Skip to main content.
October 25th, 2004

Fodor on Analytic Philosophy

The Red Sox have caused a dearth of posting here – though we’re somehow still setting daily records for visits. (Is correlation causation? Does content drive away readers?) In any case the blog isn’t looking anywhere near as sickly as my email box (or my diet or sleep patterns or blood pressure etc) so maybe it shouldn’t be the priority. But I did want to post one thing in response to something Jerry Fodor wrote to Brian Leiter

bq.. OK, fair enough; it’s never really so that one size fits all. On the other hand, I do think that there are a couple of theses that major US and UK philosophers have more or less agreed about (mostly implicitly, to be sure) over the last fifty years or so, and that have largely shaped the landscape of philosophical discussions. Since I think both theses are wrong, I feel strongly about getting them out in the open where they can be jumped up and down on.

The first is semantic pragmatism: the idea that intensional content is to be explicated as some sort of `know how’ , hence in epistemic terms. The typical avatar of this view is the thesis that concept possession is something like knowing how to evaluate inferences whose validity turns on the concept, and/or knowing how to sort things that the concept applies to. Peacocke is perhaps the current paradigm, but it’s hard to think of anyone since Wittgenstein (indeed, since Dewey) who doesn’t hold it. In my view, it’s entirely misguided. To have the concept C is to be able to think about Cs as such. Confusing epistemology with semantics has damned near ruined the philosophy of mind and the philosophy of language for as long as I can remember. (I have gone on about this in an article called `Having concepts, a brief refutation of the 20th Century,’ in a recent issue of MIND AND LANGUAGE. I expect I’ll still be going on about it for the foreseeable future.

The second is the methodological doctrine that philosophy does (or should) procede by the method of `semantic ascent’; that is by translating metaphysical questions (eg. how does perception work) into questions of conceptual analysis (`how do we use the word `see’; or `what is the concept of seeing’. The translation is supposed to underwrite the (putative) a priority of philosohical theses, and the (putative) fact that philosophy is a game that anyone can play (`you don’t need to be a psychologist to understand how seeing works; we ALL have the concept SEE (and/or we ALL know how to use the world)). In fact, I doubt that typical interesting philosophical claims should (or even can) be treated as theses about the analysis of concepts. I am deeply moved by the reflection that no concept has ever been analyzed by any analytic philosopher, however hard analytic philosophers have tried. This does suggest that their methodological assumptions are due for reconsideration.

To be sure, the scene isn’t as clear as I’ve been making out. In the UK, you get these views in their pristine condition (Dummett is the parade case); in the US, they are often combined with a strong commitment to naturalism. Quine’s a case; he’s a self-anounced semantic pragmatist, so he really ought to believe in conceptual analysis. But he’s a naturalist, so he’s dubious about the a priori. The dialectical fix was to speak (not of analysis but) of canonical representation (representation in a canoncial notation) and then procede to do just the kind of philosophy that everybody else did. A similar story could be told about Davidson, but to hell with it.

Who among the living counts as an analytic philosopher by these jaundiced criteria? Not me, for sure. But practically everybody in Australia; Peacocke (see above), McDowell, Brandom, Travis (when he isn’t being simply a nihilist), everybody in cognitive science without exception. And so forth. You needn’t aim; just pull the trigger and you’ll hit one.

Your question deserves a fuller answer than this; but that would be an article; or a book; or a lifetime; and I’m depressed enough already. To say nothing of too old.

Let’s ignore the question for a minute of whether semantic pragmatism and semantic ascent are anything like a good analysis of ANALYTIC PHILOSOPHY. (Well except parenthetically – I’m entirely in agreement with Jason and Ralph in Brian’s comments section that they’re not.) There’s a kind of theory here that’s of some sociological interest. Is it true that the views are as widely held as Fodor says they are in places he discusses.

I can’t speak for the UK, but it certainly isn’t true in Australia. In particular, what Fodor calls semantic ascent is rather frowned upon in the most influential Australian circles, at least as I understand what people are doing. When we talk about intrinsic properties, for instance, what we care about is which properties are really intrinsic to an object, not what is thought to be intrinsic by most folks, or what the folk concept of INTRINSICNESS is.

There are two complications here.

First, we tend to take folk intuitions (suitably molded and interpreted) as a decent starting point to metaphysical intuitions. But that’s not because we think they’re constitutive of the truth. Rather it’s for the simple reason that Frank Jackson gives in his Locke Lectures – it could hardly be better to start with something unintuitive.

Second, we tend to think some allegedly metaphysical discoveries are really discoveries about concepts. So the two-dimensionalist project was, in part, to explain away divergences between necessity and a priority as curious facts about the way we talk using names and natural kind terms. (Funnily enough, Fodor now seems to agree at least about natural kind terms.) That’s not because there isn’t anything more to metaphysics, but because it’s the right thing to say about this case.

To make the point most clearly, consider the widely held (in Australia) belief that vagueness is a semantic rather than metaphysical phenomenon. (I suppose the youngsters think it’s an epistemic phenomenon – that’s OK for now, it’s metaphysical vagueness that’s the bogeyman.) The view that there’s no metaphysical vagueness certainly isn’t the crazy view that there’s no vagueness in the terms we use to talk about things. Rather it’s the sensible (and true) view that the vagueness is entirely a function of the representations (and perhaps our relations to them) and not a feature of things in themselves. Someone who holds the bundle of views Fodor calls ‘analytic philosophy’ couldn’t as much as make this distinction, but in fact most Australian philosophers not only make the distinction but have strong views about which side is correct.

The claim about semantic pragmatism is tricky and frankly I don’t know enough about what my compatriots think in order to make any informed comments. But it’s very odd that Fodor would think that Australia, for years the home of good old-fashioned revisionary metaphysics, and in particular of realist metaphysics, would be a paradigm of a philosophical tradition one of whose two pillars is translating metaphysical questions into semantic ones.

Posted by Brian Weatherson in Uncategorized

1 Comment »

This entry was posted on Monday, October 25th, 2004 at 10:41 pm and is filed under Uncategorized. You can follow any responses to this entry through the comments RSS 2.0 feed. Both comments and pings are currently closed.

One Response to “Fodor on Analytic Philosophy”

  1. djc says:

    Funny, I thought that that was the only thing that Fodor was right about. The Jackson-style Canberra plan pretty clearly gives semantic ascent a major role in metaphysics: witness the appeal to conceptual analysis in solving the “location problem”. Semantic ascent may not be crucial in fundamental metaphysics (i.e. questions about what’s in the fundamental supervenience base, presumably fundamental physical properties on the paradigmatic Canberra view), but in non-fundamental metaphysics (i.e. questions about things that supervene on the fundamental), conceptual analysis is pretty central. Even in the case of intrinsicness, I’d have said that what Langton and Lewis (say) are doing is pretty clearly a sort of conceptual analysis, and so is the sort of thing that Fodor is counting as semantic ascent.