[Press & Publications] Lost Cities of the Sahara [Archaeology]



A team of archaeologists and geographers, directed by Professor David Mattingly of the School of Archaeological Studies at the University, has just returned from Libya, where deep in the Sahara they have discovered new evidence about the Garamantes - a mysterious desert people of Greco-Roman date (broadly 500 BC AD 500).

The work of the project is funded by the London-based Society for Libyan Studies, the British Academy and the universities of Leicester, Newcastle and Reading, with additional support from the oil exploration company LASMO GML, within whose concession area the fieldwork takes place.

The earlier work of another British archaeologist, Charles Daniels, is being prepared for publication in parallel with the new studies, thanks to a major grant from the Leverhulme Trust, with researchers in post at the universities of Leicester and Newcastle.

Inhabiting a region that had already been for several thousand years a hyper-arid desert environment, with negligible rainfall, elevated summer temperatures and blistering expanses of barren sand and rock, the Garamantes have long been an enigma.

They were depicted by Roman sources as ungovernable nomadic barbarians, who raided the settled agricultural zone and cities of the Mediterranean littoral. Following up earlier work by Daniels, the current project allows a different picture of the Garamantes to be drawn. Archaeological evidence shows them to have been a complex and urbanised society, with a strong emphasis on oasis agriculture a picture far removed from the shiftless nomads of our ancient sources.

Professor Mattingly said: We have recorded several urban centres, numerous villages and extraordinarily dense cemeteries in the Garamantian heartlands. They had a rich material culture, incorporating many aspects from the Mediterranean and Egyptian worlds, as well as some elements from sub-Saharan Africa. I see them as a lost civilization one of the earliest Saharan states and it is clear that they were amazingly influential in a series of developments in the central Sahara during Classical Antiquity.

Our archaeological data point to them being responsible for the evolution of urbanism in this region, for the introduction of writing, the horse, the camel and wheeled transport, for the development of sophisticated irrigated agriculture, and for the initiation of long-range trans-Saharan trade routes. One of the keys to their success was their ability to tap subterranean water sources to irrigate their fields. We have mapped thousands of kilometres of underground water channels (similar to the Persian qanats and known locally as foggaras).

The Anglo-Libyan team has been carrying out excavations at Germa in the Libyan Fezzan, long known to be the site of the Garamantian capital, Garama. The excavations have confirmed the extraordinary status of that location and allowed the Garamantian heyday to be set in the broader context of changing (and generally poorer) conditions in the post-Garamantian caravan town that overlay it, and which endured until the 1930s.

One of the most important discoveries of the most recent campaign of excavations has been the identification of what appears to be an early mosque of the Islamic phase of the site. Another exciting discovery has been the retrieval of fragments of Roman tile from Garamantian levels.

Mattingly comments, We've only recovered a few fragments so far from a late Garamantian dumping level, but the material unmistakably comprises hypocaust tiles and flue tiles. The implications are amazing: someone has carried this material 1000 km across the Sahara to construct a Roman style bath building presumably a special commission for the Garamantian king. This would be by some margin the most southerly Roman bath ever found if we're lucky enough to find the building itself in our final season next year.

In addition, the project has been surveying the Garamantian heartlands in the Wadi al-Hayat area of southern Libya, effectively a narrow band of oasis agriculture circa 150 km long, by 2-3 km wide, sandwiched between a vast sand sea and a barren rock plateau. Among the most extraordinary discoveries has been the relocation and planning of two 'lost' cities of the Garamantes and numerous smaller settlements. One of these sites, Gasr esc-Scheraba, previously recorded as an Islamic period site, was incorrectly mapped about 60 years ago and the mistake has been repeated on every subsequent map of the region.

David Mattingly explains, When we finally relocated it we found it was over 20 km away from its mapped position! The maps show its location as being in the edge of the sand sea, which has never made sense for a site of this size it is well over 50 ha in area and shows signs of regular planning. Once mapped in its correct position, the significance of the site becomes much clearer. It lies astride one of the main east-west communications lines where it meets a north-south route linking the Garamantian capital with this area.

