Mulhall On Williams Beginning

DOI: 10.1111/j.1468-0378.2008.00297.x

‘Hopelessly Strange’: Bernard Williams’ Portrait of Wittgenstein as a Transcendental Idealist Stephen Mulhall

Recent discussions of the idea that there might be some illuminating relation between Wittgenstein’s later philosophy and that of Kant’s critical project have tended to take off from Bernard Williams’ highly influential paper ‘Wittgenstein and Idealism’ (a lecture he delivered in 1973 and published in 1974.)1 It is not often noted that this paper is, in effect, a critical review of Peter Hacker’s Insight and Illusion, published in 1972—a pathbreaking commentary on Wittgenstein’s writings, early and late, which explicitly organizes itself around the idea that Wittgenstein’s and Kant’s approaches to philosophy were importantly analogous.2 And it is very rarely noted that Hacker was beaten to this particular insight by Stanley Cavell, who—in a famously devastating 1962 review of one of the first introductory books on Wittgenstein’s later philosophy (by David Pole)—explicitly related Wittgenstein’s idea of grammar to Kant’s idea of transcendental knowledge.3 Why should these matters of bibliographical priority be of any interest to us? First, they confirm that the idea of a comparison between Wittgenstein and Kant is not some relatively recent development in Wittgenstein scholarship, but rather something effectively coeval with it—as if unavoidably engendered by any serious study of Wittgenstein’s later work. Second, they raise the possibility that Williams’ understanding of Wittgenstein might be importantly inflected, to its detriment as well as its benefit, by Hacker’s sophisticated and highly influential but nevertheless contestable reading of Wittgenstein. Third, they might prompt us to recall that both Hacker’s and Cavell’s ways of understanding this comparison are each rather more subversive than supportive of that developed by Williams. By the time he published a revised version of Insight and Illusion (in 1986), Peter Hacker had come to find this putative comparison so much more misleading than illuminating that he re-structured the whole of his book so as to downplay it, and revised its subtitle accordingly. And in his 1962 review essay, Cavell no sooner identified the connection he saw between grammatical investigations and transcendental knowledge than he emphasized important differences between Wittgenstein’s and Kant’s ways of handling the very idea of such insights. Fourth, Hacker’s and Cavell’s qualms give us more than enough reason to contest Williams’ (in fact rather half-hearted) attempts to argue that Wittgenstein’s later work actually embodies a version of the first-person plural transcendental idealism whose basic structure it is the primary purpose of his paper to identify. And fifth, they also give us grounds for doubting the European Journal of Philosophy ]]]:]] ISSN 0966-8373 pp. 1–19 r 2008 The Author. Journal compilation r Blackwell Publishing Ltd. 2008, 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA.

Stephen Mulhall

2

intelligibility of the very idea of transcendental knowledge as Williams develops it that are rather different from those he presents himself. This matters because, in my view, although he never exactly says as much, the conclusion of Williams’ paper implies that his attempts to saddle the later Wittgenstein with a commitment to transcendental idealism are in the service of constructing an objection to those who might wish to take Wittgenstein’s later conception of philosophy with any seriousness.4 1. Williams on the Early Wittgenstein Williams begins his approach to Wittgenstein’s later philosophy by specifying the sense in which he takes it that his early philosophy explicitly embodied a version of transcendental idealist thinking in its treatment of solipsism; for he intends to argue that, in his later work, Wittgenstein does not so much dispense with its three central features as transpose them into a first-person plural key. These three Tractarian ideas are: [T]hat the limits of my language are the limits of my world; that there could be no way in which those limits could be staked out from both sides—rather [they] reveal themselves in the fact that certain things are nonsensical; and . . . that the ‘me’ and ‘my’ which occur in those remarks do not relate to an ‘I’ in the world, and hence we cannot conceive of it as a matter of empirical investigation . . . to determine why my world is this way rather than that way, why my language has some features rather than others, etc. Any sense in which such investigations were possible would not be a sense of ‘my’, or, indeed, perhaps, of ‘language’, in which the limits of my language were the limits of my world. (Williams 1982: 146) We can certainly see why this conjunction of ideas might be regarded as a species of transcendental idealism. The position they define is idealist insofar as it is an expression of solipsism, and transcendental insofar as the limits at issue are not empirical, and so not amenable to empirical investigation, and yet not what one might call transcendent either (not, that is, putative claims about a supra-empirical domain of entities beyond all possible knowledge). What Williams is most concerned to note is that it is an implication of this conjunction of ideas that the insights it embodies are in fact inarticulable. ‘We cannot in any straightforward sense say that there is, or that we can believe in, or accept, a metaphysical, transcendental self . . . ; for neither what it is, nor that it is, can be said, and attempts to talk about it or state its existence must certainly be nonsense’ (Williams 1981: 146). Or, as Williams expresses it elsewhere, in the only sense in which we could meaningfully say that there is a transcendental self, that statement would be false (cp. Williams 1981: 153). And it is the reflexive problem that this creates for any such transcendental idealism that Williams will want to use in criticizing the first-person plural version of this position that he claims to detect in Wittgenstein’s later work—what he describes in the conclusion to his r

The Author 2008. Journal compilation

r

Blackwell Publishing Ltd. 2008