Syntactic Weight Versus Information Structure in Word Order Variation

Abstract

The focus of the present paper is word order variation within languages. There is a widespread view in the literature that the selection among truth-conditionally equivalent word order variants permitted by the grammar is determined primarily by considerations of “information structure”, i.e. pragmatic-semantic notions such as predictability or importance (Givon 1983, 1988), agency, definiteness (Jacobs 1988), etc. It is generally agreed that syntactic weight or length is also relevant in performance, but the extent of this relevance is usually seen as being limited to a handful of structures that are particularly difficult for processing, such as center embeddings and the positioning of finite clause complements (cf. e.g. Grosu & Thompson 1977, Dryer 1980).

I have benefited greatly from discussions with a number of individuals while conducting the research reported here, especially Kaoru Horie and Stephen Matthews at USC, and the members of the Constituent Order Group of the European Science Foundation Programme in Language Typology, especially Katalin Kiss, Beatrice Primus, Anna Siewierska and Maria Vilkuna. Valuable feedback was also received from linguists at the Free University of Berlin, when portions of this material were presented there in October 1989. None of the individuals mentioned necessarily agrees with all of my conclusions. I would also like to acknowledge, with gratitude, financial support from the following sources that has made possible the ongoing research project from which the results reported here are derived: a summer stipend from the National Endowment for the Humanities (FT-34150); a small grant from the European Science Foundation; and a grant from the University of Southern California Faculty Research and Innovation Fund (FRIF).