22.1 Introduction and background
The Middle Stone Age (MSA), a period associated with shifting temporal and geographical boundaries, existed between ~300 and 22 kyr BP. Since the 1990s, an increase in research focus on the MSA by several local and international groups has resulted in a remarkable diversity of perspectives on the MSA in South Africa, albeit predominantly on MIS 4 phenomena, which is discussed in some detail in this chapter.
In southern Africa, there are relatively few sites that record the Early Middle Stone Age (EMSA), dating to between ~300 and 130 kyr BP and within marine isotope stages (MIS) 8–6 (Table 22.1). Of these sites, few have been examined in detail and even fewer can be located securely in time. Rising sea levels during Pleistocene interglacials obliterated traces of occupation on the coast, and most preserved sites are therefore found in the interior of South Africa, for example at Florisbad, Border Cave, Bundu Farm, Lincoln Cave, and Wonderwerk Cave (Fig. 22.1). Pinnacle Point is currently the only coastal MIS 6 site. A somewhat clearer picture emerges in MIS 5 (130–85 kyr BP) as a greater number of sites with more extensive geochronometric control is known. Research efforts historically converged on the Cape coastal region, but many of the data-rich early late Pleistocene sites occur inland, for example in the Orange River area, at the Cave of Hearths, Border Cave, Sibudu, Melikane and Rose Cottage Cave. There is an increase in archaeological visibility in glacial MIS 4 (~75–58 kyr BP). Recent insights into the Still Bay and Howiesons Poort techno-complexes come from Blombos Cave, Diepkloof, Klasies River, Umhlatuzana, Sibudu Cave, Rose Cottage Cave and Klein Kliphuis (Henshilwood, Reference Henshilwood2012). MIS 3 may have been less densely populated than MIS 5 and 4, but not significantly so (Mitchell, Reference Mitchell2008). A number of research projects specifically investigating MIS 3 occupation in Lesotho, Western Cape and KwaZulu-Natal also promise a deeper understanding of this period and the beginnings of the Later Stone Age (~43–22 kyr BP). Thus, the geographical context of available MSA sites changes over time in response to climate, demography, technology and cultural practices, as well as the nature of recent research activities.
Table 22.1 The South African and Lesotho Stone Age sequence (after Lombard et al., Reference Lombard and Haidle2012)
| Approximate time period (kyr BP) | Marine Isotope Stage (MIS) association | Associated techno-complex | Alternative nomenclature |
|---|---|---|---|
| 40–18 | 3–2 | Early Later Stone Age | Late Pleistocene microlithic |
| 40–20 | 3–2 | Final Middle Stone Age | MSA IV at Klasies River, MSA 4 generally |
| 60–45 | 3 | Sibudu | late MSA; post-Howiesons Poort, MSA III at Klasies, Sibudan, MSA 3 generally |
| 66–58 | 4–3 | Howiesons Poort | |
| 77–70 | 5a–4 | Still Bay | |
| 105–77 | 5c–a | Mossel Bay | MSA II at Klasies River, MSA 2b generally, Pietersburg, Orangian |
| 130–105 | 5e–d | Klasies River | MSA I at Klasies river, MSA 2a generally, Pietersburg |
| 300–130 | 8–6 | Early Middle Stone Age |
Fig. 22.1. Map of South Africa with sites mentioned in the text.
The MSA fossil record traces a slow mosaic-like development of morphologically human modern traits. The ~250 kyr Florisbad calvarium and the Hoedjiespunt fossils represent populations immediately ancestral to Homo sapiens sapiens that diverged ~200 kyr BP (Lombard et al., Reference Lombard, Schlebusch and Soodyall2013). The relatively large MIS 5 fossil collection from Klasies River on the southern Cape coast (Figs. 22.2, 22.3) shows primarily modern morphological traits, with a retention of archaic characteristics, while the ~36 kyr Hofmeyr skull, a surface find from the Eastern Cape, is fully modern but still outside the morphological range of modern Khoe-San (Dusseldorp et al., Reference Dusseldorp, Lombard and Wurz2013). The MSA encompasses the population history of anatomically modern Homo sapiens and their immediate predecessors, and one of the major issues is identifying when behavioural and cognitive ‘modernity’ developed. In this chapter, snapshots are provided into the aspects that most prominently influenced the formation of the South African MSA archaeological record, including technologies, subsistence and site maintenance behaviour, and how these may link to cognitive evolution.
