Constituent order in coordinate constructions : a processing perspective

Abstract

This dissertation investigates constituent order in coordinate constructions in English on several levels, ranging from compound constituents to the order of phrases. Previous studies in linguistics addressed this problem predominantly in fixed expressions, so-called frozen binomials, e.g., law and order. In psycholinguistics on the other hand works dealt with processing constraints relevant for linearization in syntax in general. The present work aims at bridging the gap between these two approaches, thereby exploring the differences and commonalities between frozen, lexicalized constructions and instances of \\\\\\\"free\\\\\\\" coordination. Several cases studies are conducted which rely on corpus data and use multi-factorial modeling techniques. The quantitative analysis tests the influences of a multitude of hypothesized influences on the ordering process, testing variables ranging from conceptual/semantic properties of to-be-ordered constituents to phonetic influences. The findings are discussed within the framework of speech production models. In doing so, results are compared to those of other order alternations, thereby addressing ramifications for speech production theory.