Journal of Archaeological Method and Theory (2022) 29:214–250 https://doi.org/10.1007/s10816-021-09513-x Tracking Occupational Intensity Using Archaeo-faunal Data: Case Studies from the Late Pleistocene in the Southern Cape of South Africa Jerome P. Reynard 1 Accepted: 17 February 2021/ Published online: 17 March 2021 # The Author(s), under exclusive licence to Springer Science+Business Media, LLC part of Springer Nature 2021 Abstract Occupational intensity is a common theme in current research and has been linked to significant demographic trends in the past. The Late Pleistocene in the southern Cape may be especially important in understanding the impacts of socio-demographic change given its association with developments in ‘modern’ human behaviour. The ubiquity of archaeo-faunal remains at Middle Stone Age (MSA) sites makes these convenient datasets for documenting site-specific occupational patterns. In this paper, zooarchaeological and taphonomic data are evaluated as proxies for occupational intensity, and occupational trends are explored in the southern Cape. Zooarchaeological and taphonomic data from three southern Cape MSA sites—Klipdrift Shelter, Blombos Cave and Pinnacle Point—are compared with previously determined higher and lower occupational levels within each site to assess the value of these proxies in tracking temporal changes in settlement intensity. The results show that, while frequencies of small mammals and larger ungulates often covary with occupational levels, these are problematic indicators because of the impact of carnivores. Similarly, faunal diversity generally corresponds well with increasing human occupations but is a problematic proxy because of the effects of animal activity. Anthropogenic bone surface modifications appear to be effective in tracking occupational patterns, with trampling a particularly useful indicator. Faunal and shellfish density, and transverse bone fracture patterns, are valuable proxies of occupational intensity at all sites. Generally, the data suggests close links between occupational intensity at these sites and marine transgressions. Evidence of increased exploitation of small game in the later MSA may imply periods of subsistence intensification possibly linked to increased demographic pressure during Marine Isotope Stage 4. Keywords Middle Stone Age . Zooarchaeology . Occupational intensity . Taphonomy . Trampling . South Africa * Jerome P. Reynard Jerome.Reynard@wits.ac.za Extended author information available on the last page of the article Tracking Occupational Intensity Using Archaeo-faunal Data: Case... 215 Introduction The Middle Stone Age (MSA, c. 300,000 years ago [ka] to c. 30 ka) was a key period of innovation and demographic change for Homo sapiens (Wadley 2015, 2021). In southern Africa, techno-complexes such as the Still Bay (from c. Marine Isotope Stage [MIS] 5b to MIS 4) and the Howiesons Poort (from c. MIS 5a to 4) are associated with novel technology and complex human behaviour (Henshilwood et al. 2002; Lombard 2011; Mourre et al. 2010; Brown et al. 2012; Bradfield et al. 2020). Many MSA sites with evidence of behavioural complexity occur in the southern Cape (Henshilwood 2012) with much research focused on why this particular period—and region—has produced such innovation (e.g., Marean 2010; Parkington 2010; Compton 2011; Mackay et al. 2014; Will et al. 2016). Demographic pressure is often argued to be a key factor in the development of behavioural complexity (Stiner et al. 1999; Shennan 2001; Henrich 2004; Powell et al. 2009; Richerson et al. 2009; Archer 2020). It has been suggested that innovative ideas and technologies are more likely to be produced, maintained and circulated in larger groups (Henrich 2004; Archer 2020) and Powell et al. (2009) have argued that this may have reached a critical threshold at c. 100 ka (see also Wadley 2021). This model has been challenged by some researchers. For example, changing resource distribution (Dusseldorp 2014) and risk reduction (Bousman 2005; Hiscock et al. 2011) is sometimes offered as a reason for increased technological complexity. Other researchers argue that there is more continuity between periods of technological complexity and periods without evidence of innovation, than is often suggested (Villa et al. 2010; Mackay 2011; Perreault et al. 2013; Sealy 2016). It is likely that demographic factors are linked to occupational intensity during the MSA in the southern Cape. At Klipdrift Shelter, there is good evidence that population sizes may have increased during the Howiesons Poort (Reynard et al. 2016a). Evidence of widespread standardised techno-complexes throughout this region implies that social networks played an important role in the distribution of these traditions (Ambrose and Lorenz 1990; Deacon 1992; Douze et al. 2018). The similarities between engraved ostrich eggshell patterns at Klipdrift and Diepkloof rockshelters, for example, suggests interactions between populations hundreds of kilometres apart (Texier et al. 2010; Douze et al. 2018). Mobility and organisational flexibility may have linked groups across landscapes (Douze et al. 2018). Expanded social networks mean that foraging groups were probably not static entities linked to specific locations but interconnected residential groups (Zhou et al. 2005; Bird et al. 2019). Fluctuating shorelines during glacial periods in the southern Cape would have affected resource distribution and influenced risk reduction strategies and population sizes. The pursuit of migratory game may have also been an important factor in contact between groups and the extension of social networks. Documenting site-specific occupational patterns may therefore highlight the role it plays in social networks and mobility patterns in the southern Cape. Exploring occupational intensity is an important aspect of current archaeological research. Studies utilising lithic data (e.g., Schlanger 1991; Shiner 2006), site formation processes (Schiffer 1987; Malinsky-Buller et al. 2011) and mobility patterns (Binford 1978, 1980) have been used to explore occupational patterns. Micromorphological studies also offer valuable insights into occupational patterns at many southern Cape sites (Haaland et al. 2021; Karkanas et al. 2015). The changing abundance of MSA sites has also been used to infer shifting occupation dynamics in the southern Cape 216 Reynard (Marean et al. 2014; Sealy 2016). Mathematical models can infer large-scale, regional occupational patterns (e.g., Schlanger 1990; Shennan 2001; Powell et al. 2009; Archer 2020) and have been used to predict forager group sizes (Beckerman 1983; Janson and Goldsmith 1995; Zhou et al. 2005). While these types of models are valuable, sitespecific data are a critical means of documenting diachronic trends in human activity. Indeed, site-specific data often informs these models. Tracking occupational intensity at forager sites can be challenging. Foraging groups are usually highly mobile, occupying locations on a seasonal basis dependent on animal migratory patterns, molluscan seasonality or plant growth cycles. The ubiquity of faunal remains may serve as a valuable proxy for site-use intensity. Faunal data link occupations with environment conditions, site formation processes and subsistence strategies. Much research has investigated how foraging behaviour is associated with mobility strategies (e.g., Binford 1978, 1980; Kelly 1992; Grove 2010; McCall and Thomas 2012; Dusseldorp 2014; Reynard and Henshilwood 2019). Taphonomic analyses also offer a critical insight into site-use intensity (e.g., Lyman 1987; Yeshurun et al. 2020), and these can inform on temporal changes within sites. The aim of this paper is to evaluate faunal data in tracking occupational patterns during the Late Pleistocene of the southern Cape. Zooarchaeological and taphonomic data are documented from three MSA sites—Klipdrift Shelter, Blombos Cave and Pinnacle Point—and compared with previously determined, accepted levels of occupational intensity within their sequences. The effectiveness of fauna data as proxies for occupational intensity is examined, and the links between these data and environmental or socio-demographic trends are discussed. The broad goal of this study is to generate new understandings on the southern African MSA, especially in light of probable demography-driven settlement processes in the southern Cape. Here, occupational intensity is defined by the number of individuals that occupied a site, the frequency of visits, and the duration of stay at a site (Munro 2004; Haaland et al. 2021). Analytical Framework Klipdrift Shelter (KDS), Blombos Cave (BBC) and Pinnacle Point (PP) (Fig. 1) were chosen for three reasons. Firstly, they occur in the same fynbos-dominated environment near the coast (see discussion below), and are relatively close to each other. This limits the effects that regional differences in vegetation may have had on faunal communities. It also allows shellfish data to be incorporated in the analyses as both a faunal proxy and a measure of marine regression. Secondly, because a key objective of this study is to assess changes in occupations through time, these are the only sites in the region with temporarily distinct, layer-specific taphonomic and zooarchaeological data. Each site also occupies relatively successive chronological periods from c. 163 to 90 ka (PP), c.110 to 70 ka (BBC) and c. 65 to 59 ka (KDS). Thirdly, these sites encompass similar techno-traditions. The region may have formed part of broader social networks which means that it may be possible to link occupational patterns between sites. Occupational patterns have also been investigated using numerous proxies at these three sites, so there is some comparable reference to site-use intensity here. At each site, faunal abundance, fauna and shellfish density, and taphonomic indicators are documented through the sequence. Various non-faunal proxies such as lithic abundance, and Tracking Occupational Intensity Using Archaeo-faunal Data: Case... 217 geoarchaeological and micromorphological data are used to corroborate apparent occupational patterns. Faunal Abundance It is expected that higher occupational phases at sites may be linked to an increased abundance of faunal remains, particularly that of larger ungulates. The theoretical aspect of this study is guided by optimal foraging theory (OFT) which predicts that people will forage the most ‘profitable’ food resources and maximise overall return rates (Kelly 1995: 83; Dusseldorp 2012; Winterhalder 2001). In particular, I incorporate the diet breadth model where prey are ranked according to their caloric value— usually measured as body size—with larger prey considered more profitable and higher ranked (Hames and Vickers 1982). Diet breadth models assume that the profitability of prey is defined by their caloric value minus search and handling cost (hunting, processing and transporting prey). It should then be expected that larger prey would be abundant in higher occupational periods in these sites, especially since large carnivores such as lions and hyenas are not prevalent at any of the sites under discussion (Badenhorst et al. 2021). Fig. 1 Map of southern Africa with Middle Stone Age sites referred to in this paper. DRS = Diepkloof Rockshelter; DK = Die Kelders; KDS = Klipdrift Shelter; BBC = Blombos Cave; PP = Pinnacle Point; KRM = Klasies River main; SI = Sibudu Rockshelter 218 Reynard While seemingly contradictory, an abundance of small fauna may also be associated with higher occupational periods at these sites. Demographic pressure associated with subsistence intensification may have played an important role in small fauna exploitation. Subsistence intensification can be framed within OFT and may be defined as the extraction of increased energy from food resources at the expense of foraging efficiency (Schoener 1974; Munro 2009: 141). Intensification occurs when over-exploitation caused by increasing population pressure (Flannery 1969; Stiner et al. 1999; Munro 2009; Jerardino 2010) or environmental degradation (Henshilwood and Marean 2003; McCall 2007) results in the depletion of higher-ranked food resources such as large herbivores. Intensification has been linked to increases in the exploitation of lowerranked small fauna such as hare, rabbits and birds (Stiner et al. 2000) and may be closely linked to higher occupations during the Later Pleistocene in southern Africa (Clark 2011; Reynard and Henshilwood 2017). Indeed, small fauna such as hyraxes, Cape dune mole rats and tortoises make up a significant portion of Late Pleistocene faunal assemblages (Steele and Klein 2009). Evidence of remote capture technology in southern Africa shows that people had the technological capabilities for effective mass collection in the Late Pleistocene (Wadley 2010; Bradfield et al. 2020; Val et al. 2020). Taphonomic evidence suggests that both animals and people probably collected small mammals at BBC, KDS and PP (Badenhorst et al. 2014; Armstrong 2016; Reynard and Henshilwood 2017). Klein (Henshilwood et al. 2001a; Klein et al. 2004; Steele and Klein 2009; Klein and Bird 2016) showed that shellfish, tortoises and small mammals decrease in size during the Later Stone Age and argues that this suggests increasing subsistence intensification from the Later Pleistocene through the Holocene. Increasing population pressure may thus have had a significant impact on small fauna populations (Munro 2004: S7). Faunal abundance may be a challenging proxy for human occupational intensity since rockshelters are also used by other animals (Brain 1981). Large fauna abundance would be influenced by factors such as carnivore/raptor collectors, bone densitymediated attrition, the environment, animal migration and human mobility patterns. Small fauna are consumed by a range of animals which could make this a complicated indicator of human occupations. Despite these challenges, both small and large game could be useful proxies in exploring settlement patterns in the MSA. Taphonomy Taphonomic processes can inform on both who occupied a site and how intensely it was used. Bone surface modifications (BSM) may suggest fluctuating periods of site use by people. BSM data can be an effective means of inferring depositional patterns at sites (Madgwick and Mulville 2015). Differentiating anthropogenic from zoogenic BSM is key to disentangling human from animal agency. Proportions of toothmarked bone to anthropogenically marked bone are generally used to discern who had primary access to prey (Binford 1981; Blumenschine 1986, 1995; Marean and Bertino 1994; Thompson et al. 2017). Higher frequencies of carnivore-ravaged bone within deposits would imply periods used by various bone-collecting predators (Brain 1981; Blumenschine 1988; Egeland et al. 2004; Thompson et al. 2017). Higher frequencies of butchery and processing marks may signify increased site-use activity. This is not to imply that increased human occupations necessarily lead to more Tracking Occupational Intensity Using Archaeo-faunal Data: Case... 219 processing intensity, but it is likely that as more people occupy a site, there is a greater chance that faunal remains would likely yield evidence of their occupations in the form of cut or percussion marks. That said, linking evidence of increased bone processing to more intensive site use may be problematic. Cut marks, for example, are not very prevalent in most Stone Age faunal assemblages. Often less than 5% of bone display cut marks (Bunn 1991). Their abundance corresponds to various factors such as the skill of the butcher, the element and the type of animal being butchered (Val et al. 2017). Actualistic data from Central African forest foragers also show that cut mark frequency may be a poor proxy for intensification practices (Lupo et al. 2013). While percussion marks are generally more common than cut marks, the frequencies of these too would vary according to what part of the animal is being processed (Blumenschine 1986). Exploring skeletal element abundance patterns is viewed as key to assessing transport decisions which may, in turn, be affected by residential intensity (Binford 1978; Yeshurun and Bar-Oz 2018). However, skeletal element abundance is significantly affected by bone density–mediated attrition which is common at MSA and Middle Palaeolithic sites (Faith and Gordon 2007; Marean and Frey 1997; Pickering et al. 2003). Because of this, skeletal element abundance is not used as a criterion in this review. Fire would have been used for cooking, warmth and protection, and played a role in socially bonding activities and rituals. It is reasonable to assume that site-use intensity would be inextricably linked to burning in general and hearths in particular. Darker strata within archaeological deposits are often the result of a palimpsest of burning events and may indicate increased human occupations (Marean et al. 2000a; Wadley 2012; Miller et al. 2013). While evidence of bone burning often suggests human occupations, in the grassland and scrub-bush (fynbos)-dominated biomes of African and sub-tropical regions, fire is a natural seasonal occurrence and is key to the ecology of these areas (Bond and Keeley 2005). Burning could also affect bone in layers above and below where hearths are constructed (De Graaf 1961). Caution must therefore be used in ascribing all burnt bone in archaeological deposits as anthropogenic. Bone breakage patterns have a complicated relationship with subsistence patterns but could inform on occupational behaviour. Spiral fractures indicate ‘wet’ bone breakage and are caused by both people and carnivores cracking open bones from recently killed animals (Myers et al. 1980; Haynes 1983; Johnson 1985). Thus, using frequencies of spiral fractures within an assemblage may not only correspond to human activities. It may be possible to infer occupational intensity through the frequencies of transverse fractures at a site. Sediment compaction has been linked to increases in transverse breaks (Villa and Mahieu 1991). Although sediment compaction in rockshelters may result from the weight of the deposit and geomorphic process such as rockfalls or roof collapse, it may also be associated with the accumulative weight of people in the site. Given the relatively small size of most predators along the coast and the fact that larger herbivores generally do not occupy coastal rockshelters, the weight of animals may not have significantly affected transverse fracture frequencies. However, some large rockshelters such as Bushmans Rockshelter (Louw 1969) and Grassridge Rockshelter (Collins et al. 2017) were also used to house domestic stock in the recent past and this may have resulted in more transverse breakage in their upper layers. 220 Reynard Trampling Trampling marks on bone may be an especially under-valued means of inferring siteuse intensity. Recognising reworked and trampled bone has long been an important aspect of micromorphological studies (Courty et al. 1989), but few studies have focused on this on macro specimens. Trampling is usually investigated to differentiate natural BSM from butchery activities since trampling marks often resemble cut marks (Behrensmeyer et al. 1986; Dominguez-Rodrigo et al. 2009). Because trampling marks are the result of site-specific pedoturbatory activity, these marks may be useful in tracking specific occupational events. ‘Trampling marks’ generally refer to lines or linear marks (Behrensmeyer et al. 1986; Olsen and Shipman 1988), but trampling can also result in a variety of BSM including pits (Reynard 2014), scratches or ‘microabrasions’ (Dominguez-Rodrigo et al. 2009) and small notches or ‘chips’ (Blasco et al. 2008). Bone sheen on bone surfaces may also be the result of trampling activity (Madgwick 2014). Fauna and Shellfish Density Occupational intensity is often assessed through the use of artefact, lithics, bone or shellfish density (e.g., WadleySchiffer 1987; Deacon 1984; Mitchell 1993; Marean et al. 2000b; Marean 2010; Reynard et al. 2016a). Shellfish may be a particularly useful proxy for both subsistence intensity and shoreline regression. MSA people often focused their subsistence strategies on coastal environments (Kyriacou et al. 2014; Klein and Bird 2016; De Vynck et al. 2019), and higher occupations at coastal sites often correspond to increased shellfish density (Marean 2010). Density values, however, may be problematic because depositional rates cannot be assumed to be constant through time (Butzer 1977; Sullivan 1984; Jerardino 1995, 2016; Stein et al. 2003). BSM frequencies, for example, may fluctuate based on depositional rates. Slow deposition rates may result in the long-term exposure of bone which could drive higher frequencies of transverse fractures and increase fragmentation rates. Moreover, taxonomic diversity may increase because deposits representing long intervals of slow sedimentation may incorporate a rich array of species over time. To address possible fluctuating depositional rates, some researchers use accumulation rates based on the thickness and time span of deposits (Stein et al. 2003; Jerardino 2016). Accumulation rates could reveal possible timeaveraging or stratigraphic conflation that may inform our ability to track occupational intensity. Given the ubiquitous use of density values from MSA sites, and to compare data to other sites, I use density values as a broad indicator of site-use intensity but also incorporate accumulation rates to examine depositional rates. Methods Layers or units from each of the sites are characterised as high, moderate or low occupational periods based on previous analyses of micromorphological, geoarchaeological and artefactual data from each site. The faunal indicators described below are compared to these levels of occupation, and the efficiency of these proxies is evaluated. Tracking Occupational Intensity Using Archaeo-faunal Data: Case... 221 Proxies for faunal abundance are calculated for each layer/unit from all three sites, including the proportions of small and larger fauna and taxonomic diversity. Because most researchers only include mammalian fauna in their analyses, small mammals are used to represent ‘small fauna’. Only identified hyraxes, Cape dune mole rats, hares and other non-carnivorous small mammals < 1 kg are included in the ‘small fauna’ category. ‘Larger ungulates’ are defined as sizes 3 and 4. To document changing faunal diversity for mammals at each site, I calculate Fisher’s alpha—a measure of taxonomic diversity incorporating richness and evenness reasonably unaffected by sample sizes (Magurran 2004)—for each layer. Specific taphonomic indicators suggestive of site-use intensity are also examined. The frequencies of percussion, cut and tooth marks, burning and trampling marks are documented for each layer at the sites. I recorded taphonomic data for KDS and BBC while taphonomic data for PP are reported by Thompson et al. (2017). Because cut, tooth and percussion marks are only recorded on mid-shaft long bone at PP, taphonomic data for KDS and BBC were adjusted in accordance with PP. Thus, only mid-shaft long bone percussion, cut and tooth marks are documented for all sites. To account for the varying effects taphonomic biases have on different animal size classes, cut, tooth and percussion marks are documented only for larger mammals (size 3 to 4). Transverse fracture frequencies are also noted and are also restricted to long bones from size 3 and 4 mammals. Because burning can result in an increase in transverse fracture patterns, transverse fractures are only recorded on non-burnt bone. Fracture outlines are used to signify breakage patterns at all sites. Trampling marks are only reported at BBC and KDS. Faunal and shellfish density (measured in counts per m 3 ) are documented per layer/unit at each site. Shellfish density is based on MNI counts per volume (m3) since these data are available from all sites (Marean 2010; Reynard et al. 2016a; Reynard and Henshilwood 2019). Deposit accumulation rates are calculated based on section drawings of stratigraphic profiles and available optical stimulated luminescence (OSL) dates. It must be noted that these data are gross estimates based on the average thickness of layers/units. Ideally, these data should be obtained from the specific quadrates from the layers/units under consideration. Timespans for layers from KDS and BBC are calculated by subtracting the age of the layer from the layer below (Stein et al. 2003). The age of layers at BBC is calculated as the average of OSL dating samples taken from each layer (Jacobs et al. 2020). For PP13B, unit timespans are based on the maximum and minimum ages reported by Marean et al. (2010). Principal component analyses (PCA; PAST® free software; Hammer et al. 2001) are used to group the layers/units within and between the three sites according to the BSM and faunal abundance indicators discussed above, and to align these with the occupational levels established by previous research. Eleven indicators (Table 1) comprising BSM frequencies (cut, tooth and percussion marks, burnt bone and trampling marks), transverse fracture frequencies, faunal and shellfish density, Fisher’s alpha and the frequencies of small and large fauna are included in the PCA to assess how these proxies influence occupational patterns within the sites. 