シュメール語の動詞接尾辞における母音の同化について

Abstract

When verbal suffixes follow verbal bases in Sumerian, the vowel e in the suffixes often assimilates to the preceding vowel u in the bases. This assimilation has traditionally been well known, but the environment in which it occurs has never been specified. In this article I have shown that the occurrence and nonoccurrence of this assimilation is determined by the nature of final consonants of verbal bases. The frequency of assimilation varies in terms of the final consonants of verbal bases. When verbs end in velar consonants (g, ŋ), or dental consonants (d, n), assimilation is hardly observed. Verbs ending in h and s show the similar tendency though the verbs of this group are comparatively small in number. On the other hand, assimilation is relatively regular in the verbs which end in bilabial consonants (m, b), liquids (r, l) and š. The assimilation is particularly conspicuous in the verbs ending in m and b, where this vowel assimilation is observable in documents recorded as early as in the third millennium B.C. The situation is quite different in the other verbs, which never show the assimilation before the second millennium B.C. The above evidence suggests that the vowel assimilation in question has its own unique inner-Sumerian history. It first occurred in the verbs ending in labial consonants, and then gradually spread to the verbs with other final consonants except for those ending in velar and dental consonants, where the assimilation is hardly observed even in the latest documents. The influence of the consonants with a labial feature on neighboring vowels is hardly unique to Sumerian. Typologically, similar phenomena are found many other languages in the world.