The Weekly Newsletter of MIT Linguistics

Experimentalist meeting - 9/20 Martin Hackl (MIT)

Speaker: Martin Hackl

Title: A Null-Theory of Haddock’s Puzzle and its Implications for the Role of Presupposition in Reference Resolution

Time: 2:00-3:00pm

Location: 32-D461


Haddock’s Puzzle presents a well-known challenge to the canonical treatment of definite descriptions according to which their use requires contextual uniqueness of the NP-restrictor of the. More specifically, as pointed out in Haddock 1987, in a situation with two hats and two rabbits, one rabbit in one of the hats while the other rabbit is not inside a hat, the definite description in (1), in which a definite DP the hat is nested inside a larger definite description is felicitous (Haddock 1987). Importantly, the utterance in (2) used as a description of the very same scene is not.

(1) The rabbit in the hat is excited. 
(2) #The excited rabbit is in the hat.

In this talk, I present a “null-theory” of Haddock’s puzzle according to which Haddock-definites are situational uniqueness definites (Schwarz 2009) in order to shed light on the role of presupposition in reference resolution. The central claim motivated by this null-theory is the constraint in (3).

(3) Constraint on Reference Resolution:     
Presupposed content of an utterance can be used for identifying the extension of referring expressions in the utterance, at-issue content cannot.

To assess and evaluate different ways of explicating (3) we are working on experimental paradigms that hopefully allow us to track often rather subtle distinctions in a systematic way. I will discuss where we are with that part of the project and look forward to comments on suggestions on how to improve our approach.