재분석과 전치사 잔류 = Reanalysis and Preposition Stranding

다국어 초록 (Multilingual Abstract)

There have been many references to reanalysis and restructuring in the EST literature. It is generally assumed that these processes are not transformational, in that there is no movement in usual sense, but rater that they induce a change in the constituent structure without necesarily changing the linear order of the string.
We have seen that the formulation of reanalysis presented in this paper may be used to give interesting analyses of some problematic syntacitic phenomena. Goodal (1985) has depended heavily on Stowell(1081)'s Structure-preserving Constraint, whereas one instead might wnat to rely on something like the Case Filter to rule out undesirable post-reanalysis structure. In addition, Goodal (1985) has assume that reanalysis is brought about by the addition of a single monostring.
It has been observed by Van Riemsdijk (1978b) that preposition stranding is a relatively marked phenommenon cross-linguistically. A natural way of deriving this result is to adopt Kayne's(1981) suggestion properly govern its object if it were to assing a θ-role assignment. Riemsdijk(1978b), Stowell(1981) suggests that in some cases, stranding is allowed by virture of Feanalysis-NP Incorporation and Particle Incorporation-conditioned by Structure-Preserving Condition(SPC).
Finally, certain serious problems remain unresoved-in particular, the apparent violations of SPC with respects to PP arguments in WH-movement and Tough-movement constructions-but in general it seems that the program of reducing all cases of stranding to reanalysis is feasible. We will leave the relation of restructuring phrase markers to PF and LF to be discussed in a later section.

목차 (Table of Contents)

• Ⅰ. 서 론
• Ⅱ. 격이론과 그 문제점
• Ⅲ. 재분석과 재구성
• Ⅳ. 전치사 잔류
• Ⅴ. 결 론
• Bibliography
• Abstract