Tuesday, November 27, 2012

Russell, Strawson and Donnellan on reference

To: Alistair L.
From: Geoffrey Klempner
Subject: Russell, Strawson and Donnellan on reference
Date: 25th November 2008 12:47

Dear Alistair,

Thank you for your email of 14 November, with your essay towards the University of London Logic BA examination, entitled, 'Reference'.

This is a useful summary of some of the main issues around the topic of reference, which takes in Strawson's criticisms of Russell in 'On Referring' and Keith Donnellan's distinction between the 'referential' and 'attributive' use of descriptions.

However, I do think it is very important to pick examination questions and attempt to answer them. Even if the attempt goes awry, the result will be more useful to you than merely an essay around a general topic like reference.

What you have written here is as good so far as it goes but it gives me very little to comment on because you are just summarising what someone might find in an introductory book or encyclopaedia article, for example.

There are two or three inaccuracies:

1. It is Russell who asserted (in his 1906 article 'On Denoting') that his theory of descriptions makes Frege's distinction between sense and reference redundant. The majority of philosophers who broadly accept Russell's analysis of descriptions would not agree with this claim.

To take just one example, Michael Dummett in his seminal paper, 'Truth' (reprinted in the collection 'Truth and Other Enigmas') offers (among a number of other things) a persuasive argument for regarding Strawsonian reference failure as a species of falsity (hence validating Russell's analysis), yet Dummett is a staunch supporter of Frege's sense/ reference distinction and its importance in the philosophy of language. (See Dummett's brilliant book, 'Frege Philosophy of Language' (1973) probably one of the top ten books by analytic philosophers in the latter half of the 20th century -- and also the book which got me through my BA exams!)

2. In your discussion of Russell's notion of 'logically proper name', the essential point is not that a logically proper name 'uniquely identifi[es] something now and for all time' but rather that it is impossible to doubt the existence of an entity referred to by such a name because it is given in immediate experience. It follows that I cannot know the meaning of logically proper name 'A' and logically proper name 'B' and not know whether or not A=B.

Gareth Evans in his book 'Varieties of Reference' offers a powerful argument for extending (in effect) the concept of a 'proper name' beyond the immediate experiential referents of 'this' and 'that'. Russell's view of the 'certainty' of direct subjective reference -- which leads to the sense datum theory of perception -- was attacked by the later Wittgenstein in his 'Philosophical Investigations' in series of paragraphs which have come to be known as the 'private language argument'.

3. The example of 'The present King of France' is Russell's and not Strawson's, although you seem to imply (although you don't actually say) that it is Strawson's own example. This would suggest to an examiner that you had not actually read Russell's 'On Denoting'. As 'On Denoting' is one of the most important philosophy articles written in the 20th century -- as well as being remarkably short -- I think that you should. You can access the article through JSTOR.

In general, I would say that there are some very gripping problems that can be raised relating to the issue of reference. For example, the question which you mention about the causal theory of reference. What is the theory, and can it be defended against criticisms? Evans defended a version of the theory but then later rejected his view in 'Varieties of Reference. Kripke in 'Naming and Necessity' is regarded as the main author of the causal theory. Does the notion of 'sense' play a valuable/ necessary role in a theory of meaning? What is the correct analysis of statements which refer to a non-existent entity? and so on.

To repeat what I said earlier: the best way to go about this is to pick an examination question from a past paper. You can write notes 'for your own eyes only' if this helps. What you should send me, however, are answers to focused essay questions which make a case -- whatever that case may be. I will then be able to tell you whether (in my opinion) you have made a good case, or how the essay might have been improved to make your case stronger.

Despite what I've said, this is a good start because at least you have covered a lot of ground. The next step is to get stuck in to a particular problem.

All the best,