Abstract

Stabler (2013)’s parser forMinimalist grammars has been shown to successfully predict various off-line processing preferences, by exploiting complexity metrics connecting syntactic structure to memory load. This approach provides a quantifiable way to test the effects of structural hypotheses on linguistic behavior, and thus can help bridge syntactic theory and processing phenomena. This paper extends the empirical coverage of the model by looking at the processing of Italian postverbal subjects, and it discusses the relevance of transparent computational models for existing approaches to sentence processing