`Else-modification' as a diagnostic for pseudosluicing
Matthew Barros
October 2012

A central issue examined in the literature on sluicing is what the elided structure may look like. The usual approach assumes that the elided TP in (1) is as in (2), where the interrogative CP consists of a Wh-question version of the antecedent: (1) Jack ate something, but I don't know what. (2) Jack ate something, but I don’t know what-i [TP Jack ate t-i ]. Such a view is consistent with the notion that ellipsis is subject to some form of syntactic identity condition (perhaps alongside a semantic one; cf. e.g. Chung 2013 for discussion). Nonetheless, alongside the possibility in (2), a growing body of work shows that the elided TP may be a cleft, departing more radically from the structure of the antecedent; a possibility dubbed pseudosluicing in Merchant (1998). (3) Jack ate something, but I don’t know what-i [TP it was t-i ]. Merchant (2001) provides 10 empirical arguments aimed at showing that all sluicing cannot be reduced to pseudosluicing. Merchant’s diagnostics are designed to rule out a pseudosluice source in a variety of contexts. In this paper, I present a new diagnostic which rules out the non-pseudosluice source, providing further support for the existence of pseudosluicing.
Format: [ pdf ]
Reference: lingbuzz/001761
(please use that when you cite this article)
Published in: MS Rutgers University, presented at NELS 43
keywords: sluicing, ellipsis, pseudosluicing, cleft, identity, semantics, syntax
Downloaded:745 times


[ edit this article | back to article list ]