Fig. 22.2. Human fossils from Klasies River. (A) KRM26909, two views of a parietal fragment of skull, Layer 37, Cave 1; (B) metatarsal, ~ Layer SMB2 Cave 1; (C) KRM 16425, fragment of frontal bone, Layer 16, Cave 1; (D) KRM 13400, corpus of mandible, Layer 14, Cave 1; (E) KRM 13400, occlusal view
Fig. 22.3. Klasies River main site, southern Cape, South Africa
22.2 Technological behaviour in the Middle Stone Age
The intensified interest in the MSA in recent years has resulted in new perspectives on bone, stone and fire-related technologies, and also on the production of shell beads and the utilisation of ochre. A relatively small number of formal bone tools dating to around 100 kyr BP are known from Blombos and Klasies River (Wurz, Reference Wurz2013). The Still Bay techno-complex at Blombos Cave (Henshilwood, Reference Henshilwood2012) and the MIS 5 to 3 layers at Sibudu yielded larger such collections (d’Errico et al., Reference d’Errico, Backwell and Wadley2012b). The bone wedges, pièces esquillées, pressure flakers, smoothers and sequentially notched pieces from Sibudu are similar to those found in the Upper Palaeolithic. There is widespread consensus that such formally shaped bone tools are innovative and reflect cognitive complexity. There is less agreement on the cognitive implications of the deliberately shaped and designed stone tools of the MSA, even though their production is just as complex as that of bone tools. The lithic toolkit of the MSA contains flakes, blades of different proportions, unretouched and retouched points, and tool types such as denticulates, scrapers, backed and bifacials. Discoidal, prepared core (including Levallois), blade and bipolar reduction methods were used. The earliest MSA assemblages indicate that important techno-cognitive milestones had been reached. For example, the engineering constraints involved in producing the standard toolkit in the MSA necessitated mental projections of at least five to ten actions ahead, indicating modern cognitive capabilities (Hiscock, Reference Hiscock2014). Using bonding agents to create multi-component tools denotes another complex cognitive strategy. The practice of hafting stone tools was most likely in place by the EMSA in South Africa (Wadley, Reference Wadley2013), perhaps even as early as 500 kyr BP (Wilkins et al., Reference Wilkins, Schoville, Brown and Chazan2012; but see Rots and Plisson, Reference Rots and Plisson2014).
Identifying chronological and regional trends in MSA lithic technology is a major challenge. There is a strong possibility that technological conventions observable on regional levels occurred prior to 100 kyr BP, but they become more apparent after this time. A number of broadly similar assemblages from the Cape dating to MIS 5b and 5c probably denote such a convention (Wurz, Reference Wurz2013). The Still Bay and Howiesons Poort are markedly distinct from other MSA phases, as they occur on a subcontinental scale within bracketed time periods (but see Porraz et al., Reference Porraz, Parkington, Rigaud, Miller, Poggenpoel, Tribolo, Archer, Cartwright, Charrié-Duhaut, Dayet, Igreja, Mercier, Schmidt, Verna and Texier2013). At most sites, the Still Bay dates to between ~80 and 70 kyr BP and the Howiesons Poort to between ~70 and 58 kyr BP (Wurz, Reference Wurz2013). There is much evidence to indicate that the Still Bay and Howiesons Poort are associated with innovative technological practices. In the Still Bay of Blombos Cave, heat-treated silcrete was pressure flaked to produce lanceolate bifacial points (Mourre et al., Reference Mourre, Villa and Henshilwood2010), and in the Howiesons Poort a soft stone (Soriano et al., Reference Soriano, Villa and Wadley2007) or perhaps a vegetal hammer was used to knap blades and bladelets which were transformed into geometric backed tools (Villa et al., Reference Villa, Soriano, Tsanova, Degano, Higham, d’Errico, Backwell, Lucejko, Colombini and Beaumont2012; Henshilwood et al., Reference Henshilwood, van Niekerk, Wurz, Delagnes, Armitage, Rifkin, Douze, Keene, Haaland, Reynard, Discamps and Mienies2014). The Howiesons