A Multiple Inheritance Analysis of the Internally-Headed Relative Clause in Japanese

Language, Information and Computation (PACLIC12), 18-20 Feb, 1998, 82-93 A Multiple Inheritance Analysis of the Internally-Headed Relative Clause in Japanese Chiharu Uda Kikuta Doshisha University, Kyoto, Japan <cuda@mail.doshisha.ac.jp> Abstract This paper proposes a mutliple inheritance type-hierarchical analysis of the Japanese "Internally- Headed Relative Clause" (IHRC) in Head-driven Phrase Structure Grammar (HPSG). It has been a long standing issue whether the IHRC shares syntactic properties with the EHRC or it is basically a clausal complement structure. This paper claims that such bipartite debate is misguided. It is undeniable that IHRC is similar to EHRC in one respect, and to clausal complement clause in another. Defining it categorically as either one only distorts the reality. It will be proposed, therefore, that IHRC is a subtype of both a relative clause and a clausal complement. The hybrid structure of IHRC is quite a marked type of multiple inheritance, and I claim that this accounts for the restrictive availability of IHRC and its unstability as an independent grammatical construction, as well as for the IHRC's intermediacy between the EHRC and the clausal complement structure. This discussion is done on Japanese alone; however, the analysis with minor modifications will accommodate the Korean IHRC as well. 1. Introduction This paper discusses the structure of the so-called "Internally-Headed Relative Clause" (IHRC) in Japanese in the constraint-based framework of Head-driven Phrase Structure Grammar (HPSG) and proposes a mutliple inheritance type-hierarchical analysis to account for the dual nature of the IHRC as well as for its restrictive availability. There has been a long-standing debate regarding whether the IHRC is to be analyzed as a relative clause or as a clausal complement (Kim 1996, Kuroda 1992, Ohara 1996, Park 1994). Semantically speaking, the IHRC apparently modifies a noun, strongly analogous to the Externally-Headed Relative Clause (EHRC). Syntactically, however, it is quite distinct from the EHRC; it does not even have a syntactic nominal head which is the target of, modification. Rather, the syntactic structure (on the superficial level at least) of the IHRC is identical with that of a clausal complement. This state of incongruity has led many researchers to assume a quasi-EHRC structure for the IHRC, positing empty categories for the missing syntactic head noun and other positions when necessary. Such relative clause analyses, however, fail to capture some differences between the IHRC and the EHRC. The clausal complement analysis, on the other hand, must explain why and how the IHRC refers to an entity involved in the event denoted by the clause (=entity-reading), while a clausal complement refers to an event as a whole (=event-reading). Park's (1994) analysis of the Korean IHRC leaves this essential question to some type of pragmatics, and Y-B. Kim's (1996) analysis contains crucial technical problems, to be shown in a note below. Thus the adequate analysis of the IHRC in Japanese is yet to be known. The present paper claims that the bipartite debate regarding the identification of the IHRC is misguided. It is an undebatable fact that the IHRC is similar to …

Content truncated to 500 words