Grant Goodall, UC San Diego

Friday, May 1st at 2 pm, Stevenson Fireside Lounge

Grammar and working memory: How experimental syntax can help us tell the difference

The use of formal experiments to measure sentence acceptability, known as “experimental syntax”, is able to capture many fine-grained grammatical contrasts, but it also captures effects that have long been thought to be extra-grammatical, such as those induced by increased cost to working memory. This ambiguity in the source of acceptability is a problem in some ways, but experimental syntax itself gives us some useful tools to address it. I show this through a series of case studies of constraints on wh-dependencies, including the role of intervening arguments, finiteness, D-linking, and islands. These cases suggest that grammatical and working memory constraints can be usefully distinguished, with the former sensitive to intervening hierarchical structure and the latter sensitive to intervening referents.