In response to Quine's rejection of the analytic-synthetic distinction, Strawson and Grice appear to reduce Quine's rejection (or skepticism) of synonymy to a rejection of meaning. What is/would be Quine's response?
It would be to argue against meanings as mental or objective entities. Grice and Strawson rely on meaning as something propositional statement "inherently" has, Quine's position, like late Wittgenstein's, is to replace reified meanings with linguistic roles in social use when interpreting language. In Word and Object, that came out about a decade after the Two Dogmas, he argued that even if "inherent" meanings existed they wouldn't make much of a difference, because due to indeterminacy of translation we wouldn't be able to communicate them.
"Critics have said that the thesis is a consequence of my behaviorism. Some have said that it is a reductio ad absurdum of my behaviorism. I disagree with this second point, but I agree with the first. I hold further that the behaviorist approach is mandatory. In psychology one may or may not be a behaviorist, but in linguistics one has no choice".
One has no choice because children learn language from overt social practice, and grow up into as much an authority on meanings as anybody else.
Grice and Strawson resurrect the analytic/synthetic distinction based on synonymy treated as a black box. Quine, they argue, makes a Socratic assumption that something not describable in simpler terms without circularity has no place in discourse. But it should be rejected, clearly some terms must be admitted as basic, and why not synonymy? It can be reliably recognized and taught to novices, so the fact that it can not be characterized without circular reference to analyticity should not be grounds for dismissing it, perhaps it latches on to something "intuitively".
However, Quine was asking for a description of analyticity-in-a-language, where language is varied, so the description has to explain analyticity as such, not just give an assurance that we will be able to tell when it comes to it. He did not contest that groups of sentences can be recognizably prescribed as analytic in particular languages. And it is unclear if reduction to synonymy can do the job even in the sense of Grice and Strawson, because recognition of synonymy may not be "reliable" or "teachable" beyond particular languages with familiar usage.
Quine's attack on the analytic/synthetic distinction is contained in a series of papers: Truth by Convention, Two Dogmas of Empiricism, Carnap on Logical Truth, and in the early chapters of Word and Object. In broad brush terms he argues:
- Attempts to establish that some sentences are true by virtue of their meanings is either vacuously true or viciously circular.
- Even sentences identified as analytic by its defenders are not actually immune to revision in the light of new data. For example, the definition of momentum was revised in the light of special relativity.
- Sentences cannot be understood one-by-one in isolation from others, but only by reference to a broader quantity of knowledge. This is Quine's holism thesis, and implies that we cannot neatly parcel up sentences and specify verification conditions for them individually, nor identify those that have no verification conditions because they are analytic.
- The concept of 'meaning' is according to Quine unclear and lacks explanatory value. This is part and parcel of his doctrine of the indeterminacy of translation, and follows from his behaviourist approach to linguistics.