Title: Differential Quantifier Scope: Q-Raising versus Q-Feature Checking *
Abstract: Divergent scope-taking and scope interaction possibilities of noun phrases have been the focus of interest ever since it became clear that the omnivorous scope-shifting rule of Quantifier Raising (QR) (May 1977, 1985) plainly both underand overgenerates. Liu (1990), Ben-Shalom (1993) and others point out that in interactions with other quantifier types certain quantifiers exhibit a smaller set of inverse scopal options than would be predicted if QR applied to them. In a series of influential studies seeking to account for the rather complex pattern of differential scope-taking options, Beghelli and Stowell (1994, 1995) and Szabolcsi (1997) propose to treat various quantifier classes as performing checking operations in quantifier-specialized functional projections in the clause. The proliferation of functional projections as descriptive devices has been a primary concern in the past decade or so, and an object of much conceptual controversy. Here I take the methodological stance that introducing functional projections as new primitives in the theory requires substantial empirical motivation. What I will demonstrate here is that in this regard Beghelli and Stowell’s/Szabolcsi’s quantifier-projection-based (or A-bar feature checking-based) approach to Q-scope is insufficiently grounded: their quantifier projections lack the necessary empirically motivation. In fact, some aspects of the model also create conceptual complications. Worse still, on closer inspection, the approach both underand overgenerates in the domain of Qinteraction. In this paper I will work with a restricted set of functional projections in the clausal domain assumed in Chomsky (1993), and demonstrate that an alternative, more conservative model incorporating QR is able to provide not only a more restricted, but also an empirically superior account of differential Q-scope. In particular, I show that independently motivated scope-affecting mechanisms interact in complex ways to yield precisely the attested scopal possibilities for the various quantifier classes. These mechanisms are existential closure, reconstruction within Achains, and QR. A repercussion of the present study is that Quantifier Raising exists at the level of narrow syntax—an assumption that has recently been repeatedly challenged, perhaps most strongly in the specialized quantifier-projections approach (cf. also Hornstein’s 1995 approach). I argue here that the QR-view is essentially correct, though the domain of its application is more restricted than commonly believed. If the analysis of Q-interaction presented here is correct, then A-reconstruction also must be available (alongside A-bar reconstruction), contra Chomsky (1995) and Lasnik (1999).
Publication Year: 2004
Publication Date: 2004-01-01
Language: en
Type: article
Access and Citation
Cited By Count: 3
AI Researcher Chatbot
Get quick answers to your questions about the article from our AI researcher chatbot