rethinking ellipsis (thesis)

supported by a training grant from the Center for Research in Language, UC San Diego Interdisciplinary Training in Language, Communication, and the Brain (NIH Grant No. 5-T32-DC0041)

Standard syntactic licensing models of verb phrase ellipsis predict grammaticality just in case a target verb phrase is structurally parallel to its antecedent. In a variety of naturally occurring contexts, however, ‘mismatched’ ellipses sound acceptable to many speakers. In my work I have identified a confound in the previous literature where syntactic parallelism and discourse coherence have both been confounded with information structure.

In a series of offline experiments I dissociate these variables and show that information structure, topic structure in particular, plays a crucial role in determining ellipsis acceptability. Follow-up studies show that the information structural effects are in fact independent effects, observed even in non-ellipsis contexts. The model I propose makes the further prediction that defective topic structures should lead to processing difficulties online before the reader has encountered an ellipsis. That prediction has been confirmed in a self-paced reading time study.

Together, these findings pose a challenge to models which attribute parallelism effects to an ellipsis-specific processing mechanism, and instead support a model where processing difficulties associated with information structure are amplified in ellipsis contexts.

The Moro Language Project

with PIs Sharon Rose and Farrell Ackerman, supported by NSF Grant No. BCS-0745973

Moro is a Kordofanian language spoken in the Nuba Mountain region of Sudan. Little linguistic work on the Kordofanian family has been undertaken, and it is one of the most poorly described language groups in Africa. Moro has recently been designated an endangered language, and The Moro Language Project seeks to document it by producing a grammar and lexicon. In addition to this descriptive project, our group has undertaken a theoretical investigation of various phonological and morpho-syntactic phenomena. My research on Moro involves the morphosyntax of filler gap constructions and the syntax and semantics of the possessive system.

A Discourse Coherence APproach to Pronoun Interpretation

with PIs Andy Kehler and Jeff Elman, supported by grant from the UCSD academic senate

For more than three decades, research into the psycholinguistics of pronoun interpretation has argued that hearers use various interpretation ‘preferences’ or ‘strategies’ that are associated with specific linguistic properties of antecedent expressions. This focus is a departure from the type of approach outlined in Hobbs (1979), who argues that the mechanisms supporting pronoun interpretation are driven predominantly by semantics, world knowledge, and inference, with particular attention to how these are used to establish the coherence of a discourse. In a series of behavioral experiments, we have developed a coherence-driven model which can outperform preference based approaches and offer an explanation for the often contradictory predictions posed by preference-based models. My research in this area has focused on effects of parallelism in the interpretation of pronouns.