Scope and Anaphoric Links in Dynamic Discourse Representation Theory

Scope and Anaphoric Links in Dynamic Discourse Representation Theory ∗∗∗∗ So-Woo Chung and Jungmin Lee Department of English Language and Literature, Sungshin Women’s University 249-1, Dongseon-dong 3-ga, Seongbuk-gu, Seoul, 136-742, Korea swchung@sungshin.ac.kr, soulmate914@naver.com Abstract. This paper analyzes anaphoric links in conditional and quantificational structures in which more than one operator occurs. To this end, we adopt the indexing system and the basic principles of Dynamic Discourse Representation Theory proposed by Chung (2008a, 2008b, 2009) and extend them to the structures in question. We also propose that a restrictor discourse representation structure of a quantification or conditional operator is an island which blocks any quantification or conditional operator inside it from introducing its own restrictor DRS out of it. This implies that a quantifier phrase can have wide scope over another quantifier or conditional operator only when the first occurs in the scope DRS of the latter. Keywords: discourse representation structures, quantificational structures, conditional structures, anaphoric resolution, scope 1 Introduction 1.1 Purpose This paper discusses anaphoric links between pronouns and their antecedents in conditional and quantificational structures, particularly where more than one operator occurs. The analysis we propose here is of pure semantics in that it does not resort to any particular syntactic theory and in that the rules and principles are defined solely in terms of lexical information and accessibility between discourse representation structures and discourse referents. 1.2 Motivation Even though quantification structures and anaphoric resolutions have been one of the central issues in Discourse Representation Theory since its birth in Kamp (1984), not many works have discussed why the coreferential readings in (1a) and (1c) are blocked. Just a few papers including Robert (1987), Asher and Wada (1989), and van Deemter (1990) have discussed it in terms of Discourse Representation Theory. Most of them, however, have attributed its grammaticality to the violation or non-violation of some kinds of syntactic principles such as Binding Conditions of the Generative Syntax, rather than accessibility between discourse referents. (1) a. *A senator who likes him1 supports every representative1. b. Every senator1 meets most representatives who support him1. c. *Most representatives who like him1 vote for every senator1. Only a couple of papers such as Chung (2008) and Kamp and Reyle (1993), to my knowledge, have tried to provide an explanation of their grammaticality in terms of accessibility. But, neither of them has discussed, in terms of accessibility and discourse representation structures, why the universal quantifier cannot have wide scope over another quantifier as in (2a) below Copyright 2009 by So-Woo Chung and Jungmin Lee 101 23rd Pacific Asia Conference on Language, Information and Computation, pages 101–109 and over the conditional operator as in (2b), consequently not being able to have the intended coreferential readings with the pronouns. They have not dealt with what blocks a coreferential reading between the universal quantifier and the pronoun in (2c), either. (2) a. *Most representatives who like every senator1 will vote for him1. b. *If every senator1 likes Mary, he1 will vote for her. c. *If he1 …

Content truncated to 500 words