To: David G.
From: Geoffrey Klempner
Subject: Frege's distinction between sense and reference
Date: 15 August 2002 11:08

Dear David,

Thank you for your letter of 4th August, with your third essay for the Philosophy of Language program, in response to the question, 'What is the point of Frege's distinction between sense and reference?'

Asking 'what is the point' is a way of asking for clarification on what exactly the distinction is and the purpose it serves. These are the two questions that you have to answer. Of course, there are other questions one might consider. Was Frege right to distinguish between sense and reference? or, if Frege's version of the sense/reference distinction cannot be defended 'as is' (e.g. because of the difficulty of accepting an indefinite hierarchy of indirect senses in cases of nested propositional attitudes) is there a version that can be defended? (such as Dummett's, for example).

What is the distinction? I have to take issue with you when you seem to imply that the distinction between sense and reference is an extension of the distinction between concept and object. Maybe this isn't what you meant when you said, 'These ideas [concerning the analogy between concepts and mathematical functions] were applied and extended in his philosophy of language'. However, it is important to see that the sense/reference distinction is a new idea and not merely a generalization or development of the concept/object distinction.

Indeed, the point should be emphasized that for Frege the sense/reference distinction applies to both concepts and to objects. A concept word can have both sense and reference, just as a proper name can.

It is true that in a statement like, 'The morning star = the evening star' the two referring expressions on the right and left of the '=' sign involve different combinations of concepts. In this case, it is clear that the difference lies in 'how one conceives of the denotation of the term'. The concept '...appears in evening' is a different concept from the concept '...appears in the morning'. Similarly, in a mathematical statement of the form 'x=y', where 'x' and 'y' are two different ways of referring to the same number such as your examples '4' and '8/2', the 'route to reference' is clearly displayed in the way concepts are combined together.

Taking this idea to its logical conclusion, however, implies a Russellian solution to the problem of the informativeness of identity statements. That is to say, whenever you have two names on either side of an '=' sign, some equivalent descriptive phrase can be given for each name which displays the two different 'ways of conceiving' of the denotation. As you correctly note, Frege did not hold this view. The crucial point of difference is that the 'mode of presentation' need not be conceptualised, i.e. articulated in words. When we take the most basic terms of the language, both terms which refer to objects (i.e. proper names) and terms which express concepts, the speaker may have no way of giving an adequate definition: they simply know how to use the terms correctly, to refer to an object, or to express a concept.

What, then, is the purpose of the sense/reference distinction?

Although in his essay 'On Sense and Reference' Frege motivates the distinction using the example of identity statements, it is clear that his aim is to account for indirect discourse. I felt that you didn't say enough here to make this convincing (although I accept that this is difficult within the word limit). How does Frege's theory explain the difference between, e.g., 'Tom believes that Mr Hyde struck the coach driver' and 'Tom believes that Dr Jekyll struck the coach driver'? For Frege's theory to work, the sense of 'Dr Jekyll' and 'Mr Hyde' must be the same for everyone. So when Tom expressis his belief, he is not referring to Dr Jekyll (=Mr Hyde) but rather referring to the commonly recognized sense of 'Dr Jekyll' or the commonly recognized sense of 'Mr Hyde'. That is how it is possible, e.g. to believe the first statement but not the second. As we have said, these 'senses' are ontologically a distinct kind of thing from concepts.

You are right to emphasize that senses are not private impressions existing in the mind of a particular speaker, but rather are shared between speakers of the language. But how exactly does this come about? By what process do we come to recognize non-physical objects called 'senses' at the same time as we learn to recognize the physical objects that populate our world? - These are some of the questions raised by Frege's theory of sense and reference.

Given the restriction on length, I think you did a good job in explaining the essence of the idea of distinguishing sense and reference. As you will have seen from Dummett's magnificent book, this is indeed an inexhaustible topic.

All the best,

Geoffrey