25 September: Mora Maldonado

Tuesday 25 September, 11:30-12:30

Sentences involving plural expressions can give rise to distributive and non-distributive interpretations (in (a) and (b) respectively):

(1) The girls ate two cookies.
(a) The girls ate two cookies each.
(b) The girls ate two cookies in total.

(2) The bags are light.
(a) The bags together are light.
(b) Each bag is individually light without the bags together being light.

In formal semantics, the derivation of these readings has been traditionally accounted for by assuming that non-distributive readings are obtained by default (as soon as the plural subject is in the predicate denotation), whereas distributive readings arise through the insertion of a covert distributivity operator D, whose meaning roughly corresponds to that of the word each in English (Link, 1987; Champollion, to appear). The D operator applies the VP to each atomic member of the plural subject.

While the availability of distributive and non-distributive readings for sentences in (1) and (2) is undisputed, one could ask whether the abstract mechanisms proposed to derive the contrast are accessed as such during sentence comprehension. I will report the results of three studies showing that non-distributive and distributive interpretations can be independently primed for both sentences in (1) and (2). Priming serves here to complement introspection on truth-value and inferential judgments, revealing the abstract mechanisms underlying semantic interpretation. Our findings suggest that the compositional operations proposed to derive the non-distributive/distributive contrast are at play during comprehension. The existence of distributive priming in absence of object covariation (e.g. 2) reveals an abstract representation of the non-distributive/distributive distinction, which is orthogonal to specific verification strategies.