초록 close

PRO의 통제를 일종의 이동으로 분석하는 Hornstein (2001)에 다음과 같은 두 가지 문제점이 있음을 살펴보았다. 첫째, PRO는 인상구문의 NP-흔적을 위해 제거되었다는 점을 들 수 있다. 둘째, 의미역기준(θ-criterion)이 폐기되어야 할 위기에 처해 있다는 점을 들 수 있다.다음 장에서는 2.2에서 살펴본 Hornstein의 문제점에 대한 이론적인 증거를 제시하기로 한다.


Control and Movement in Infinitival Constructions Hwang, Choon-shikHornstein (2001)'s proposal to control in terms of movement has two serious problems as follows: First, PRO is eliminated in favor of trace, and second, the θ-Criterion must be abandoned.The purpose of this study is to give the systematic explanations of Control and Movement in the infinitival constructions. To do so, I differentiate between raising constructions and control constructions, analyzing Culicover & Jackendoff (2001), Martin (2001), and Carnie (2002).Following Chomsky and Lasnik (1995), I assume that PRO has null Case, where null Case is licensed by non-finite Ts. Collins (1997) postulates two types of null Case (TPRO and Tnull) and relies on the assumption that checking of null Case can be asymmetric to allow ECM and successive cyclic movement to satisfy Last Resort. I also assume that there are two types of null Case feature. Here, I refine the Collins' types and suggest that all the infinitival non-finite T have TPRO/trace; TPRO represents features which check features of PRO in control construction and Ttrace represents traces which license A-movements in raising construction. In terms of TPRO, we can find it possible to raise control construction in matrix verb, while in terms of Ttrace, find it possible to raise construction if the accusative Case is externally checked by the matrix verb.