PRO Gate and movement

Hirohisa Kiguchi University of Maryland at College Park 0. Introduction There has been a controversy concerning the approach to grammars. The issue is whether to take a representational approach or derivational approach. To capture syntactic phenomena, the former exploits the notion of chains and the latter uses histories of movement. Brody (1995) who advocates a representational version of the Minimalist Program argues that if both the notion of chain and movement are constrained by c-command, one of them must be discarded because there would exist too much redundancy. Sideward movement proposed by Nunes (1995), which de-couples movement from c-command supports a derivational approach, for the notion of chain would not cover this operation. But Nunes (1995) also argues that the notion of chain is necessary at the end, to license the result of movement. Thus, his system remains a hybrid one. In this paper I will examine PRO gate phenomena first discussed in Higginbotham (1980). One of the main interests of this phenomenon is that I believe its correct analysis argues against having a chain at all. (1) illustrates the phenomenon: