The Relationship between Eye Gaze and Verb Agreement in American Sign Language: an Eye-tracking Study

The representation of agreement is a crucial aspect of current syntactic theory, and therefore should apply in both signed and spoken languages. Neidle et al. (2000) claim that all verb types in American Sign Language (agreeing, spatial, and plain) can occur with abstract syntactic agreement for subject and object. On this view, abstract agreement can be marked with either manual agreement morphology (verb directed toward locations associated with the subject/object) or non-manual agreement (eye gaze toward the object/head tilt toward the subject). Non-manual agreement is claimed to function independently as a feature-checking mechanism since it can occur with plain verbs not marked with overt morphological agreement. We conducted a language production experiment using head-mounted eye-tracking to directly measure signers’ eye gaze. The results were inconsistent with Neidle et al.’s claims. While eye gaze accompanying (manually/morphologically) agreeing verbs was most frequently directed toward the location of the syntactic object, eye gaze accompanying plain verbs was rarely directed toward the object. Further, eye gaze accompanying spatial verbs was toward the locative argument, rather than toward the object of transitive verbs or the subject of intransitive verbs as predicted by Neidle et al. Additionally, we found a consistent difference in the height of directed eye gaze between spatial and agreeing verbs. Gaze was directed lower in signing space for locative marking than for object marking, thus clearly distinguishing these two argument types. Plain verbs occurring with null object pronouns were not marked by gaze toward the location of the object and always occurred with an overt object topic. Thus, Neidle et al.’s analysis of null objects as licensed by agreement (manual or non-manual) was not supported. Rather, the data w This work was supported by a grant from the National Science Foundation awarded to Karen Emmorey and the Salk Institute (Linguistics program: BCS0216791). We would like to thank Shannon Casey, Diane Lillo-Martin, John Moore, Maria Polinsky, and two anonymous reviewers for their helpful comments on earlier drafts of this paper. This paper also benefited from discussions with Grant Goodall, Irit Meir and Adam Shembri. Additionally, thanks go to Shannon Casey for sharing her data on classifying verb types. We are grateful to Samuel Hawk, Helsa Borinstein, and Stephen McCullough for valuable help conducting the experiment. Finally, we are especially grateful to all of the Deaf participants who made this research possible. Natural Language & Linguistic Theory (2006) 24: 571–604 Springer 2006 DOI 10.1007/s11049-005-1829-y