222 Reynard Regional Background The sites under review occur within the Greater Cape Floristic Region (GCFR) (Marean et al. 2014). This region has a temperate climate and is dominated by fynbos, a sclerophyllous, evergreen shrub (Bergh et al. 2014). Rainfall regimes shift from winter-dominated in the west to year-round on the southern coast and summerdominated in the eastern part of the GCFR (Chase and Meadows 2007). The region sits at the edge of the Agulhas Bank, a broad, shallow continental shelf which would have transformed into a vast expansive coastal plain during marine regressions during glacial periods (Compton 2011; Marean et al. 2020). The GCFR has an abundance of unique endemic plants (Bergh et al. 2014), geophytes (Proches et al. 2005), shellfish (Jerardino and Marean 2010; Langejans et al. 2012) and small and large fauna (Klein 1980; Skinner and Chimimba 2005; Faith 2011), making this area ideal for foraging communities (De Vynck et al. 2019). The rockshelters discussed here are caverns of limestone-rich Bredasdorp Group formations wave-cut from shear zones within Table Mountain Sandstone (Malan 1989). In the following section, the site background, and faunal and taphonomic data for each site, are described. I report the levels of occupational intensity determined by previous studies for each layer/unit. Non-faunal comparative proxies of occupational intensity and sediment accumulation rates for each layer are also noted. Finally, a PCA is used to group layers/units from each site based on various faunal indicators. Klipdrift Shelter Klipdrift Shelter (KDS) is approximately 150 km east of Cape Town in the De Hoop Nature Reserve. Noteworthy finds include engraved ostrich eggshell and worked ochre (Henshilwood et al. 2014). The deposits at KDS were dated by OSL to between c. 52 ka for the uppermost layers (PAN/PAO) and c. 72 ka for the lowest deposits (PE). Most of the analysed archaeological materials were recovered from between layers PAY near the top (dated to c. 60 ka) to the lower layer, PCA (c. 64 ka) (Henshilwood et al. 2014). Formal tools in layers PAY to PCA are typical of the Howiesons Poort with segments, backed and notched tools, borers and retouched blades. Quartzite, quartz and silcrete are the most common raw materials (Douze et al. 2018). The KDS sequence is sometimes grouped into three phases: an upper phase (PAY and PAZ), a middle phase (PBA/PBB, PBC and PBD) and a lower phase (PBE and PCA). These groupings are loosely based on taxonomic composition and stratigraphy (Henshilwood et al. 2014; Reynard et al. 2016b). Equids, angulate tortoises, Raphicerus and hyraxes are the most common taxa at KDS. Eland, alcelaphines (wildebeest, hartebeest and blesbok/bontebok) and lagomorphs also occur. Most of the ungulate taxa are grazers suggesting that the environment was most likely dominated by grasslands, especially in the middle phase. Fisher’s alpha values peak in layers PBA/PBB, PBD and PCA (Table 1). Small fauna is abundant in the upper layers, with equids and alcelaphines more common in the middle layers. The lowest Howiesons Poort layer, PCA, has proportionally more larger bovids (size 3 and 4) (Fig. 2). There is little evidence of carnivore activity which suggests that humans were the main accumulators (Reynard et al. 2016b). The presence of burning, Site/layer Occupation level Taphonomy Trans. frac. Density Burn Cut Percussion Tooth Trampling Fauna Faunal abundance Shellfish Diversity Small mammal Large ungulate Klipdrift PAY Low 0.000 0.078 0.000 0.500 0.250 0.126 12960 15 1.21 0.508 0.115 PAZ Moderate 0.444 0.124 0.222 0.777 0.000 0.138 13797 46 3.49 0.333 0.424 PBA/PBB High 0.155 0.120 0.094 0.500 0.000 0.352 24869 168 6.45 0.056 0.381 PBC High 0.378 0.190 0.143 0.595 0.000 0.217 60279 423 3.47 0.212 0.397 PBD High 0.467 0.236 0.091 0.606 0.000 0.169 36585 141 4.64 0.200 0.295 PBE Low 0.142 0.005 0.286 0.429 0.000 0.133 20537 126 3.18 0.063 0.250 PCA Moderate 0.393 0.108 0.071 0.857 0.000 0.216 10101 80 5.69 0.110 0.537 0.283 0.123 0.130 0.609 0.036 0.193 25590 143 4.02 0.212 0.343 Mean values Blombos CA Moderate 0.100 0.160 0.091 0.273 0.273 0.107 78812 3220 7.48 0.111 0.365 CB High 0.453 0.274 0.177 0.177 0.118 0.230 63833 4290 3.54 0.111 0.556 CC High 0.395 0.167 0.186 0.233 0.163 0.320 77670 2550 4.93 0.191 0.393 CD Low 0.125 0.433 0.143 0.143 0.000 0.106 66148 1987 3.32 0.277 0.200 CF High 0.500 0.113 0.073 0.244 0.122 0.160 40210 6627 3.92 0.107 0.202 0.315 0.229 0.134 0.214 0.135 0.185 65334 3735 4.638 0.159 0.343 Mean values Tracking Occupational Intensity Using Archaeo-faunal Data: Case... Table 1 Faunal and taphonomic data from Klipdrift Shelter, Blombos Cave and Pinnacle Point 13B. Taphonomic data in frequencies. Densities in counts per cubic metre. Diversity = Fisher’s alpha. Small mammal and large ungulate abundance in frequencies. Occupational levels based on previous analyses (KDS: Reynard et al. 2016b, BBC: Reynard and Henshilwood 2019, PP13B: Marean 2010). References for data in ‘Methods’ section Pinnacle Point 13B High 0.220 0.442* 0.109 0.449 0.055 nd 8427 4533 6.29 0.020 0.265 Upper DBS Moderate 0.210 nd 0.064 0.379 0.136 nd 4174 695 5.95 0.074 0.482 LBGS Low 0.240 0.257* 0.072 0.289 0.157 nd 1682 33 9.86 0.037 0.519 223 SBS/URS 224 Table 1 (continued) Site/layer Occupation level Taphonomy Trans. frac. Density Burn Cut Percussion Tooth Trampling Fauna Faunal abundance Shellfish Diversity Small mammal Large ungulate LRS Moderate 0.260 0.410* 0.076 0.288 0.046 nd 3016 2789 5.25 0.100 0.300 LC-MSA Lower Moderate 0.190 0.568* 0.064 0.320 0.064 nd 6380 643 16.36 0.000 0.462 0.224 0.419* 0.077 0.345 0.092 – 4736 1739 8.742 0.046 0.406 Mean values *Magnetic susceptibility (excluded from combined principal component analysis) Reynard Tracking Occupational Intensity Using Archaeo-faunal Data: Case... 225 and cut and percussion marks on the Cape dune molerat, hyrax, hare and carnivore remains indicate that humans also accumulated these taxa. Taphonomic data suggest that the middle phase was the most intensely occupied period at KDS (Table 1). Burning is common throughout the sequence but spike in the middle layers (Reynard et al. 206b). Percussion marks are high throughout and pulse in PAZ and PCA, two moderately occupational layers. Cut marks are highest in PBE and lowest in PAY. Carnivore tooth marks only occur in PAY. Faunal density—in terms of both weight and the number of specimens (NSP)—is highest in the middle phase, PBA/PBB, PBC and PBD (Table 1). Shellfish density is also highest in those layers. There is a significant correlation between faunal NSP/m3 and shellfish MNI/m3 (rs = 0.857; p = 0.014) through the sequence. The proportion of grazers is also significantly correlated to bone density (kg/m3) and shellfish MNI/m3 (rs = 0.857; p = 0.014). Trampling data in the form of bone abrasion was documented at KDS. Given that burning sometimes results in sheen on bone surfaces, burnt bone was removed from calculations of abrasion. Abraded bone is more common in the middle layers and less so in the upper layers. There are significant relationships between the proportion of abraded specimens versus bone density (kg/m3) (rs = 0.786; p = 0.036), grazing ungulates (rs = 0.964; p < 0.001) and the frequency of all cut marks (rs = 0.893; p = 0.007) through the sequence (Reynard et al. 2016a). In fact, for all specimens, cut mark frequencies are also significantly correlated to bone density (kg/m3) (rs = 0.786; p = 0.0362). Bone fragments are generally smaller in PBC and PBD (Reynard et al. 2016b). Unlike in any other layers, transverse breakage is more common than spiral breakage in PBD suggesting more occupational events in PBD and the layers above it. Accumulation rates are difficult to estimate accurately given the lack of chronologically stratified dates through the sequence (Table 2). The KDS sequence is tightly constrained within relatively short timespans. The top three Howiesons Poort layers (PAY, PAZ and PBA/PBB) have statistically similar dates of c. 60 ka, while the lower layers (PBC to PCA) centre around c. 64 ka. PAY has similar dates to overlaying layers 100% 7 90% 62 80% 14 48 8 58 73 70% 23 60% 50% 8 106 57 40% 22 71 30% 48 31 20% 11 31 10% 42 2 7 0% PAY PAZ PBA/PBB Small mammal PBC Smaller bovid PBD PBE 15 PCA Larger ungulate Fig. 2 Faunal proportions at Klipdrift Shelter. NISP in columns (data from Reynard et al. 2016b) 226 Table 2 Accumulation rates at Klipdrift Shelter, Blombos Cave and Pinnacle Point 13B Layer/unit Mean OSL age Shoreline distance (km) Deposit timespan (ka) Estimated deposit thickness (cm) Accumulation rate (cm/ka) Klipdrift PAY 60 < 10 −1 15 PAZ 59 < 10 0 5 − 15 0 PBA/PBB 59 < 10 7 9 1.3 PBC 66 > 12 −1 9 −9 PBD 65 > 12 −1 10 − 10 PBE 64 > 12 0 3 0 PCA 64 > 12 8* 7 0.9 CA 73 3–11 0.5 4 8 CB 73.5 3–11 0.5 3 6 CC 74 3–11 1 8 8 CD 75 3–11 1 10 10 CF 76 2–4 6* 15 2.5 93 (98–91) 1–2 7 30 4.3 Blombos Pinnacle Point 13B SBS/URS Upper DBS 97 (102–91) 2–3 11 12 1.1 LBGS 100 (134–94) 0–1 40 45 1.1 LRS 110 (114–106) 1–2 8 5 0.6 LC-MSA Lower 162 (174–153) 5–6 21 30 1.4 Fisher et al. 2010; Henshilwood et al. 2014; Reynard et al. 2016a Fisher et al. 2010; Henshilwood et al. 2011; Jacobs et al. 2013; Jacobs et al. 2020; Haaland et al. 2021 Marean et al. 2010; Jacobs 2010; Karkanas and Goldberg 2010 Reynard *Based on OSL ages for underlying PBE layer of 72 ka at KDS and underlying CG layer of 82 ka at BBC References Tracking Occupational Intensity Using Archaeo-faunal Data: Case... 227 suggesting that depositional rates may have been relatively rapid during this later period. Non-faunal Proxies The upper layers (PAY and PAZ) contain lighter, grittier sediment with little visible material but with some evidence of hearth features. PAY appears to be a distinct geogenic event that may have occurred rapidly (M. Haaland, pers. comm. 6 Jan. 2021). The stratigraphy of the middle layers (PBA/PBB, PBC and PBD) consists of alternating organic-rich, dark, grey and white lenses with numerous artefacts, bone and shell visible in profile (Henshilwood et al. 2014). Much of the faunal remains were recovered from these darker layers. Lithic data show significant changes in raw material patterns and blank production in the middle layers (Douze et al. 2018). Shellfish patterns also show a turnover in PBC from a dominance of Dinoplax gigas to Haliotis midae. PBE is a grey, ashy layer with more combustion features in PCA. No micromorphological analyses were conducted, but the data show good evidence for increased occupational intensity in the middle phase at KDS, with lower occupations in the upper layers (PAY and PAZ). Principal Component Analysis A PCA for KDS indicates that the first and second components explain 42% and 19.8% of the total variance respectively (Fig. 3). Except for tooth marks and small mammals, all proxies show positive loadings for component 1. Burning, tooth and trampling marks and fauna and shellfish density have positive loadings for component 2. The PCA groups the middle layers (PBA/PBB, PBC and PBD) in the 1st quadrant suggesting that these layers share common faunal signals. PAY and PBE show negative values along axis 1, with PAY particularly isolated from the other layers. Fig. 3 Principal component analysis of layers from Klipdrift Shelter 228 Reynard Blombos Cave Approximately 45 km east of KDS is BBC, a rockshelter about 35 m above sea level and 100 meters from the current shoreline. The MSA layers are divided into four phases. The uppermost M1 phase, dated by OSL to c. 74 ka (Jacobs et al. 2013), consists of layers CA down to CD. The upper M2 phase beneath that is composed of the CF layers and is dated to c. 76 ka (Jacobs et al. 2020). The lower M2 phase (layer CG) is dated to c. 82 ka (Jacobs et al. 2020) while the bottom M3 phase (layers CH to CP) dates to over c. 100 ka (Henshilwood et al. 2011). The M1 and upper M2 phases are linked to the Still Bay techno-complex (Henshilwood et al. 2001a). These phases contain Still Bay bifacial points, engraved ochre plaques, perforated shell beads, formal bone tools, engraved bone and a possible ochre drawing (Henshilwood et al. 2001a, 2001b, 2002, 2004, 2018; d’Errico and Henshilwood 2007). The lower M2 phase is generally less dense than the M1 and is considered a low-occupation phase (Henshilwood et al. 2001a). The pre-Still Bay M3 phase has yielded an ochreprocessing workshop (Henshilwood et al. 2011). Because layer-specific taphonomic assessments have only been conducted in the Still Bay layers, only the M1 and upper M2 phases are used in this study. Hyrax are very common at BBC while animals associated with scavenging or denning such as honey badgers and jackals are relatively rare (Henshilwood et al. 2001a). Small mammals are prevalent in CD while larger ungulates are most common in CF (Fig. 4). On the whole, larger ungulates are not as common at BBC as they are at the other sites under review. Ungulate remains show that the environment from 80 to 70 ka was browse-dominated, with evidence of more grazers during the > 100 ka phase (Badenhorst et al. 2016; Reynard and Henshilwood 2019). Seal remains are also quite common, particularly in the M1 and M3. Fisher’s alpha values are generally high in layers CA, CC and CF (Table 1). Shellfish density at BBC correlates well with marine regression palaeoscape models (Fisher et al. 2010). They suggest marine regression during the Still Bay period (M1 phase)—with a closer shoreline in the early Still Bay (layer CF)—and the shoreline closest during the CI layers of the M3 phase (Henshilwood et al. 2001a; Fisher et al. 2010). Taphonomic data show that tooth marks are highest in CA, dropping significantly in CD (Table 1). Butchery marks pulse in CB and CC and are lowest in the early Still Bay (CF). Burning is highest in CD with percussion marks spiking in CA, CC and CF. In the mid-Still Bay, bone fragments are smaller and transverse fractures are more common in CB and CC (Reynard and Henshilwood 2018). Faunal density is highest in CA and CC (Table 1). Animal-induced modification and transverse fractures are high in CF. Trampling modification and bone abrasion are more common in CC and CF. Indeed, trampling marks and abrasion are significantly correlated in the Still Bay sequence (rs = 0.900; p = 0.0370) (Reynard and Henshilwood 2018). BBC shows relative depositional consistency of dune and marine sands throughout the sequence. OSL ages are bracketed in MIS 5a and generally occur within a chronological sequence from c. 77 (the oldest layer in CF) to c. 73 ka in CA (Table 2). Deposit thickness varies greatly due to roofspall, but the upper layers (CA and CB) appear relatively thin, while the lower CD and CF layers are thicker deposits. Estimated accumulation rates range from relatively slow deposition during CF (c. 2.5 cm/ka) to rapid deposition during CD (c. 10 cm/ka). Tracking Occupational Intensity Using Archaeo-faunal Data: Case... 229 100% 90% 80% 13 23 34 33 25 70% 60% 34 50% 116 40% 35 33 30% 15 20% 18 10% 16 7 5 CA CB 18 0% Small mammals CC Small bovid CD CF Large ungulate Fig. 4 Faunal proportions at Blombos Cave. NISP in columns (data from Reynard and Henshilwood 2019) Non-faunal Proxies Micromorphological and geoarchaeological data suggests that occupational intensity was likely greatest in Layer CI in the c. 100 ka M3 phase. Here, there is more evidence of ash-rich deposits associated with thick combustion features with highly reworked and trampled sediment (Haaland et al. 2021). Layer CF and CC also show evidence of increased site use, although not as intense as in CI. The data indicate that CF experienced numerous periods of human occupation, followed by multiple phases of abandonment. Layer CF and CC display signs of more intense reworking and trampling events that indicate that these were the most occupied periods in the Still Bay period. In contrast, layer CD generally shows more evidence of human abandonment with no anthropogenic material in sub-layer CDB (Haaland et al. 2021). Principal Component Analysis The PCA for BBC shows that the first and second components contain 40.6% and 28.2% of the total variance respectively (Fig. 5). Except for burning, cut marks, faunal density and small mammal frequency, all other indicators display positive loadings for axis 1. Shellfish density and transverse fractures are the only two indicators that have negative loadings on axis 2. Unlike KDS, no discernible groupings are noted. CA and CC show strong signals for most proxies while CD displays the most negative value along axis 1. Pinnacle Point PP lies approximately 80 km east of BBC and about 15 m above the present sea level. It has yielded significant archaeological finds, including some of the earliest evidence of shellfishing (Jerardino and Marean 2010), ochre processing (Marean et al. 2007) and microlithic technology (Brown et al. 2012). The site consists of a series of rockshelters 230 Reynard Fig. 5 Principal component analysis of the Still Bay layers from Blombos Cave including Pinnacle Point 5-6 (PP5-6), Crevice Cave, Pinnacle Point 9, Pinnacle Point 13B (PP13B), and Pinnacle Point 30 (PP30), a brown hyena den. PP13B contains abundant archaeological finds and is the focus of this study. PP13B does not preserve continuous stacked sediments as ‘layers’ but rather as horizontal samples of occupations (Marean 2010: 427). The deposits at PP13B have an extensive age range. Stratigraphic units with evidence of human occupation date from c. 163 ka in LCMSA Lower to c. 90 ka in SBS/URS (Jacobs 2010). The majority of lithic material is composed of local quartzite while fine grained material such as silcrete is relatively rare (Marean 2010). Blades, points and quadrilateral flakes are common and retouch is quite rare (Thompson and Marean 2008). There is relatively early evidence of microlithic technology at PP5-6 at c. 71 ka (Brown et al. 2012), and fire may have been used to heat-treat tools as early as 162 ka in PP13B (Brown et al. 2009). The available faunal sample is small but shows that ungulates comprise almost 98% of the faunal assemblage from PP13B. Springbok and wildebeest are the most abundant taxa, and grey rhebok, hartebeest and Raphicerus are also common (Rector and Reed 2010). Generally, ungulate size classes are relatively evenly distributed suggesting a diverse range of ungulates in the assemblage (Thompson 2010). Fisher’s alpha is highest in the LC-MSA Lower with diversity also high in LBG Sand (Table 1). Overall, there is no significant change in ungulate representation through the sequence indicating similar paleoenvironmental conditions through time. However, large ungulates are more common in the MIS 6 deposits while size 1 bovids dominate the MIS 5 units (Marean 2010). The prevalence of grazing ungulates such as wildebeest and hartebeest does not concord with typical fynbos fauna. Indeed, Raphicerus and hyrax—which generally dominate most southern Cape MSA palaeoscapes—are not particularly common at PP13B suggesting the presence of grasslands and more open habitats here. Although small mammals are included in the PP13B analysis, they are relatively rare with trace amounts of hyrax, dune molerat and springhare (Rector and Reed 2010). Generally, carnivores and raptors were not major bone accumulators suggesting that the faunal assemblages were mostly the result of human activities (Thompson 2010). Tracking Occupational Intensity Using Archaeo-faunal Data: Case... 231 Taphonomic data generally correspond to changing occupations as represented by density and magnetic susceptibility data. At PP13B, there is evidence that some small mammals were processed by people (Thompson 2010). For larger mammals, percussion marks are very common (Table 1). Percussion and cut marks are more abundant in SBS/URS units (Thompson et al. 2017), which concurs with peaks in shellfish, lithic and faunal densities (Fig. 6). Similarly, the proportion of tooth-marked and gastricacid-etched bone is lowest in those layers. Bone breakage patterns in the various deposits generally do not correspond well with higher occupational phases. Rightangled breaks on large ungulates are prevalent in the LC-MSA Middle unit and less common in the upper DB Sands (Thompson 2010). Non-faunal Proxies Micromorphological analyses, shellfish, fauna, ochre and lithic densities, and archaeomagnetic data (‘magnetic susceptibility’), have been used to infer occupational intensity at PP13B (Fig. 6). Magnetic susceptibility is linked to burning intensity and hearth production. SBS/URS contain high densities of shell, lithics, bone and high magnetic susceptibility values. Peaks of lithic, faunal and shellfish densities and heavily trampled and reworked sediment suggest high occupations here (Marean 2010). Accumulation rates appear highest in these units (4.3 cm/ka) but are generally consistent through the sequence (Table 2). Upper DB Sands is a dark brown, organic-rich, gritty layer with less bioturbation. Occupations are considered moderate in these units (Karkanas and Goldberg 2010; Marean 2010). LBG Sand are light brown/grey sediments with inputs of aeolian sand and roofspall. The high geogenic content suggests this was a low occupational period. LRS is a loose matrix containing roofspall with combustion features and accumulations of bone and shell. Micromorphological Fig. 6 a Faunal; b lithic and c shellfish densities, and d magnetic susceptibility at Pinnacle Point 13B (data from Marean 2010) 232 Reynard evidence of reworked and trampled bone suggests moderate occupations in this unit. At 0.6 cm/ka, accumulation rates are lowest here (Table 2). LC-MSA Lower is characterised by multiple lens of heavily burnt material containing bone, shell and lithics. Micromorphological analyses indicate moderate reworking of these sediments by trampling or hearth cleaning (Karkanas and Goldberg 2010). Principal Component Analysis Burning was replaced by magnetic susceptibility in the PCA for PP13B. The PCA indicates that the first and second components explain 46.6% and 33.3% of the total indicators respectively (Fig. 7). Cut and percussion marks, magnetic susceptibility and fauna and shellfish density show positive loadings for component 1, with tooth marks and small mammals displaying high negative values. Magnetic susceptibility, percussion marks, faunal density, small mammals and especially diversity show positive values for component 2. Units at PP13B are particularly dispersed with no obvious grouping discernible. SBS/URS display especially high positive values along axis 1, with LC-MSA Lower showing the highest positive value on axis 2. Figure 8 shows a combined PCA for all sites. The first and second principal components explain 40.2% and 17.4% of the total variance respectively. Based on the PCA and a hierarchical cluster analysis, layers/units were clustered into nine groups. Generally, these groups correspond to the three sites. Loadings for component 1 is dominated by trampling marks and diversity, with large ungulates also showing relatively high positive values. Burning, small mammals and faunal density yield high negative loading values for PC1. PP13B units show strong positive values for axis 1 with the highest value corresponding to LC-MSA Lower. Component 2 has strong positive loadings for faunal and shellfish density and tooth marks, with percussion marks, cut marks and small mammals yielding high negative values. Most BBC layers have high positive values on axis 2, with PCA and PAZ showing the highest negative values. Fig. 7 Principal component analysis of selected stratigraphic units at Pinnacle Point 13B Tracking Occupational Intensity Using Archaeo-faunal Data: Case... 233 Fig. 8 Principal component analysis of layers/units from Klipdrift Shelter, Blombos Cave and Pinnacle Point 13B. Squares = Klipdrift Shelter; circles = Blombos Cave; triangles = Pinnacle Point 13B Discussion Diversity Faunal diversity corresponds well with higher occupational phases at KDS and BBC (Table 3) but is a problematic indicator for human occupations because