1. W.V. Quine, Word and Object, M.I.T. Press, Cambridge, Mass., 1960.
2. Accepting, that is, the interpretation of his remarks that is discussed above.
3. Elsewhere, Quine states that “the learning of these wholes (sentences) proceeds largely by an abstracting and assembling of parts” and that “as the child progresses, he tends increasingly to build his new sentences from parts” (p. 13). For consistency of interpretation, we must suppose that this refers to “analogical synthesis”, since the three methods enumerated are intended to be exhaustive. If something else is intended, then the scheme again reduces to vacuity, until the innate basis for the “abstracting” and “assembling” is specified.
4. It is interesting that Russell, in his Inquiry into Meaning and Truth, Allen and Unwin, London, 1940, with his concept of real logical form and of logical words as expressing a mental reality, does appear to presuppose a structure that would avoid at least these very obvious problems. But a discussion of Russell’s quite intricate and interesting approach to these questions, though a useful undertaking, is impossible within the scope of this paper.
5. The reasons for this choice would take us too far afield, into a much more general consideration of Quine’s thesis, developed later in the book, about the scheme of discourse that one must use in “limning the true and ultimate structure of reality” (p. 221), and in describing “all traits of reality worthy of the name” (p. 228).