Volume 37, Issue 2 -

Edward Keenan coined the term “existential- have ” for have -sentences containing a relational noun in object position that present a definiteness effect (DE) similar to the one in there be -sentences. We begin this paper by showing in detail that the DE in these sentences is in fact different from the one found with there be -sentences. We then explain how these contrasts reflect differences in the semantics of the two sorts of sentences that we have independently argued for in previous work. We will specifically challenge two assumptions that are frequently made about the definiteness effect in have -sentences: (1) that it is related to any version of the so-called “weak”/“strong” distinction that has been used to characterize the effect in there be -sentences; and (2) that it is limited to relational nouns like handle and follows from treating such nouns as two-place predicates. Finally, we show how our account is superior to other accounts that have been offered of the definiteness effect in have -sentences.

Sentences with the verb exist and with a lexical DP in subject position show no definiteness effect. This suggests that the definiteness effect is keyed in English to the presence of expletive there . More strongly put, a definiteness effect is invariably found whenever expletive there (or a counterpart of it in other languages, whether pronounced or not) is present. This effect may in some languages be limited to the case of unstressed pronouns. Expletive there and its counterparts originate within the associate (in a way that accounts for the expletive being there , and not then ). In part, this conclusion is driven by anti-homophony. Freeze noted that English is exceptional in having expletive there in subject position. The proposal will be that this is related to English allowing preposition-stranding under A-movement. The definiteness effect itself results from a blocking effect, probably not specific to existential sentences, that certain determiners such as the impose on the movement of expletive there from its DP-internal position up to a sentential subject position.

Despite previous studies (cf. among others Huang 1987. Existential sentences in Chinese and (in)definiteness. In Eric J. Reuland & Alice G.B. Ter Meulen (eds.), The representation of (In)definiteness , 226–253. Cambridge, Mass.: MIT Press; Li, Yen-Hui Audrey. 1990. Order and constituency in Mandarin Chinese . Dordrecht: Kluwer; Li, Yen-Hui Audrey. 1998. Two types of existential sentences. Illinois Papers in Linguistics 26. 175–191; Pan, Haihua. 1996. Imperfective aspect zhe , agent deletion, and locative inversion in Mandarin Chinese. Natural Language and Linguistic Theory 14(2). 409–432), the defining characteristics of existential sentences in Chinese (including potential equivalents of locative inversion in English) have remained controversial. This is shown to be due to the failure to acknowledge the existence of two different constructions, the existential construction (ExC) ‘Ø V DP’ where a sentence-initial phrase indicating location (PlaceP) is not required, on the one hand, and the locative construction (LoC) with an obligatory PlaceP, on the other: ‘PlaceP V DP’. Only the ExC can serve as a diagnostic context for unaccusative verbs, whereas the LoC allows for a wide range of verbs, including a subset of unergative verbs. Furthermore, two types of LoC need to be distinguished, depending on the type of aspect (perfective aspect - le vs imperfective aspect -zhe ), giving rise to different semantics. Both have, however, in common that the PlaceP occupies the subject position (SpecTP), not the topic position, and that it is merged in SpecTP, not moved there, as evidenced by the systematic lack of a corresponding source structure with the PlaceP in postverbal position.

This paper proposes an interface account of existential sentences, in which the examination of the semantic, morphosyntactic, discourse and prosodic properties of these and related constructions is aimed to explore the similarities and differences with other types of IS-marked copular structures. In particular, a structural parallelism is proposed between existentials and clefts, as well as between (inverted) locatives and (inverted) pseudoclefts. In the analysis of existential constructions, the investigation of the Definiteness Effect reveals the need for a distinction across there -sentences; in particular, the interface properties of the relevant structures suggest that different analyses should be provided for existential and presentational there -sentences, which present crucial formal asymmetries at various levels of analysis.

In this work we deal with two structures that have a very similar pragmatic function in Italian and have been claimed to have similar semantic and syntactic properties, namely clefts and left peripheral focus. Since Chomsky (1977. On wh-movement. In Peter W. Culicover, Thomas Wasow & Adrian Akmajian (eds.), Formal Syntax , 71–132. New York: Academic Press.) they have been both considered as instances of A’-movement and should therefore behave alike. Here we investigate their prosody and their syntax on the basis of three experimental studies and show that while the prosodic patterns found are indeed very similar, their syntax is less homogenous than expected if we apply general tests that have been traditionally used to distinguish A- from A’-movement. In particular, we will discuss three of these tests, namely parasitic gaps, weak crossover and anaphoric binding and show that the two constructions yield quite different results. We analyse the differences within the framework of featural relativized minimality originally proposed in Rizzi (2004. Locality and the left periphery. In Adriana Belletti (ed.), Structures and Beyond: The Cartography of Syntactic Structures 3 , 223–251. Oxford: Oxford University Press.) and subsequent work. On this basis, we conclude that there is no one to one match between prosodic and syntactic properties, since we observe differences in the syntactic behaviour of the two constructions that do not surface in the prosodic patterns. Indirectly, this study sheds new light on the interface between prosody and syntax and is a confirmation of a modular theory of the components of grammar: some specific syntactic properties have no reflex in other components of grammar and can only be detected through purely syntactic tests.