Heritage Cases

In October 2018, we conducted fieldwork at the site of Olieboomspoort Shelter, in the Waterberg (Limpopo Province, South Africa). The project focuses on the Middle Stone Age (MSA) layers from the site and one of the key aims is to provide a better chronological framework for these layers. The only existing chronological indication is an age greater than 36 ka BP (Mason, 1962) provided by the radiocarbon technique in the 1950’s (therefore, beyond the range of the method). In October, we have re-opened the main excavation surface, which upper part (Later Stone Age) was previously excavated (van der Ryst, 2007). Inside, we have started a small excavation (2 m2) directly into the MSA layers. Faunal preservation proved to be relatively good and faunal remains include dental material. We have sampled two equid (Equus sp.) teeth for Electron Spin Resonance (ESR) dating. Here, we are applying for an export permit to send these two teeth to one of the world leading research laboratories in ESR dating (Australian Research Centre For Human Evolution, Environmental Futures Research Institute, Griffith University, Australia), in order for them to be dated by two of our scientific collaborators on the project, Professor Rainer Grün and Dr Mathieu Duval. This project focuses on one of the few large rock shelters in the Waterberg, Limpopo, preserving rich archaeological deposits: Olieboomspoort Shelter. R. Mason was the first to excavate the site in the 1950’s, revealing a ~2 meter deep sequence preserving Later, Middle, and Early Stone Age layers (Mason, 1962). Subsequent excavations were undertaken by M. van der Ryst in 1997 (van der Ryst, 2007) and focused exclusively on the Later Stone Age (LSA) units. While the latter have been well studied and well dated, very little is known on the Middle Stone Age (MSA) layers, beside an age greater than 33 ka (since falling beyond the range of radiocarbon dating) and the attribution of the lithics they yielded to the poorly known ‘Pietersburg’ techno-complex. This multi-disciplinary project, involving various specialists (of lithics, faunal and archaeobotanical remains, geoarchaeology, and dating), as well as south African students, aims at: (1) obtaining a new, finely excavated sample of archaeological material from the MSA layers, (2) providing a better chronological framework those layers, and (3) investigating the formation processes at play in the accumulation of the archaeological deposits. In southern Africa, the vast majority of MSA sites where signs of emergence of modern behaviours amongst Homo sapiens have been identified are located along or very close to the coastal lines of the Atlantic and Indian oceans (e.g., Diepkloof Rock Shelter, Pinnacle Point, Blombos Cave, Klasies River Mouth and Sibudu Cave). East Africa is another key-region where multiple lines of evidence for the development of ‘complex’ human behaviours have been found at several sites. Relationships and interactions between these two regions are, however, poorly understood, as very few sites in between are known and/or have undergone modern excavations and benefitted from modern dating techniques. Field work conducted in the past three years by myself and others (Porraz et al., 2015, 2018; de la Peña et al., in press) at the sites of Bushman Rock Shelter and Mwulu’s Cave in Limpopo, constitute several attempts to compensate for that lack of knowledge regarding the nature of MSA behavioural traits in the interior of southern Africa generally, and in the East-to-South Africa corridor particularly. This recent work has brought back to life an almost forgotten techno-complex, initially defined by R. Mason using the lithic implements recovered from Cave of Hearths, in Limpopo: the ‘Pietersburg’. While, at the time, more than a dozen sites were considered to contain Pietersburg artefacts – including notably Border Cave, Cave of Hearths, Mwulu’s Cave, Bushman Rock Shelter and Olieboomspoort –, this techno-complex suffered from the limits posed by radiocarbon dating and could not be placed precisely within the southern African MSA. Furthermore, unlike the Still Bay and the Howiesons Poort, the Pietersburg is lacking lithic “fossiles directeurs” and was defined only on the basis on varying proportions of tool types rather than thorough technological descriptions (Mason, 1962; Sampson, 1974). This eventually led the Pietersburg to fall into oblivion after the 1970’s and to be excluded from syntheses on the southern African MSA. Olieboomspoort, located in the Waterberg area, Limpopo, about 10 km south of Lephalale, is one of the few sites containing evidence for intermittent human occupation from as far as the Early and Middle Stone Age (chronology unknown) and up until the end of the Later Stone Age (last 2000 years). R. Mason’s excavation of the MSA layers yielded a rich lithic assemblage that he attributed to the Pietersburg industry, alongside one of the richest ochre assemblage known for any MSA sites in southern Africa (Watts, 1997). While MSA techno-complexes such as the Still Bay and the Howiesons Poort are now well dated (e.g. Jacobs et al., 2008), the only ages available for the Pietersburg come from a single site (i.e., Border Cave: Grün et al., 2003). Thus, one of the key underlining goals of the recent projects undertaken in Limpopo (Bushman Rock Shelter and Mwulu’s Cave), including Olieboomspoort, is to refine the Pietersburg’s chronology. The second main aspect of this project relates to the nature of archaeological deposits in southern African. Unlike other regions of the world, southern African rock shelters tend to contain well preserved, long archaeological sequences, composed of multiple layers, resulting from a combination of anthropogenic activities (e.g., combustion, knapping, butchery, beddings, ash dumping, site maintenance, etc), geogenic processes and bioturbation. Classical excavations tend to focus on the description of long sequences, which require “vertical” digs. Given the excavation methods used today, progress is very slow (a few cm of sediments excavated per day, generally less than half a meter excavated by campaign), forcing the archaeologists to work on a limited area - often only a couple of square meters. Consequently, questions related to spatial distribution of activities within the shelter, nature of the relationship between the different stratigraphic units observed (e.g., contemporaneous or chronologically distinct? linked to post-depositional processes?), and, broadly speaking, to the formation processes involved in the accumulation of archaeological deposits cannot be addressed.