Binding and PRO in the SPEC of NP

Abstract

This article is an attempt to analyze some facts concerning around the binding theory under a new assumption based on the A/A′-division of the subjects of NP suggested in Kim(l987). In section 1, the motivation and explanatory power of the division is briefly reviewed, followed by a new interpretation and assumption of the structure of NP. In section 2, the problem of positing PRO in the SPEC of NP is discussed. The theories for the PRO in the SPEC of NP and those against the PRO in the SPEC of NP are carefully examined to draw the conclusion that PRO must be posited in the SPEC of NP. Especially in section 2.2., it is argued that the intermediate projection N' of an NP is actually acting as a maximal projection and should be understood as such at least for a proper analysis of binding phenomena in English NPs. The new assumption of the structure of NP not only enable us to analyze data without any extra conditon or cost to the standard binding theory, but also gives a unified explanation of previously problematic failure of complementary distribution between anaphors and pronominals.