Adam Zachary Wyner (Jerusalem):
Adverbials and VP-Ellipsis

Donnerstag, 11.30 Uhr

In this paper, I discuss the behavior of manner and eventive adverbials in elided constructions, showing some problems for extant theories and an alternative which better explains the observations. With manner adverbials, we can interpret (1) as in (2), where the manner Adverbial appears in both clauses; or we can interpret (1) as in (3), where the manner Adverbial does not appear in the second clause; however, we can not interpret (1) as in (4), where the passionate manner is attributed to the "combination" of Kim's kissing and Robin's kissing.

(1) Kim kissed Sandy passionately, and Robin did too.
(2) Kim kissed Sandy passionately, and Robin kissed Sandy passionately.
(3) Kim kissed Sandy passionately, and Robin kissed Sandy.
(4) #That Kim kissed Sandy and Robin kissed Sandy was passionate.

In contrast, with eventive adverbials, we do not interpret (5) to mean that each of the kissings was stupid, as in (6), but Kim's kissing Sandy was stupid without making a committment to the stupidness of Robin's kiss as in (7) or that the combination of Kim's kissing and Robin's kissing was stupid as in (8).

(5) Stupidly, Kim kissed Sandy, and Robin did too.
(6) That Kim kissed Sandy was stupid, and that Robin kissed Sandy was stupid.
(7) That Kim kissed Sandy was stupid, and Robin kissed Sandy.
(8) That Kim kissed Sandy and that Robin kissed Sandy was stupid.

I argue that the syntactic and semantic distribution of manner and eventive adverbials is relatively free; furthermore, it is not the case that manner adverbs adjoin only to VP and eventive adverbs adjoin only to IP. Thus, the "identity-of-relations" analysis of VP-ellipsis resolution, where one "copies" the VP from the source clause to the target, can not in and of itself explain the observations above. In addition, I present some of the semantic properties of eventive adverbials; we see that, in contrast to manner adverbials, eventive adverbials are "inert" with respect to a variety of "sentence level" operations. I account for the ellipsis observations using Unification (Dalrymple, Shieber, and Pereira (1991)), for it provides an algorithm to determine the value of the ellided phrase based on semantic, not syntactic considerations. Where the algorithm operates on a sentential expression, I predict that it can't apply to eventive adverbials but can apply to manner adverbials. Thus, I account for the distinct behavior of manner and eventive adverbials based on the semantic properties of adverbials in a semantic account of ellipsis resolution.

zum Programm der AG 8
zur alphabetischen Übersicht der Abstracts
zur zeitlichen Übersicht