Toward a unified theory of wh-in-situ and islands

Abstract

In a number of wh-in-situ languages, only causal wh-adverbs corresponding to ‘why’ exhibit island effects. The widely adopted explanation for this by Tsai (On economizing the theory of A-bar dependencies, 1994a, Nat Lang Linguist Theory 12(1):121–175, 1994b, in: Del Gobbo, Hoshi (eds) UCI working papers in linguistics, University of California, Irvine, 1999, J East Asian Linguist 17(2):83–115, 2008) is to draw a distinction between wh-nominals and wh-adverbs, where the former are unselectively bound by an operator in their scope position, while the latter must raise at LF, thereby inducing island violations. The result is a hybrid approach where wide scope of wh-words is derived by two distinct mechanisms. In this paper, I show that the island-sensitivity of wh-adverbs can be captured in a unified approach based on the observation that island sensitivity correlates with the adjunction height of the adverb. The resulting approach allows us to treat all in situ wh-phrases alike and leads to a unified theory of wh-in-situ which does not rely on LF movement. Furthermore, it will be shown how the analysis extends to other to donkey anaphora, wh-islands and A-not-A questions.