Syntax and Prosody of either...or... Sentences

Speaker
Danfeng Wu
Affiliation
MIT
Date
Fri February 26th 2021, 1:30 - 3:00pm
Location
Online (Zoom)

Please jiayi.lu [at] stanford.edu (email ) for Zoom link.

Abstract:

Either in either…or… sentences has a relatively free distribution, as was observed by Larson (1985), Schwarz (1999), Han and Romero (2004), den Dikken (2006), and others. For example, if the apparent disjunction is rice or beans in (1)-(2), either can appear as the sister of this apparent disjunction (1), or higher than the sister of the disjunction (2a-c).

(1) John will eat either rice or beans.

(2) a. John will either eat rice or beans.

      b. John either will eat rice or beans.

      c. Either John will eat rice or beans.

I propose that either originates inside the disjunction phrase, and later moves to be the sister of the disjunction phrase. In the meantime, ellipsis may delete material in the second disjunct, and make the disjunction seem smaller than it actually is. (2a-b) is derived by this ellipsis:

(3) a. John will either eat rice or eat beans.

      b. John either will eat rice or will eat beans.

      c. Either John will eat rice or he will eat beans.

In this talk, I will focus on the ellipsis part of the analysis, and present arguments for ellipsis, one of which comes from a prosodic / phonetic experiment. This study is thus part of a larger program of creating a new domain of argumentation in syntax: I believe that we can draw evidence for syntactic claims not only from traditional sources, but also from prosodic experiments.

The prosodic study of either…or… sentences not only informs us about the syntactic analysis of these sentences, but also sheds light on the question of mapping between syntax and prosody. Specifically, it informs an apparently counterintuitive question: is silent material (specifically elided material) represented in prosodic structure? Preliminary results suggest that elided material is represented in prosody, despite having no phonological content. This supports a derivational view of syntax-prosody mapping that requires particular ordering of operations, so that at the point of prosodification, elided material has not been fully deleted yet.