Sulla relazione storica tra il giapponese del periodo Nara e i dialetti ryukyuani contemporanei.

This thesis aims to investigate and prove the historical relationship between jōdai tokushu kanazukai from Nara period literary works and vocalic systems within contemporary Ryukyuan dialects. Ryukyuan dialects are quite conservative and show several features of Old Japanese, which disappeared in mainland dialects, and their language history may illuminate our understanding of mainland Japanese linguistic changes.

 The discovery that certain syllables which are now pronounced identically were given systematic distinctions in terms of different characters in the Nara period works, such as Kojiki (712), and Man’yōshū (759), led to the hypothesis that in addition to a and u, there were, in Old Japanase, two types of i, e and o. The two series of the latter have traditionally been designated by the class names kō and otsu. The theory of the eight-vowel system is widely accepted but, in recent times, many scholars have provided different theories regarding its validity. Hattori Shirō (1976) and Matsumoto Katsumi (1984) proposed systems of five and six vowels for Old Japanese, in which the distinction of the two series of syllables with high vowels is due to the presence or absence of palatalization or labialization of the consonants. With this thesis I intend to further elaborate on this topic, investigating the phonological system of Ryukyuan dialects which have not been examined in detail yet. I will take into consideration dialects of the five subgroups (Amami, Okinawa, Miyako, Yaeyama, Yonaguni), but I will focus on the dialect of Kakeroma (Amami), as a representative of the northern Ryukyuan dialect group, and the dialect of Hateruma (Yaeyama), as a representative of the southern Ryukyuan dialect group. The linguistic data useful for this research have been collected through fieldworks in the Ryukyuan archipelago. After a rigorous analysis of the phonological system of these two dialects, a comparative study will be carried out. Through a comparison between cognates of the Kakeroma and Hateruma dialects, and Old Japanese, it will be possible to verify, whether the theory of three distinct syllables in proto-Japanese of Hattori (1978-79) can be also applied to other Ryukyuan dialects. Also, as part of the comparative analysis, the chain of diachronic change will be reconstructed, in order to show which kind of evolution stands behind the discrepancy among different Ryukyuan dialects.