Volume 28, Issue 1

This article aims at focusing on a set of phenomena related to the syntax and semantics of bare count nominals. We will show that bare count nominals unmarked for number (BNs) can occur both in object argument position and in predicate position, and can form denominal verbs. This is because, in syntax, BNs can only occur as internal arguments of relational categories (V, P) to which bare nouns can move and conflate at some point during the syntactic derivation (Hale & Keyser, Prolegomenon to a theory of argument structure, MIT Press, 2002). On the other hand, BNs – being property denoting expressions – can neither occur as internal arguments of unaccusative and complex transitive verbs nor as external arguments due to a structural constraint on subjects / specifiers (Kallulli, The comparative syntax of Albanian. On the contribution of syntactic types to propositional interpretation, University of Durham dissertation, 1999): they must be properly licensed by appropriate functional categories. The novelty of this article is to relate the occurrence of bare nominals with the argument structure position in which they may occur at a syntactic level of representation, and to explain the relationship between argument structure and the interpretation of bare nominals.

This paper investigates the syntactic structure of the English get -passive construction and explores the implications it holds for theories of Agentivity, the determination of the syntactic status of External arguments, and the nature of Control. Four previous analyses of get -passives are examined, these being the two variants of the verbal passive analysis put forth respectively in Quirk et al. (A grammar of contemporary English, Longman, 1972) and Haegeman (Lingua, 66: 53–77, 1985); the Control approach suggested in Lasnik and Fiengo (Linguistic Inquiry 5: 535–571, 1974); and the currently accepted adjectival passive analysis put forth in Fox and Grodzinsky (A-chains in children's passive and get as an unaccusative (raising) verb, 1992, Linguistic Inquiry 29: 311–332, 1998). By reintroducing and adding to the empirical base, it is shown that this construction is three-way structurally ambiguous, possessing a verbal passive, an adjectival passive, and a Control structure.

Under what conditions is ellipsis possible? Fiengo and May (Indices and identity, MIT Press, 1994), Fox (Economy and semantic interpretation, MIT Press, 2000), and Rooth (Ellipsis redundancy and reduction redundancy, 1992), among many others, conclude that syntax plays a role in ellipsis, at least in part. Merchant (The syntax of silence: sluicing, islands, and the theory of ellipsis, Oxford University Press, 2001: p19-25) challenges this view, and argues that ellipsis in general, or sluicing in particular, is conditioned purely by semantics, providing arguments against syntactic accounts. More recently, this view is challenged by Chung (Sluicing and the lexicon: the point of no return, Ms. University of California, 2006) and Merchant himself (Voice and ellipsis, Ms. University of Chicago, 2007, Linguistic Inquiry 39: 169–179, 2008b), who provide evidence for the relevance of syntax. This article shows that Merchant's (The syntax of silence: sluicing, islands, and the theory of ellipsis, Oxford University Press, 2001) arguments do not constitute a real challenge to a syntactic account. In so doing, it is also demonstrated that syntax does play a role in ellipsis. We then develop a syntactic isomorphism condition on ellipsis that accounts for observed facts.