Poort (Figs. 22.4, 22.5) is the most intensely researched techno-complex in South Africa, and now appears to consist of several phases. Furthermore, flake production and notched tools form an important part of the wider technological repertoire at, for example, Klasies River, Diepkloof, Pinnacle Point, Sibudu (Wurz, Reference Wurz2013) and Klipdrift Shelter (Henshilwood et al., Reference Henshilwood, van Niekerk, Wurz, Delagnes, Armitage, Rifkin, Douze, Keene, Haaland, Reynard, Discamps and Mienies2014). Finding commonalities in the lithics from different areas after ~58 kyr BP becomes problematic, but unifacial points occur in many assemblages, for example at Sibudu and Umhlatuzana. Typical MSA artefacts occur in various regions until 22 kyr BP, but at Border Cave, small flakes in quartz and San-like bone tools dating to 44–42 cal kyr BP are described as the earliest evidence of the Later Stone Age (LSA) in that region (d’Errico et al., Reference d’Errico, Backwell, Villa, Degano, Lucejko, Bamford, Higham, Colombini and Beaumont2012a; Villa et al., Reference Villa, Soriano, Tsanova, Degano, Higham, d’Errico, Backwell, Lucejko, Colombini and Beaumont2012).
Fig. 22.4. Artefacts from the Howiesons Poort levels at Klasies River. (A) KRM28942, silcrete backed artefact, Cave 1A, Layer 12; (B) KRM 31421, silcrete backed artefact, Cave 1A, Layer 20; (C) Silcrete notched backed artefact and blade section, Cave 1A, Layer 18; (D) Quartzite blades, Cave 2 surface; (E) Quartzite core, Cave 1A, E50 CP8; (F) KRM 26733, engraved midshaft bovid fragment, Cave 1A, Layer 19; (G) KRM 42160 bone point, Cave 1A, Layer 19
Silcrete was sometimes heated to enhance flakeability (Brown et al., Reference Brown, Marean, Herries, Jacobs, Tribolo, Braun, Roberts, Meyer and Bernatchez2009), but the particular procedures used to accomplish this are still under debate. Field-based experiments illustrate that controlled management of the heating process, involving burying specimens in sand below a fire, is important to achieve this (Wadley and Prinsloo, Reference Wadley and Prinsloo2014). A possibility is also that silcrete nodules were heated in areas surrounding open fires or directly on slightly cooled embers (Schmidt et al., Reference Schmidt, Porraz, Slodczyk, Bellot-Gurlet, Archer and Miller2013). Another procedure that required careful and controlled heating is the production of compound adhesives for hafting, recorded at >70 kyr BP at Sibudu and Rose Cottage (Wadley, Reference Wadley2013). This evidence favours the hypothesis that MSA peoples were able to intentionally control heat in different ways to achieve desired results. These technically advanced processes have been tied to modern levels of cognitive complexity (Wadley, Reference Wadley2013).
MSA technologies that provide more direct links to cultural ideology involve ochre and beads. Their early occurrence is key in supporting the hypothesis that by 100 kyr BP human lives were structured by objects that acted as symbols and social mediators (McBrearty and Brooks, Reference McBrearty and Brooks2000; Henshilwood and Marean, Reference Henshilwood and Marean2003). The 100 kyr-age containers at Blombos Cave were used to mix and store ochre-rich compounds and these, as well as the iconic ~77 kyr geometrically engraved ochre, are benchmarks for complex cognition in the MSA (Henshilwood et al., Reference Henshilwood, d’Errico, van Niekerk, Coquinot, Jacobs, Lauritzen, Menu and García-Moreno2011; Wadley, Reference Wadley2013). Engraved ochres dating back to ~100 kyr BP at Blombos, Klasies River and Pinnacle Point (Wurz, Reference Wurz2013), and the MIS 4 shell beads and engraved ostrich eggshell from Diepkloof and Klipdrift Shelter, are widely regarded as indicators of modern cognition (e.g. Henshilwood, Reference Henshilwood2012; Wadley, Reference Wadley2013; Henshilwood et al., Reference Henshilwood, van Niekerk, Wurz, Delagnes, Armitage, Rifkin, Douze, Keene, Haaland, Reynard, Discamps and Mienies2014). However, these inferences on modern cognition are under debate (see Henshilwood and Dubreuil, Reference Henshilwood and Dubreuil2011).
