¹ Kripke, SaulWittgenstein on Rules and Private Language (Oxford: Blackwell, 1982), 37Google Scholar. Cf. 11 and 23f.

² Gibbard, Allan ‘Meaning and Normativity’, in Villanueva, Enrique (ed.), Philosophical Issues 5: Truth and Rationality 1994, 95–115Google Scholar. The quoted passage occurs on p. 100.

³ Boghossian, Paul, ‘The Rule-Following Considerations’, Mind 98, 1989, 513.Google Scholar

⁴ Even where correctness of use is not, as in the quotation from Boghossian, identified with truth or warrantedness of application, uses which are applications are commonly used at least to illustrate correctness of use. See, for example, McCulloch, Gregory, The Mind and Its World (London and New York: Routledge, 1995), 100Google Scholar, Loewer, Barry, ‘A Guide to Naturalizing Semantics’, in Hale, Bob and Wright, Crispin (eds), A Companion to the Philosophy of Language (Oxford: Blackwell, 1997), 108–26Google Scholar and Horwich, Paul, Meaning, (Oxford: Clarendon Press, 1998)CrossRefGoogle Scholar, 92ff.

⁵ See Wright, Crispin, Wittgenstein and the Foundations of Mathematics (London: Duckworth, 1980)Google Scholar, ch. 2 and McDowell, John, ‘Wittgenstein on Following a Rule’, Synthese 58, 1984, 325–63CrossRefGoogle Scholar. The notion also seems to figure in Boghossian's discussion at the point at which he emphasizes that normativity has to do with a relation between meaning something by an expression at some time and the use of that expression at that time. (‘The Rule-Following Considerations’, 513). This suggests that the use may or may not be in keeping with the meaning and would be correct if it is and incorrect if it is not.

⁶ For details, see note 4.

⁷ Op. cit. 674.

⁸ It might be suggested that it cannot be merely knowledge of the conditions for true application of ‘arthritis’ which commits one to avoiding the use Fred makes of the term. This might be said on the grounds that since Fred and the doctor both know that ‘arthritis’ is given true application to a condition if and only if the condition is arthritis, what the doctor has and Fred lacks cannot be accounted for in terms of knowledge of the conditions for true application of the term. The right response here, I think, is that while both Fred and the doctor have knowledge of the conditions of true application of the term, the doctor, unlike Fred, appreciates what this knowledge commits him to. This appreciation amounts to the fact that the doctor knows enough about arthritis to know that unless a condition is due to inflammation of the joints it is not arthritis. We should resist the tempting thought that Fred does not really know the conditions for the true application of the term, since that makes it hard to see how it can be that he has some grasp of the meaning of the term.

⁹ As well as being plausible in itself, this view avoids familiar regresses deriving from thinking that all conformity to a rule is a matter of following an instruction in the way that one follows a recipe.

¹⁰ Recall the distinction between expressive and interpretative uses made in section 2.

¹¹ I look further into issues surrounding intention and instrumental reasoning in ‘Normative Reasons and Instrumental Rationality’ in Bermúdez, José Luis and Millar, Alan (eds) Reason and Nature (Oxford: Oxford University Press, forthcoming)Google Scholar. John Broome's work on practical reasoning drew my attention to the importance of distinguishing between what one ought to do and what one is committed to doing. See his contribution to the volume just cited.

¹² Meaning, 92f. Horwich appears to conceive of normativity as Boghossian does but with the proviso that the central explanatory consideration is about correct, in the sense of true, application.

¹³ Confusingly, at least in relation to my preferred terminology, he says that (5) implies (6), but appears to mean by that only that if (5) is true then (6) is as well.

¹⁴ Meaning, ch. 8.

¹⁵ It should be stressed that I sympathize with those who are sceptical about establishing that meaning (or the having of propositional attitudes) is intrinsically normative on the basis of considerations about correct, in the sense of true, application (or belief). In addition to Horwich, see Heal, Jane, ‘The Disinterested Search for Truth’, Proceedings of the Aristotelian Society 88, 1988/1989, 97–108CrossRefGoogle Scholar, Papineau, David, ‘Normativity and Judgement’ Proceedings of the Aristotelian Society, Supplementary Volume 73, 1999, 17–43CrossRefGoogle Scholar, and Dretske, Fred, ‘Norms, History, and the Constitution of the Mental’, in his collection, Perception, Knowledge and Belief: Selected Essays (Cambridge: Cambridge University Press, 2000), 242–57. My point here has been that normativity theorists, and their critics, should look elsewhere. So far as linguistic meaning is concerned, the crucial considerations are about use which is correct in the sense of being in keeping with meaning.CrossRefGoogle Scholar

¹⁶ Closely parallel considerations apply with equal force if an aiming-at-truth account of normativity is developed in terms of aiming at truth in assertion and if the strategy is detached from the deflationist aims of Horwich.

¹⁷ Work relating to this paper began in 1997 during a research leave part of which was spent as a Visiting Fellow at Clare Hall, Cambridge, and as a Visiting Scholar in the Faculty of Philosophy at Cambridge University. I am grateful to the College and the Faculty for facilities provided, and to Ross Harrison, Jane Heal, Isaac Levi, and Hugh Mellor for stimulating discussion during that stay. The leave was supported by a Research Fellowship from the Mind Association and by a grant from the Carnegie Trust for the Universities of Scotland. I am grateful to both of these organizations. I have benefited enormously from discussion with colleagues at Stirling and with the audience at the lecture based on ideas and arguments in this paper. Thanks are due to José Bermúdez, Bob Hale, Peter Sullivan, Neil Tennant, and Tim Williamson for helpful and encouraging comments.