LETTER FROM THE FIELD: IN SEARCH OF LOST MAR'ASH
By Donald S. Whitcomb, Research Associate (Associate Professor)
The Oriental Institute, and the
Department of Near Eastern Languages and Civilizations
The University of Chicago
(This article originally appeared in The Oriental Institute News and Notes, No. 171, Fall 2001, and is made available electronically with the permission of the editor.)
The cow moved nearer to the tent. The village of Kelibishler in late June is a buzz of agricultural activity; the winter wheat is almost all harvested and the cotton is beginning to come along. Little attention can be paid to the archaeologists gathered to study the ancient history of the plains south of Kahramanmarash in southeast Turkey near the Syrian border. This has been the project of Elizabeth Carter for the last six years, well after she took her doctorate and left the halls of the Oriental Institute to teach in the more favored weather at the University of California, Los Angeles. Southern Turkey has long been a focus for the Oriental Institute and its students, from the early days of Alishar Höyük and the Braidwoods to the salvage work of Guillermo Algaze and the more recent work of Asl1han Yener and Tony Wilkinson, now working in the Amuq.
Many archaeologists drawn north for the salvage campaigns crossed the modern border between Syria or Iraq into Turkey with little concern for the line drawn on maps, their focus being the continuities and variations of ancient cultural areas. Thus the Kurban Höyük Project drew on the great fund of information from north Syria, and specifically the Tabqa dam project, in attempting to understand its Bronze Age materials. Likewise, Algaze was able to date a ninth-century caravanserai at Kurban Höyük based on information from Raqqa and other Islamic sites in Syria. What is recognizable in Islamic artifacts becomes much more complicated when one tries to apply historical contextualizations. Thus a move from Aleppo in north Syria to Kahramanmarash, about 200 km to the north, crosses interesting historical borders.
The Oriental Institute archaeological project at Hadir Qinnasrin in north Syria is intended to study the development of an early Islamic city, a new foundation which became the provincial capital for all of north Syria until the mid-ninth century. The excavations of last year revealed two phases of the early Islamic period: the earlier reflected the late Byzantine context of north Syria with its great cities of Antioch and Apamea, as well as the so-called Dead Cities immediately northwest of Qinnasrin; the latter phase presents the shift to the east with the growing influence of the Abbasids in Mesopotamia, illustrated in the close connections with Raqqa in the ninth and tenth centuries. Qinnasrin was the administrative center of the north. Among the special responsibilities this entailed was the organization of the yearly campaigns into the Byzantine territories of Anatolia. However, archaeologists have rarely paid serious attention to interactions to the north, across the border as it were.
North of Qinnasrin was the Thughur, as the border zone was known in early Islamic times. The Arabic term al-thughur is interesting, with an implied meaning of the gaps between teeth. Here were great teeth - the imposing ranges of the Anti-Taurus and Taurus Mountains leading onto the high plains of the Anatolian plateau (see map). As the Muslim conquests moved northward, the Thughur system became a series of fortified bases established near the gaps or passes onto the plateau, about 60 km from one another. The central position was the site of Mar'ash (as Kahramanmarash was then known), which was established during the Umayyad caliphate under Mu'awiya, who also founded Qinnasrin. Indeed Mar'ash seems to have functioned as a forward base and was garrisoned from Qinnasrin. Such a garrison might be considered as a misr or ribat, two terms for fortified settlement. On the other hand, there was a clear intention to establish urban centers and fortified cities were attempted, repeatedly. Only after the mid-tenth century would Muslims and Byzantines resort to isolated castles as their main centers in this region.
The Mar'ash region presents a splendid opportunity to observe the extension of Qinnasrin to the north and to seek another early Islamic urban foundation. The UCLA survey has recorded more than 250 archaeological sites of all periods and the first task was to distinguish those with Islamic (or medieval) occupations. There were many sites which held clear diagnostic sherds, especially sgraffiato glazed wares indicative of the Middle Islamic (Anatolian Late Byzantine) period. Others had less well- studied common ceramics indicating Middle and Late Islamic occupations. These corpora of artifacts present intriguing problems, and with very few archaeologists working on them, there is a strong temptation to impose some order on these materials. Gradually, however, diagnostics of an Early Islamic (Anatolian Middle Byzantine) period became clear, especially with stylistic similarities to north Syrian ceramics. These ceramics include glazed wares with cut designs (noted at Kurban and other sites) and the yellow-glaze family (defined by Watson at Raqqa), as well as common red brittle wares and cream-buff wares, analogous to "white kitchen wares" in Anatolia. All of these types date from the eighth and ninth centuries.
Perhaps as interesting is the pattern of distribution of archaeological sites with this corpus of ceramics. Sites in the vicinity of modern Kahramanmarash, with its ancient citadel, yielded no indication of early Islamic activity, nor did the upper portion of the Aksu and its tributaries. About 20 km south of the modern city is the very large site of Danishman (site 55), which some think is the classical site of Germanikeia. Further south is the `Amq al-Mar'ash, the marshy lowlands similar, on a smaller scale, to the Amuq farther to the south. The Aksu River comes from the east at this point and passes a wide, fertile plain now known as the Narl1 valley. Early Islamic settlements seem concentrated only within this valley, as small farmsteads and late phases of sites on the periphery of the valley. Given the intensive, modern irrigation system and cultivation, only a percentage of the original settlements are still extant (and even these are rapidly disappearing).
No urban center of the Umayyads or Abbasids has been located; yet a remarkable pattern can be seen. The Muslims avoided the classical center and even older town and citadel of Mar'ash and preferred the wide pasturage of plains above marshy ground. A strange parallel seems evident with Qinnasrin, when the Islamic conquerors avoided the classical town of Chalcis and even older urban center of Aleppo (also about 25 km to the north) and selected a new town beside the Matkh marshes of the Qoueiq River. The implication of these very preliminary observations is that the Muslim settlers did not see themselves as crossing a border line into a new region but intended to duplicate a settlement pattern already familiar in Syria and perhaps elsewhere. Of course other lines do exist, such as the one supporting my tent at Kelibishler, which collapsed from the perfidious cow.
Donald S. Whitcomb is Research Associate (Associate Professor) of Islamic Archaeology in the University of Chicago's Department of Near Eastern Languages and Civilizations. He has led excavations at Quseir al-Qadim, Egypt, and Aqaba, Jordan. Don is currently field director of excavations at Hadir Qinnasrin, Syria, which is being excavated by the Institut français d'étude arabes de Damas and the Oriental Institute.
Revised: April 28, 2011