This book is a collection of articles on Japanese grammars written in a framework broadly construed as constraint-based. Having evolved from a progress report on the Japanese phrase structure grammar (JPSG) project, it implements ideas from recent developments in phrase structure grammar formalism, such as head-drive phrase structure grammar (HPSG). Part I gives an overview of JPSG, introducing fundamental assumptions and discussing mostly syntactic and semantic aspects of Japanese in the constraint-based formalism, as well as the implementation of a parser based on the grammar. The papers in Part II discuss mostly pragmatic aspects of Japanese. Innovative concepts are introduced in Part I, such as the concept of cost, which plays a central role both in the treatment of quantifiers and phonology, and is used extensively in the treatment of conditionals in Part II. Stress is also placed on the role of discourse participants; the analysis of relativization adopted in Part I is partly based on zero pronominals, whose interpretation is left to pragmatics. This issue is further pursued in Part II and may serve as one possible avenue to explore pragmatics.

This book is a considerable revision and extension of my thesis for The Ohio State University completed in 1981: A Phrase Structural Analysis of the Japanese Language (Gunji 1981a). The book discusses some of the major grammatical constructions of Japanese in a version of phrase structure grammar called Japanese Phrase Structure Grammar (JPSG), which is loosely based on such frameworks for phrase structure grammar as Generalized Phrase Structure Grammar (GPSG) and Head-driven Phrase Structure Grammar (HPSG). Particular emphasis is placed on the binding and control of pronouns (both implicit - "zero" - and explicit ones, including reflexives) in complementation structures (chapter 4) and adjunction structures (chapter 5). Even though this book started as a revision of my 1981 thesis, the resultant book has few traces of my thinking then. The 1981 thesis was closely related to an early version of GPSG, which was then at a very preliminary stage, and I had only a few preprints of papers by Gerald Gazdar and others to read. GPSG itself has evolved during the past. several years, culminating in a book published last year (Gazdar, Klein, Pullum, and Sag 1985), which differs from the early theory in many ways.

This collection of papers reports our attempt to sketch how Japanese grammar can be represented in a constraint-based formalism. Our first attempt of this nature appeared a decade ago as Japanese Phrase Structure Grammar (Gunji 1987) and in several papers following the publication of the book. This book has evolved from a technical memo that was a progress report on the Japanese phrase structure grammar (JPSG) project, which was conducted as an activity of the JPSG Working Group at ICOT (Institute for New-Generation Computing Technology) from 1984 to 1992. JPSG implements ideas from recent developments in phrase structure grammar formalism, such as head-driven phrase structure grammar (HPSG), (see Pollard & Sag 1987, 1994) as applied to the Japanese language. The main goal of this project was to state various grammatical regularities exhibited in natural language in general (and in Japanese in particular) as a set of local constraints. The book is organized in two parts. Part I gives an overview of developments in our framework after the publication of Gunji (1987), introducing our fundamental assumptions as well as discussing various aspects of Japanese in the constraint based formalism and summarizing discussions of the JPSG Working Group during the above-mentioned period. Naturally, in the period after the publication of the above book, our discussion was centered on topics not covered in the book.