In lieu of an abstract, here is a brief excerpt of the content:

SYNTACTIC MARKEDNESS AND THE DEFINITION OF SENTENCE TOPIC Alice Davison University ofIllinois Sentence topics are linguistic constituents, with syntactic and semantic properties which single them out for a linking function in the process of relating a sentence to its discourse context. This paper explores the relation between the linguistic properties of sentence topics and their use as discourse links in sentence processing. It proposes a set of criteria for distinguishing relatively weak or strong topic NP's, based on syntactic and semantic/pragmatic properties. Syntactically defined topics include subjects and those in 'marked' NP positions, where the surface features define grammatical function in an ambiguous or indirect way. Implications of this proposal are drawn for several languages, and for various models of language processing.* It is a common observation that corresponding active and passive sentences differ in what they appear to be 'about'—what they contribute to a discourse— even though they describe the same state of affairs (Ziff 1966); e.g., (1)a. A tiger chased a tourist. b. A tourist was chased by a tiger. The difference is compatible with a difference of context, so that the sequence of 2a followed by la is somehow more 'connected' than 2b followed by la, and so on: (2)a. Wild animals are dangerous even in nature parks, b. Visitors should not be careless in where they go. But the context does not determine what will be the syntactically marked topic of sentences like la-b; and ifthe context is vague, the difference between these sentences may persist entirely independently of the context: (3)An unfortunate incident occurred last week. So the difference between la and lb must result from internal properties of the sentences—in fact, solely from differences in their syntactic surface structure . The purpose of this paper is to bring together and to relate some facts about (a) sentence structure, (b) the representation of syntax in grammars of different types, (c) sentence processing, and (d) the notion of sentence topic—which will normally be referred to below simply as 'topic'. Up to now, topic has been * The research reported here was supported by the National Institute ofEducation under contract No. HEW-NIE-C-400-76-0116 to the Centerforthe Study ofReading, University of Illinois, Champaign -Urbana. I am indebted to the Reading Center for providing a climate in which this proposal could be worked out in various forms. For helpful discussion and criticism which has been very useful, I would like to express gratitude to the late Adrian Akmajian and to David Dowty, Nina Garrett, Georgia Green, Gabriella Hermon, Adrienne Lehrer, George McConkie, Jerry Morgan, Ellen Prince, Ivan Sag, and Carlota Smith. Others are hereby thanked for specific contributions mentioned in the body of the paper. In addition, I would like to thank Paula Chen Rohrbach for helpful discussion of sentence topics in Chinese, and Atsushi Fukada for the same in Japanese. Flaws and imperfections are solely mine. 797 798LANGUAGE, VOLUME 60, NUMBER 4 (1984) discussed for the most part independently ofthese other matters, although there is growing interest in relating topic to processing. But the notion of topic, while amenable to discussion in a very general way, has been elusive when writers on particular languages have tried to define it narrowly for a particular language (Kuno 1973, Tsao 1979 being signal exceptions). In this paper, I propose that topic in English can be discussed in fairly specific and definable terms, provided its discussion is restricted to the contribution of syntactic structure and lexical/pragmatic properties of NP's. Many contextual and variable factors can define a topic; but in the absence of stronger contextual factors which define it in a particular context, certain constituents of a sentence are perceived as more salient or marked than others, and these are topics. This perception is based on the speaker's knowledge of the rules of syntax of the language in question, and the range of lexical information which helps to pick out the referent of an NP. What makes active and passive sentences different in English and other languages is clear. What is less easy to define is what general properties of surface structure are responsible for defining topics...


Additional Information

Print ISSN
pp. 797-846
Launched on MUSE
Open Access
Back To Top

This website uses cookies to ensure you get the best experience on our website. Without cookies your experience may not be seamless.