Is there any determinate structure of meanings to ask for the origin of? Isn't any operative 'system' at most one axiomatization of items produced without reference to it? Does the system enter into the production of meaning, or only its retrospective organization (which is, though, more meaning?)
One could say that this whole question is bad, that there is no definite combinatorial space we inhabit, let alone a fixed logical space of possibilities. Quine's arguments about indeterminacy of translation and inscrutability of semantics try to show that there is no definite answer to the question which such system is in effect right here and now, let alone which system might be active with historical and cultural others.
Logical spaces, and formalized languages with grammatical restrictions on combinatorial space, these may be useful interpretative devices at times, but they are fictions. If they were real, Quine argues, they would sit there like Plato's forms, large abstract entities, collections of propositions and/or concepts. But, Quine asks, *which* abstract entities? How could we ever individuate them? And how assign them? His arguments about the indeterminacy of translation and the inscrutability of reference try to show that there is no definite answer to the question about which such system is in effect right here and now, let alone with presumed historical and cultural others. Davidson argues that there is really no such thing as language, in the sense of one of these great abstract systems.
In this case, the logical spaces would be at best like tangent lines coming off a curve. As we move in our career we might project our motion at any point into a vision of a system which regulates what people are saying and doing, but there are many lines projectible from a curve, and they change constantly. There are no unique systems, semantic or syntactical, which our practice can be shown to uniquely embody. We revise our talk in response to the world, and all those wonderful Sellarsian systems and Platonic abstract forms wither away, except as interpretive tools.
But what is it to say they can be used as interpretive tools? How are translation manuals and T-theories related to the old spaces? How are they individuated? If you talk about theories and generalizations at all, does this problem recur?
(c) David Kolb, 1 August 2001