Mismatch under Why -stripping in Japanese

This paper examines the identity condition on clausal ellipsis based on previously unnoticed data from Japanese why -stripping. We show that why -stripping in Japanese exhibits a mismatch similar to verb phrase ellipsis in English observed by Stockwell (2020. Contrast and verb phrase ellipsis: Triviality, symmetry, and competition . Los Angeles, CA: University of California PhD dissertation) and van Craenenbroeck and Johnson (2025. Sloppy symmetry under ellipsis. The Linguistic Review 42(3). 443–463). These authors point out that the mismatch under verb phrase ellipsis in English is allowed only for symmetric predicates; for example, John can meet Mary, but Mary can’t meet John . / John should meet Mary, but they can’t meet . This study shows that the mismatch under why -stripping in Japanese is also limited to cases involving symmetric predicates; for example, John-to Mary-ga kekkonsita ‘John and Mary got married’ – naze John-ga? ‘Why John (nominative)?’ We then demonstrate that why -stripping in Japanese is not merely the Japanese counterpart of verb phrase ellipsis in English but it involves clausal ellipsis. As for why -stripping in English, Weir (2014. Why -stripping targets voice phrase. In Proceedings of NELS , vol. 43, 235–248) claims that it is not a type of clausal ellipsis but rather an ellipsis of Voice Phrase and that it lacks the TP-layer. We contend that Weir’s TP-less analysis cannot be extended to why -stripping in Japanese. We also argue that why -stripping in Japanese is not always derivationally related to either a copular structure with a pronominal subject or a cleft, contrary to Nakao et al.’s (2013. On the structure of Japanese ‘why’ stripping. In Proceedings of GLOW in Asia IX , 199–212) claim. We finally argue that the mismatch under why -stripping in Japanese is best analyzed in terms of the v P-level argument-structure identity with the antecedent, which is in line with the proposal that clausal ellipsis, such as Sluicing, can be licensed based on the identity of some proper subdomain of the elided clause and its antecedent (Anand et al. 2025. The domain of formal matching in sluicing. Linguistic Inquiry 56(2). 353–373; Rudin. 2019. Head-based syntactic identity in sluicing. Linguistic Inquiry 50(2). 253–284; Stockwell. 2025. Be and the domain of matching in ellipsis . ms. Ulster University).