it also reflects animal activity. Predator dens generally have higher levels of diversity than archaeofaunal assemblages (Henschel and Skinner 1990; Hayward et al. 2006). PP30, for example, is significantly more taxonomically diverse than PP13B (Rector and Reed 2010; Reynard and Wurz 2020), yet, at PP13B, faunal diversity is also high in low occupational units (Marean 2010). While there is some expectation that lower occupations correspond to higher carnivore activity, lower diversity values at KDS during lessintense occupational phases suggest that this was not always the case. Indeed, human and animal seasonal movements may mean that diversity values likely reflect overlapping and corresponding site use between humans and other animals. The frequency of small mammals is inversely correlated with diversity through all sites (rs = − 0.717; p = 0.001), suggesting that changing diversity is more likely driven by medium-sized and larger herbivores. There is a weakly significant correlation, for example, between large ungulates and diversity (rs = 0.463; p = 0.061). Small Fauna Small mammal abundance pulse in both higher and lower occupational periods. At KDS, for example, small mammals are common in both low occupational PAY and the more intensely occupied PBC and PBD layers, while at BBC they are more abundant in low occupational CD. Given that these taxa are frequently exploited by cave-dwelling raptors (Cruz-Uribe and Klein 1998), they may not be a particularly good proxy for occupational intensity. Yet taphonomic analyses—key in differentiating humans from animal accumulators—show that small fauna were regularly exploited by people during 234 Table 3 Summary of faunal and taphonomic data and occupational intensity for the sites under review Site Proxies High occupations Klipdrift Layers/periods Middle phase (PBA/PBB, PBC, PBD) Shelter Diversity High Low occupations References Upper phase (PAY, PBE) Henshilwood et al. 2014; Reynard et al. 2016a, 2016b; Douze et al. 2018 Low Small fauna Common in PBC and PBD Most prevalent in PAY Larger fauna Common in PBA/PBB and PBC, most prevalent in PCA Less common in PAY Faunal density High Low Shellfish density High Low Taphonomy Percussion marks and transverse fractures more common, burning Percussion marks and transverse fractures less (esp. calcined bone) more common, anthropogenic marks common, more tooth marks, anthropogenic marks common on small fauna, trampling modification high rare on small fauna, trampling modification low Other proxies Darker, more organically rich layers; more ochre and very high lithic density in PBA/PBB, more engraved ostrich eggshell Blombos Layers/periods Layers CI*, CF and CC, possibly CA? Cave Diversity High in CA and CC Lighter-coloured, inorganic, sandy layers; less ochre and lowest lithic densities in PAY; less engraved ostrich eggshell Layers CG*, CD Low Small fauna Small mammal abundance low Small mammals common Larger fauna Larger ungulates common from CA to CC Least prevalent in CD Moderate Shellfish density High in CI, moderate in CF, low in CC Low Taphonomy Cut marks high in CB and CC, low in CF; percussion marks high Cut and percussion marks low, transverse fractures less in CA, CC and CF; trampling more common in CB and CC; common, bone abrasion low, less trampling marks, transverse fracture more common; bone abrasion high burning more common, tooth marks lowest Other proxies %Retouched lithics high in CC, more micromorphological Geogenic/sandy, coarse-grained micro-facies in CD trampling and reworking in CC, more worked bone in CC and and CG; less anthropogenic deposit in CD; lower CF artefact abundance in CG Reynard Faunal density High in CC, moderate in CF Henshilwood et al. 2001a; Reynard and Henshilwood 2019; Haaland et al. 2021 Site Proxies High occupations Pinnacle Layers/periods SBS/URS Point Diversity Moderately high 13B Small fauna Not common Larger fauna Least common Low occupations References LBGS Karkanas et al. 2010; Marean et al. 2010; Thompson 2010; Thompson et al. 2017 High Not common but slightly more small mammal More common Faunal density High Low Shellfish density High Low Taphonomy Cut and percussion marks more prevalent, less tooth marks and gnaw marks Lower faunal density, less anthropogenic marks, more tooth and gnaw marks Other proxies Higher lithic and ochre density, extensive micromorphological trampling and reworked sediment Lower lithic and ochre density; high geogenic sediment, little reworked sediment *No fauna analysed from these layers Tracking Occupational Intensity Using Archaeo-faunal Data: Case... Table 3 (continued) 235 236 Reynard the Late Pleistocene in the southern Cape (Badenhorst et al. 2014; Armstrong 2016; Reynard and Henshilwood 2017), so pulses in small fauna abundance in high occupational layers at these sites may suggest subsistence intensification periods. This is particularly evident in the Howiesons Poort at KDS which may be linked to environmental change or technological developments then (Reynard and Henshilwood 2017). Moreover, in comparing taphonomic analyses of small mammals from PP5-6 and Die Kelders 1, it seems that small prey were less exploited during the earlier MIS 5 (PP5-6) than later MIS 4 periods (Die Kelders) (Armstrong 2016). Because layer-specific tortoise data were not available for all sites, these were not included in this review. Yet frequencies of tortoise to small mammal may be good indicators for intensification. Tortoises are common at all sites, except at PP13B which has a richer tortoise assemblage. This richness may be environmentally driven or a product of taphonomic bias. Tortoises may be particularly susceptible to overexploitation because they are easily caught and have slow growth rates (Thompson and Henshilwood 2014). Due to their high caloric value and low handling cost, tortoises are considered high-ranked game (Stiner et al. 2000). Because tortoises are highly ranked prey, with periods of intensification, we should expect to see increases in small mammals as tortoise abundance decreases (Stiner et al. 2000). This appears to be the case at BBC where the frequency of tortoise abundance is significantly and inversely proportional to the frequency of small mammal abundance through the Still Bay sequence (rs = − 0.900; p = 0.037), although it must be noted that small mammals here include size 1 bovids. While it is possible these changes may reflect taphonomic or environmental biases, it could indicate periods of intensification, particularly in high occupational CC (Reynard and Henshilwood 2018). At KDS, tortoise remains drop slightly in the more intensely occupied middle phase, corresponding to an increase in small mammals. There is also a weakly significant, inverse correlation between the proportion of tortoise remains and the frequency of abraded specimens through the sequence (rs = − 0.679; p = 0.094). Since bone abrasion may signify an increase in human activities at sites, this may imply that as occupations intensified, tortoises were overexploited. It would be interesting to examine tortoise and small mammal data from earlier MSA periods to explore whether these tortoise/small game ratios extend beyond the Still Bay and Howiesons Poort periods. However, non-mammalian taxa such as birds and fish are critical in assessing whether intensification models hold true. Larger Fauna Increased frequencies of larger ungulates cannot be confidently linked to higher occupational layers (Table 1). The exception is KDS. Here, the abundance of equids and alcelaphines in the middle phase corresponds to a change in raw material procurement patterns with a prevalence of quartz and a drop in silcrete during this period (Douze et al. 2018). Most of the ostrich eggshells were also recovered in the middle layers, suggesting an increase in the use of containers to transport water during that period (Henshilwood et al. 2014). Increases in occupational intensity here may therefore be linked to more extensive foraging ranges. Both larger residential groups and increased logistical mobility are associated with expanded foraging ranges. More intensified site use may also be a result of sites being visited more often or for longer periods of time. Tracking Occupational Intensity Using Archaeo-faunal Data: Case... 237 Given that large fauna abundance also reflects environmental changes, these data underscore the challenges of using faunal abundance to differentiate mobility patterns and to track occupational patterns. At BBC, environmental conditions likely had the most significant impact on changes in large fauna abundance. Although occupational intensity fluctuates from ~ 100 ka, there is a general decrease in the abundance of large fauna from this time (Thompson and Henshilwood 2014) indicating that fluctuations in larger fauna abundance are more likely driven by environmental conditions than occupational patterns (although these may be linked). Changing habitats would have affected large herbivory abundance and human settlement patterns. Expanding grasslands during periods of marine regressions, for example, would have attracted larger, more gregarious game to the region, and may have stimulated increased occupations at these sites. Increases in frequencies of larger bovids during periods of occupational intensity in the southern Cape may indeed speak more to the links between ameliorated environmental conditions and increased site use than occupational intensity per se (cf. Marean et al. 2014; Reynard and Henshilwood 2019). Taphonomy Percussion marks are significantly and inversely correlated with tooth marks through all sites (rs = − 0.491; p = 0.046) (Table 1) which suggests that carnivore activity increases in layers with lower human occupations. The utility of processing marks, however, does vary in each site. At PP13B and KDS, percussion and cut mark frequencies appear to provide good indicators of increased human site use, with both indicators prevalent in higher occupational layers (Table 1). At BBC, percussion mark frequencies also track occupations with high frequencies in the more intensely occupied CA, CC and CF layers. Magnetic susceptibility—associated with burnt stratigraphic units—is a good indicator of site use at PP13B and corresponds with increases in faunal, lithic and shellfish densities (Herries and Fisher 2010; Marean 2010). At BBC, burnt bone does not correlate well with occupational pulses and is more common in the lower occupational CD period and rare in high occupational CF. Fracture patterns may be useful indicators of occupations. At all sites, transverse breakage frequencies are greater in layers showing more intense occupations. Transverse fractures are also more common in the higher occupational layer 10 at Die Kelders 1 compared to lower occupation layer 11 (Marean et al. 2000b). At Klasies River, taphonomic analyses of the Howiesons Poort shows that the more intensely occupied CPx4 layer yielded more transverse fractures than the less occupationally intense YSx5 layer (Achieng 2019). An abundance of transverse fractures could also be the result of other taphonomic processes such as an increase in weathered bone (although weathering is relatively low at these sites). Using fracture patterns as a proxy, therefore, necessitates a detailed understanding of the taphonomic context of an assemblage. Trampling Trampling is a good proxy of site-use intensity and corresponds well with other indicators of occupational intensity at BBC and KDS (Discamps and Henshilwood 2015; Reynard et al. 2016a; Reynard and Henshilwood 2018). At Klasies River, 238 Reynard trampling marks are significantly more common in the more intensely occupied CPx4 layer than the lower occupational YSx5 layer of the Howiesons Poort assemblage (Achieng 2019). One issue though could be equifinality with trampling marks difficult to differentiate from some BSM (Domínguez-Rodrigo et al. 2009, 2012; Backwell et al. 2012; Fernández-Jalvo and Andrews 2016). Trampling abrasion may be the result of continuous trampling events over extended periods, and could be a better indicator of occupational intensity than lines, pits or other types of trampling marks (Reynard and Henshilwood 2018). At both KDS and BBC, bone abrasion tracks occupational intensity well. The CPx4 layer in the Howiesons Poort at Klasies River also had significantly more abraded