English VP ellipsis: ellipsis site, licensing condition and identity condition. A literature review

Abstract

Verb Phrase ellipsis is the syntactic occurrence in which the elements within the verb phrase–the verb and its complements–are elided (van Craenenbroeck, 2014). Although this linguistic phenomenon can be encountered in various natural languages (Hedberg, 2015), the challenging questions it poses have located English Verb Phrase ellipsis in the spot of several researchers in the field of theoretical linguistics (Merchant, 2012). From the beginning of Generative Grammar in the late 1950s, ellipsis has been widely researched, with specific interest in Verb Phrase ellipsis and Sluicing (Merchant, 2012). Most work on Verb Phrase ellipsis has endeavoured to provide answers to questions related to the elided information, the context, and the relationship with the antecedent (Merchant, 2012). Nonetheless, the majority of studies have presented proposals that differ from one another, yielding the aforementioned questions open for debate. Consequently, it is the aim of the present dissertation to analyse a wide range of studies regarding this phenomenon in an attempt to review the literature on English Verb Phrase ellipsis. To be more specific, this paper explores the three principal queries on the topic: (i) the ellipsis site, (ii) the licensing condition, and (iii) the identity condition. Firstly, the nature and the structure of the ellipsis site will be examined following the VP proform approach (Rooth, 1981; Klein, 1987; Hardt, 1993; Hardt, 1999) and the PF- deletion approach (Sato, 2013; van Craenenbroeck, 2014; Aelbrecht & Harwood, 2015). Secondly, the licensing condition will be described in terms of an overt realisation of the inflectional node (Sag, 1976; van Craenenbroeck, 2014) or the presence of the [E] feature in the licensing head (Merchant, 2005; Merchant, 2012; Hedberg, 2015). Thirdly, the identity relationship between the antecedent and the ellipsis site will be explained via a syntactic approach (Ross, 1969; Sag, 1976; Hankamer & Sag, 1976; Hankamer, 1979; Lasnik, 1995; Sato, 2013; Murphy, 2018), a semantic approach (Hardt, 1993; Fiengo & May, 1994; Hardt, 1999; Hedberg, 2015) and a hybrid approach (Chung, 2006; van Craenenbroeck, 2008; Merchant, 2013). In short, the existence of diverse approaches supported by empirical data hinders the prevalence of one approach over the rest.