To: Christian M.
From: Geoffrey Klempner
Subject: Reference: between subjectivity and objectivity
Date: 25th September 2008

Dear Christian,

Thank you for your email of 16 September with your University of London Logic essay entitled, 'Reference - between subjectivity and objectivity'.

There are some good things in this essay, and I like the idea that there is room for compromise between a Kripkean and Fregean account of reference, and also the implication that we should not be looking for a cut-and-dried 'theory of reference' but rather be aware of the somewhat messy way that reference to objects actually works in natural language, a point which Donnellan's article first brought to prominence.

However, there is a confusion in your use of the terms 'internalism' and 'externalism' which undermines your case to some extent.

First of all, what are internalism and externalism? Both the internalist and the externalist are prepared to embrace a realist view of the external world, as adumbrated in Frege's essay, 'The Thought: a logical inquiry' where Frege goes to some lengths to explain what he means when he says that our thoughts refer to things outside is in the world, rather than to ideas in our own minds. (It must be remembered that Frege was writing at a time when idealism was a the dominant philosophy.)

Given a realist view of reference, however, the question is still wide open which of our terms -- or whether any of our terms -- function as rigid designators in Kripke's sense.

It is important to recognize that internalism is not just the common sense idea that what we refer to or talk about depends on our knowledge. It is ideological. The internalist believes, for principled reasons, that there is no way in which we can secure (realist) reference to things out there in the world except via our knowledge. In Putnam's much quoted formula, 'Meaning is in the head'. Rorty, in his excellent book 'Philosophy and the Mirror of Nature' characterizes this view as the belief that knowledge is a process of 'mirroring' reality, and that your language serves as the mirror, a view which Rorty vehemently attacks.

Kripke, and other writers in a similar vein such as Putnam, Evans and McDowell, also attack principled internalism. It could be argued, however, that Kripke goes too far in the opposite direction, creating a genuine problem about how our thoughts can reach thousands of years back into the past, along the slender thread of his 'chain of communication'.

Rorty is an anti-realist about truth, which has consequences for any view of reference. That alone should suffice to show that we can't equate externalism with realism. If anything, externalism naturally invites (although it does not entail) an anti-realist view of meaning and truth, as any reading of Wittgenstein's Philosophical Investigations makes apparent.

Michael Dummett, in his Appendix to the chapter on Sense and Reference in 'Frege Philosophy of Language' (Duckworth 1973) attacks Kripke, or, rather defends Frege against Kripke's arguments, although Dummett also argues for an anti-realist view of meaning and truth, claiming support for his views from the later Wittgenstein. On the other side, McDowell (see his seminal paper 'The Sense and Reference of a Proper Name', Mind 1977) defends a realist view of meaning and truth and externalism, along broadly Davidsonian (Donald Davidson 'Truth and Meaning') lines.

I don't normally go to such great length in sketching the background when I review an essay, however, in your case because you raise such fundamental issues you need to 'go to the source' in order to form an adequate picture.

Let's ask, how good are Kripke's arguments? Viewed as an attack on ideological or principled internalism, I think they are successful. The only question then is what exactly is the alternative to principled internalism, what an adequate theory of proper names should be.

Dummett made the point in his Appendix that Kripke's criticisms of the cluster of descriptions theory can be easily met by making a Russellian 'scope distinction'. It is possible that the teacher of Alexander did not teach Alexander. No problem with that, so long as you get the scope of the description operator right.

The only real bone of contention concerns existence. If none of the things we believe about Aristotle turned out to be true, would that mean that it would be true to say that 'Aristotle did not exist'? As philosophical conundrums go, this is about as gripping as the Medieval question, 'How many angels can dance on the head of a pin?' -- Say what you like, as long as you understand what it is that you are saying.

Is Dummett right? Yes, in his continued insistence on the importance of the distinction between 'full understanding' and 'acting like a tape recorder'. This pushes aside a lot of the irrelevant discussion provoked by Putnam's notion of the 'division of linguistic labour'. The real issue is how one can formulate a theory of what it is to 'fully understand', to know what one means, without embracing an internalist description theory. Gareth Evans attempts this in 'Varieties of Reference' (OUP).

You raise the question of twins. This is actually a very good way of provoking our intuitions on the question of internalism vs externalism. Suppose you go to a party and meet a corporate tax lawyer who is interested in philosophy, but you don't catch his name. Later, you tell another friend who is thinking of doing the UoL BA about the man you met. Unknown to you, the tax lawyer (let's call him Francois) has a twin brother Jean who is also a tax lawyer interested in philosophy and was also at the party. What makes your thought, a thought about Francois rather than a thought about Jean? In the hubbub, no-one else noticed the two of you discussing Kripke in the corner of the room. An internalist would say that there is no determinate object of your thought: your definite description, 'The man I talked to about philosophy at the party' fails to refer. An externalist would say that the twin you refer to, the one *you mean* is the twin you actually *met*. Anyone who remains mystified about how this consequence could follow is a principled internalist.

On the question of existence, a natural move by an externalist would be to deny that a thought is expressed, in cases where there is no object out there in the world. 'No object, no thought.' An internalist will always insist that a thought was expressed, and look for a suitable definite description in order to give a content to the thought. If you think you are thinking about something, and your thoughts are not logically incoherent, then you must be thinking about something says the principled internalist.

All the best,

Geoffrey

P.S. I should have made clear: All the philosophers I am discussing are 'realist' about the existence of objects outside us. This goes equally for the 'realist about truth and meaning' and the 'anti-realist about truth and meaning'.