function seems to be normal in people with mental illness before treatment. In Whitaker’s words:

"Prior to treatment, patients diagnosed with schizophrenia, depression, and other psychiatric disorders do not suffer from any known “chemical imbalance.” However, once a person is put on a psychiatric medication, which, in one manner or another, throws a wrench into the usual mechanics of a neuronal pathway, his or her brain begins to function…abnormally.”

Carlat refers to the chemical imbalance theory as a “myth” (which he calls “convenient” because it destigmatizes mental illness), and Kirsch, whose book focuses on depression, sums up this way: “It now seems beyond question that the traditional account of depression as a chemical imbalance in the brain is simply wrong.” "

As Angell also implies in the above, some have gone so far as to claim, not merely that these common pathologies are instantiated in terms of neurotransmitter imbalances, but that the general cause of these pathologies is some kind of largely genetically determined imbalance of this kind. My view, which I think ought to be uncontroversial, is that some cases of mental illness can only be explained on the biochemical or neurophysiological level - those cases of sparked by no identifiable social or experiential cause must be sparked instead by some unknowable web of chemical causes, something to do with the chemical environment and the way it affects the expression of certain genes - but that many other cases of mental illness clearly do warrant an experiential and social explanation in combination with a 'lower-level' one. In fact, that puts my case too weakly: I think it would be completely softheaded to say that the cause of the mental illness of someone who develops PTSD after going to war, or someone who develops psychosis after years of abusing drugs, or someone who develops depression after the death of a loved one, or some young girl who develops anorexia after fixating on the body shape of a famous model, or some teenage boy who develops anxiety and suicidal tendencies after consistent bullying and social ostracism, is merely something internal and chemical. This is evidently not only softheaded, but false. The relevant, explanatory level of causation in these cases is the social level. To be clear, this is not to say that we should not talk about brain chemistry and genetics if we want a thorough explanation for the development of psychopathology in these cases. I do not believe this, because, clearly, some people have a greater predisposition to developing these illnesses than others, and, clearly, some people may experience more severe derangement after certain traumatic events than others would. The explanation for these differences lies in differences in genetics and brain differences, so such things evidently are always at least a little relevant to a complete exposition of the causes of a mental illness. But the point is that being overly reductionistic about this is totally idiotic.

I suppose I'm sceptical of the assumption of modern psychiatry (I don't want to generalise - I don't know how dominated by this attitude psychiatry is) that the mental pathologies "depression" and "anxiety", defined vaguely in terms of behaviours and habits in response to certain social environments (if they weren't defined vaguely, there wouldn't be borderline cases of depressed and anxious people, though there clearly are, and there are very different degrees of depression and anxiety also), would each reduce to their own simple chemical signature. This seems highly unlikely on the face of it; I would have thought the more likely scenario would be a constellation of different chemical correlates, with each depressed or anxious person having a pattern differing in a multitude of subtle ways.

Here's the thing: we cannot ignore the massive fissure between the social and the neurobiological. Don't be a Churchland.