On pages 39 and 40 of “Truth and Progress”, Rorty writes:
“We pragmatists think that philosophers who view the defense of ‘our realistic intuitions’ as an important cultural or moral imperative are held captive by the picture of getting closer to a fixed goal. As an initial step in breaking free of this picture, we suggest following Davidson in abandoning what he calls ‘standard ideas of language mastery.’ Then one will think of such mastery as involving ‘no learnable common core of consistent behavior, no shared grammar or rules, no portable interpreting machine set to grind out the meaning of an arbitrary utterance.’ [Rorty has a footnote here, which I put after this quote -BDM] Dropping these standard ideas makes it very difficult to take seriously the idea of human beings as portable representing machines that incorporate a priori knowable input-output functions. The idea of such a machine lies behind both Wright’s notion of Cognitive Command and his Kripkean suggestion that language, meaning, truth, and knowledge might all collapse together if, horribile dictu, it should turn out that there is no fact of the matter about what we have been meaning by ‘addition’. But the skepticism described by Kripke’s Wittgenstein holds no terrors for those who follow Davidson in abandoning the whole idea of ‘rules of language.’…If, as good Darwinians, we want to introduce as few discontinuities as possible into the story of how we got from the apes to the Enlightenment, we shall reject the idea that Nature has settled on a single input-output function that, incarnated in each member of our species, enables us to represent our environment accurately.”
Footnote: “Davidson, ‘A Nice Derangement,’ 445. Dropping these ideas also makes it very difficult to get excited about Wittgenstein’s rule-following argument. For freedom from these ideas permits one to see it as simply a version (adapted to the needs of those who still take the notion of ‘rules of language’ seriously) of a generic argument against the existence of any relation that is both natural (i.e., not simply a product of contingent human practices) and noncausal. That is the sort of relation which representationalists are constantly invoking: For the Sellarsian version of this argument, see Brandom, Making It Explicit, and my ‘Robert Brandom on Social Practices and Representations,’ in this volume.”
I have not (yet) read Davidson’s essay, but here is the Stanford Encyclopedia of Philosophy description of it:
“In papers such as ‘A Nice Derangement of Epitaphs’ (1986), Davidson addresses just this point [“that ordinary speech is full of ungrammatical constructions” -BDM], arguing that while linguistic understanding does indeed depend upon a grasp of the formal structure of a language, that structure always stands in need of modification in the light of actual linguistic behaviour. Understanding a language is a matter of continually adjusting interpretative presuppositions (presuppositions that are often not explicit) in accord with the utterances to be interpreted. Furthermore, this calls upon skills and knowledge (imagination, attentiveness to the attitudes and behaviour of others, knowledge of the world) that are not specifically linguistic and that are part of a more general ability to get on in the world and in relation to others — an ability that also resists any formal explication. In ‘A Nice Derangment of Epitaphs’, Davidson puts this point, in provocative fashion, by claiming that ‘there is no such thing as a language’ (adding the immediate qualification ‘not if a language is anything like what many philosophers and linguists have supposed’). Put less provocatively, the essential point is that linguistic conventions (and in particular linguistic conventions that take the form of agreement over the employment of shared syntactic and semantic rules), while they may well facilitate understanding, cannot be the basis for such understanding.”
Rorty ftw. As far as I can tell Rorty and Dennett are in complete agreement on this issue. Reading the literature on the rule-following paradox has made me much more sympathetic to pragmatism, which, as you’ll recall, I was fairly hostile to over the summer. That said, I still feel like my doubts have not been placated completely, probably because I find it very hard to give up the representationalist worldview. It is clear that, even if rules in the philosophical sense don’t exist, something-like-rules exist and are invoked all the time. How do we, both psychologically and sociologically, follow, understand, and interpret rules? Maybe this is putting the burden of proof on the wrong people – the pragmatist philosopher’s job is just to say, “there is no such thing as intrinsic meaning, and go to a psychologist if you want to understand functional meaning”. Still, the solution of “just give up the idea of language expressing truth conditions and the problem goes away” seems a little underdeveloped to me. Stay tuned for more updates.