Thursday, May 5, 2011

Dummett's argument for a global anti-realist theory of meaning

To: James D.
From: Geoffrey Klempner
Subject: Dummett's argument for a global anti-realist theory of meaning
Date: 12 April 2002 12:46

Dear Jim,

Thank you for your e-mail of 28 March, with your fourth essay for the Philosophy of Language program, in response to the question, 'Give a critical account of the main features of Michael Dummett's argument for a global, anti-realist theory of meaning.'

As a second-year undergraduate, Dummett's brilliant book, 'Frege: Philosophy of Language' made a deep impression on me. Later, as a graduate student at Oxford in the late 70's early 80's., I attended some of Dummett's seminars. My doctoral thesis, 'The Metaphysics of Meaning' included a critical examination of Dummett's account of the form of an anti-realist theory of meaning'.

Yet, in all those years of study, I can honestly say that I have never *really* understood how, on Dummett's view, an anti-realist theory of meaning could actually be constructed.

I suspect that Dummett does not know either. In an interview which he gave for a religious program around that time, he speculated that it might be possible to use the anti-realist account of meaning as a premise in an argument for the existence of God, just as over 200 years ago Bishop Berkeley had argued for God's existence on the basis of the idealist theory that 'to be is to perceive or be perceived'.

In other words, what Dummett was saying that it was inconceivable that things could really be the way the anti-realist philosopher of language describes them, without God in the picture to preserve the objectivity of truth.

In your essay, you have given a reasonable summary of the main steps of Dummett's argument. You rightly stress the intuitive plausibility of the idea that 'meaning is use': 'Perhaps one could say that a good "theory of meaning" would provide a balance between what understanding a language is, with what one could know of a language.'

The key move in Dummett's argument is the requirement that understanding of meaning should be *manifested* in linguistic practice. In learning language, we learn to follow rules of various kinds, and there is no way of telling whether we have succeed in learning a rule or not, or whether we have cottoned on to the meaning of a particular word or phrase, other than by how we actually use the word or phrase in question in concrete situations.

When it comes to statements which do not have 'effective decision procedures', such as statements about the past, the only thing we can do in a learning situation is show that we grasp the meanings of the terms of the statements in question. So, in one sense we 'understand' the statement, but in another sense we don't.

In Intuitionist mathematics, which does not allow the law of excluded middle, certain mathematical propositions which can be proved in Classical mathematics are not accepted as theorems because the proof involves illicit steps. You mention the connection with Dummett's work on mathematics. Clearly, his intention is to provide an analogous 'intuitionist' view of empirical discourse about the world. The problem is that in empirical discourse we don't discover things by 'proving' them a priori, but by looking at the world, or by formulating theories and testing them. So there are in fact no knowledge claims that a realist philosopher of language would make about the world that an anti-realist philosopher of language would deny. That is why it is so difficult to see what the anti-realist is actually saying, and what would be the actual consequences of adopting the anti-realist view.

My feeling is that Dummett is looking in the wrong place for an explanation of the metaphysical 'visions' of the realist and the anti-realist. He thinks that there has to be something that follows if you are an anti-realist, e.g. about the past, namely, that you hold a different 'theory of language', and a different logic. But Dummett has never got beyond saying what such a theory 'might be like'. He has never actually given the theory, or shown how it works.

The alternative is to accept that the anti-realist about truth, like Berkeley, is not talking about language but about the world. The anti-realist has a particular metaphysical view of the nature of reality which is different from the realist view. Philosophical argument can explore this difference, but it cannot be expressed as a clash between different theories of meaning.

All the best,

Geoffrey