Past work and drafts

I work mostly on 'anti-lexicalist' or 'neo-constructionist' syntax, with a special interest in Hagit Borer's approach, the 'Exo-Skeletal Model' (XSM). The theory is closely related to the more mainstream program initiated by Distributed Morphology, but differs in some details. In particular, DM often allows a semantic component that manipulates syntax (in 'alloseme selection'). I am attracted by the idea that the syntax builds up unambiguous formulae, interpreted mechanistically by the semantic system, completely independent of world-knowledge and context. This is related to philosophical concerns about semantic minimalism.

I also work on quantification in cognition at large, following much work that began at UMD. This psycholinguistic approach to quantification lends itself to a picture of meanings that are firmly 'inside the head', providing, again, mechanistic, context-insensitive instructions for thought. Following Paul Pietroski, I think of the meaning of a sentence as something like a recipe for a thought. It makes you think of the world in one way, rather than another. This also interacts with polysemy in interesting ways, as recipes can be neutral between certain things, but explicit about others.

In summary, the way I hope the world turns out to be is as follows: the syntax generates formulae that, without further editing, are assigned an interpretation part-by-part and by rule. Semantics is invariant, blindly composing the output of the syntax, not paying attention to world-knowledge or context. Much of the work traditionally assigned to composition semantics, then, belongs with syntax or with pragmatics.