Earticle

Based on the Chomsky(2005)'s proposals, this study aims at accounting for the A'-movement in French croire+inf. constructions which has been considered a unique and exceptional phenomenon. Extending conclusions from previous work in 2001 and arguing for a system in which the phase status of C and v* is increasingly strengthened, Chomsky(2005) makes three claims to explain the extraction from subject: first, C and v*, not T or V, are phase heads; second, phase heads have both Edge- and Agree-features, that is, they trigger both A'- and A-movement; third, Edge- and Agree-features of a phase head apply in parallel at the end of the phase. Making sure that various analyses of these constructions result from the non-uniformity of derivation between English believe-verb and French croire-verb constructions, we adopt the proposals of Chomsky(2005) in order to explain the complicated structural status of A'-movement. After observing that the A'-movement of subject in infinitival does not trigger the agreement of the past participle in matrix verbal phrase, a standard French grammar rule, we insisted that this subject be analyzed as a real subject of infinitival, but not as an object of "croire". This observation led us to a first conclusion that the infinitival of "croire" is a real CP phase in which all the computational operations are accomplished with the exception of successive-cyclic movements. Applying the three proposals of Chomsky(2005) to the A'-movement in these constructions, we verified that the parallel application of A'- and A-movement can account for their interrogative structure in an effective manner. Finally, we concluded that the [Tense] feature of infinitival influences the structural distinction between event- and stative- infinitival which determines the grammaticality of the whole sentence.

Young-min Lee (2006), Infinitival complement clauses in English. The aim of this paper is to analyze the syntactic and semantic characteristics of infinitival complement clauses in English based on the types of matrix verbs. I argue that control infinitivals are [+tense, -finite], EPP, person, and number features whereas ECM and raising infinitivals are [-tense, -finite], EPP and person features. Control infinitivals have unrealized future tense and allow eventive predicates since they are tensed. In contrast, ECM and raising infinitivals don't have unrealized future tense and allow stative predicates since they are not tensed. Perception bare infinitivals select a single clausal constituent as their complement. The temporal interpretation of perception bare infinitivals must be identical to that of the matrix clause and the highest degree of finiteness correlates with the highest degree of indirectness.