Speakers: Peter Graff & Jeremy Hartman (MIT)
Title:Constraints on Predication
Time: Thursday, November 18, 12:30-1:45pm
Location: 32-D461
In this talk we propose a novel semantic universal “Myopia”, which we argue constrains the denotations of lexical predicates in natural language. We generalize that for any lexical predicate P, the only entities relevant to determining the truth of a formula involving P are the entities identified by P’s arguments. Roughly, a predicate P is “myopic” if the truth of a formula involving P in a universe consisting exclusively of the individuals identified by the arguments of P is always identical to its truth value in the full domain of individuals. We show that this constraint not only excludes many denotations of natural language determiners previously excluded by Conservativity (Keenan and Stavi 1986), but also generalizes to predicates of other lexical categories, including nouns, verbs, adpositions and adjectives. We support the psychological reality of our typological observation with evidence from an on-line artificial determiner learning study (N=454), showing that both Conservativity and Myopia are active in determiner learning. Conservative determiners are easier to acquire than non-conservative myopic determiners, which in turn are easier to acquire than non-conservative non-myopic determiners, showing that typological universals may have a more or less profound impact on word learning.