•  
  •  
 

University of Pennsylvania Working Papers in Linguistics

Abstract

This paper considers properties of the Japanese Accusative wh-adjunct 'nani-o (what-Acc)' (Kurafuji, 1996, 1997; Ochi, 1999) in sentences such as "Kare-wa nani-o sawai-dei-ru no? (lit. What is he making a noise?)". Although the Accusative wh-adjunct 'nani-o' is usually translated in the same way as 'naze (why)', there are a number of differences between them: (i) 'Nani-o' has an animacy restriction on the subject, (ii) it has some special speaker’s inference, and (iii) it is incompatible with sluicing (Ochi, 1999). We will explain the properties (i) and (ii) by claiming that Accusative wh-adjuncts are base-generated in a functional projection FP, which is related to speaker's illocutionary force. We attribute the property (iii) to Fox and Lasnik's (2003) parallelism condition on sluicing; because Accusative wh-adjuncts are base-generated in a different position from other reason adjuncts, they do not satisfy parallelism with the corresponding adjunct in the antecedent clause. By clarifying the syntactic positions of the two types of reason adjuncts, we attempt to contribute to the typological study of adjuncts.

Share

COinS