22.3 Subsistence behaviour and site maintenance
For most of the MSA, both the interior and coastal areas of South Africa were productive environments with grasslands that could support large herbivores (Brink, Reference Brink1999). Pleistocene assemblages contain relatively more grazers than those from the Holocene (Klein, Reference Klein, Deacon, Hendey and Lambrechts1983) and are often characterised by a higher ungulate richness (Faith, Reference Faith2011). Grassland taxa include plains zebra (Equus quagga), wildebeest (Connochaetes spp.), giant wildebeest (Megalotragus priscus) and springbok (Antidorcas spp.), while the presence of browsers such as duiker (Sylvicapra grimmia), Cape grysbok/steenbok (Raphicerus spp.) and kudu (Tragelaphus strepsiceros) indicates mosaic environments with closed vegetation in some areas and periods. Available data are too sparse to understand subsistence behaviour in the EMSA, but at Pinnacle Point Cave 13B (PP13B) in MIS 6, large and very large ungulates (size 3 and larger) such as the extinct giant buffalo (Megalotragus priscus), buffalo (Syncerus antiquus) and eland (Tragelaphus oryx) were selected prey. At a number of MIS 5 sites, e.g. Blombos Cave, Diepkloof Rock Shelter, Florisbad, Klasies River, Pinnacle Point Cave 13B and Ysterfontein, relatively more grazers than browsers occur. As in MIS 6, the majority of prey consists of the highest ranked species, large and very large ungulates (Clark and Kandel, Reference Clark and Kandel2013). It is a distinct possibility that some of the retouched and unretouched points of the MIS 6 and 5 assemblages were used as part of thrusting spears, but this hypothesis needs further investigation.
In MIS 4 a shift in faunal subsistence patterns occurs with a marked increase in the exploitation of small mammals and small (size 1) ungulates, which is an indication of widened dietary breadth (Clark and Kandel, Reference Clark and Kandel2013). A decline in dietary breadth occurs in MIS 3, with a return of the dominance of large and very large-bodied bovids (Clark and Kandel, Reference Clark and Kandel2013). Technological innovations relating to the MIS 4 dietary shift may include remote capture methods such as snares (Wadley, Reference Wadley2013) and bow and arrow technology at Sibudu (Lombard and Phillipson, Reference Lombard and Phillipson2010); procedures that imply complexity in cognition (Lombard and Haidle, Reference Lombard and Haidle2012). A commonly discussed theme in the MSA, particularly regarding MIS 4, is whether there is a cause-and-effect relationship between cultural and environmental change. It is telling that, at Sibudu, changes in lithic technology from Howiesons Poort to post-Howiesons Poort precede changes in the environment as reflected by the proportional representation of habitat-sensitive species (Clark, Reference Clark, Clark and Speth2013). Whether prey choice in the MSA and LSA, particularly the Holocene, reflect hunting skill is another issue of interest. Even though less dangerous species were sometimes exploited in the MSA compared to the LSA (e.g. Klein and Cruz-Uribe, Reference Klein and Cruz-Uribe1996), similar types of ungulate prey were targeted (Faith, Reference Faith2008). Resource availability and perhaps ranking of food sources rather than capabilities may have affected subsistence behaviour (Deacon, Reference Deacon, Mellars and Stringer1989). Holocene and MSA hunter-gatherers occupied different niches, as environmental changes at the Pleistocene-Holocene transition compelled an expansion of dietary breadth and more intensive exploitation of food sources such as shellfish and tortoise (Clark and Kandel, Reference Clark and Kandel2013).
