Habuba on the Tigris: Archaic Nineveh Reconsidered

Abstract

FOR some time now archaeologists have been puzzled by the obvious dichotomy between the presumably contemporaneous late fourth-millennium levels at the neighboring sites of Nineveh and Tepe Gawra in the Upper Tigris area.' This puzzlement stems from the fact that the extensive connections of Nineveh with Protoliterate Mesopotamia are unparalleled at Tepe Gawra.2 The striking differences cannot be attributed to sampling differentials, as Tepe Gawra has one of the broadest exposures of any site of this period in the region. Attempts at solving this discrepancy have ranged from suggestions that Tepe Gawra has an unrecognized gap corresponding to the Early Protoliterate period,3 to arguments that both sites represent functionally distinct types of settlements.4 Tepe Gawra exhibits an indigenous Late Chalcolithic culture which is clearly oriented in the direction of the northern Syro-Mesopotamian steppe and beyond to the Taurus highlands. The glyptic and ceramic assemblages of levels XI-VIII of the Gawra sequence have close parallels in that direction. The distinctive stamp seal tradition of those levels at Gawra, so different from the cylinder seal tradition of Nineveh,5 is now readily paralleled in the impressive glyptic repertoire found in the context of the Temple and Palace structures of period VI A at Arslan Tepe in Malatya.6 The chaff-faced ceramic assemblage which characterizes the Late Chalcolithic levels of the Gawra sequence has a wide distribution along an arc ranging from coastal and northern Syria to northern Mesopotamia and up to the Taurus highlands. Within this broad area of distribution, the closest ceramic parallels to the Gawra assemblage are found in the Keban/ Altinova region of the Taurus. This is