Description:

Property noise and ambiguity resolution: The case of stubborn distributivity

Plural predications (e.g., the boxes are heavy) are common sources of ambiguity in everyday language, allowing both distributive and collective interpretations (e.g., the boxes each are heavy vs. the boxes together are heavy). In this talk, I focus on the role of context in ambiguity resolution and its consequences for our semantic theories. I address the key phenomenon of “stubborn distributivity,” whereby certain predicates (e.g., big, tall) are claimed to lack collective interpretations altogether. I first validate a new methodology for measuring the interpretations of plural predications. Using this method, I then analyze naturally-occurring plural predications from corpora, searching for evidence of a distinct class of predicates that resist collective interpretations. Finally, I systematically manipulate the predication context, showing that both the predictability of properties and the knowledgeability of the speaker affect disambiguation. These results suggest a pragmatic account of how ambiguous plural predications are interpreted. In particular, stubbornly distributive predicates are so because the collective properties they name are unpredictable, or unstable, in most contexts; this unpredictability results in a noisy collective interpretation, something speakers and listeners recognize as ineffective for communicating efficiently about their world. To implement this model of utterance disambiguation, I formalize the semantics and pragmatics of plural predication within the Bayesian Rational Speech Act modeling framework. I conclude with a discussion of connections to a broader set of linguistic phenomena, including cross-linguistic adjective ordering preferences.