The surface pottery shows that the site originated in the Garamantian period and continued into the early Islamic phase. The project made another important discovery in the Wadi al-Hayat to the east of Germa, when they located another Garamantian urban site surrounded by a substantial defensive wall, with external towers. Once again, the densely packed interior of the site showed signs of careful planning of the urban space.

In addition to the larger urban sites, the project has recorded numerous smaller nucleated settlements or villages. At a site called Saniat Gebril close to Germa, Dr John Hawthorne, the Newcastle Leverhulme researcher, has made a special study of the industrial activity. To judge from the material excavated by Daniels in the 1970s and recovered by further surface collection this year, the site was at various times involved in large-scale jewellery manufacture (principally ostrich eggshell and carnelian stone beads), metal-working and textile production. Although there is lots of evidence of imports from the Mediterranean world, especially in Garamantian tombs, the craft activity at Sanait Gebril shows that the Garamantes maintained their own traditions of material culture alongside.

Another important aspect of the work has been the recording of the many Garamantian cemetery sites, which reveal an extraordinarily varied mortuary tradition. There are over 100,000 visible burials along the escarpment slope that fringes the southern side of the Wadi al-Hayat and other less visible cemeteries existed in the centre of the valley. Dr David Edwards, the Leverhulme Research Fellow working at Leicester on the site gazetteer, explains, There is a great variety of funerary monuments, reflecting the different contacts of the Garamantes. For instance, the Garamantes incorporated both Greco-Roman style mausolea and pyramid cemeteries, perhaps inspired by Garamantian knowledge of peoples to the east and south-east. The pyramids are mostly quite small scale, generally between 2 and 4 m high, but some of the cemeteries contain over 100 monuments. Although many burials have been robbed, some of the tombs are still well-preserved often with offering tables and stele shaped like hands placed on their east side.

The project has also assembled the first comprehensive pottery typology for the region, spanning the late Neolithic to the relatively recent past. John Dore, the Newcastle co-ordinator of the project, comments, We've examined 10,000s of sherds, putting together the evidence from older excavations which dealt primarily with Garamantian levels, with the newer work which has focused more on the medieval levels, and also integrating the material collected by surface sherding at many sites. The surface collections fill in some gaps and also allow us to start to pick out regional differentiation in pottery supply within the Garamantian zone. Conventional ceramic dates are being supplemented by radiocarbon dating of some sites and excavated contexts, in an attempt to peg the relative chronology more closely to absolute dates.

With over 500 Garamantian sites now recorded, and many susceptible to dating for the first time, a reappraisal of this early Libyan state can be made on the basis of concrete evidence. The picture that emerges is of a powerful Saharan polity, employing a wide range of material culture and architectural styles to reinforce a pronounced social hierarchy. Faunal analysis shows that there were animals in the diet, notably sheep/goat, but it is clear that pastoralism lagged far behind sedentary agriculture in this desert kingdom. The main question the researchers are now grappling with is when and why did Garamantian civilization come to an end. According to David Mattingly it is possible that there may be a lesson in it for the present, We think that the very scale of Garamantian irrigated agriculture may have had a long-term impact on the aquifer they were tapping into. If the water level fell below a certain point, their foggara systems would have ceased to operate efficiently and necessitated a switch in irrigation technique. Certainly by the later middle ages all the foggaras appear to have been abandoned in favour of small-scale garden cultivation based on wells. The cultivated area clearly shrank to a fraction of its maximum extent in the Garamantian period. So the Garamantes made the desert bloom, but through their over-exploitation of the non-renewable groundwater sources they may have contributed to their own downfall.

The work of the project can be followed on our website at http://museums.ncl.ac.uk/garamantes/feztop.htm


[Leicester University] [*] Administration [*] Press & Publications
Information supplied by: Barbara Whiteman
Last updated: 26 July 2000 11:57
University Administration Web Maintainer

This document has been approved by the head of department or section.
If you are an authorised user you may edit this document through your Web browser.