ORIENTAL INSTITUTE DHAMAR PROJECT
NOTES ON THE 1996 FIELD SEASON
By T. J. Wilkinson, 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. 154, Summer 1997, and is made available electronically with the permission of the editor.)
The Oriental Institute Dhamar Project completed its third field season by Thanksgiving 1996. Although all seasons have been brief-around five weeks per season-the results have been remarkably productive. One does not often have the opportunity to go to a virtually pristine archaeological region and build up a cultural and environmental sequence for thousands of years. Prior to our fieldwork, the high plains of Yemen were largely unknown but for the remains of the Himyarite state, which enjoyed control of the region between approximately the first century BC and the sixth century ad. This period was the subject of the earlier campaigns initiated by Professor McGuire Gibson during the 1970s and 1980s, when Ray Tindel (now Registrar of the Oriental Institute Museum) was effectively resident curator of the museum at Zafar, a small village that is all that remains today of the capital of the Himyarite state.
Our objective during the field campaigns of the 1990s was to build upon Tindel's archaeological and historical foundations concerning the Himyarite heartland and extend them chronologically back to the origins of complex states in the highlands. Furthermore we attempted to provide an environmental framework in order to demonstrate under what conditions early terraced agriculture was developed. Thus we wanted to know not only who the first "proto-Himyarites" were, but also what type of environment they experienced and how and when they managed to develop a viable agricultural economy. Yemen, being in southwest Arabia, is within the area that has provided compelling evidence in the form of relict lakes for a period of moist, verdant conditions between 9,000 and 6,000 years ago. Little however was known of conditions on the high plateau, and it was therefore necessary to examine numerous stratigraphic sections to develop a full natural stratigraphic sequence in parallel to the cultural sequence (see "Investigations in Yemen: Progress Report" in The Oriental Institute Annual Report 1995-1996).
We must thank a number of private contributors to the Oriental Institute for making the 1996 campaign possible. Without such donations fieldwork would have been impossible. We are also grateful to the American Institute for Yemeni Studies (AIYS) which provided, through its offices and hostel in Sanªa and through the expert guidance of its director, Dr. Noha Sadek, an administrative foundation for the ongoing fieldwork. The AIYS must also be thanked for providing a grant that covered the expenses of Chris Edens and his excavations, which provided a major contribution to the season's success. In addition, we should thank the General Organization of Antiquities and Museums, Sanªa, and its director, Dr. Yusuf Abdullah, for speedily providing permits and administrative assistance for fieldwork.
Early Towns and Terraced Agriculture
Probably the most exciting aspect of our earlier field seasons was the discovery of a number of large sites of nominally Early Bronze Age date. I say nominally because our dating evidence was by ceramics alone. Such sites were dated by reference to small sites surveyed to the northeast of our region by a team of Italian archaeologists under the directorship of Dr. Alessandro de Maigret. Their discoveries, during the early 1980s, marked a turning point in the archaeology of Yemen because they were the first to recognize a coherent archaeological signature for this period in southwest Arabia. However, their sites-in the semiarid mountain valleys to the southeast of the capital Sanªa-were small rather straggling agglomerations of sub-rectangular houses and compounds that were little more than villages.
In contrast, the settlements that our surveys were documenting on the relatively verdant plateau were large coherent settlements of rectangular houses, some streets, and even, in the most noteworthy case, an outer defensive wall. Such defenses were of course supplementary, Yemenis through the ages being more likely to adopt hilltops for occupation for reasons of both defense and status.
These sites, although yielding sufficient pottery to furnish a Bronze Age date for their occupation, were not dated absolutely. Therefore, building on the foundations laid by Professor McGuire Gibson during the 1995 field season, we decided to initiate small-scale excavations in order to define the architecture and stratigraphy of two sites and to obtain pottery in good stratified contexts in association with charcoal. This charcoal could then be subject to radiocarbon assay, thereby providing an "absolute" date for the sites in radiocarbon years before present ("BP"). Excavations were undertaken by Christopher Edens, an expert in Arabian prehistory, who with his wife (surveyor Julie Edens), took on two sites, Hamat al-Qa (site DS 101) and al-Sibal (DS 66) for detailed investigation. Of these al-Sibal had already provided two radiocarbon dates as a result of Professor Gibson's investigations in 1995, but we felt that such a chronological picture required strengthening, not only by more dates, but also by an enlarged pottery collection.
Hamat al-Qa was partially mapped, and small soundings were placed in representative parts of the site. In villages of this part of Yemen, public relations is important. By public relations I do not mean reception desks, attended by cologne-scented, power-dressed receptionists; rather I mean that labor should be carefully selected so that an equitable number of people from each affiliated village, family, or tribal subdivisions is selected. With the help of our representatives Khalid al-Ansi and Ali Sanabani, as well a number of influential local sheiks, we managed to progressively iron out labor problems and rationalize our work force into functional units that provided not only valuable archaeological results but also some excellent colleagues.
