Read on Twitter
Thanks again to @davidadger for commenting on my “general linguistics” paper. I take the point that Minimalism may not mean abandoning the natural-kinds approach, but may reduce the posited natural kinds, while preserving the earlier insights. 1/5 https://twitter.com/davidadger/status/1350817508136275969
But there are not many examples of successful reduction, and broad cross-linguistic work of a Minimalist bent (like Daniel Harbour’s) is very rare. Most generative syntax is language-particular, and thus contributes only a single datapoint to the overall endeavor. 2/5
Language-particular analyses may indeed be used to test general theories (as @davidadger says), but we must be clear about the distinction betw. general theories and particular theories. For some reason, @davidadger continues to use “theory” as a synonym of “general theory”. 3/5
Why do @davidadger and I differ – because I see a grammar as “a theory of behaviour”? (like Quine, and as opposed to Chomsky, who regards a grammar as a theory of a natural object?) No, I see a grammars as theories of systems of social conventions. 4/5
Social conventions do not figure in Chomskyan thinking at all, but mental grammars presuppose them. Children base their acquisition not on an amorphous “input”, but are highly sensitive to the social significance of what they hear. And aren't conventions natural objects, too? 5/5
