This book examines the role of syntax in theories of sentence comprehension, and argues for a distinct processing component which is devoted to the recovery of syntactic structure and which utilizes the contrasting types of information found within a Government-Binding grammar. Paul Gorrell contrasts the primary relations (dominance and precedence) and secondary relations (case assignment, theta-role assignment, etc.) in a phrase-structure tree, and shows how this computational distinction of information types is reflected in the internal structure of the parser, which consists of two subcomponents: a structure builder (responsible for creating nodes in a tree and positing primary relations between them), and a structure interpreter (responsible for analyzing the tree in terms of secondary relations). This model can also predict gardenpath phenomena in the processing of verb-final clauses.