Genitive particles, historical change, and grammar: Issues in Japanese and broader implications.
- 459 p.
Source: Dissertation Abstracts International, Volume: 62-08, Section: A, page: 2742. Adviser: Charles J. Quinn, Jr.
Thesis (Ph.D.)--The Ohio State University, 2001.
Old Japanese (OJ), a variety of Japanese in the eighth century, had three major genitive particles, ga, no, and tu. This dissertation attempts to shed new light on various issues concerning historical changes undergone by genitive particles in Japanese, and to provide broader implications for general linguistics. It examines not only historical data, but also modern dialect variations, as well as a wide range of crosslinguistic data, such as English, Chinese, Korean, Tibeto-Burman languages, and Australian languages. To list several key arguments: (i) the distribution of GEN ga, no, and tu was not determined solely based on their linguistic features, but the role allocation involved historical contingency, including social factors; (ii) the origins of GEN ga, no, and tu are remaining issues, but it seems plausible that GEN ga came from the demonstrative ko/ ka ‘this’, GEN no from the copula *nu, and GEN tu from either the copula *tu or the demonstrative *to ‘that’; (iii) based on the observation of GEN-marking of subjects in OJ, it can be claimed that each clause is “nouny” to a different degree, and that the different degree of clausal nouniness is a result of what I call “category management” in the domain of the sentence; (iv) despite the unidirectionality hypothesis in grammaticalization theory, the development of the conjunctive ga and other crosslinguistic data show that change from subordination towards parataxis is possible as long as two morphological and syntactic conditions are met; and (v) GEN ga, no, and tu all took the same developmental path in different dialects (i.e. genitive > pronominal genitive > bound pronominal > nominalizer > sentence particle), and the first two steps in this development do not fall into grammaticalization. One recurring theme throughout this dissertation is how grammars are formulated for competing forms, structures, and categories, and how that competition projects to language change. In conclusion, I claim that grammar is formulated based on speech production; i.e. grammar is not autonomous. I also claim that categories, such as “noun” and “verb”, and “subject” as well, are not given in grammar, but rather can only be first induced from speech production.