Hendrick (2011)

• Hendrick, R. 2011. 'Some Breton Indefinites', Andrew Carnie (éd.), Formal Approaches to Celtic Linguistics, Cambridge Scholars Publishing, 95-114.

Abstract:

 The study of indefinite expressions has been a major driving force in semantics (cf. Kamp 1981 and Heim 1982) as well as the syntax-semantics interface (cf. Diesing 1992). It is commonly thought that the English indefinite a(n) does not itself introduce a quantifier semantically, unlike every and that phrases like a linguist are semantically variables restricted by the description 'linguist'. Further, it is believed that the English indefinite is semantically provided a default quantifier in a privileged structural domain, e.g. VP, recalling the suggestion by Chomsky (2005) to chunk clauses into VP and TP as special phases. Whether this similarity is coincidental or has some principled basis is one of the basic questions that this article tries to answer.
 A second concern of this article involves using Breton to document cross-linguistic variation in indefinites. There is valuable evidence that languages vary in how quantifiers are restricted semantically (cf. Bach et al. 1995) and in how indefinites are provided with default quantifiers (cf. Chung & Ladusaw 2003). The literature also contains some intriguing conjectures that at least some of the variation closely tracks syntactic variation (e.g., Jelinek 2000). Many analyses of verb initial languages posit that the arguments of V remain within the VP. Suppose this is true and that the structure of a Breton sentence like Ne gar ket Yann Mona involves V-raising. This allows us to pose the second central question that this paper addresses. Do indefinites and expressions of quantity behave differently in a natural language like Breton where an argument of V is not forced to move out of VP to occupy specifier TP? Preliminary evidence suggests that this should be a fruitful domain of inquiry. Negative Polarity Items (NPIs) have been treated as a special subclass of indefinite expressions having an additional structural licensing requirement (cf. Ladusaw 1996). Specifically, NPIs must be c-commanded by a negative expression. Some Breton NPIs such as that in Den ebet ne gar Mona are superficially at odds with the classic view of NPIs (as noted in Hendrick 2000) since they violate the standard c-command licensing requirement. Should we resolve this conflict by modifying the classic view or by appeal to its interaction with some other property of Breton?
 

Références

• Bach, Emmon, Angelika Kratzer, & Barbara Partee. 1995. Quantication in Natural Languages. Dordrecht, the Netherlands: Kluwer.
• Chomsky, Noam, 2005. On phases.
• Chung, Sandra & William Ladusaw 2003. Restriction and Saturation, Linguistic Inquiry Monograph 42. Cambridge, Mass.: MIT Press.
• Diesing, Molly. 1992. Indefinites. Cambridge, Mass.: MIT Press.
• Heim, Irene. 1982. The Semantics of Definite and Indefinite Noun Phrases in English, Dissertation. University of Massachusetts, Amherst.
• Jelinek, Eloise. 2000. 'Predicate raising in Lummi, Straits Salish', The Syntax of the Verb Initial Languages, Andrew Carnie & Eithne Guilfoyle (éds.), 213-233. Oxford, UK: Oxford University Press.
• Kamp, H. 1981. 'A theory of truth and semantic representation', J. Groenendijk, T. Jannsen, et M. Stokhof, éds., Formal Methods in the Study of Language. Mathematical Center, Amsterdam. 277-321.
• Ladusaw, William. 1996. 'Negation. a notion in focus', , Heinrich Wansing (éd.), Perspectives in Analytic Philosophy, Berlin: Walter de Gruyter.