Abstract:

This dissertation investigates the diachronic development of personal pronouns in Japanese which is known for its large pronominal inventory. Japanese pronouns come from two major sources, lexical nouns and spatial expressions such as demonstratives. Through examinations of historical texts, this study show how the framework developed in recent studies of grammaticalization and semantic change such as Lehmann (2002 [1995]) and Traugott & Dasher (2005) can shed light on the nature of pronouns in Japanese, a language which is claimed to be 'pronoun-less' by some (e.g. Kanaya 2002). Though the development of some forms was influenced by extralinguistic factors such as a change in a social structure and literary movement, overall stories of Japanese pronouns show remarkable similarities to one another.

This dissertation is organized as follows. Chapter 1 outlines the objectives and methods of this study, describing the nature of texts and periodization employed. Chapter 2 presents the relevant literature regarding deixis, Japanese pronouns, grammaticalization, and semantic change. Chapter 3 investigates the development of personal pronouns that arose from lexical nouns. Pragmatic inference and the shift from the world of the described event to that of the speech event are crucial in their development: e.g. first person watakushi 'private' was often used with the notion of ooyake 'public' to express the contrast with one's own private matters and those of public nature such as one's social obligations, which led to the interpretation of the former as 'personal', 'individual', and eventually as the speaker. Chapter 4 discusses pronouns that are said to have developed from demonstratives. Contrary to the common assumption in Japanese linguistics, this dissertation views the so-called demonstrative-based pronouns, not as semanticized personal pronouns, but simply as demonstratives used in place of personal pronouns. Textual evidence suggests that demonstratives are the common source of third person pronouns only and that they give rise to first and second person pronouns only in limited contexts. Chapter 5 deals with the shift of person categories. Besides the cross-linguistically common shift from the third person to first/second person (e.g. from nouns to pronouns), some have claimed that the shift between the first and second person occurred in Japanese pronouns (e.g. Whitman 1999). This dissertation shows that there are at least four types of shift, the result of which may or may not lead to semanticization, and each of which is motivated by independent principles. Finally, Chapter 6 discusses the direction for future research by placing the findings of this dissertation in cross-linguistic context.