다중 wh-구문의 우위효과에 관한 연구

On the Superiority Effect in Multiple Wh-questions

This paper attempts to propose a way to the superiority effect under the ECP. Wh-questions without resort to the superiority effect under the ECP. Wh-movement in syntax appears obligatory in English. This is assumed to be due to the property of the wh-operator, which needs to bind a variable in its scope. Since multiple Specs of CP are not available at S-S in English, there must exist wh(s)-in situ in multiple wh-questions. The superiority effect under the ECP has been assumed to be able to explain the different grammaticality of multiple wh-questions(Lasnik & Saito 1984). Despite of many trials to explain asymmetries among wh-elements, there still remain several problems. In this paper I propose three things. First, the superiority effect under the ECP, which conflates lexical government and antecedent government, is in lack of generality and that often makes if difficult to find out which wh-element moves to the Spec of CP prior to the others. Second, following Reinhart(1981), Lasnik & Saito(1992) and Kuno & Takami(1992), I propose that there should be S” whose Spec accommodates topic and sentential adjunct wh-expressions like ‘why’ and ‘how’ at the base. Finally, it is proposed that with the base-generated Spec of S”, we can explain which wh-phrase moves prior to the others. Thus, the ill-formed sentences are not ruled out by the Superiority Effect but by the hierarchical order of base-generated position in their structures.