Reconstruction, Beck Effects and Wh-phrases in Situ

RECONSTRUCTION, BECK EFFECTS AND WH-PHRASES IN SITU* Keiko Y oshida Department of English Language and Literature, Waseda University, Tokyo, Japan E-mail: yoshidak@mn.waseda.ac.jp ABSTRACT This paper argues that wh-in situ is interpreted in its original position. Focusing on the ambiguity of how many sentences, it shows that the semantic behavior of wh in-situ can be explained without postulating LF (phrasal) movement. It is then pointed out that the fact that wh-phrases in situ lack anti-reconstruction effects follows directly from this non movement approach. The paper also shows that certain intervention effects discussed by Beck do not necessarily present a problem to this approach. It is suggested that the intervention effects are due to a constraint on movement of `wh- features', but not of `wh-phrases'. 1. INTRODUCTION This paper discusses the interpretation of a wh-phrase that remains in situ. Huang [9] proposes that wh-phrases in situ move at LF (Logical Form) just like overtly moved wh-phrases. Under his analysis, we expect that wh-phrases left in situ are interpreted in the same way as those moved to the Spec of CP already at S-structure. However, there is some difference in the interpretation of these two types of wh-phrases. Examining relevant sets of data from English and Japanese, I argue for the position that Reinhart [16], [17] take; in situ wh-phrases do not move to the Spec of CP at LF. Throughout the paper, I assume a modified version of Karttunen [10]'s semantics of Question. A question meaning is associated with a set of possible answers as proposed in Hamblin [6], but not with a set of true answers as seen in Karttunen. The meaning of (la) is represented as in (lb). It is roughly equated with a set of propositions of the form 'x came' where x is a student. When John, Bill and Mary are the students in the world, (1c) is the relevant set for the question in (la). A possible answer is a subset of the set of propositions that the question denotes when each member of the subset is true in the world. (1)� a.� Which student came? b. Ap3x [student (x) A p = Acame (x)] c. {John came, Bill came, Mary came}� where John, Bill, Mary are the students 2. RECONSTRUCTION AND OVERT/COVERT DISTINCTION In this section, we will see that wh-phrases left in situ are interpreted differently from overtly moved wh-phrases. Rizzi [18] shows that a wh-phrase that has moved across a weak island cannot be reconstructed to a position inside the island. If there is covert movement of wh-in-situ, we expect that the anti-reconstruction effects are also observed in wh-in-situ constructions. However, wh-in-situ in *This paper owes much to the lectures of Veneeta Dayal and Satoshi Tomioka at the LSA Summer Institute in 1997. I would like to express my thanks to them here. I am also grateful to John Whitman, Brent de Chene, David Pesetsky and Tomoyuki Yoshida for their valuable comments and suggestions. the weak island is interpreted in the position where it …

Content truncated to 500 words