Introduction

The hyper-arid Namib Desert extends approximately 2000km along the south-western coast of Africa, mainly within Namibia, and reaches up to 200km inland (Figure 1). Almost half of the Namib is a vast erg, or sand sea, with some of the highest dunes on earth (Figure 2). The remainder is a complex array of gravel desert, mountain ranges and ephemeral river systems flanked by narrow ribbons of riparian woodland. Archaeological investigations have been carried out intermittently for almost a century, and the mid-Pleistocene to Holocene sequence is well known in outline (Reference MitchellMitchell 2002).

Following Namibian independence from South African mandate in 1990, the pace of development and mineral exploration in the Namib has intensified. With the discovery of several major uranium resources, the impact of mining activity now poses a severe threat to the archaeological record and to the continued existence of important archaeological landscapes. Although the National Heritage Act of 2004 provides for archaeological impact assessment of such developments, there is insufficient capacity to assess, document or mitigate potential damage to archaeological sites. However, the archaeological importance of the Namib is recognised in the mining sector and compliance is largely voluntary.

Setting and approach

Figure 2.

Mixed terrain in the southern Namib.

Among the most remarkable archaeological features of the Namib is the wealth of rock art associated with some of the larger granite mountain areas. Excavations at some of these sites have documented the Holocene archaeological record of hunter-gatherer settlement and the transition to nomadic pastoralism over the last 1000 years (J. Reference KinahanKinahan, 2001). There is also exceptionally well preserved evidence of early colonial contact and coastal trade (J.H.A. Reference KinahanKinahan 2000). The earlier record of human occupation is not as well understood; deep stratified deposits are extremely scarce, and most of the Pleistocene material occurs as surface lag deposits (Reference VogelsangVogelsang 1998).

The project reported here arose as the number of local archaeological assessments for mining projects in the Namib so increased that archaeological survey became continuous over very large areas. Results from one assessment formed the basic predictive model for the next, so improving the precision of field survey. This allows reconstruction of archaeological settlement systems in relative entirety, rather than as partial distributions falling within mining leases, road corridors, or other arbitrary limits. More complete knowledge of archaeological resources in the Namib has made it possible to assess the significance of impacts, and mitigation needs, on an archaeologically realistic spatial scale.

The Namib Desert Archaeological Survey is a project of the Namibia Archaeological Trust (est. 1991) and has so far covered approximately 5000km² of previously unexplored, or superficially explored, ground, adding more than 1000 documented sites to the known record for the country. As a large regional-scale investigation, the survey is possibly unique in Africa, being entirely sustained through archaeological contract work required by mining and other clients for purposes of legal compliance. The survey is also independent of national institutions in Namibia.

We use terrain analysis of cumulative site data and survey design is weighted accordingly. All survey is done on foot, and sites are documented according to standard criteria of type, setting, extent and cultural affinity. Large sites including stone hut remains are surveyed by differential GPS. Site significance and vulnerability are assessed using separate interval scales specially developed for these projects. Survey results are generated in GIS format and as a descriptive report.

Now, by combining the results of more than one hundred piecemeal archaeological surveys, we have widened our focus from client-centered compliance, to a more research-oriented investigation of the Namib region.

Prospects

The extreme aridity of the Namib, and the fact that this ecosystem is almost entirely rainfall dependent, means that archaeological data are also proxy climate data. The project is occupied with detailed reconstruction of human cultural adaptations to the desert environment, but we are also concerned with improving the resolution of the sequence, especially for the last 10 000 years. While the Namib was probably depopulated during the Last Glacial Maximum, intensive re-occupation is most noticeable from the mid-Holocene (Figure 3). Furthermore, the record of occupation varies north and south of 23° S.L. This may be due to the fact that northerly parts of the Namib benefit from southward migration of the Intertropical Convergence Zone, whereas the southern desert is subject to an irregular winter rainfall regime.

Figure 3.

Mid-Holocene site remains in the northern Namib.

Figure 4.

Late pre-colonial glass trade bead assemblage.

Our regional-scale distribution data for the mid-Holocene to Recent period suggest regular episodes of activity in the central and northern Namib, targeting areas with reliable water as well as plant food and animal resources. These are also the areas with the highest concentrations of rock art and the greatest wealth of evidence for complex subsistence technology. The first evidence for the introduction of pottery, metallurgy and domestic livestock to the Namib is associated with these sites, as well as glass trade beads (Figure 4). Rock shelters are being evaluated for stratified sequences (Figure 5), and samples retrieved for radiocarbon dating wherever possible (Figure 6).

Figure 5.

Typical Namib rock shelter test excavation.

Figure 6.

Mid-Holocene to Recent radiocarbon data from sites north (red) and south (green) of 23° S.L.

By contrast, the southern Namib record indicates a lower intensity of occupation with at least one extended hiatus. It is comparatively poor in rock art, and the technological acquisitions that formed the basis of the nomadic pastoral economy were much later in their arrival.