bone than layer YSx5 (Achieng 2019), and Madgwick (2014) also used abrasion to infer occupational events at seven Late Bronze Age/Early Iron Age sites in Britain. The advantage of abrasion is that it is easy to discern, requiring only a hand lens (Fig. 9). Furthermore, all bone Fig. 9 Examples of trampling abrasion (a) and trampling marks (b) from Blombos Cave. Bar = 1 cm (a), 2 mm (b) (photos by J.P. Reynard) Tracking Occupational Intensity Using Archaeo-faunal Data: Case... 239 fragments over 1 cm2 can be assessed for bone polish (Madgwick 2014) resulting in increased examinable datasets (Reynard and Henshilwood 2018). Equifinality is still an issue since trampling polish may resemble abrasion caused by post-depositional movement, wind, burning, water wear and acid-etched sheen (Brain 1967; Nicholson 1993; Boschian and Saccá 2010; Malinsky-Buller et al. 2011; Fernández-Jalvo and Andrews 2016). Another issue may be bone preservation with lower levels of trampling likely affected by badly preserved bone cortical surfaces (Clark 2019, p. 22). It must be noted that a lack of trampled bone does not necessarily imply lower occupations. Site maintenance, for example, may have resulted in bone fragments being disposed of in hearths, or outside activity areas, with less chance of being trampled. Moreover, changing frequencies of BSM may be the result of fluctuations in sedimentation rates which will be discussed later. Fauna and Shellfish Density Fragmentation data appear to track occupational intensity well. Increases in bone fragmentation are generally more evident in higher occupational layers at all three sites (Table 1). Higher faunal densities and more severely fragmented assemblages often occur in intensely occupied Howiesons Poort phases at sites such as Diepkloof Rockshelter (Steele and Klein 2013) and Sibudu (Clark 2017). At Die Kelders, faunal density is significantly greater in the higher occupational layer 10 than in layer 11 (Marean et al. 2000b). Faunal density is also higher in the MSA II lower layers at Klasies River (Reynard et al. in prep.), a period associated with the MIS 5c high sea level stand (Van Andel 1989; Carr et al. 2016). Shellfish density is significantly correlated to fauna density through all three sites when shellfish weight per volume is used (rs = 0.853; p < 0.0001) which suggests close links between coastal settlement and occupational intensity. The close connection between occupations and shoreline is seen in the M3 phase at BBC where the highest occupational layer at BBC is considered to be CI which also yields the densest shell midden (Jacobs et al. 2006; Haaland et al. 2021). Accumulation Rates Depositional rates may have had significant effects on fragmentation and taphonomic data. Indeed, the PCA for all layers show that most layer groupings are associated with their corresponding sites suggesting that local depositional histories had a significant impact on settlement patterns at each site (Fig. 8). At KDS, taphonomic data show remarkably similar correlations so it is possible that site formation processes here may have affected these trends. Deposits at KDS have been severely truncated by rising sea levels resulting in a talus with a seaward slope of over 38° (Henshilwood et al. 2014). PCAs for layer PAY at KDS show its uniqueness among most other site layers because of low anthropogenic inputs (Figs. 3 and 8; Table 3). Given its relative thickness and the similar OSL ages between the layers above and below, it may suggest high depositional rates for PAY which could account for low taphonomic frequencies and density values in that layer. Still, there is little non-faunal evidence for intensive occupations in this period. These data suggest that KDS may have had two major occupational phases: earlier, more intensive occupations followed later by possibly 240 Reynard more sporadic ones. It is also possible that the severe truncation of the deposits may have resulted in some mixing of the sediments. Additional geoarchaeological and micromorphological data from this site could reveal more detailed information on settlement patterns. At BBC, accumulation rates are especially complicated by significant variance in deposit thickness between—and within—spatial units across the site. Based on my general estimations, depositional rates appear quite high in the low occupational CD which may have impacted density values and taphonomic frequencies in this layer. Slow depositional rates are apparent in the high occupational CF. The significant variation in accumulation rates and levels of occupation between these layers suggests a link between depositional and occupational intensity at BBC. Layer CD shows some interesting patterns and may reflect an environmental shift in the Still Bay sequence. It has significantly more small mammals and size 1 bovid remains than other layers, and shows significantly different skeletal element profiles than any other layer (Reynard and Henshilwood 2019). The earlier Still Bay (CF and CD) also has significantly lower ungulate diversity values than the later phase (CA to CC) suggesting possible shifts in moisture availability between these periods (Reynard and Henshilwood 2019). Layer CD may therefore reflect a period of changing environmental conditions at BBC. Deposits at PP13B have accumulated over an extensive period, yielding an over 80,000-year sequence. The site was excavated within natural lens—later grouped in stratigraphic aggregates—but in different locations of the cave. It may therefore be challenging to assess accumulation rates sequentially. Accumulation rates are generally lower in LRS (the low occupation unit) and higher in SBS/URS (the most intensely occupied unit). It is unlikely that high frequencies of occupational indicators were affected by lower sedimentation rates at PP13B which suggests that density-related proxies of occupations may be accurate reflections of site-use intensity. It must be emphasised again that these accumulation calculations are gross estimates of depositional patterns at these sites. For all sites, layer thickness may vary considerably (even within quadrates) and OSL ages have broad age ranges that may not accurately reflect the mean dates attributed to layers/units. Occupational Trends in the Southern Cape Occupational intensity data support arguments that suggest the coast was a focus of population aggregation during the MSA (Table 3) (e.g., Oppenheimer 2009; Parkington 2010; Marean 2010; Marean et al. 2014). The KDS deposits show two phases of occupations constrained within about 5000 years in MIS 4. While high shellfish densities suggest that the earlier period may be linked to closer sea levels, it is also likely that the shoreline would have been quite far away at c. 64 ka during MIS 4–based sea level regression models (Table 2; Fisher et al. 2010). The later occupational phase was probably more ethereal with less site activity. At BBC, the most intense occupations also occur when the shoreline was closest in layer CI dated to c. 100 ka (Haaland et al. 2021). Occupations also pulse in the early Still Bay (CF)—with this period possibly involving longer-term site use—and in the middle Still Bay (CC) with more frequent occupational events evident here (Reynard and Henshilwood 2018; Haaland et al. 2021). Proximity to the coast is therefore not always associated with increased occupations. At PP, for example, there is good evidence of high occupations when Tracking Occupational Intensity Using Archaeo-faunal Data: Case... 241 shellfish density is low, implying these continued even during marine regressions (Fig. 6). Generally, higher occupations occur in units dated to MIS 5d and early MIS 5e when the coastline is relatively close (Karkanas and Goldberg 2010; Marean 2010). Table 3 summarises faunal indicators at the three sites and shows how these proxies are not always linked to occupational intensity. Indeed, faunal and other data reveal complex pulses of occupational intensity within—and between—these sites. At PP13B, we see an early onset of occupations during glacial MIS 6. These tend to be minimal, probably related to the distant shoreline at that time. In MIS 5, occupations are moderate with evidence of high faunal and lithic densities when the coastline is between 1 to 2 km away (Marean 2010). The most intense occupations appear to occur between 100 and 90 ka. From c. 90 ka, increased dune activity sealed the cave, with occupations only resuming at c. 40 ka (Jacobs 2010). It is not unreasonable to imagine population movements between dispersed PP groups and BBC during this phase. Occupational trends at BBC and KDS may point to inter-site population movements. The lowermost anthropogenically sterile layer at KDS (PE) dates to c. 71 ka— contemporaneous to when occupations cease at BBC (Henshilwood et al. 2001a, 2014)—suggesting population shifts during MIS 4 from BBC to KDS c. 45 km away (Reynard and Henshilwood 2017). This is not to claim that groups occupying each site were bound to these site-specific locales, but it does pose scenarios where the three sites were demographically linked and shared extensive social and residential networks. If we disregard the effects of shoreline fluctuations on coastal settlement, there appears to be a general trend towards increased exploitation of low-ranked resources—and higher occupations—through the Late Pleistocene in the southern Cape. Faunal data from these three sites suggest that occupations may have become more intense from earlier to later MSA periods, with other research showing increased demographic pressure especially evident in the LSA (Steele and Klein 2009; Sealy 2016). Conclusion Zooarchaeological and taphonomic data are an important—but complex—means of tracking occupational intensity. This study suggests that small mammal frequencies may not be effective proxies for occupational intensity in southern Cape sites. Despite overwhelming evidence that Late Pleistocene people here exploited small game, the contemporaneous use of rockshelters by carnivores and humans makes it challenging to disentangle human and animal agency. The ratio of small mammal to tortoise may be useful in tracking periods of intensification here. While an abundance of larger ungulates often reflects more human occupations, carnivore contributions also make this indicator problematic. Faunal diversity is generally higher in intensely occupied layers but also high in predator dens, making this too a challenging indicator for human occupations. Taphonomic analyses are therefore critical in contextualising these indicators as occupational proxies. BSM frequencies generally track occupations well at these sites. Covarying percussion and tooth mark ratios confirm that increased carnivore activity could be a useful indicator of less intense human occupations. Burning and transverse fractures also occur more frequently in higher occupational layers. Increased cut mark frequencies often corresponded to increased site-use activity. Trampling marks and abrasion appear to be good proxies for 242 Reynard occupational intensity and correspond well with geoarchaeological and micromorphological data. Bone abrasion is also a valuable and simple means of assessing increased site use. The advantage of using trampling modification is that all faunal remains, not just identified specimens, can be analysed, increasing sample sizes. A key issue may be different methods used to assess BSM. Research has highlighted the value of standardising BSM analyses (Fernández-Jalvo and Andrews 2016; Thompson et al. 2017; Bertran et al. 2019), which would allow for more accurate data comparisons. Higher faunal densities correspond to increased site use at all sites and frequently correlate well with other proxies. This is probably because activities such as trampling, burning and faunal processing are linked to more fragmented bone. These activities could also be associated with higher frequencies of transverse fractures. Yet an awareness of depositional rates is critical to assessing the utility of density data. Given the short timespan, sediment accumulation rates suggest rapid deposition over two major occupational phases at KDS. Possible high depositional rates in low occupation PAY could account for lower faunal and taphonomic indicators in that layer. At BBC, depositional rates