The first evidence for shellfish in MSA diets is at 164 kyr BP at Pinnacle Point Cave 13B, although only in very low quantities. It became a more substantial, but never dominant, element of subsistence behaviour for the remainder of the MSA at coastal sites in the southern and western Cape. Proximity to the sea strongly influenced shellfish collecting strategies: these were aimed at optimising distance to the resource and their yield (Langejans et al., Reference Langejans, van Niekerk, Dusseldorp and Thackeray2012). Fruits, seeds, nuts and geophytes must have also formed a substantial part of the MSA diet, but palaeobotanical data are limited compared to faunal and shellfish evidence. Nonetheless, at Sibudu, the analysis of plant remains such as burnt grass, sedges, charcoal, carbonised seeds, pollens, phytoliths and plant residues are the source of the most informative new perspectives on MSA behaviour (e.g. Sievers and Muasya, Reference Sievers and Muasya2011) and climate (Bruch et al., Reference Bruch, Sievers and Wadley2012). It has, for example, been determined that sedges and other grasses were used for bedding at Sibudu at ~77 kyr BP. This is the oldest example of this practice yet identified (Wadley et al., Reference Wadley, Sievers, Bamford, Goldberg, Berna and Miller2011), and becomes more prevalent in the post-Howiesons Poort levels. It has also been shown that wood and leaves were used for their insecticidal properties. Aromatic leaves at Sibudu were laid on the sedge bedding, and in the post-Howiesons Poort, Tamboeti wood smoke may also have been used for this purpose (Wadley, Reference Wadley2013). Recent research on pyrotechnology through micromorphological and experimental methods reveals unexpected levels of behavioural sophistication (Goldberg et al., Reference Goldberg, Miller, Schiegl, Ligouis, Conard and Wadley2009; Bentsen, Reference Bentsen2014) in the Howiesons Poort and post-Howiesons Poort. For example, hearths were maintained and altered, ash dumps formed, and bedding was burnt.
22.4 Discussion and summary
The clearest pattern in the MSA across southern Africa, in particular during MIS 4, is the apparent increase in bone tools and ornaments, and the homogeneity of stone tool technology. This may be a reflection of larger population sizes, or interactions between populations across southern Africa (Mackay et al., Reference Mackay, Stewart and Chase2014, but see Klein and Steele, Reference Klein and Steele2013). It may also be an artefact of research intensity, as the trends discussed here rely on results from relatively recent investigations from a handful of sites. It is a priority to expand the database on the MSA in order to more fully understand landscape utilisation, demographic patterns, and the geographical extent of technological conventions and innovations. Whether homogeneous developments in stone and bone tool technologies occurred on a subcontinental scale, as smaller-scale regional pulses, or as idiosyncratic expressions, will largely remain unknown until more data are available. The development of innovative behaviours in the MSA, especially in MIS 4, has been linked to climatic changes (e.g. Ziegler et al., Reference Ziegler, Simon, Hall, Barker, Stringer and Zahn2013). However, neither the resolution of palaeoenvironmental (Mackay et al., Reference Mackay, Stewart and Chase2014) nor archaeological data is sufficient to determine which occurred first.
Among the crucial evolutionary events in the early MSA are the development of anatomical modernity and complex stone tool reduction and hafting procedures. In MIS 5 the pace of cultural change as reflected in bone and stone tool, ochre and engraving technologies seems to increase, culminating in MIS 4. It is also mostly in MIS 4 that an increase in diet breadth is recorded, while faunal and shellfish exploitation patterns generally remain stable. Recent advances in experimentation and lab analysis have also provided insight into MSA cultures and practices, including weaponry, pyrotechnology, site maintenance, ochre, and plant use. These advances are deepening our understanding of the MSA and the relationship of MSA peoples to their local environment and resources, and to regional-scale Quaternary climate and environmental changes.
