Whamit!

The Weekly Newsletter of MIT Linguistics

Experimentalist Meeting 12/4 - Sherry Yong Chen (MIT), Christine Soh (UPenn), Athulya Aravind (MIT)

Speaker: Sherry Yong Chen (MIT), Christine Soh (UPenn), Athulya Aravind (MIT)
Title: Intermediate Wh-Copies in a Non-Wh-Copying Language
Time: Friday, December 4th, 2pm – 3:30pm

Abstract: A core issue within current theories of movement is to explain the conditions that drive and constrain the phonological realization of copies left by movement. Although the moved item syntactically and semantically occupies two positions, in languages including English only the highest copy is pronounced, while all lower copies are deleted. At the same time, multiple copy spell-out is attested in many languages, and is also found in the grammars of children acquiring English, a language where we otherwise do not find such phenomena. English-acquiring preschoolers produce long-distance wh-questions with an extra medial wh-word, as in (1) (Thornton 1990, Lutken et al. 2020), a result taken by some to be indicative of the realization of intermediate copies in children’s grammar:

(1) a. Who do you think who is in the box?
b. What do you think what she brought?

In this work-in-progress, we ask whether English-speaking adults respond differently to violations of copy spell out rules from other kinds of ungrammaticality, to see if there is continuity between adult and child grammars in their treatment of intermediate copies. We will present two experiments where we compare adults’ behavioral responses to sentences like (2a) and (2b). Neither is a well-formed sentence of English and both have a corpus frequency of 0. (2a), however, can be construed as a syntactically well-formed structure that violates constraints on copy spell-out, in contrast to (2b) whose infinitival complement does not provide an intermediate landing site for movement.

(2) a. Who did the consultant expect who the new proposal had pleased?
b. Who did the consultant expect who the new proposal to have pleased?

Pilot results are less-than-promising. We’d like to reflect on some of our operationalization assumptions in light of these results, and would welcome comments and suggestions on how to improve our approach.