Tuesday 23 June 2009

Rationality, interpretation and inconsistency

One of my underlying philosophical commitments is to the idea that there’s something right about what Rorty described in Davidson’s work as the ‘philosophy of language of the field linguist’. So the idea that radical interpretation is a good perspective from which to think about meaning (although some care is needed to avoid a Quinean sideways on view of it), and that it connects belief and meaning holistically to the demands of rationality even whilst that notion resists codification, are both in my conceptual index. In picking out the philosophy of language of the field linguist, I also mean to leave on the shelf the framework of the Tarskian truth theory with its paratactic analysis of ‘that’ and dark talk of same-saying etc.

Still even this gives me something to worry about in its application within philosophy of psychiatry. If rationality is at the heart of interpretability and if that is at the heart of mentality, what are we to make of the apparently mental and apparently non-rational psychopathological experiences and states?

One person who has a particular line on this is Lisa Bortolotti. Admirably steeped in Davidson, she’s mounted a kind of immanent critique. I’ve just been reading her 2003 paper ‘Inconsistency and interpretation’ in which she aims to make trouble for the central role of rationality in interpretation given (hardly surprising) empirical claims about the inconsistency in beliefs of real people (including college students, for heaven’s sake!).

Bortolotti considers two moves Davidson can make. One is to claim that whilst obvious inconsistencies are ruled out by the Constitutive Principle of Charity, unobvious ones may not be. So Davidson can argue that if beliefs are not ‘active’ or attended to then they may be held despite being inconsistent. Because the distinction between active and unactive isn’t a logical or semantic distinction I’ll ignore this bit of the paper and, in any case, Bortolotti doesn’t claim very much from it.

In the next section, she offers a key distinction between ‘conformity with’ and ‘subscription to’ norms of rationality. If we assume that radical interpretation is tied to the first (as a methodological and thus a constitutive thesis) there are problems with real cases of inconsistency. But if we retreat to subscription, we can defend actual inconsistency at the ground level (with it regained at a higher level of subjects correcting attended to inconsistency). But then, if subscription doesn’t imply in general conformity, it is no help in radical interpretation. Further, Davidson seems to say it isn’t a matter of evidential connection. So Bortolotti concludes that if there is no connection between subscription and conformity then the link between rationality and interpretation in radical interpretation is lost.

If this is what is going on in the paper, then Bortolotti’s rejection of an empirical link between subscription and conformity does a great deal of work. Surely what Davidson rejects is the idea that it is merely a contingent empirical result that thinkers/speakers subscribe to the norms of rationality? (He explains his claim saying by: ‘it is only by interpreting a creature as largely in accord with these principles that we can intelligibly attribute propositional attitudes to it, or that we can raise the question whether it is in some respect irrational.’) But that is consistent with the idea that it is contingent that a particular ‘system’ subscribes to the norms of rationality and hence (given some other assumptions about their behaviour) is interpretable. So there can be an evidential link between behaviour and subscription to rational norms.

In other words, it is a priori / necessary that interpretation requires rationality rather than being an empirical result. But that doesn’t rule out the idea that subscribing (in the anodyne sense) to a norm explains (only partially, obviously) success. One of the reasons why darts players score triple 20s is that that is what they aim at. But the idea that triple 20 is an aim of the game goes beyond a mere descriptive conformity (even if it may rationally be inferred from behaviour by those non-Quineans with eyes to see!).

Despite this, I do have some qualms about the Davidsonian picture. What seems attractive is that he connects rationality to mentality (akin, in their different ways, to Dennett, McDowell and Jane Heal). But whether our grasp of the notion of rationality so relied upon goes beyond our capacity to find subjects interpretable seems to me rather unlikely. If so, there are significant limits to how elucidatory the connection is.

(The last point is developed here.)