Researcher: Dr Huw Barton

The 1950s excavations by Charles McBurney in the great Haua Fteah cave, situated on the northern shore of the Gebel Akhdar ('Green Mountain') in northeast Libya, revealed a deep (14m) sequence of human occupation from the Graeco-Roman to Middle Palaeolithic periods. As a result, it is commonly recognised as the most important prehistoric site in North Africa. Occupation in the cave may well go back 200,000 years.

In 2007 a renewed programme of archaeological and geomorphological investigation began (The Cyrenaica Prehistory Project), directed by Professor Graeme Barker (University of Cambridge), with the objective of improving understanding of the cave's occupation sequence and, combined with fieldwork in the landscape, of the history of landscape change and human responses to it. This research project involves renewed archaeological investigation within the Haua as well as extensive surface survey of the archaeology and geomorphology outside the cave. Staff from the University of Leicester are involved in that aspect of the project as well as the future functional analysis (usewear and residues) of stone tools recovered from the Haua stratigraphy.

Geoarchaeological Survey

The main aims of the geoarchaeological survey conducted in 2009 were to characterise the archaeology and geomorphology of three main landscape regions - Gjebel, pre-desert, and desert - and to establish new baseline data before conducting a large-scale systematic investigation of the offsite archaeology of Cyrenaica. The ultimate objective of this study is to correlate the Palaeolithic surface archaeology from these landscape regions with the deeply stratified chronology of the Haua Fteah.

While the cave is a remarkable temporal record of the archaeology in Cyrenaica it lacks the larger spatial dimension within which its users would have hunted and gathered throughout the Pleistocene and Holocene periods. Through a detailed technological analysis of stone tools within and outside the cave, coupled with palaoenvironmental reconstructions, we seek to address human adaptive strategies during several key periods of modern human evolution over the last 200,000 years.

Archaeological survey recovered evidence of human occupation in every part of the landscape. There was a distinct predominance of MSA materials in the pre-desert and desert regions, and weathering patterns indicate that we may have evidence for early and late MSA here. Typical artefacts include large bifacially reduced cores, disc cores and radial cores manufactured on locally available chert cobbles. Levallois elements are also widespread but not common.

One site, CPP-8009, contains a dense scatter of knapping debris that is largely related to the production of levallois flakes from a local chert source. There is also some evidence for later use of earlier flaked material by people producing small blades. A few small bifaces were recovered but there was no evidence for the manufacture of handaxes in any of the regions surveyed in 2009.

As might be expected, palaeolakes formed definite foci for multiple periods of human occupation and while no good exposures suitable for excavation were discovered in this season, this will be a continued aim of future survey. Lithic scatters recovered from the alluvial fans and mid-reaches of the drainage channels in the pre-desert were all dominated by MSA assemblages.

The weathering of artefacts suggests long-term or periodic exposure but not for the fluvial transport of material along within the wadi. It seems most likely that wadi gravels provided a source of flakeable chert for middle Pleistocene populations of hominins. The source of these cherts is widespread throughout the entire survey region anywhere that bedrock limestone is found outcropping. Petrographic analysis may help define local sources and possibly identify broad patterns of raw material transport; however, the sheer availability of chert may complicate this.

Clear evidence for the LSA in the pre-desert and desert was rare. Only three sites produced artefacts characteristic of this time period. Two of these contained blakes typical of Epi-palaeolithic assemblages from the Fezzan and the Oranian within the Haua. One site in the extreme south of the study area produced a single bifacial flaked piece that is characteristic of the Neolithic. While several sites were located with relatively large scatters of stone fireplaces (often >20); few stone artefacts were recovered, and none characteristic of any time period, ie. no bifacial arrow heads, no ground stone, and no egg shell beads.

• The 2008 fieldwork was funded by the Society for Libyan Studies and the Leakey Foundation, and the project is also grateful to the NERC/AHRC ORADS committee for funding to date a suite of radiocarbon samples at the Oxford Radiocarbon Laboratory.