On Vowel Reduction in Aramaic

Abstract

THE QUANTITY OF AVAILABLE TEXTS in and studies on the various ancient and modern Aramaic dialects has only recently reached the point where a comprehensive, comparative historical grammar could honestly be written. Mindful of our celebrant's call for such a project, the following remarks are directed at a central problem in it-the correlation of observed textual data with historical reconstructions based on the comparative method.' If asked to identify characteristic features common to all of the Aramaic dialects, most Semitists would probably cite first the merger of the proto-Semitic spirants with their corresponding stops, the reflex of proto-Semitic d as 'ayin, and the conditional change resulting in the reduction of short vowels in open, unstressed syllables. The initial working assumption of the historical linguist in such cases is that the development of common features of this sort took place before major dialect divisions occurred. For these Common Aramaic developments, however, alternative hypotheses could and indeed have been suggested in consonance with the somewhat disconcerting fact, long recognized by classical comparative linguistic theory, that related languages tend to develop independently along similar (and frequently identical) lines.2 Until recently, those Aramaicists favoring the working assumption of common development could rest content in their understanding that the distinct Aramaic dialects emerged only in the late classical period (First Century C.E. and later) and that the period of common development could have extended to just prior to that time. But recent work has demonstrated that distinct dialects are apparent at least 500 years earlier,3 thus necessarily raising the latest possible date for the origin of the Common Aramaic features, perhaps to a period deemed unacceptable on the basis of the textual evidence. If so, we would need to have recourse to the alternative explanation of independent development. This paper is devoted to an analysis of the evidence for the dating of the reduction of unstressed short vowels in open syllables, a development that, as we shall see, has been dated as early as the Eighth Century B.C.E. and as late as the Third Century C.E. In marked contrast to the case of the merger of stops with spirants, wherein substantial direct evidence may be adduced from the Aramaic texts themselves,4 the ancient, unvocalized consonantal texts are only mute witnesses to the development of vocalic change. We must search elsewhere for our evidence. The primary evidence that has been used to solve this problem is of two kinds: external, i.e., the appearance of Aramaic forms, in transliteration or as loanwords, in datable texts in other languages and scripts; and internal, i.e., the appearance within Aramaic of observable changes that depend for their own development on the prior reduction of unstressed short vowels. Yet a third body of evidence, whose relevance to the topic at hand was first pointed out by J. Blau,' was analyzed by me in the course of a study on the