UCL Working Papers in Linguistics 15 (2003)
What’s in an island?
HANS VAN DE KOOT & ERIC MATHIEU
The primary aim of this paper is to argue that a unified approach to strong and weak islands is misguided: strong and weak islands are different phenomena, to be accounted for at different levels of representation. Our central claim is that a theory of locality that represents these islands in one level cannot capture the pattern of island sensitivity exhibited by the various grammatical dependencies. The relevant empirical difficulties are easily overcome in an approach that treats strong islands as syntactic and weak islands as scope-related. We then attempt to answer the question whether scope is represented at LF or later. In order to maintain the view that scope is an LF phenomenon, one has to demonstrate that strong islands are not due to properties of LF but have their origin in the order of syntactic operations in a derivational syntax. We show that such derivational theories of strong island phenomena have major shortcomings. This strongly suggests that scope is not an LF phenomenon.