Beyond the Drip-Line: A High-Resolution Open-Air Holocene Hunter-Gatherer Sequence from Highland Lesotho

Article excerpt



Rockshelters loom large in the archaeology of hunter-gatherers worldwide--and southern Africa is no exception. Highly visible and easy to locate, they typically preserve well-stratified deposits rich in material culture and organic remains. Yet they capture only a fraction of people's activities, most of which took place beyond the drip-line. Landscape-oriented research (e.g. Sampson 1985) confirms that at many times and in many places people camped in the open, just like recent Kalahari Bushmen. Away from the coast, however, where shell middens provide obvious contexts for archaeological investigation (e.g. Jerardino & Yates 1997), few open-air sites of Holocene age have attracted serious excavation. In areas like the Karoo, the semi-arid interior of western South Africa, they are often deflated, with bone and macroplants poorly preserved, while elsewhere dense vegetation may render their discovery difficult. Rare exceptions have received only limited investigation.

This paper summarises work at an open-air site that, unusually, combines excellent faunal preservation, high quality spatial patterning and good temporal resolution over multiple occupation episodes. Likoaeng is significant for what ir tells us about hunter- gatherers in the Maloti-Drakensberg mountains of Lesotho, bur also because ir addresses wider debates, including the exploitation of aquatic resources (freshwater fish), shifts in seasonal focus linked to climate change, and forager acquisition of domestic livestock. Such debates have broad relevance since issues of intensification using resources of the kind that Hayden (1990) terms r-selected, the impact of global climatic pulses like the late Holocene Neoglacial (e.g. Jerardino 1995), and the relations between hunter-gatherers and farmers (Spielmann & Eder 1994) are topics far from unique to southern Africa. For Bushman rock art the critical role of ethnohistoric observations of painted sites in the very part of the Maloti- Drakensberg mountains where Likoaeng is located (Lewis-Williams 2003) also means that archaeological excavations there can help establish something of the historical dynamics of the region's late Holocene forager societies and thus the reliability with which late nineteenth-century comments can, or should, be generalised across time and space.

Stratigraphy and sequence

Likoaeng (29[degrees]44'08"S, 28[degrees]45'47"E; 1725m asl) lies in Lesotho's eastern highlands at a confluence where a small stream joins the Senqu (Orange) River (Figures 1 & 2, upper). The site was discovered as a result of a flood event that had cut through the deposit and left ir open to erosion: no artefacts were visible on the surface. Excavations in 1995 and 1998 proceeded stratigraphically to a depth of 4.5m, decreasing from an original 30[m.sup.2] to a smaller 3.5[m.sup.2] trench (Figures 2, lower & 3). Spatial control was maintained by employing 0.25[m.sup.2] quadrats and three-dimensionally recording significant finds; all sediment (except that from culturally sterile layers and the 1995 component of Layer I) was sieved through a 2mm mesh.

Likoaeng's stratigraphy is relatively straightforward (Figure 4), consisting of occupation levels separated by episodes of flooding from the river. The lie of the layers, sloping down from west to east, suggests they were part of the talus of a now buried rockshelter ar the far western edge of the excavated area. Several of the later occupation strata (e.g. Layers III, V and VII/IX) were clearly laid down very rapidly, facilitating in situ preservation of bone, features and the spatial patterning of human activities. Combined with clear stratigraphic alternation of occupation and non-occupation levels, 18 radiocarbon determinations make Likoaeng one of southern Africa's most precisely resolved late Holocene hunter- gatherer sequences (Table 1). When calibrated (McCormac et al. …