The relationships of the Eastern Neo-Aramaic dialects.

• Author: Fox, Samuel Ethan

Introduction

IN RECENT YEARS THERE HAS BEEN an increase in scholarly activity in the field of Neo-Aramaic and a corresponding growth in our knowledge of the range of dialects which make up NENA (Northeast Neo-Aramaic).(2) In light of this, it is now possible to achieve a somewhat clearer view of the whole range of dialects, their interrelations, and their history. NENA dialects were, up to 1915, spoken by Christians and Jews in an arc of mostly mountainous territory stretching about 550 kilometers across parts of southeast Turkey, northern Iraq, and western Iran. The majority Muslim population of the same area spoke Arabic, Turkish, and above all, Kurdish. Since 1915 the Neo-Aramaic (NA) speaking people have been uprooted from their traditional territory, and successive upheavals have left very few of them in their traditional homes. Southeast Turkey has become all but empty of Neo-Aramaic speakers, and their villages throughout the area are largely deserted and in ruins. Significant numbers now live in northeast Syria, the cities of Iraq and Iran, in Israel, and in a growing world-wide diaspora. At the same time the related NA dialects, Turoyo and Mandaic, have also come under severe pressure.

Background of the Modern Dialects

Since Noldeke's (1868, xxxv) pioneering work on NA we have known that NENA is not a descendant of Syriac, but of an unattested sister dialect. Not having Proto-NENA before us, in seeking to understand the background of the modern dialects we turn to the three earlier dialects of Eastern Aramaic which we do possess: Syriac, Babylonian Talmudic (BT), and Classical Mandaic, Syriac is the dialect of Edessa (present-day Urfa) around the third to seventh centuries A.D. BT reflects the language of the Jewish academies of Sura, Nahardea, and Pumbedita, along the Euphrates in central Mesopotamia. Classical Mandaic was the written language of the Mandaeans of southern Mesopotamia around the third to seventh centuries A.D. Before the Arab conquest, Aramaic was the dominant language in Palestine, Syria, and Mesopotamia. Between those points in time and space about which we know something, there must have been a whole continuum of dialects about which we have no knowledge at all (Boyarin 1981). The Aramaic of the Mesopotamian plain (and whatever portion of the surrounding mountains that spoke Aramaic) must have varied from place to place, over time, and very likely also confessionally. To what extent, then, do the very great differences between NENA dialects carry on these older dialect divisions? Is it possible to trace all the dialects of NENA to a single distinctive source? The dialects of NENA exhibit an amazing diversity. In view of this, might the dialect groups in NENA continue divisions between unattested earlier dialects? Should they be described as a distinct unit or might they better be seen as independent parts of a spectrum of modern dialects extending from Turoyo, through NENA, and on to Mandaic?

Difficulties

One of the many problems that contribute to the difficulty of determining the history of NENA is the mobility of its speakers. The calamities of the last seventy-five years have left few of them where they were at the beginning of the twentieth century. We know little of their earlier movements, but local tradition (Krotkoff 1982, 103) may suggest depopulation and resettlement. Maclean (1895, xiv) mentions a village in the plain of Urmi which seemed to be a colony of Tiari speakers. Sabar (1976b, xxviii) also points to the number of Jews in Zakho whose family names suggested origins elsewhere in Kurdistan. Conditions in the area have often been unsettled. For example, a serious incident of pillage and plunder in 1842 (Joseph 1961, 64) and an extended period of chaos at the end of the nineteenth century (Joseph 1961, 116-17) resulted in the dislocation of many NENA speakers. It is possible that some very different dialects, which in the early years of this century were spoken in close proximity, may have evolved in locations quite remote from each other.

Another factor complicating the situation is the differential influences of the various co-territorial languages. The dialect differences within Kurdish are very considerable, and to their impact can be traced some of the variation among NENA dialects. For example, the Jewish Neo-Aramaic of Halabja, like the Kurdish of the area, does not distinguish a present and future tense, but where Kurdish maintains such a distinction, so does Neo-Aramaic. In Azerbaijan, Turkish has exerted strong influence on the Jewish dialect (Garbell 1965b), and the dialects of the plain of Mosul show a powerful influence from Arabic not seen elsewhere.

Thomason and Kaufman (1988, 78-91) suggest a correlation between the types and intensity of contact between languages and the kinds of influence which result. The types of changes which are found in Neo-Aramaic seem to fall most easily into their category 3 or 4, which are characterized by heavy lexical borrowing, as well as limited borrowing of structural elements. This pattern is indicative of borrowing, rather than language shift. That is, it points to a population which has always spoken Aramaic, but whose language has evolved under pervasive influence from other communities, rather than one which has adopted Aramaic while retaining features of its original tongue.(3)

Yet another difficulty is our limited knowledge of the modern dialects and their taxonomy. The only existing dialect classifications are those of Maclean (1895, xii-xv), Cohen (1972) and Hoberman (1988). The first two are restricted to the Christian and Jewish dialects, respectively, and neither makes clear the criteria on which it is based. Hoberman (1989, 6-7) summarizes and extends the classifications of Maclean and Cohen. Hoberman (1988) is an excellent study of pronouns and pronominal affixes covering a range of dialects and incorporating extensive comparative reconstruction; however, these data support only a few firm dialect groupings.

Some important factors hold back progress in this area. On the one hand, there are many undocumented dialects. The most urgent task facing specialists in NA is gathering texts and grammatical descriptions of the many dialects which are on the verge of extinction. Odisho (1988, 19-24) describes the swift retreat of the older dialects in the face of a new koine formed in the cities of Iraq. The least fieldwork in the area is sufficient to make clear how many dialects have already vanished. On the other hand, our information is sketchy and insufficient even for many known dialects. Under these conditions, Macuch suggests (1990, 223), we can hardly write a comparative grammar of NENA, and certainly, at the current state of our knowledge, such a work could be neither complete nor satisfying. Still, we should be able to cast some light on the essential questions which we have raised by comparing a range of NA dialects to each other, to the other modern dialects, and to their predecessors.

Selection of Dialects

We have selected eleven different dialects of NENA for our comparison. They have been picked as representatives of each of the well-defined groups of dialects now known to us. As one of our main goals is reviewing the boundaries of the NENA unit, we will include both of the other modern Eastern Aramaic dialects, Turoyo and Mandaic. On the other hand, we will omit the dialects of Aramaic spoken in Malula and two other Syrian villages as clearly...