A BALIKH PROSPECT
By Tony J. Wilkinson, Research Associate (Associate Professor)
The Oriental Institute
The University of Chicago
(This article originally appeared in The Oriental Institute News and Notes, No. 147, Fall 1995, and is made available electronically with the permission of the editor.)
To Sir Max Mallowan the Balikh Valley in northern Syria (near present day Raqqa) was something of a backwater, so one may wonder why I spend most of my summers baking in the relentless heat in a bare bleak valley devoid of trees and most wildlife. For example, during the summer of 1994 I was only able to allocate less than one week of my time to the idyllic grandeur of the Taurus Mountains north of Adana, Turkey where Professor Aslihan Yener of the Oriental Institute is conducting research on tin extraction and processing; I spent the remaining six to seven weeks in the Balikh Valley. Despite their obvious differences, however, similarities can be found and here I will attempt to illustrate one theme that links the two areas: that of agricultural technology, and the ability of human communities to harness it to their own advantage.
By agricultural technology, I mean the introduction of some innovation that enables the inhabitants of an area to produce more food from a given piece of land, or to take in new land that could otherwise not be cultivated. Such innovations might include the construction of terraced fields to increase the area of cultivation, digging canals to introduce water to otherwise dry and unproductive land, or the introduction of manuring to improve crop yields and to stem the decline of soil fertility.
The area of the Kestel Mine and the related Early Bronze Age site of Göltepe is nothing less than majestic (fig. 1), however even majesty has its penalties, and in the case of the Kestel/Göltepe area this takes the form of a rather restricted cultivable area. Therefore when this area experienced a minor boom as a result of the mining of tin and its processing in the third millennium b.c., it arguably was incapable of supplying sufficient food from its immediate vicinity to provide the needs of the expanding mining community. Fuel that was necessary for smelting the ore, on the other hand, could have been supplied by woodland that once grew in this moist and relatively cool upland. However, because such resources are finite, it is likely that the huge demands for smelting, firesetting, and domestic purposes would have resulted in the rapid depletion of any tree cover. The task in 1994 was therefore to start an assessment of the area of Kestel in order to determine where the inhabitants might have grown their food and what evidence there might be for former woodland, or for its depletion.
One basic geographical principle relevant to mountainous areas is that when the population rises (for whatever reason) and if people are to remain in the area, it is necessary to expand the area of fields in order to provide food to supply that population. If that is impossible because of the dearth of fertile lowlands for example, then it is necessary to construct fields in order to increase the cultivable area. Thus in order to ensure survival, a technological innovation had to be introduced into the area. Consequently in the region of Kestel it is no surprise that such fields and terraces are abundant. They take a variety of forms, sometimes simply appearing as a patchwork of field boundaries, at other times forming distinct steppe-like benches (fig. 2). Elsewhere, where the amount of "fieldstone" was excessive, the field interiors also include numerous mounds of stones that must have been created during initial field clearance (fig. 3). The distribution of ancient fields and terraces mapped in 1994 is indicated on the perspective drawing generated by John and Peggy Sanders of the Oriental Institute Computer Laboratory, which shows terraces and ancient fields as a gray tone and sites of various dates in black (fig. 4). The distribution of these fields up to elevations of 2,000 meters (around 6,500 feet) is at least as high, if not higher, than present day mechanically-aided cultivation, a fact which testifies to the high demand for food in the past. Although thus far it is only possible to date these fields by their associated sites, our fieldwork is benefiting from techniques developed by myself and McGuire Gibson in Yemen where similar terraced fields are well developed. For example, in highland Yemen our field work shows that test excavations can reveal buried soils or sediments that have accumulated behind the terrace walls. Such deposits can then be dated either by the contained artifacts, or by radiocarbon dating of contained charcoal or by other radiometric methods (for more information see "Oriental Institute Investigations in Yemen 1994," by McG. Gibson and T. J. Wilkinson, Oriental Institute Annual Report 1993-1994 , pp. 62-68). At present, though it is only possible to provide a very general range of time for their use, it seems likely that some will prove to be contemporary with the early Bronze Age mining activity. Probably the best dated examples at present occur near a small Roman site close to Göltepe, where the terraces can be dated by associated pottery scatters to the Roman period. Elsewhere, a final date will only be obtainable if the fields (or stone mounds) are excavated and datable charcoal (or pottery) can be recovered. In general the terraced fields are restricted to the rolling uplands that surround the mining area. They only just overlap on to the higher areas of mountain limestone beyond (see the higher terrain on the perspective illustration fig. 4) and it is possible that at least a portion of this mountainous terrain was once wooded. Indeed some areas to the northeast of Kestel still exhibit a sparse scatter of scrub woodland and in one particular soil section a thin lens of charcoal hints that woodland clearance and associated soil erosion may once have taken place in this area thereby exposing large areas of bare limestone slopes. Although the above observations are incomplete and sketchy, I would argue that the area of Kestel/Göltepe well illustrates how in the past technology was employed to extend the food supply base for the occupants of an important mining area.
Not only is there a total contrast between the scenery of the Balikh Valley and that of Kestel/Göltepe, but also the resources that frame human development are different as well. Therefore in the valley, apart from the world weary and somewhat apathetic sheep, it is difficult to recognize natural resources. There is, it is true, an extensive area of potentially fertile soil, but this is rather limited by available water: rainfall is only sufficient to support crops in the northern part of the valley (located mainly in present day Turkey). On the other hand the meager flow of the Balikh River cannot be stretched infinitely to supply irrigation water.
