When does ellipsis occur, and what is elided?

Abstract

This dissertation is concerned with how elliptical sentences are generated. To be specific, I investigate when and in what module ellipsis occurs, and what is elided as a result of ellipsis. With regard to the first research question, I propose that XP ellipsis occurs as soon as all the featural requirements of the licensor of XP ellipsis are satisfied during the derivation, rather than in the other modules. An important consequence of this proposal is that the point of XP ellipsis can vary depending on the derivational point where all the featural requirements of the licensor are satisfied in narrow syntax. Concerning the second research question, I suggest that ellipsis is a syntactic operation that eliminates phonological feature matrices of lexical items inside the ellipsis site, preserving the formal feature matrices. Segmental content (i.e. phonological features) is inserted into the phonological feature matrices when lexical items are sent to PF after Spell-out. This insertion does not apply to lexical items whose phonological feature matrices are eliminated, since there is no appropriate venue which segmental content is inserted into. Thus, they are not pronounced. This implies that even though narrow syntax cannot look into the information of the segmental content inside the phonological feature matrices, it can make reference to the phonological feature matrices in lexical items. This proposal is supported by the fact that elements whose phonological feature matrices have been eliminated can take part in further formal operations that occur after ellipsis, since they still contain formal features. However, unlike the other lexical items, elided interrogative wh-phrases do not seem to participate in formal operation occurring after ellipsis. In order to resolve this puzzle, I suggest a prosodic requirement questions must obey, adopting and modifying Richards’ (2016) Contiguity Theory.

Standard English copular phrase ellipsis is mainly used to develop the present theory of ellipsis. Cross-linguistic evidence from Indian Vernacular English, Belfast English, Korean, Farsi, British English, and Dutch data is also provided to argue that the present theory of ellipsis is not restricted to English.