Case, Coordination, and Information Structure in Japanese

Case, Coordination, and Information Structure in Japanese * Akira Ohtani aband Mark Steedman a aSchool of Informatics, University of Edinburgh, 2 Buccleuch Place, Edinburgh, EH8 9LW, Scotland, United Kingdom bFaculty of Informatics, Osaka Gakuin University, 2-36-1 Kishibe-minami, Suita, Osaka 564-8511, Japan {aotani, steedman}@inf.ed.ac.uk Abstract. This paper investigates the nature of Japanese argument cluster (Steedman 2000b). Based on Combinatory Categorial Grammar, a type-raising analysis of case particles which captures some aspects of the information structure in Japanese is discussed, including contrastive interpretation of coordination, wh-constructions, and some theme and rheme-related grammatical phenomena. These observations offer further support for the study of syntax, semantics, and phonology interface and the earlier analysis of English information structure. Keywords: argument cluster, Japanese case particles, information structure, Combinatory Categorial Grammar (CCG), coordination, wh-constructions, multiple-ga, theme, rheme 1. Introduction Steedman (2000b: p.172) accounts for a number of facts about “non-constituent” coordination in Japanese by allowing type-raised subject and object NPs in Japanese to combine not only by forward application to the verb (>), as in (1b) below, but also by forward-composition (>B), as in (1c) below, under the framework of Combinatory Categorial Grammar: (1) * We are indebted to Takeo Kurafuji and an anonymous PACLIC21 reviewer for their invaluable comments on an earlier version of this paper. Our thanks also go to Yoshiyasu Shiraii, the president of Osaka Gakuin University. All remaining inadequacies are our own. Copyright 2007 by Akira Ohtani and Mark Steedman 365 The derivation in (1b) is isomorphic to a standard phrase structure analysis, as in (2a) below. However, the derivation in (1c) allows the NPs to compose non-standard constituent cluster, as shown in (2b): (2) In this paper, we discuss the nature of non-standard constituent cluster in Japanese. Based on Combinatory Categorial Grammar (CCG) (Steedman 1996, 2000b, Steedman and Baldridge 2007), we propose a type-raising analysis of case particles which captures some aspects of the information structure in Japanese/Korean. 1 This account of the information structure of argument clusters offers further support for the earlier analysis of English information structure. 2. A CCG Analysis of Gapping 2.1.Gapping Revisited One motivation for the non-standard structure in (2b) is the tendency of argument clusters to act like constituents under coordination, and the relation of this phenomenon to the base order of constituents across SOV, VSO and SVO language and/or construction (Ross 1970). Ross's generalization follows as a theorem from the axioms of CCG, as illustrated by Japanese gapping construction like (3a) and the possibility of the non-standard constituent illustrated in (1c) as its part, as in derivation (3b): (3) The possibility of semantically surface-compositional syntactic derivations like this is one of the main theoretical attractions of CCG. The availability of two different derivations in (1b) and (1c) for sentences like (1a) allows us to consider the possibility that they are semantically and pragmatically distinct in some way. There are some specific properties of gapping that are interesting in this respect. See below: (4) In English, the second occurrence of the verb visited in (4) can …

Content truncated to 500 words