Notes by RBJon

The Semantic Tradition from Kant to Carnap

by J. Alberto Coffa

Part IThe Semantic Tradition
Chapter 1Kant, analysis, and pure intuition
Chapter 2Bolzano and the birth of semantics
Chapter 3Geometry, pure intuition, and the a priori
Chapter 4Frege's semantics and the a priori in arithmetic
Chapter 5Meaning and Ontology
Chapter 6On Denoting
Chapter 7Logic in Transition
Chapter 8A logico-philosophical treatise
Part IIVienna, 1925-1935
Chapter 9Schlick before Vienna
Chapter 10Philosophers on relativity
Chapter 11Carnap before Vienna
Chapter 12Scientific idealism and semantic idealism
Chapter 13Return of Ludwig Wittgenstein
Chapter 14A priori knowledge and the constitution of meaning
Chapter 15The road to syntax
Chapter 16Syntax and truth
Chapter 17Semantic conventionalism and the factuality of meaning
Chapter 18The problem of induction: theories
Chapter 19The problem of experience: protocols

LSL = Logical Syntax of Language [Carnap37]


The "primary topic" of this book is the philosophy of the Vienna Circle in the decade from 1925 to 1935. The topic is addressed in the second of the two parts of the book, the first of which sets the context by describing the progression of the "semantic tradition" through the nineteenth and into the twentieth centuries.

The introduction puts this account of the semantic tradition in context. The author discerns three major currents in epistemology in the nineteenth century, distinguished by their attitudes toward the a priori. These traditions were:

which denied that there were any a priori truths
which explained the a priori using "pure intuition" (the Copernican turn)
the semantic tradition
which is characterised by a particular concern for meaning, and which accepts the a priori while rejecting Kant's explanation of how a priori knowledge is possible.

Part 1 - The semantic tradition

Chapter 1 - Kant, analysis, and pure intuition

Chapter 17 - Semantic Conventionalism and the Factuality of Meaning

In his introduction Coffa says that Carnap is not so hot on the underlying philosophical issues as he is on "formal level philosophical issues", and that though LSL exhibits "a certain attitude towards semantic matters" he fails to articulate that attitude clearly.

I find myself disagreeing with Coffa about the substance and merits of Carnap's position here, even though I have not read LSL!

I sketch here what I can remember of my disagreement (having failed to take notes while reading).

Factuality of Meaning

The most important point I recall is the one about "factuality of meaning" which gives the title its meaning. Here Coffa more or less openly insists that Carnap must accept factuality of meaning "at the second level" without there appearing to be any evidence that he does. I think by this Coffa means that though meaning at the level of object languages is conventional, in order to be able to give rules prescribing such conventions one must have a non-conventional semantics for the language in which these prescriptions are made.

I doubt that this is something which Carnap would have accepted, and it is not something which I myself accept. There is a problem of semantic regress which does require an answer, but I would not myself answer this problem with the supposition that there is a language somewhere whose semantics is determined absolutely rather than being a matter of convention, which is the best sense I can make of Coffa's "factuality of meaning" thesis.

A Point About Syntax

There is a discussion in this chapter of Carnap's claim that empirical questions could be dealt with syntactically. At least I think its in this chapter, but I can't find it right now. Anyway Carnap appears to say that it is not just questions of L-truth (logical truth) which can be settled syntactically (i.e. formal proof) but also of P-truth (synthetic claims). This Coffa rejects.

There is a question here about what Carnap meant. To me it seems obvious that he is saying that once a physical theory has been formalised, the consequences of the theory can be obtained syntactically. i.e. once the P-rules have been laid down, P-truth becomes a matter of syntax. He is not saying that you can decide syntactically whether the theory is true, i.e. he is not saying either that the P-rules themselves can be decided syntactically, or even that the P-rules are a matter of convention.

The Principle of Tolerance

I think Coffa's objections to this principle largely follow from his belief in the "factuality of meaning". Another source is Coffa's inability to credit that certain philosophical positions held by Carnap could be simply "proposals" rather than beliefs about matters of fact. I however, in relation to the examples raised by Coffa, did not have a problem regarding these as being appropriate for consideration as proposals.

For example:

"When he explained that the problem of foundations and other philosophical questions were 'at bottom (im Grund) syntactical , although the ordinary formulation of the problems often disguises their character' (LSL p331), Carnap was inadvertently expressing the non-conventional character of his convictions"
I don't myself accept this. Carnap has adopted a very general position about the character of philosophy, but does not feel obliged to reflect this position in a wholesale rewording of talk about philosophical problems. This is analogous to his position on pseudo-object sentences. Its not quite the same because in his explanation of how pseudo-object sentences are to be rendered in the formal mode as talk about syntax, he does not talk about the possibility that they be rendered as proposals about conventions or methods rather than as statements about syntax. However, if challenged I think it likely that he would claim that his assertions about the talk about foundations are themselves also proposals, or are covered by his previous proposal about the nature of philosophy.

It is true that Carnap seems sometimes more dogmatic about these things than is strictly consistent with making a proposal, but again, if challenged I think he would moderate his dogmatism rather than his principle of tolerance.

UP HOME © RBJ created 1995/5/19 modified 2009/02/01