Skip to main content

Rebekah Baglini

UC San Diego

Monday, December 1st at 2 PM in AP&M 4301. Rebekah Baglini (UC San Diego) will be presenting.

Cross-linguistic variation and the semantics of states

Linguists and philosophers often mention states in characterizing the referential properties of certain lexical items. But different languages use different syntactic categories to encode these meanings, leading to systematic variation in the shape of stative constructions. The following data from English exemplifies the three primary strategies for expressing stative meanings attested cross-linguistically: non-dynamic verbs (1), adjectival predicates (2), and certain abstract mass nouns or roots (3) (Baker 2003).

(1) VERBAL: Sam hungers. (2) ADJECTIVAL: Sam is hungry. (3) NOMINAL: Sam has hunger.

Statives seem to be bound by some common properties across languages; for example, in every language, there is a core set of statives called property concepts which are the prototypical means for expressing gradable meanings (e.g. Sam is very hungry; Sam is hungri-er; Sam is the hungriest) (Dixon 1982). Yet there is at present no established basis for unifying stative meanings model theoretically. This prompts the following question: can stative meanings can be formally captured as a natural class across syntactic categories?

I approach this question by examining cross-linguistic variation in the morphosyntax of stative constructions. This variation, I argue, provides important clues to identifying the structures which underlie stative meanings universally. I draw heavily on my ongoing fieldwork on the Senegambian language Wolof, a language which exemplifies two different strategies for constructing statives expressing gradable property concepts (concepts like tall,expensive, and happy which are prototypically associated with adjectives): some Wolof property concepts are lexicalized as stative verbal predicates, while others are lexicalized as mass nouns. Wolof's dual system allows for direct comparison of expressions containing these stative lexemes to evaluate the semantic consequences of their distinct predication patterns (direct predication as in (1) and possessive predication as in (3)). The data I discuss in this talk points us towards a unified treatment of statives as a natural class of meanings, with variation in lexical denotations falling into a predictable range based on morphosyntactic properties of particular languages.