One of the advantages of radiocarbon dating is that in areas where pottery sequences are not well known, the results can erode preconceptions. In this case the site that we thought was probably of distinctly third millennium BC date-Hamat al-Qa-proved to comprise occupations dating between the final quarter of the third millennium (Early Bronze Age IV in the Palestinian chronology) and the first half of the second millennium BC. Al-Sibal, on the other hand, has now provided dates ranging from the mid-third millennium BC to roughly the third quarter of the second millennium. Although it is clear that more radiocarbon determinations are required, it seems that urbanization on the high plateau may not be contemporary with the communities located by de Maigret in the semiarid regions to the northeast. Rather they appear to follow on from that phase of occupation that was dated to around the middle of the third millennium BC. However, because there are remarkable ceramic similarities between the Khawlan groups and those of Hamat al-Qa and related sites, it seems that we are dealing with the same ceramic tradition, but one which shifted, or became focused on the plateau, rather later in the third millennium BC.
Such sites were not investigated in isolation but were shown to form part of a remarkably dense scatter of Bronze Age settlements on the plateau. It was even possible to recognize places subordinate to Hamat al-Qa. These provided identical pottery but were rather smaller and again were primarily defensive. Agriculture in the form of relict systems of terraced fields was also evident. The faint traces of such fields clung to the hillside below both Hamat al-Qa and its subordinate satellite settlement, Hait al-Ahmar (DS 192). The area can even boast industry in the form of obsidian extraction sites. Although the region has long been known as an obsidian source, in 1996 for the first time we discovered (of course on a mountain top) a Bronze Age site replete with obsidian-working debris. Thus it seems that the inhabitants were gathering chunks of obsidian from the base of hill slopes, carrying them to specialized hilltop sites, and working them by striking blades and flakes from carefully prepared cores. The relation between the Bronze Age occupations and obsidian tool preparation are therefore unequivocal.
Iron Age Towns
I do not wish to give the impression that all of our efforts in 1996 were expended on Bronze Age sites. It is true, however, that at times they seemed difficult to avoid. Nevertheless, major advances were made in our studies of the pre-Himyarite "Iron Age" settlements, namely those that are loosely contemporaneous with the Sabaean kingdoms of the desert fringes to the north. In the previous season we had defined this culture by reference to its distinct pottery assemblage found on numerous large hilltop and valley bottom sites. Its chronological status has been approximated by radiocarbon determinations on samples obtained from two of Professor Gibson's 1995 soundings. These confirmed a first millennium BC date for such sites. The 1996 season not only provided more sites, but also two that were equipped with outer defensive walls. Such settlements, with their massive defensive walls of roughly-dressed stones, monumental gateways, and square interval towers, although known in other parts of Yemen, have hitherto not been recorded from the high plateau. Two walled sites are now known, and both have been mapped at a general level. From their surface pottery, both sites can be shown to belong to both the Iron Age and the Himyarite periods, and our preliminary investigations suggest that there may well have been continuity of occupation between the late Iron Age and Himyarite periods.
That the foundation of Himyarite towns lie within the Iron Age is also supported by a deep stratigraphic sequence recorded from the Himyarite town of Ribat Amran. There, deep excavations showed that prior to the Himyarite occupation, earlier occupations were present in stratigraphic sequence well back into the Iron Age (ca. mid-first millennium BC).
With regard to agriculture, our environmental investigations are providing abundant supporting data for the agricultural systems that were associated with the early towns of the Yemen plateau. In addition to the terraced fields associated with the Bronze Age sites, we now have radiocarbon dates associated with soils and terraced walls of the Iron Age (between 400 and 100 BC). These determinations show that agriculture on the valley floors was well under way by the Iron Age. Furthermore, a radiocarbon-dated soil behind a large valley floor terrace wall near the head of the Wadi Shalalah indicates that valley floor terracing was in existence between about 400 and 100 BC.
Dhawran: A former capital of a unified YemenAfter the decline of Himyar in the late seventeenth century ad, the leader of the Qasimis, Mutawakkil al-Ismaªil (1644-76) moved his capital to Dhawran to the east of Ma'bar where it was the capital of all Yemen for a brief period. This capital underwent a decline in the eighteenth and nineteenth centuries and dwindled into oblivion in the twentieth century, partly as a result of the 1982 earthquake. One of our first objectives upon arrival in our headquarters at Dhamar (in fact before we had unpacked our pots, pans, and smoking jackets) was to investigate this former capital that, according to rumor, clung to a hillside one hour's drive to the northwest of Dhamar. Here was a large town, now entirely in ruins, dominated by an extraordinary mosque, ablution, and mausoleum complex (see illustration). The recording of this major monument will therefore provide one of the multitude of tasks allocated for the forthcoming 1998 field season.
To conclude, as a result of our investigations we can now confidently say that the Yemen highlands were well populated back to the third millennium BC and the inhabitants were living in towns as early as 2000 BC. Although many more campaigns are needed before we can confidently demonstrate that there has been continuity of occupation throughout the last four or five millennia, this seems to have been extremely likely from existing field evidence. Thus we now feel that the Oriental Institute can claim to have made a significant step in defining the development of civilization in Arabia by discovering and dating what may well be the earliest true towns of the Arabian peninsula.
After training in high Arctic geomorphology, T. J. Wilkinson became interested in the archaeology of the Near East. He worked as a free-lance archaeological consultant for several years and became Assistant Director of the British Archaeological Expedition to Baghdad, Iraq. He has been Co-Director of the Dhamar Project with Professor McGuire Gibson since 1990 and joined the Oriental Institute as a Research Associate in 1992.
Revised: February 7, 2007