Thursday, March 13, 2014

Donald Davidson on radical interpretation

To: Christian M.
From: Geoffrey Klempner
Subject: Donald Davidson on radical interpretation
Date: 22nd November 2011 12:38

Dear Christian,

Thank you for your email of 13 November, with your essay for the University of London BA Philosophy of Language module, in response to the question, 'What role does the radical interpreter play in Davidson's account of meaning?'

This is a question which begs for an account of what it is to offer a 'theory of meaning' (cf. Dummett 'What is a Theory of Meaning? (I)' in Guttenplan 'Mind and Language', 1975, and 'What is a Theory of Meaning? (II)' in Evans and McDowell 'Theories of Meaning', 1976). Between Dummett and Davidson, to take just two philosophers who have been involved in this debate, there is a huge gulf concerning the appropriate aims of a theory of meaning. What is it to represent a speaker's knowledge in the form of an explicit theory (if that is the right way to put the question)?

Considerations about rule following, which you refer to at the end of your essay, are very much at the forefront of Dummett's argument (especially in WTM II) regarding the requirement that knowledge of meaning be 'manifested' in linguistic practice. (See also McDowell's seminal article, 'Truth Conditions, Bivalence and Verificationism' in Evans & McDowell, op. cit.)

Davidson's original idea is beautifully simple, and it is not so hard to see how he could have inspired an entire generation of graduate philosophy students (as it happens, the generation just before me) to follow the example set by 'On Saying That' in seeking to tame 'difficult' semantic idioms in a Tarski-style compositional analysis. That's an aim worth pursuing in itself, for the illumination it provides, quite apart from the question which Davidson pursued into the nature of radical interpretation.

However, I do accept the point you make here, that radical interpretation looks like, or is presented in a way which seems to fill a lacuna in Davidson's original account in 'Truth and Meaning'.

There is a subtle point to be made here (exploited my McDowell) concerning just who we imagine to be doing the 'radical interpreting'. An alien from the planet Zog, or just you and me? Just as Davidson's Tarski-style analysis shows meanings rather than offering a reductive analysis (because it's compositionality that we're after, nothing else) so the work of the radical interpreter in 'making sense' assumes a great deal (a point you effectively make, but in a way that looks like a criticism of Davidson) about our common nature and culture.

It is something quite different to seek the kind of reductive, behaviouristic account of meaning of the sort which Quine is interested in. Quine (along with Dummett) is McDowell's main target in 'Truth Conditions, Bivalence and Verificationism'.

There's a good, sophisticated account of the kind of 'holistic' theory Davidson is seeking in Christopher Peacocke's 'Holistic Explanation', 1980. There is a circle, as you describe, of beliefs, intentions and meanings which offers no single point where one can 'break in'. The idea that one starts with sentences that 'the natives' would readily assent to (like the sentence 'Snow is white' in the presence of snow) is sadly deficient if taken literally as a recipe for radical translation, indeed misses the point about holism, that there are no privileged entry points. We just make assumptions and test them, then form new assumptions in the light of those results and test those, and so on, a process which in principle can carry on indefinitely.

Why is this interesting? why focus on radical interpretation when in the real world we rarely find ourselves in a position of needing to slow down in this way?

From a broad perspective, what Davidson is seeking is an account of 'meaning as use', just as Wittgenstein sought. (What's so interesting about the debate is that Dummett thinks he is doing this too!) It is in human behaviour, including linguistic behaviour, that the facts about meaning are to be found, and the superstructure of semantic interpretation rests loosely on the hard, nitty gritty facts of bodies and vocal chords in motion.

So the question really comes down to: is Davidson's story of the radical interpreter the best way to approach this? As you observe, Davidson adjusted his theory as time went on. But the key point, it seems to me, is to comprehend this 'looseness of fit', a point which Wittgenstein also effectively makes in his considerations regarding rule following.

If we can't adopt the original suggested methodology of 'identifying T-sentences' (like 'snow is white' in the presence of snow), what is left of the suggestion that 'there is' an appropriate procedure for the radical interpreter? What is to be gained from even raising methodological questions at this point? This is where I am sceptical.

So I am tempted to say that Davidson is relying here on a perceived connection between his original account in 'Truth and Meaning' and considerations on radical interpretation, which just does not exist in reality.

Grade: 70

All the best,

Geoffrey