A Pause

Before going on with Quine's arguments, let's stop here a moment and see what's going on.

1. Quine is arguing that if we follow our analysis of what Kant has in mind when he talks about meanings we first notice that he is not talking about Bedeutung--reference-- whether of singular terms (naming) or of general terms (extension). Instead, he is talking about the Sinn, sense of a sentence and its terms.
2. Quine then says that the concept of sense arose from the Aristotelian doctrine of essences, as a mechanism for moving essences from the world into language.

3. Once this occurs, we quickly come to see that while theory of meaning is not concerned with extensions, it also cannot be concerned with meanings, but must concern itself with synonymy and with object essences as preserved in Sinn as what grounds or secures synonymy and analyticity.

4. And once again we find ourselves back to analyticity. We've abandoned meaning as a way of explicating analyticity, and have replaced it with synonymy which we hope is clearer--not involving any metaphorical notion of containment.

   Synonymy can be defined as follows:
   

   Essentialism:

   A and B are synonymous iff the A's and B's have the same essence.

   Linguistic:

   A and B are synonymous iff A and B have the same sense.
   

Quine doesn't say the first, and really he can't mean it, but it seems to follow from his explanation of the origin of sense. If we look at the Linguistic definition and place it in the context of Quine's explanation of where senses come from, it seems that the Essentialism account must be correct as well.

The problem is that while the Linguistic account is linguistic, it simply dumps us back into the question of meaning which we were trying to avoid; but the Essentialism account dumps us into the world, i.e. takes us out of language. On this treatment of synonymy it seems clear that the business of the theory of meaning not only doesn't concern meaning, as Quine has already told us, but that its business seems to be ontology and essentialism. Now this can't be right.

I believe that the problems arises at Step 2, where Quine gives his account of the origin of sense in essentialism. While it may indeed account for some treatments of sense, and while it does make a good story about senses, how they work in the world, the aboutness of language, it isn't necessarily true in general, or even on anything more than an metaphorical level. There are, after all, theorists of meaning who accept the distinction between sense and reference but who do not tie sense to essences. Conventionalists and meaning as use theorist locate the ground of meaning in convention, agreement, social practices rather than in essences. For them there is no reason to appeal to an Essentialism account of synonymy, nor is there anything questionable about the appeal to sense to explain synonymy. but I'll leave this here and return to Quine.

To be fair, Quine doesn't really commit meaning theorists to accepting essentialism. However, if the goal is to eliminate the ``obscure intermediary entities'' (meanings) from the discussion, it's hard to see what else he can say.