In a previous post I urged the importance of distinguishing matters of causal explanation from matters of causal constitution. What I have in mind is that the answer to 'why is he angry?' is not typically to be answered by reference to what happens in his brain, but rather by reference to what happens in his life. Such reference essentially refers to what the angry person says when he gives his reasons for feeling angry. The exceptions however are important. Thus when there is no answer (or no decent answer) to be had from him as to his reasons, then matters to do with changes in bodily constitution, injury, overwhelm, diet etc. are now 'released' to play the role of reasons. What they are 'released' from is their mereological subordination to reasonable reaction (since here, in such exceptional cases, we don't have to do with reasonable reaction). It isn't of course that constitutional matters - bodily mechanisms and bodily changes - suddenly spring onto the scene when reason goes down, but that they can now play a role in that discourse which has to do with reason provision i.e. which has to do with answering 'why?' questions about actions and reactions. What I claimed in that post too was that reasons which appeal to our psychology (dynamic motivational reasons that is - one's which cite the motivated deployment of defences, the motivated avoidance of anxiety and pain, etc.) become available (are 'released') when the provision of ordinary personal reasons (my reasons for acting, the one's I avow) give out. And I suggested too that subpersonal reasons (hormonal, neurophysiological and neurotransmitter anomalies etc) only become available as explanations when both the personal and the psychological reasons fail.
So what I am specifically writing against here is that form of thought which is so keen to demonstrate its all-encompassing scientifically and philosophically unprejudiced prowess that it fails to do adequate justice to the 'helpful prejudice' that is already embedded within our explanatory practices - to the 'prejudice' which says we've gotta start at the top (with my reasons) and only work down (to neural anomalies) when those forms of explanation at the top fail. I am writing against the throw-it-all-into-the-mix bio-psycho-socio-spirituo explanatory models of mental illness, for example. I am claiming that our being constituted by matters chemical, biological, psychological, social, spiritual, etc., and the dependency of our healthy minds on bodies that politely function to support them, do not mean that observations at lower levels of description (how a certain hormone affects amygdala functioning for example) could play a role in explaining why someone acts as they do unless the levels of personal and of psychological reasons fail us. If someone says 'but why can't we have both?' they are still, I think, in the grip of an unhelpful picture, a picture which without warrant transposes stories from a context of inquiry into matters constitutional (about the processes and structures which constitute or subtend our psychological functioning) into the context of inquiry into matters of reason ('why did he wave his arm?' is not helpfully answered by mentioning the tightening of his muscles).
What I want to add here is that this all has implications for rational treatment of psychological disturbance. If I am crazy because my neurotransmitters are de trop then it will not make sense to treat me psychologically. But the reason for this is that what it is for neurotransmitter levels to be de trop is something which has to determined relative to the (un)availability of reason-giving or dynamic motivational explanations for an emotional reaction. For it is perfectly conceivable (even if not actual) that the same level of neurotransmitter activation may obtain in the context of a healthy grief reaction as in the context of an endogenous depression. But in the former case there is no meaning to be had for 'de trop'. This, by the way, gives the lie to those evolutionary accounts of 'neuropsychiatric disorders' (a stupid term which begins ('neuro') where it often shouldn't even end) which want to define what counts as excessive or as misfiring or what-have-you in relation to proper function, where 'proper function' is given in evolutionary terms: is it performing the task it evolved to perform? Such accounts start in the wrong place - from the bottom up - whereas we ought to start from the top down - from the availability or otherwise of personal-level reasons for action. Sure, 'excessive activation' buggers up our psychological functioning, but what makes such activation count as excessive is that what it is doing is not playing its part in the mechanics of what on quite independent grounds we find to be rationally intelligible reactions, but is rather causing reactions which, again on quite independent grounds, we find to be rationally unintelligible. And that, I imagine, should come to all but the excitable theory mongerer as something of a relief - for it might have felt like an exciting intellectual project, but it was surely always rather a desperate long shot, to try and milk normativity out of evolution, and to milk the concept of something like a natural kind so hard that it would explain how our unscientific concept of mental illness, used with normative aplomb by many a scientifically illiterate person, could contain such hidden intricacies.
Back to treatment. The point I want to make is that whilst, with an eye on matters merely constitutional, it might seem to make a breezily unprejudiced kind of sense to say 'well, we could treat the problem with a talking cure, with exercise, or with a pill; all that's happening here is that we're intervening on different levels', in fact this really is far too breezy for its own (and the patient's) good. For a rational treatment, surely, is one which tries to address the reasons why someone is distressed. And if they are distressed because they are grieving their father, or alternatively if they are differently, anxiously/depressively, distressed because their repressed anger at the father leads to aborted grieving, then even if any of this distress or instead this anxious depression is partly (what-shall-we-call-it) 'realised' in a statistically-abnormal-for-them level of dopamine (or whatever), then 'treating' the abnormal dopamine levels will not be treating the cause of the problem. (Where what 'the cause' is is coextensive with the reason, and not with the constitutional mechanics.) ('Treating' goes in inverted commas because it can't really be called treatment if what is 'treated' isn't itself a problem.) In fact, far from being helpfully unprejudiced, such an approach tramples all over the humanity of the patient. For being given extra, or being caused to have reduced, neurotransmitters may reduce the healthily-grieving person's sadness but, since their sadness is their grief, and since what they need to do is to get used to the fact that their father is dead, and since 'grieving' is the name of that process of getting used to his being dead, then to reduce sadness is to interfere with their reality contact itself. Which, er, is not normally what we take for the goal of psychiatric treatment. Or, if we imagine that we have to do with the patient whose depression obstructs healthy grief, then having agitation removed will remove the clue that here we have to do with conflicted feelings (anger and loss) and will leave them in emotional limbo (welcome to zombie land). Only if we had a situation, as yet here un-described, in which the reason why the patient was struggling to grieve was not because of grief's intrinsic hardness nor because of their neurotic ambivalence, but because they have too much dopamine (or whatever), would it make any rational sense to treat them using drugs.
NB the different responses to Prozac: 'I feel numb now' vs 'now I feel far more myself'. Even if they had, pre- and post-treatment, the same level of (say) serotonin as each other, only the second patient had, we might rationally infer, been suffering a serotonin imbalance, and only patient's who had a serotonin imbalance could meaningfully be said to be being treated with Prozac, regardless of the efficacy of the treatment.