Sunday, 22 June 2008

Just Because

In this post I want to focus on the reasons why it can be so hard to accept the non-causal character of the psychological 'because'. In the previous post I claimed that (popular expositions of) the James-Lange theory of emotions fail to distinguish between cause-providing and intelligibility-providing uses of 'because', and that this is why they appear to be radical challenges to our everyday understanding - rather than the nonsense that they really are. Now I have to report that my internal interlocutor - who is steeped in the intellectual vices of mentalism, reductionist naturalism, and scientism, and who has a bit of a metaphilosophical or philosophically reflexive blind spot - has already lodged an objection to what I wrote there.

His objection is as follows:

It isn't good enough to simply say that intelligibility-conferring uses of 'because' in psychological contexts are different from efficient-cause-providing uses in mechanical or other such contexts. Such an account simply begs the question. This is because it fails to provide us with a proper, adequate (substitute your preferred philosophical terminological jargon at this point) 'metaphysics' of the relation between explanans and explanandum. As soon as we try to reflectively understand the 'psychological because', then we realise that it must after all be a 'causal because'. Without such an analysis we are left without any reflective understanding of the kind of explanation on offer.

Bill Child

I ought to say that I suspect my inner interlocutor has been re-reading Bill Child's Causality, Interpretation and the Mind. When I myself return to his text, this is what I find he says regarding the 'psychological because' (p. 92):

We have two options. Either we say that the link between reason and action is sui generis, a basic relation; in which case there is no further understanding to be had and the relation remains a mystery. Or we give some kind of analysis of the relation; and what could that be but a causal analysis?

The point of Donald Davidson's causalist argument, we are told (p. 96), is

that we need to understand the 'because' in 'She Phi-d because she believed that p'. ... [W]e must appeal to causation in order to understand the metaphysics of the relation between reason and action. ... [T]he understanding we get from seeing that reason explanation is a form of causal explanation is a reflective understanding of the metaphysics of this form of explanation; we understand what sort of explanation it is, and how reasons explain actions.
The Wittgensteinian distinction between reasons and causes must, we are informed, be chucked out the window as another unfortunate superstition from those not-so-good old days and the pre-Davidsonian era of RKP 'little red books'.

Well, perhaps that is just how things look if we abandon metaphilosophical reflexivity so we can at least get on with the good old business of metaphysics. But my own suspicion is that what looks like an absence of 'metaphysical understanding' for the 'psychological because' is due not to any faillure in the anti-causalist construal of reasons, but because of tacit and unwarranted prior metaphysical and metaphilosophical commitments which skew the contest from the outset.

I leave it to others (Julia Tanney, Severin Schroeder) to demolish the Davidsonian idea that we cannot get enough of an explanatory fix on the relation between reasons and their explananda unlesss we supplement rational with causal relations. (The basic moral of their story was that - so long as we take care to individuate reasons for actions as we should (by appealing to what people tell us with respect to why they acted, for example), and so long as we do not try to insist on reason explanations possessing a greater degree of determinacy than in fact they do possess (a degree possessed by causal explanations, of course, which nicely rationalises the causalist's analysis), then - the rationale for introducing causality into the analysis is far weaker than the causalist supposes.) What I want to focus on here is Child's idea that the non-causalist owes us an account of the kind of understanding offered by the 'psychological because'. According to Child, this is what the causalist provides - when we accept that reasons are causes, then we understand 'how reasons explain actions'. As Child says

The point of the argument for a causal view is not that we must appeal to causation in order to tell whether S phi-d for one reason rather than another [well, actually, that was the point of Davidson's argument]; it is, rather, that we must appeal to causation in order to understand the metaphysics of the relation between reason and action.'

The principle idea I wish to focus on is that quoted above:

Either we say that the link between reason and action is sui generis, a basic relation; in which case there is no further understanding to be had and the relation remains a mystery. Or we give some kind of analysis of the relation; and what could that be but a causal analysis?
And what I want to question is the idea that saying that a 'psychological because' is basic, sui generis - of its own model, not to be modelled on something else, not to be understood in terms of something else - is to render it a mystery. For why should some thing always be intelligible along the lines of some other thing? Why shouldn't there be several different kinds of things in the universe - perhaps more than are dreamed of in the philosophies of (say) the reductive metaphysical naturalist - and why shouldn't one of these 'things' be reasons?

The question needs to be turned back against the causalist. So, what if someone were to say 'But we need a reflective metaphysical understanding of the kind of relation which is being described as a 'causal explanation', and what could provide the form for this be other than a reason explanation?' We can't just suppose that causality is a basic relation, sui generis. For that would be to make a mystery out of it. It must be modelled according to some other relation. And what other relations are there - other than relations of rationality and intelligibility?

Now clearly the Wittgensteinian non-causalist would not want to say that. They would most likely be quite happy with the idea that there are different kinds of 'becauses', and see their job as 'teaching us the differences' between these. When we learn our metaphysical lessons, we do not provide explanatory accounts of these different kinds of 'because'. We do not render them perspicuous in the terms of some pre-understood super-language. (Recall the phantasies behind the very idea of a 'theory of meaning'...) We just track the differences 'from within', note the different ways of going on in different language-games.

Is it mystery-mongering to indulge in positing (or less prejudiciously, acknowledging) sui generis phenomena? I think it depends on what we mean by a 'mystery'. Whether or not we feel it to be a problem that something cannot be explained will depend on whether we feel it is something which will baffle us unless it is explained - whether we feel it is something for which an explanation is owing, something that ought to be explained. And whether we feel it is something which ought to be explained will be a function of the prior theoretical commitments we have made.

It is only when we have (perhaps tacitly) decided, ab initio, to accept some particular range of phenomena as paradigms of intelligibility, as that to which all else must conform if it is to be intelligible, that phenomena outside this range start to present themselves as unintelligible or as mysterious in a troubling way. And of course it is perfectly legitimate to invoke some set of phenomena as paradigms of intelligibility in this way - otherwise we would never find anything mysterious, and take everything as its own model (which would be to evacuate all content from the ideas of a 'model' and of an 'explanation').

All I am requesting is that we own our metaphysical 'prejudices'. As Nietzsche wrote, metaphysical systems of the past have all too often been nothing other than the personal confessions of their authors, unconscious and involuntary memoirs. What I am suggesting is just that these 'confessions' become conscious and voluntary - that we attempt to excavate, and then explicitly lay down on the table in the light of day, our 'prejudices' regarding what we find intuitively compelling, or unproblematically sui generis.

What I suspect was underlying Child's feeling that it is simply mystery-mongering of the anti-causalist to urge that reason explanations are sui generis was a tacit commitment to a naturalistic metaphysics (his view that actions are construable as happenings or events is one other such manifestation of this). Reductive naturalism (unlike the relaxed naturalism of John McDowell) finds no place 'in the natural world' for phenomena which are not intelligible along the model provided by the natural sciences. And what is the model that is to be used to render the natural sciences intelligible? It is part of the metaphysical outlook of naturalism that this question is not to be asked. They, perhaps, are allowed to be their own model.

What the Wittgensteinian is asking is that the same courtesy be extended to the 'psychological because'.