appear to be lowest in the high occupational early Still Bay (CF) and higher in the low occupational CD layer suggesting links between occupational intensity and accumulation rates there. Increased accumulation rates in low occupation CD at BBC may have affected density and frequency values in that layer and may reflect shifting environmental conditions at BBC at that time. At PP13B, accumulation rates are relatively constant through the various units suggesting that density values correspond well to occupations. These accumulation rates, however, are gross assessments based on published section profiles and broadly estimated OSL age ranges. More informed analyses—especially at the quadrate level—may reveal refined depositional histories. Correlations between shellfish, faunal and artefact data from the sites under review support previous studies that show close links between occupational intensity at these sites and marine transgressions. Still, evidence of increased site activity during MIS 4 suggests that, even though this period is linked to increased marine regressions, human occupations at these sites continued—in some cases, increased—at this time. There is abundant evidence for increased occupational intensity in the later Howiesons Poort compared to the earlier Still Bay in this region (Reynard and Henshilwood 2017), suggesting links between demographic pressure and technological development. Archaeo-faunal data have a complex relationship to site-use intensity. Factors such as equifinality, sampling strategies and zoogenic BSM during high occupational periods could also affect the utility of faunal proxies and bias evidence of human occupational activity. Although layer-specific samples are often small, they are useful in highlighting depositional trends at these sites. Despite the shortcomings of some faunal indicators, these data—on the whole—are valuable as proxies if the context of the site is understood. Well-defined, layerspecific taphonomic data—combined with multiple, reliable proxies—show the significance of faunal data in tracking occupational intensity. Acknowledgements I thank two anonymous authors for thought-provoking and insightful comments. Their suggestions contributed immensely to the manuscript. I also thank Magnus Haaland for his helpful advice. Regular grants from the Centre of Excellence in Palaeosciences (CoE-Pal) and the Palaeotological Scientific Trust (PAST) are acknowledged. Data and Material Availability Data was obtained through published studies. Tracking Occupational Intensity Using Archaeo-faunal Data: Case... Code Availability 243 Not applicable. Funding The author was funded by a South African Department of Science and Technology (DST)National Research Foundation (NRF) Thuthuka grant (grant no. 129689) and an Enabling Grant through the Diversifying of the Academy from the University of the Witwatersrand. Any opinion, finding, conclusion or recommendation expressed in this article is that of the author, and the NRF does not accept any liability in this regard. Declarations Conflict of Interest Not applicable. References Achieng, P. (2019). A taphonomic analysis of the large ungulate mammal fauna from the Howiesons Poort sequence at Klasies River site. Unpublished Masters Dissertation. University of the Witwatersrand: Johannesburg. Ambrose, S. H., & Lorenz, K. G. (1990). Social and ecological models for the Middle Stone Age in southern Africa. In P. A. Mellars (Ed.), The emergence of modern humans (pp. 3–33). Edinburgh: Edinburgh University Press. Archer, W. (2020). Carrying capacity, population density and the later Pleistocene expression of backed artefact manufacturing traditions in Africa. Philosophical Transactions of the Royal Society B, 376(1816), 20190716. Armstrong, A. (2016). Small mammal utilization by Middle Stone Age humans at Die Kelders cave 1 and Pinnacle Point site 5-6, western Cape Province, South Africa. Journal of Human Evolution, 101, 17–44. Backwell, L. R., Parkinson, A. H., Roberts, E. M., d'Errico, F., & Huchet, J. B. (2012). Criteria for identifying bone modification by termites in the fossil record. Palaeogeography, Palaeoclimatology, Palaeoecology, 337, 72–87. Badenhorst, S., van Niekerk, K. L., & Henshilwood, C. S. (2014). Rock hyraxes (Procavia capensis) from Middle Stone Age levels at Blombos Cave, South Africa. African Archaeological Review, 31(1), 25–43. Badenhorst, S., Van Niekerk, K. L., & Henshilwood, C. S. (2016). Large mammal remains from the 100 ka Middle Stone Age layers of Blombos cave, South Africa. South African Archaeological Bulletin, 71, 46– 52. Badenhorst, S., Ezeimo, J., van Niekerk, K. L., & Henshilwood, C. S. (2021). Differential accumulation of large mammal remains by carnivores and humans during the Middle Stone Age in the Eastern and Western Cape, South Africa. Journal of Archaeological Science: Reports, 35, 102752. Beckerman, S. (1983). Optimal foraging group size for a human population: the case of Bari fishing. American Zoologist, 23(2), 283–290. Behrensmeyer, A. K., Gordon, K. D., & Yanagi, G. T. (1986). Trampling as a cause of bone surface damage and pseudo-cutmarks. Nature, 319(6056), 768–771. Bergh, N. G., Verboom, G. A., Rouget, M., & Cowling, R. M. (2014). Vegetation types of the Greater Cape Floristic Region. In N. Allsopp, J. F. Colville, & G. A. Verboom (Eds.), Fynbos: ecology, evolution and conservation of a megadiverse region (pp. 1–25). London: Oxford University Press. Bertran, P., Todisco, D., Bordes, J. G., Discamps, E., & Vallin, L. (2019). Perturbation assessment in archaeological sites as part of the taphonomic study: a review of methods used to document the impact of natural processes on site formation and archaeological interpretations. PALEO Revue d'Archéologie Préhistorique, 30(1), 52–75. Binford, L. R. (1978). Nunamiut ethnoarchaeology. New York: Academic Press. Binford, L. R. (1980). Willow smoke and dogs’ tails: hunter gatherer settlement systems and archaeological site formation. American Antiquity, 45(1), 4–20. Binford, L. R. (1981). Bones: ancient men and modern myths. New York: Academic Press. Bird, D. W., Bird, R. B., Codding, B. F., & Zeanah, D. W. (2019). Variability in the organization and size of hunter-gatherer groups: foragers do not live in small-scale societies. Journal of Human Evolution, 131, 96–108. 244 Reynard Blasco, R., Rosell, J., Peris, J. F., Cáceres, I., & Vergès, J. M. (2008). A new element of trampling: an experimental application on the Level XII faunal record of Bolomor Cave (Valencia, Spain). Journal of Archaeological Science, 35(6), 1605–1618. Blumenschine, R. J. (1986). Carcass consumption sequences and the archaeological distinction of scavenging and hunting. Journal of Human Evolution, 15(8), 639–659. Blumenschine, R. J. (1988). An experimental model of the timing of hominid and carnivore influences on archaeological bone assemblages. Journal of Archaeological Science, 15(5), 483–502. Blumenschine, R. J. (1995). Percussion marks, tooth marks and experimental determinations of the timing of hominid and carnivore access to long bones at FLK Zinjanthropus, Olduvai Gorge, Tanzania. Journal of Human Evolution, 29(1), 21–51. Bond, W. J., & Keeley, J. E. (2005). Fire as a global ‘herbivore’: the ecology and evolution of flammable ecosystems. Trends in Ecology & Evolution, 20(7), 387–394. Boschian, G., & Saccá, D. (2010). Ambiguities in human and elephant interactions? Stories of bone, sand and water from Castel di Guido (Italy). Quaternary International, 214(1-2), 3–16. Bousman, C. B. (2005). Coping with risk: Later Stone Age technological strategies at Blydefontein Rock Shelter, South Africa. Journal of Anthropological Archaeology, 24(3), 193–226. Bradfield, J., Lombard, M., Reynard, J., & Wurz, S. (2020). Further evidence for bow hunting and its implications more than 60 000 years ago: results of a use-trace analysis of the bone point from Klasies River Main site, South Africa. Quaternary Science Reviews, 236, 106295. Brain, C. K. (1967). Hottentot food remains and their bearing on the interpretation of fossil bone assemblages. Scientific Papers of the Namib Desert Research Station, 32, 1–11. Brain, C. K. (1981). The hunters or the hunted? An introduction to African cave taphonomy. Chicago: University of Chicago Press. Brown, K. S., Marean, C. W., Herries, A. I., Jacobs, Z., Tribolo, C., Braun, D., Roberts, D. L., Meyer, M. C., & Bernatchez, J. (2009). Fire as an engineering tool of early modern humans. Science, 325(5942), 859– 862. Brown, K. S., Marean, C. W., Jacobs, Z., Schoville, B. J., Oestmo, S., Fisher, E. C., Bernatchez, J., Karkanas, P., & Matthews, T. (2012). An early and enduring advanced technology originating 71,000 years ago in South Africa. Nature, 491(7425), 590–593. Bunn, H. T. (1991). A taphonomic perspective on the archaeology of human origins. Annual Review of Anthropology, 20(1), 433–467. Butzer, K. (1977). Geomorphology of the Lower Illinois Valley, as a spatial-temporal context for the Koster Archaic Site (p. 34). Springfield, IL: Illinois State Museum Reports of Investigations No. Carr, A.S., Chase, B.M., & Mackay, A. (2016). Mid to Late Quaternary landscape and environmental dynamics in the Middle Stone Age of southern South Africa. In B. Stewart & S.C. Jones (Eds.), Africa from MIS 6-2 (pp. 23-47). Springer, Dordrecht. Chase, B. M., & Meadows, M. E. (2007). Late Quaternary dynamics of southern Africa’s winter rainfall zone. Earth Science Review, 84(3-4), 103–138. Clark, J. L. (2011). The evolution of human culture during the late Pleistocene: Using fauna to testmodels on the emergence and nature of “modern” human behaviour. Journal of Anthropological Archaeology, 30(3), 273–291. Clark, J. L. (2017). The Howieson's poort fauna from Sibudu cave: documenting continuity and change within Middle Stone Age industries. Journal of Human Evolution, 107, 49–70. Clark, J. L. (2019). The Still Bay and pre-Still Bay fauna from Sibudu cave: taphonomic and taxonomic analysis of the macromammal remains from the Wadley excavations. Journal of Paleolithic Archaeology, 2(1), 26–73. Collins, B., Wilkins, J., & Ames, C. (2017). Revisiting the Holocene occupations at Grassridge Rockshelter, Eastern Cape, South Africa. South African Archaeological Bulletin, 72, 162–170. Compton, J. S. (2011). Pleistocene sea-level fluctuations and human evolution on the southern coastal plain of South Africa. Quaternary Science Reviews, 30(5-6), 506–527. Courty, M. A., Goldberg, P., & Macphail, R. (1989). Soils and micromorphology in archaeology. Cambridge: Cambridge University Press. Cruz-Uribe, K., & Klein, R. G. (1998). Hyrax and hare bones from modern South African eagle roosts and the detection of eagle involvement in fossil bone assemblages. Journal of Archaeological Science, 25(2), 135–147. Marean, C. W., Bar-Matthews, M., Fisher, E., Goldberg, P., Herries, A., Karkanas, P., Nilssen, P. P., Thompson, E. (2010). The stratigraphy of the Middle Stone Age sediments at Pinnacle Point Cave 13B (Mossel Bay, Western Cape Province, South Africa). Journal of Human Evolution 59(3-4):234–255. Tracking Occupational Intensity Using Archaeo-faunal Data: Case... 245 d’Errico, F., & Henshilwood, C. S. (2007). Additional evidence for bone technology in the southern African Middle Stone Age. Journal of Human Evolution, 52(2), 142–163. De Graaf, G. (1961). Gross effects of a primitive hearth on bones. South African Archaeological Bulletin, 16(61), 25–26. De Vynck, J. C., Difford, M., Anderson, R., Marean, C. W., Cowling, R. M., & Hill, K. (2019). The resilience to human foraging of intertidal resources on the south Cape coast of South Africa and the implications for pre-historic foragers. Quaternary Science Reviews, 235, 106041. Deacon, J. (1984). The Later Stone Age of southernmost Africa. Oxford: British Archaeological Reports International Series 213. Deacon, H. J. (1992). Southern Africa and modern human origins. Philosophical Transactions of the Royal Society of London. Series B: Biological Sciences, 337(1280), 177–183. Discamps, E., & Henshilwood, C. S. (2015). Intra-site variability in the Still Bay fauna at Blombos Cave: implications for explanatory models of the Middle Stone Age cultural and technological evolution. PloS ONE, 10(12), e0144866. https://doi.org/10.1371/journal.pone.0144866. Domínguez-Rodrigo, M., De Juana, S., Galan, A. B., & Rodriguez, M. (2009). A