Conférence de Marijke De belder, "On the syntax of countability"

[faustista]

Marijke De Belder donnera la conférence ci-dessous le mercredi 11 mars à 14h, à l'UFR de linguistique de paris 7, au 30 rue du Château des Rentiers (premier étage, salle 127) (Mo. Olympiades).

ON THE SYNTAX OF COUNTABILITY / Marijke De Belder

Count readings can be subdivided into kinds and units. Consider the following examples. (1) shows a kind reading, (2) shows a unit reading.

(1) three dogs, viz. the poodle, the jack russel and the parson terrier.
(2) three dogs, viz. Fido, Laika and Lassie.

I characterize kinds as concepts that do not occupy time or space and can be ordered taxonomically. Units, on the other hand, do have a stretch in time or space and are traditionally ordered in a lattice. In this talk I discuss the syntax of this second count reading, the count unit reading. In order to derive units, I assume a more fine-grained structure of the countability domain in the noun phrase.	
Firstly, I propose an additional functional projection –SizeP – that is involved in the syntactic derivation of units. I argue that the Dutch diminutive and the semi-lexical noun stuk are a realization of the head Size°. I further show that the Dutch pre-determiner universal quantifier heel  `whole' (e.g. heel de appel `the whole apple', lit. whole the apple) is only licit if Size° is present in the structure. 
Secondly, I extend this analysis to other Germanic languages. I show that Norwegian shows parallel behavior to Dutch despite the absence of a diminutive in its lexicon. For German I propose that the features for number (sg/pl) and size are collapsed on one single head. Finally, for English I argue that the feature Size is absent. 
Thirdly, I propose a new semantic analysis of units. I define the concept of a unit as a supremum on a closed scale that expresses an ordered set of sums of atoms. 

(A diagram was omitted here)

This supremum should be understood as a natural terminal point. For the verb the syntax and semantics of natural terminal points are well known (Dowty 1979, Vendler 1967 and subsequent literature). Moreover, research on absolute adjectives (i.e. adjectives that can be modified by completely, e.g. completely pure vs. *completely long) indicates that terminal points play a role outside of the verbal domain as well (Kennedy 2006). As such, the analysis of units as concepts with natural terminal points in the NP constitutes a new addition to this cross-categorial taxonomy.
>From a theoretical perspective, I reach the following conclusions. Firstly, units can be syntactically derived. They are a product of syntax, and do not stem from the lexicon (cf. Borer 2005). Secondly, languages select a subset of features from a universal set of features (Chomsky 1965) and evidence for a certain feature should be found for each language and each construction independently (cf. Iatridou 1990). Thirdly, cross-linguistic variation stems from differences in the functional lexicon (cf. Borer 1984).

(Pour plus d'info, écrire à Noam Faust, faustista at yahoo.com)

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/parislinguists/attachments/20090306/832dc9f7/attachment.htm>