@article{oai:nagoya.repo.nii.ac.jp:00027926,
 author = {中川, 直志 and Nakagawa, Naoshi},
 journal = {IVY},
 month = {Oct},
 note = {In the GB framework, tough constructions as in (1) have been derived by the null operator movement as in (2). (1) John is easy to please. (2) John is easy [CP Opi [IP PRO to please ti]]. However, it is wellknown that the above analysis has some counter examples listed as follows. A. The subject of infinitival clauses cannot appear. (3) *This bedi is impossible [CP [c for] your letter to be under ti]. B. The null operator movement cannot be applied to the subject of the complement of ECM verbs. (4) *Billi is hard to believe [ti to be insane]. It must be noted here that the subject of small clauses can be moved. (5) Jonesi is hard [Opi to consider [ti incompetent]]. Furthermore, the object of control verbs can be moved, too. (6) Johni is easy [Opi to convince [ti to drive a car]]. C. The null operator movement cannot be applied to the object in finite clauses. (7) *The Presidenti is hard to believe [CP [c that] he met ti]. D. The null operator movement cannot be applied to the NP contained in another NP. (8) *Johni is fun to see [pictures of ti]. In this paper, adopting the null operator movement analysis, I will make clear the property of tough constructions which can account for the gramamaticality of the sentenses in (5)-(6) and the ungrammaticality of the sentense in (4). In addition, I will make a brief discussion on the phenomenon in (8). The basic framework of this paper is that of Chomsky (1993). It requires that a feature assigned to a functional head must be checked by the same feature in the Spec-head or head-head relation. This requirement will play the crucial role in this paper.},
 pages = {157--169},
 title = {Tough 構文の派生について},
 volume = {29},
 year = {1996}
}