The Typology of Expletive THERE

In this article I propose an alternative account of expletive THERE based on the theory that formal features may move overtly at PF. Specifically, I argue that the «inadequacies» of expletive THERE alluded to in Chomsky (1993: 33) are the result of the fact that THERE is simply a collection of formal features, an «alternative» spell-out of the Case and Agreement features of a lexical NP and has no independent category (NP) features. In other words, expletive THERE is an instance of Overt Move F, following Blight (1998) and Roberts (1998), without «generalized pied-piping» in the sense of Chomsky (1995). I depart from the proposals of Blight (1998) and Roberts (1998) however in maintaining that Overt Move F is motivated by the need to satisfy economy principles, specifically, the Principle of Economy of Projection (EOP) proposed in Speas (1995). The rest of this article is organized as follows. In Section 2, I present cross-linguistic evidence which shows that expletive THERE occurs only in languages which mark for both morphological (subject-verb) Agreement and morphological or structural Case. In section 3, I consider two recent analyses of THERE, Chomsky’s (1993, 1995) LF-Affix analysis and Lasnik’s (1992, 1995) Partitive Case analysis, and show that both not only encounter empirical and conceptual problems, but also fail to account for the crosslinguistic distribution of expletive THERE. Section 4 introduces the Overt Move F analysis of expletive THERE in relation to English and shows how such movement is driven by the need to satisfy the EOP. In addition, I show how the cross-linguistic distribution of expletive THERE is readily explained by the fact that THERE is a collection of formal (Agreement and Case) features. The proposed analysis has three immediate advantages over previous analyses. First, it provides an adequate account of expletive THERE in English without positing a new class of morpho-syntactic objects («phrasal» affixes) as proposed in Chomsky (1993). Second, it readily accounts for the Case and Agreement properties of expletive-associate pairs while also accounting for the chain-like properties of expletive-associate pairs, a robust observation which is lost under Lasnik’s Partitive Case analysis. Third, it provides a principled explanation for the fact that languages without both morphological Agreement and morphological or structural Case do not have expletive THERE. Since THERE is an «alternative» spell-out of both the Agreement and Case features of an associate NP, the Overt Move F analysis readily