Heritage Cases

This application seeks support for radiocarbon dates to refine the chronology of Buffelskloof and Kangkara rock-shelters, both of which are key sites in the formulation of the southern African Later Stone Age (LSA) sequences (J. Deacon 1984). They are thus critical for broader technological comparisons across the sub-continent and for understanding change and variability within this sequence. Yet, the deposits at both sites were last dated in the 1970’s and 80’s, a fact that substantially undermines the importance of their lithic records for interpretations of broader technological changes during the LSA in southern Africa. Thus, an improved chronology for Buffelskloof and Kangkara has implications for wider interpretations of In attempting to establish a framework for technological and typological change for southern Africa’s Stone Age, Lombard et al. (2012) have drawn attention to the inadequacy of many individual site chronologies. Several of these chronologies are decades old and precede the introduction of methodological advances that have transformed the field of radiocarbon dating (AMS dating, rigorous pre-treatment strategies, improved calibration curves). The poor quality of many radiocarbon dates (evident in their large errors) puts inter-site technological comparisons on an uncertain footing. The southern African Middle Stone Age (MSA) has recently witnessed a florescence of chronological research (using techniques such as OSL and thermoluminescence), particularly aimed at investigating the concurrence in technological change between sites and the relationship of these technological changes to regional and global climate shifts (e.g. Jacobs et al. 2008). There has, however, been little new research directed at the innovative technological changes observed during the Late Pleistocene Later Stone Age (LPLSA), particularly among the rich collection of already excavated sites. Fortunately, the problem of dating is markedly more tractable for the LSA than for the MSA, given the applicability of radiocarbon methodologies in this period. Buffelskloof and Kangkara are key southern Cape LSA sequences, with evidence of human occupation from ca. 25 ka until the proto-historic period (H.J. Deacon 1995). Notwithstanding a long history of research (Opperman 1978; J. Deacon 1984), their chronologies are in desperate need of revision, with substantial gaps and uncertainties. Many of the sites’ levels are currently delimited by only a single conventional radiocarbon date and several of the existing dates were measured on materials now known to be unsuitable for radiocarbon dating. In addition to this proposal for Buffelskloof and Kangkara, comprehensive dating programmes are currently underway at Boomplaas, Nelson Bay Cave and Byneskranskop 1 in the southern Cape (Loftus), Klipfonteinrand on the West Coast (Mackay) and Sehonghong in the Lesotho highlands (Pargeter). Together, these projects will provide an improved chronological framework from which to begin exploring the interaction of human culture, climate and geography during the Late Pleistocene in southern Africa. Site Backgrounds, Materials and Methods Buffelskloof was excavated by Opperman between December 1974 and January 1975. Opperman chose the site because of its visible stratigraphy and the potential it provided to complement observations from the nearby (c. 40km) Boomplaas Cave. This project is concerned with the lowermost units in Opperman’s excavation. The lowest stratigraphic unit (HE2) is a ± 0.5 m-thick ashy deposit that has produced three dates. Two from a hearth at the base of HE2 indicate an occupation before the LGM (22,800 ± 850 BP (UW-339) and 22,575 ± 270 BP (Pta-1807)), while no more than 5 cm above these a third from another hearth dates to 11,875 ± 115 BP (Pta-1805), presumably indicating a prolonged period of non-deposition. Opperman (1978) nevertheless does not differentiate between any of the artefacts from HE2, indicating that that they are “typologically homogeneous” and attributable to the Albany Industry (i.e. the Oakhurst Complex). Given that unit HE1 produced an early Holocene age (8960 ± 80 BP; Pta-1484) just 10 cm above Pta-1805, we suspect that the youngest of the three HE2 dates may be contaminated. Opperman (1978: 24) indicates that bone preservation (mammal and tortoise) was good throughout the deposit and all the dates were run on charcoal, at least some of which was identified (Deacon 1979). Proposed dates: The dating of HE1 and HE2 are uncertain and those already obtained have large standard errors. We propose two dates on HE1 materials and three new dates on HE2 materials. Kangkara is a large rock-shelter excavated by H.J. Deacon and colleagues in 1972. Robberg-attributed assemblages come from the two basal stratigraphic units, Members BD and BBD (Deacon 1984). A single charcoal sample was dated from a hearth at the interface of BBD with the underlying MSA unit BHL. Two chemically different fractions produced ages of 12,550 ± 110 and 12,330 ± 130 BP (Pta-782). The MSA levels – studied by Volman (1982) – are undated. Member BD is also undated. Deacon (1984: 72) notes that charcoal was scarce and that most of the dates from the site were thus run on unidentified bone fragments recovered in sieving. Identified taxa in BD are few, but include single specimens of dassie, red hartebeest and suid. Possible dates: We propose four new dates two each from layer BD (bone) and BBD (charcoal). Methods The Buffelskloof and Kangkara assemblages currently housed at the Iziko Museum in Cape Town and will be sampled and selected for export to the Oxford University’s Radiocarbon Accelerator Unit (ORAU) by Dr. Alex Mackay in May 2017. Charcoal nodules or unidentified bone samples (approx. 1-2 g) will be taken from the four excavated contexts mentioned above and in Table 1. The ORAU is a leading laboratory for the processing of high resolution AMS dates and is one of the few centres currently performing the ABOx-SC pre-treatments necessary for this project. Emma Loftus (Oxford University) will be processing the charcoal samples and running the pre-treatment procedures. Loftus will use ORAU’s acid-base-acid extraction method for charcoal with graphitised samples dated on the Oxford Radiocarbon Accelerator Unit's HVEE AMS system. The greater sensitivity of AMS systems permits measurement of considerably smaller samples than required for conventional beta-counting measurements, and typically produces more accurate and precise dates. The AMS technique and improved pre-treatment strategies should substantially refine the two sites’ chronologies. Conclusions Emerging perspectives on southern Africa’s Late Pleistocene archaeological record show that technological trends have long and complex pathways with possible links to ecological and geographical variability and population demography (Mackay et al. 2014). Re-evaluating the Buffelskloof and Kangkara chronologies will contribute to a much needed temporal framework for testing the roles of adaptation and cultural transmission in the uptake and spread of these technological systems. Coupled with newly emerging archaeological and chronological data from other LPLSA sites in southern Africa, these data may ultimately also help test models of population structure in the region (e.g. Mitchell 2010; Schlebusch et al. 2013). Furthermore, southern Africa’s LPLSA is uniquely positioned to offer insights into the operation of analogous technological systems in older periods and thus of human behavioural evolution more broadly. Understanding the chronological context for this variation can only improve archaeological models of Pleistocene human behavioural variability in Africa and elsewhere. References Deacon, J. 1984. The Later Stone Age of Southernmost Africa. Oxford: British Archaeological Reports. Jacobs, Z., et al. 2008. Ages for Middle Stone Age innovations in southern Africa: implications for modern human behaviour and dispersal. Science 322:733-735. Mackay, A., Stewart, B., Chase, B.M. 2014. Coalescence and fragmentation in the late Pleistocene archaeology of southernmost Africa. JHE 72:26-51. Lombard, M., et al. 2012. South African and Lesotho Stone Age sequence updated (I). SAAB 67:120–144. Opperman, H., 1978. Excavations in the Buffelskloof rock shelter near Calitzdorp, southern Cape, SAAB, 18-38. Schlebusch, C.M., et al. 2012. Genomic variation in seven Khoe-San groups reveals adaptation and complex African history. Science, 338:374–379. Volman, T.P. 1982. The Middle Stone Age in the Southern Cape. PhD thesis, University of Chicago.