스페인어 부가의문문의 도출에 대한 통사적 분석

Abstract

The goal of this paper is to deal with the derivation of Spanish tag questions, which have received little attention in generative linguistics. First, although scholars have been assuming tag questions to be the products of ellipsis, we point out that in Spanish, predicate ellipsis constructions and tag questions show a contrast with respect to the appearance of modal verbs, and therefore argue that it is not clear whether Spanish tag questions are derived from predicate ellipsis. Second, we argue that attempting to account for the derivation by adopting Culicover(1992)''s bi-clausal analysis faces several empirical problems. As an alternative, we provide a mono-clausal analysis of the construction. We argue that its basic structure is CP and that its surface structure is derived by moving the polarity particle to C and the host sentence to the CP-adjoined position respectively. We show that this mono-clausal analysis can simply account for why neither personal pronouns nor verbs can appear in tag: because both are included within the lower copy of the moved host sentence that is deleted at PF. Finally, based on the analysis of the writing survey conducted on learners, we suggest what should be particularly emphasized to learners when teaching them about tag questions.