Language users must constantly choose between different linguistic forms that refer to the same entity, such as 'he', 'Henry' or 'the man.' How do speakers choose which form to use? In spoken languages, this choice is largely determined by grammatical and information structural factors such as the previous grammatical role of the entity and whether the entity is the topic under discussion. This dissertation research asks whether pronoun usage in sign languages follows grammatical or gestural rules. American Sign Language (ASL) pronouns, which are points, are thought to get their meaning from ad hoc associations between entities and the space the signer points to. A point towards a space associated with a man named Henry is equivalent to the pronoun 'he' in English. By contrast, a pointing sign towards a space where no entity has been previously associated is predicted to be meaningless. This view suggests that ASL pronouns are primarily spatial associations between entities and signing space, which are not necessarily chosen based on principles of grammar or information structure. Two experiments test whether ASL pronominal forms follow the grammatical patterns of spoken languages. The experiments investigate how ASL signers refer to entities in contexts that vary in spatial associations and the number of possible entities being talked about.
This research will inform our understanding of whether choice of referring expressions is guided by universal or language specific principles. This knowledge can directly contribute to improving education of deaf children who on average have much lower literacy skills than hearing children. By developing a model for how ASL creates referential links between sentences, the project will provide the means to help signing children to understand the parallel processes in English text.