Our work in the Balikh Valley is being conducted jointly with a team directed by Dr. Peter Akkermans of the Netherlands National Museum at Leiden. We are benefiting enormously from the expertise of Peter, who is director of excavations at Tell Sabi Abyad (located in the Balikh near Tell Hammam et-Turkman) and who undertook the main archaeological survey of the valley in 1983. Fieldwork is also made easier by the work of Dutch soil surveyors who mapped the area's natural resources a number of years ago, and whose data base provides a wealth of information for archaeological research. In practical terms, in addition to sponsorship from the Oriental Institute, we have received funding from the National Geographic Society who have provided an enduring interest in that academic borderland between geography and archaeology. All of these previous studies have supplied us with essential comparative data. Rather than resurvey the area, we are examining aerial photographs and other resources for the ancient infrastructure of the region: traces of early canals, cultivation, roads, quarries, as well as changing water resources that can help in the elucidation of landscape changes over the past ten thousand years or so. Although the area is poorly endowed in terms of mineral resources, the agricultural potential could be progressively extended by the introduction of technology, primarily water engineering, designed to tap the modest flow of the Balikh River.
Our 1994 field season concentrated on a number of interrelated issues, but by chance (and I must admit that opportunism does occasionally raise its seductive head in archaeology) we encountered a number of deep trenches that, geomorphologically-speaking, satisfied our wilder cravings.
During the season we aimed to excavate several sections across a rather grand canal that might have been an Old Babylonian (early second millennium) feature that had once supplied the inhabitants of Tell Hammam et-Turkman (fig. 5) with water. By so doing this canal would then have deprived the people of Tuttul (Tell Bi`a at the confluence of the Balikh with the Euphrates) of irrigation water. Archaeologists often posit dates on the basis of a few scattered and undistinguished sherds, and we are glad to say that our persistence, and the excavation skills of the site supervisor Fokke Gerritsen (a former student in the Department of Near Eastern Languages and Civilizations at the University of Chicago), resulted in a rather well-dated sequence for this canal. These results complemented the rather less successful achievements of our 1993 drilling program, which although providing evidence for a diverted course of the Balikh, proved to be rather frustrating to conduct in the field (fig. 6, fig. 7, and fig. 8). Thus excavation within the canal (fig. 9) showed that the canal had slowly filled in during the Islamic and Byzantine periods. At the base of the sedimentary sequence, where a shelly gravel deposit eloquently testified to the former rather vigorous flow of the canal, the pottery was Parthian-Hellenistic (referred to by modern wisdom as Hello-Parthian), which instantly killed the romantic assumption of an Old Babylonian canal. The late date, which was confirmed by a fifth-sixth century radiocarbon date for the final flow phases of the canal, although rather a surprise, turned out to be entirely logical in terms of the history of settlement of the region, which included a well-developed phase of Hellenistic settlement. This settlement phase was contemporaneous with the construction of the canal, and presumably must have benefited from the introduction of irrigation waters.
The crucial data derived from a section across the Balikh River, exposed some twenty kilometers downstream by engineers who were constructing a new canal from Lake Tabqa, which showed a broad channel cut into the alluvium of the Balikh adjacent to the substantial Bronze Age pile of Tell es-Seman (fig. 10). Here rather than being a true canal, detailed leveling by our surveyors Eleanor Barbanes (of the Department of Near Eastern Studies, University of California-Berkeley) and Gregg Munson (a student in the Department of Near Eastern Languages and Civilizations at the University of Chicago) showed that the channel base extended below the river level. This and other un-canal-like features therefore suggested that this channel was in fact an early course of the Balikh rather than an artificial canal.
Of particular interest is that the sediments contained within this channel demonstrated that the river must have flowed with some vigor in the Bronze Age, was sluggish (but flowing) in the second and early first millennium b.c., and was essentially dry by the Hellenistic period. In other words, just as the new canal (i.e., our excavated canal upstream) came into use the Balikh River became dry. Of course this was not necessarily a sudden event because the gradual decline of flow that occurred during the second and first millennium b.c. can be matched by a gradual increase in the number of sites in the region. Because this increase appears to indicate a gradual increase in population it follows that there must have been a steady increase in the demand for irrigation water through the first millennium b.c. Similarly, earlier fieldwork had shown that around the sixth to eighth centuries a.d. extensive new canal systems were dug to supply the growing rural and urban populations that developed when Raqqa was achieving its prime in the early Islamic period. In other words, it seems that as population increased from a sparse scatter of small towns in the Early Bronze Age through a vigorous phase of Neo-Assyrian rural settlement, the flow of the Balikh was gradually reduced by localized abstraction for irrigation. In Hellenistic times, the major canal was excavated, and this resulted in a new rash of settlements within the irrigated zone and the final drying out of the Balikh River. Then as settlement increased still further, a new and even longer canal network was introduced, thus severing the flow of the Hellenistic canal. We can therefore conclude from the 1994 field results that in this area an apparently natural increase in population necessitated the progressive development of water extraction technology that resulted in canals of increasing scale through time.
Therefore in areas that are so vastly different as the Taurus Mountains and the semiarid steppe of northern Syria, archaeology can provide a similar perspective, in this case that the introduction of agricultural innovations can increase food production often in the face of a reluctant environment. Although the innovations are very different in each area, it is evident that increasing population and demand for food, or growing affluence of the inhabitants, can make such innovations worthwhile.
After training in Canada 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 Iraq, Baghdad. He joined the Oriental Institute as a Research Associate in 1992.
Revised: February 7, 2007