B4: "Processing at the Syntax-Discourse Interface" (Lyn Frazier)

Mon 14.15-15.45, Tue 16.15-17.45, Thu 9.15-10.45, Fri 11.15-12.45
Room: 0.01


This course will focus on two questions. The first concerns how Not At Issue (NAI) content is processed, in particular content in appositive relative clauses, parentheticals and expressives. My colleagues and I have argued that NAI content expresses a distinct speech act and that speech acts generally organize working memory during sentence processing (Dillon, Clifton and Frazier 2014, Dillon et al., submitted). Evidence supporting this claim will be presented. We will also explore its consequences for the issue of how syntax constrains interpretation: where is compositional interpretation tightly constrained by the syntax, and where do content-driven rather than form-driven inferences largely determine interpretation, e.g., the interpretation of expressives, and of the relation between NAI and At Issue content

The second main question to be addressed is how syntactic form influences the identification of focus. We will go over psycholinguistic evidence from many experimental techniques (phoneme monitoring, acceptability judgments, eye movements during reading) and from different empirical domains (e.g., ellipsis, corrections) suggesting that pitch accent alone does not determine focus in language processing, even in a language like English. The consequences of focus identification for related notions like main assertion/At Issue content and for identification of a potential implicit Question Under Discussion (QUD) will also be addressed. The discussion of focus will end with a look at how the processor's assumptions about focus and contrast interact to organize discourse (Repp, Frazier and Hemforth 2015).

The final part of the course will pick up, and perhaps pull together, strands of the course concerned with accommodation of unstated information. Discussion will include the role of focus in accommodation (Moulton et al., submitted), accommodation of the speakers epistemic state (Clifton and Frazier, submitted), but also accommodation more generally (Singh et al. 2015). The course will end with a discussion of implications of the work for theories of processing and theories of grammar, and a discussion of unsolved puzzles.


Lecturer: Lyn Frazier, UMass Amherst