VP-Ellipsis Is Not Licensed by VP-Topicalization

Abstract

Based on the observation that the constraints on VP ellipsis (VPE) closely match those on VP topicalization, Johnson (2001) proposes a movement account for VPE: in order for a VP to be deleted, it first has to undergo topicalization. Our paper shows that, although attractive, making VPE dependent on VP topicalization is problematic because VPE and VP topicalization are not distributionally equivalent. While VP topicalization targets the left periphery and consequently is subjected to constraints on movement, VPE is not so restricted.The paper outlines some alternatives for capturing the observed parallelism in the licensing of VP topicalization and VPE.

French adverbial clauses, rescue by ellipsis and the truncation versus intervention debate

“…Johnson (2001) and Aelbrecht and Haegeman (2012) raise the same issue with respect to English VP Topicalization vs. VP ellipsis. We will only discuss some examples of this distributional asymmetry and refer to reader to the papers cited for a full discussion.…”

“…Having rejected the VPT derivation of VPE, Aelbrecht and Haegeman (2012) consider a number of alternative derivations for VPE that ensure that the domains where VPT is excluded are compatible with VPE. The scenarios they envisage have in common the assumption that the crucial locus for the derivation of VPE is not the left periphery but, rather, the middle field.…”

Stative versus Eventive Predicates and vP-internal Structure

“…I'm liking this more and more), in which case it tolerates do so ellipsis. 10 See Aelbrecht and Haegeman (2012) for a counter-proposal to Johnson (2001 Our friendly amendment to the "topicalization plus deletion" analysis is that what is traditionally described as "VP" ellipsis can in fact target at least two distinct XPs: the lower part of the verbal projection (that is, VP), or the upper part of the verbal projection (that is, vP).…”

Semantic Identity and Deletion

“…As the same reviewer suggests, however, it might be possible to recast at least part of the voice mismatch phenomena in terms of minimality/ locality. First, Johnson (2001) argues that the derivation of VP-Deletion involves VPTopicalization (see also Aelbrecht and Haegeman (2012)). Second, it has been proposed in the literature that discourse factors such as focus and (contrastive) topic, to which deletion is sensitive, be encoded in terms of features or syntactic positions (see, for example, Rizzi (1997Rizzi ( , 2004).…”