I get asked a lot about the role appeal to language plays in philosophy, so I thought I'd write a little bit about that. According to standard lore, in the mid-Twentieth century, there was a movement called ordinary language philosophy. The ordinary language philosophers (or at least some of them) thought that all (most? some?) philosophical problems could be dissolved by appeal to ordinary language. The ordinary language philosophers turned out to be wrong. First, almost all the philosophical problems were not in the end problems about language. Once we were perfectly clear about language, the philosophical problems remained. Secondly, the way in which ordinary language philosophers appealed to language was misguided. Often, they appealed to "ordinary use", without any sense of the distinction between what was part of the linguistically determined truth-conditions of an utterance and what was merely implicated as a function of general conversational maxims. Paul Grice came around and made the relevant distinctions, and it became a lot harder to appeal straightforwardly to facts about use to draw conclusions about meaning (this was genuine progress in understanding communciation, progress that is threatened by advocates of now fashionable 'use theories of meaning').
In recent years, something like ordinary language philosophy has returned, but in a much different (and more modest) form. My friend Michael Glanzberg at UC Davis calls this technical language philosophy. First, instead of appealing to ordinary use, the new sort of philosopher of language appeals to facts about linguistic structure, fully informed by all the progress linguists, logicians, and philosophers have made in understanding language over the last forty years. Secondly, such appeals are typically made in the service of advancing our understanding of a philosophical problem, rather than seeking to "dissolve" it (it was never clear anyway how one "dissolved" a philosophical problem). So one might (for example) argue, as Kit Fine has recently done, that the statue is not identical to the lump of clay that constitutes it, on the ground that the statue is valuable, and the lump of clay is not, and there is no linguistic ground for treating "is valuable" as anything but a simple predication. Or one might argue (for example) as Tim Williamson and I have recently done, that knowing how is a species of propositional knowledge, since constructions such as "know how to ride a bicycle" are linguistically parallel to constructions like "know who to call in case of a fire". Since the latter are best treated linguistically as a species of propositional knowledge, so are the former.
On this model of the philosophy of language, appeal to language has two roles. First, it is a source of insight into possible analyses of philosophically important notions. Secondly, it is one among several ways to test the consequences of a philosophical view. If, for example, your philosophical view predicts that "is valuable" is not an ordinary predication, or "knows how to ride a bicycle" is linguistically quite different from "knows who to call in case of a fire", or that knowledge claims are context-sensitive, that's something we can now evaluate.
A second and more foundational role grappling with particular linguistic constructions has played in philosophy is in sharpening our understanding of the nature of representational content (Josh Dever's comment on a previous post makes this point too). That is, attention to particular linguistic constructions has genuinely advanced our understanding of how to represent content. In the mid-Twentieth century, there was some dim grasp of different models of meaning, e.g. that one could construe a discourse as fact-stating, or non-fact stating. But now we have a much better sense of what these models of meaning look like. Certain models of representational content (truth-conditional semantics, intensional semantics as descended from the philosopher Richard Montague) have proven very helpful for linguists in understanding how language works, and computer scientists in implementing these models. This provides some empirical evidence for these approaches to representational content. Secondly, as I've indicated in an earlier post, intensive investigation of certain linguistic constructions has raised the possibility of very different conceptions of meaning, of the non-fact stating variety. Philosophers such as Allan Gibbard who have intensively investigated some of these constructions (e.g. indicative conditionals) have done so because of their connections to these alternative conceptions of meaning, which they have then gone and applied in other domains (in Gibbard's case, to ethical language).
In my view, it was the intensive focus on modals, conditionals, and counterfactuals that gave us the serious original models of representational content that are now operative throughout philosophy, from philosophy of mind to metaethics. As a result, debates about the nature of content take on a much sharper and clearer focus, and progress is more easily possible.
-Jason Stanley
Recent Comments