Friday, July 27, 2012

Frege's puzzle about identity

To: Graham C.
From: Geoffrey Klempner
Subject: Frege's puzzle about identity
Date: 25 January 2008 12:36

Dear Graham,

This is my feedback on your University of London Logic essay, in response to the question, 'Does Frege's puzzle about identity show that there is more to the meaning of a proper name than its reference?'

I will say right out that this is an excellent piece of work. I would have liked to have seen a bibliography, but apart from that you have written a model essay in response to this question. If you can keep this up, that puts you in a very select group of three UoL students, well ahead of the rest of the field.

However, my job is to pick holes, where I can!

The question is quite specific, asking whether Frege's puzzle about identity is sufficient grounds for positing Sinn in addition to Bedeutung. It is clear from the original article, 'Uber Sinn und Bedeutung' that Frege does not rest his case on this example alone. So statements to the effect that there is 'no need' for Sinn need to bear this in mind.

Frege remarks at one point in the article that to 'know' the Bedeutung one would have to know everything about the entity referred to, 'but to this knowledge we never attain'. This is part of his broader argument, aided by the identity puzzle, that knowing the Bedeutung is necessarily knowing it via a mode of presentation, through a particular way, by a route or however you want to put this.

Mathematics gives very nice examples of this, e.g. 2+3=6-1. Here the route to reference is crucial to understanding what knowledge this equation expresses: what happens if you perform the functions + and - on particular natural numbers. Frege gives an example from geometry to illustrate the same principle.

More about this in a minute.

Another point to make is that the question does not say, 'succeed in showing that a proper name has Fregean Sinn in addition to its reference'. Although you show awareness of this, your case is centred on the peculiarness of Sinn. Anything else which is not Sinn is merely Vorstellung, and thus irrelevant.

Couldn't there be something which qualified as 'more' which is neither Sinn nor Vorstellung?

That depends to a significant extent on what is meant by 'its reference'. Here we do need to use English because the question largely turns on the nuances of the expression 'having a reference'. In his 'Principles of Logic', Mill claimed, notoriously, that the meaning of a name IS simply the object named. That is one meaning of 'having a reference'. There is nothing to consider except whether an object corresponds to the name or not.

However, it is from Frege that we learned to raise the question what it is for a speaker to know the meaning of a term, or to be fully competent with its use, a question Mill did not consider. (Dummett in his book 'Frege: Philosophy of Language' is very good on this.) You say at one point, 'it seems more intuitively correct to say that what is grasped is the Bedeutung.' That may be the right thing to say, but Frege's response would be one of incredulity. Of course one doesn't 'grasp the thing named', not even if you pick it up and hold it. You grasp it in a particular way, through a route etc.

What is weird about names in natural language is that there is nothing corresponding to the precise functional definition of 'mode of presentation' which you find in mathematics. Natural numbers are defined by the successor function, and so on. What you need to say is that you are not denying that there is an interesting account to be given of what it is to 'grasp the Bedeutung' of a name in natural language. This is what, e.g. Evans attempts to do in his 'Varieties of Reference'. There is something to say here, a theory to give. It is not sufficient to rest content with Mill.

I fully agree with you about the artificiality of examples like 'Hesperus=Phosophorus' and 'Afla=Ateb' (Frege's example of the mountain seen from different sides). I like the idea that an error analogous to misnaming has taken place when someone doesn't know that two individuals which they refer to are in fact one and the same individual, so that an assertion of an identity statement is more like correcting an error than stating a fact.

However, I do also have a bit of a suspicion that pushing this point too hard is driven by ideology rather than a desire to account for our knowledge of natural language. Consider Police work, where a criminal is identified, say, 'The Ripper', and various suspects considered as possible candidates for identity. Ideally, we would like an account of proper names to take into consideration the fact that we can deliberately 'name' an individual fully aware that we don't know (in a sense) know enough to be able to NAME them.

We don't need Sinn to do this. But we need something, a philosophical account of proper names and how they work.

I thought your discussion of 'Italy is Italy' was on the borderline of relevance. Wittgenstein remarks at one point in the Investigations where he is discussing the problem of private language, about the mistaken idea that we have a 'paradigm' of identification in the identity of a thing with itself, that 'War is War' is not intended as an example of the law of identity.

All the best,