Syntax of Why-in-situ: Merge Into [SPEC,CP] in the Overt Syntax

Abstract

This paper proposes that ‘why’ in wh-in-situ languages (Korean, Japanese, and Chinese) is directly merged into [Spec,CP] of the clause it modifies. This proposal not only captures long-standing issues regarding the peculiarity of ‘why’, as opposed to other wh-phrases, but also accounts for previously unnoticed asymmetries among why-constructions. In particular, I argue that due to its initial merge position, ‘why’ in an interrogative clause is licensed with external merge while ‘why’ in a declarative clause must undergo LF-movement. This argument is supported by the non-uniform behavior of ‘why’ with respect to the Intervention Effect in Korean and Japanese (cf. Beck and Kim 1997) and is further confirmed by the question-marker drop phenomenon in Japanese. Under this proposal, a puzzling divergence between Chinese and Korean/Japanese in why-constructions is reduced to the fact that Chinese disallows A′-scrambling. The proposal also captures a syntactic parallelism between ‘why’ in wh-in-situ languages and ‘why’ in wh-fronting languages, like Italian and Irish. Among the theoretical consequences of this paper is a demonstration that a subject may scramble (cf. Saito 1985) and that string-vacuous scrambling is responsible for judgment variations concerning the Intervention Effect.