new protocol to differentiate trampling marks from butchery cut marks. Journal of Archaeological Science, 36(12), 2643–2653. Domínguez-Rodrigo, M., Pickering, T. R., & Bunn, H. T. (2012). Experimental study of cut marks made with rocks unmodified by human flaking and its bearing on claims of∼ 3.4-million-year-old butchery evidence from Dikika, Ethiopia. Journal of Archaeological Science, 39(2), 205–214. Douze, K., Delagnes, A., Wurz, S., & Henshilwood, C. S. (2018). The Howiesons Poort lithic sequence of Klipdrift Shelter, southern Cape, South Africa. PloS ONE, 13(11), e0206238. https://doi.org/10.1371/ journal.pone.0206238. Dusseldorp, G. L. (2012). Studying prehistoric hunting proficiency: applying optimal foraging theory to the Middle Palaeolithic and Middle Stone Age. Quaternary International, 252, 3–15. Dusseldorp, G. L. (2014). Explaining the Howiesons Poort to post-Howiesons Poort transition: a review of demographic and foraging adaptation models. Azania: Archaeological Research in Africa, 49(3), 317– 353. Egeland, C. P., Pickering, T. R., Dominguez-Rodrigo, M., & Brain, C. K. (2004). Disentangling Early Stone Age palimpsests: determining the functional independence of hominid-and carnivore-derived portions of archaeofaunas. Journal of Human Evolution, 47(5), 343–357. Faith, J. T. (2011). Ungulate community richness, grazer extinctions, and human subsistence behaviour in southern Africa’s Cape Floral Region. Paleogeography, Paleoclimatology, Paleoecology, 306(3-4), 219– 227. Faith, J. T., & Gordon, A. D. (2007). Skeletal element abundances in archaeofaunal assemblages: economic utility, sample size, and assessment of carcass transport strategies. Journal of Archaeological Science, 34(6), 872–882. Fernández-Jalvo, Y., & Andrews, P. (2016). Atlas of taphonomic identifications: 10001+ images of fossil and recent mammal bone modification. New York: Springer. Fisher, E. C., Bar-Matthews, M., Jerardino, A., & Marean, C. W. (2010). Middle and Late Pleistocene paleoscape modeling along the southern coast of South Africa. Quaternary Science Reviews, 29(11-12), 1382–1398. Flannery, K. V. (1969). Origins and ecological effects of early domestication in Iran and the Near East. In P. J. Ucko & G. W. Dimbleby (Eds.), The domestication and exploitation of plants and animals (pp. 73–100). Chicago: Aldine. Grove, M. (2010). Logistical mobility reduces subsistence risk in hunting economies. Journal of Archaeological Science, 37(8), 1913–1921. Haaland, M. M., Miller, C., Unhammer, O., Reynard, J. P., Ligouis, B., Mentzer, S., & Henshilwood, C. S. (2021). Geoarchaeological investigation of occupation deposits in Blombos Cave, South Africa indicate changes in site use and settlement dynamics in the southern Cape during MIS 5b-4 (94 – 72 ka). Quaternary Research. https://doi.org/10.1017/qua.2020.75. Hames, R. B., & Vickers, W. T. (1982). Optimal diet breadth theory as a model to explain variability in Amazonian hunting. American Ethnologist, 9(2), 358–378. Hammer, Ø., Harper, D. A., & Ryan, P.D. (2001). PAST: Paleontological statistics software package for education and data analysis. Palaeontologia electronica, 4(1), 9. Haynes, G. (1983). Frequencies of spiral and green-break fractures on ungulate limb bones in modern surface assemblages. American Antiquity, 48(1), 102–114. Hayward, M. W., Henschel, P., O'Brien, J., Hofmeyr, M., Balme, G., & Kerley, G. I. H. (2006). Prey preferences of the leopard (Panthera pardus). Journal of Zoology, 270(2), 298–313. 246 Reynard Henrich, J. (2004). Demography and cultural evolution: how adaptive cultural processes can produce maladaptive losses – the Tasmanian case. American Antiquity, 69(2), 197–214. Henschel, J. R., & Skinner, J. D. (1990). The diet of the spotted hyaenas Crocuta crocuta in Kruger National Park. African Journal of Ecology, 28(1), 69–82. Henshilwood, C. S. (2012). Late Pleistocene techno-traditions in southern Africa: a review of the Still Bay and Howiesons Poort, c. 75–59 ka. Journal of World Prehistory, 25(3-4), 205–237. Henshilwood, C. S., & Marean, C. W. (2003). The origin of modern human behavior: critique of the models and their test implications. Current Anthropology, 44(5), 627–651. Henshilwood, C. S., Sealy, J. C., Yates, R., Cruz-Uribe, K., Goldberg, P., Grine, F. E., Klein, R. G., Poggenpoel, C., van Niekerk, K., & Watts, I. (2001a). Blombos Cave, Southern Cape, South Africa: preliminary report on the 1992 – 1999 excavations of the Middle Stone Age levels. Journal of Archaeological Science, 28(4), 421–448. Henshilwood, C. S., d’Errico, F., Marean, C. W., Milo, R., & Yates, R. J. (2001b). An early bone tool industry from the Middle Stone Age at Blombos Cave, South Africa: implication for the origins of modern human behaviour, symbolism and language. Journal of Human Evolution, 41(6), 631–678. Henshilwood, C. S., d’Errico, F., Yates, R., Jacobs, Z., Tribolo, C., Duller, G. A. T., Mercier, N., Sealy, J. C., Valladas, H., Watts, I., & Wintle, A. G. (2002). Emergence of modern human behaviour: Middle Stone Age engravings from South Africa. Science, 295(5558), 1278–1280. Henshilwood, C., d’Errico, F., Vanhaeren, M., van Niekerk, K., & Jacobs, Z. (2004). Middle Stone Age shell beads from South Africa. Science, 304(5669), 404. Henshilwood, C. S., d’Errico, F., Van Niekerk, K. L., Coquinot, Y., Jacobs, Z., Lauritzen, S. E., Menu, M., & García-Moreno, R. (2011). A 100,000-year-old ochre-processing workshop at Blombos Cave, South Africa. Science, 334(6053), 219–222. Henshilwood, C. S., van Niekerk, K. L., Wurz, S., Delagnes, A., Armitage, S. J., Rifkin, R. F., Douze, K., Keene, P., Haaland, M. M., Reynard, J., & Discamps, E. (2014). Klipdrift Shelter, southern Cape, South Africa: preliminary report on the Howiesons Poort layers. Journal of Archaeological Science, 45, 284– 303. Henshilwood, C. S., d’Errico, F., van Niekerk, K. L., Dayet, L., Queffelec, A., & Pollarolo, L. (2018). An abstract drawing from the 73,000-year-old levels at Blombos Cave, South Africa. Nature, 562(7725), 115–118. Herries, A. I., & Fisher, E. C. (2010). Multidimensional GIS modelling of magnetic mineralogy as a proxy for fire use and spatial patterning: evidence from the Middle Stone Age bearing sea cave of Pinnacle Point 13B (Western Cape, South Africa). Journal of Human Evolution, 59(3-4), 306–320. Hiscock, P., Clarkson, C., & Mackay, A. (2011). Big debates over little tools: ongoing disputes over microliths on three continents. World Archaeology, 43(4), 653–664. Jacobs, Z., Geoffrey, A.T., Duller, A G., Wintle, Henshilwood, W. C. S. (2006). Extending the chronology of deposits at Blombos Cave, South Africa, back to 140ka using optical dating of single and multiple grains of quartz. Journal of Human Evolution, 51(3):255–273. Jacobs, Z. (2010). An OSL chronology for the sedimentary deposits from Pinnacle Point Cave 13B – a punctuated presence. Journal of Human Evolution, 59(3-4), 289–305. Jacobs, Z., Hayes, E. H., Roberts, R. G., Galbraith, R. F., & Henshilwood, C. S. (2013). An improved OSL chronology for the Still Bay layers at Blombos Cave, South Africa: further tests of single-grain dating procedures and a re-evaluation of the timing of the Still Bay industry across southern Africa. Journal of Archaeological Science, 40(1), 579–594. Jacobs, Z., Jones, B. G., Cawthra, H. C., Henshilwood, C. S., Roberts, R. G. (2020). The chronological, sedimentary and environmental context for the archaeological deposits at Blombos Cave, South Africa. Quaternary Science Reviews 235:105850. Janson, C. H., & Goldsmith, M. L. (1995). Predicting group size in primates: foraging costs and predation risks. Behavioral Ecology, 6(3), 326–336. Jerardino, A. (1995). The problem with density values in archaeological analyses: a case study from Tortoise Cave, Western Cape, South Africa. South African Archaeological Bulletin, 50(161), 21–27. Jerardino, A. (2010). Large shell middens in Lamberts Bay, South Africa: a case of hunter–gatherer resource intensification. Journal of Archaeological Science, 37(9), 2291–2302. Jerardino, A. (2016). Shell density as proxy for reconstructing prehistoric aquatic resource exploitation, perspectives from southern Africa. Journal of Archaeological Science: Reports, 6, 637–644. Jerardino, A., & Marean, C. W. (2010). Shellfish gathering, marine paleoecology and modern human behavior: perspectives from cave PP13B, Pinnacle Point, South Africa. Journal of Human Evolution, 59(3-4), 412–424. Tracking Occupational Intensity Using Archaeo-faunal Data: Case... 247 Johnson, E. (1985). Current developments in bone technology. In M. B. Schiffer (Ed.), Advances in archaeological method and theory (Vol. 8, pp. 157–235). New York: Academic Press. Karkanas, P., & Goldberg, P. (2010). Site formation processes at Pinnacle point Cave 13B (Mossel Bay, Western Cape Province, South Africa): resolving stratigraphic and depositional complexities with micromorphology. Journal of Human Evolution, 59(3-4), 256–273. Karkanas, P., Brown, K. S., Fisher, E. C., Jacobs, Z., & Marean, C. W. (2015). Interpreting human behavior from depositional rates and combustion features through the study of sedimentary microfacies at site Pinnacle Point 5-6, South Africa. Journal of Human Evolution, 85, 1–21. Kelly, R. L. (1992). Mobility/sedentism: concepts, archaeological measures, and effects. Annual Review of Anthropology, 21(1), 43–66. Kelly, R. L. (1995). The foraging spectrum: diversity in hunter-gatherer lifeways. Washington D.C.: Smithsonian Institution Press. Klein, R. G. (1980). Environmental and ecological implications of large mammals from Upper Pleistocene and Holocene sites in southern Africa. Annals of the South African Museum, 81, 223–283. Klein, R. G., & Bird, D. W. (2016). Shellfishing and human evolution. Journal of Anthropological Archaeology, 44, 198–205. Klein, R. G., Avery, G., Cruz-Uribe, K., Halkett, D., Parkington, J. E., Steele, T., Volman, T.P & Yates, R. (2004). The Ysterfontein 1 Middle Stone Age site, South Africa, and early human exploitation of coastal resources. Proceedings of the National Academy of Sciences, 101(16), 5708-5715. Kyriacou, K., Parkington, J. E., Marais, A. D., & Braun, D. R. (2014). Nutrition, modernity and the archaeological record: coastal resources and nutrition among Middle Stone Age hunter-gatherers on the western Cape coast of South Africa. Journal of Human Evolution, 77, 64–73. Langejans, G. H., van Niekerk, K. L., Dusseldorp, G. L., & Thackeray, J. F. (2012). Middle Stone Age shellfish exploitation: potential indications for mass collecting and resource intensification at Blombos Cave and Klasies River, South Africa. Quaternary International, 270, 80–94. Lombard, M. (2011). Quartz-tipped arrows older than 60 ka: further use-trace evidence from Sibudu, KwaZulu-Natal, South Africa. Journal of Archaeological Science, 38(8), 1918–1930. Louw, A. W. (*1969). Bushman Rock Shelter, Ohrigstad, eastern Transvaal: a preliminary investigation, 1965. South African Archaeological Bulletin, 24, 39–51. Lupo, K. D., Fancher, J. M., & Schmitt, D. N. (2013). The taphonomy of resource intensification: zooarchaeological implications of resource scarcity among Bofi and Aka forest foragers. Journal of Archaeological Method and Theory, 20(3), 420–447. Lyman, R. L. (1987). Archaeofaunas and butchery studies: a taphonomic perspective. In M. B. Schiffer (Ed.), Advances in archaeological method and theory (Vol. 10, pp. 249–337). New York: Academic Press. Mackay, A. (2011). Nature and significance of the Howiesons Poort to post-Howiesons Poort transition at Klein Kliphuis rockshelter, South Africa. Journal of Archaeological Science, 38(7), 1430–1440. Mackay, A., Stewart, B. A., & Chase, B. M. (2014). Coalescence and fragmentation in the late Pleistocene archaeology of southernmost Africa. Journal of Human Evolution, 72, 26–51. Madgwick, R. (2014). What makes bones shiny? Investigating trampling as a form of bone abrasion. Archaeological and Anthropological Sciences, 6(2), 163–173. Madgwick, R., & Mulville, J. (2015). Reconstructing depositional histories through bone taphonomy: extending the potential of faunal data. Journal of Archaeological Science, 53, 255–263. Magurran, A. E. (2004). Measuring biological diversity. Oxford: Blackwell Publishing. Malan, J. A. (1989). Lithostratigraphy of the Wankoe Formation (Bredasdorp Group). Pretoria: Department of Mineral and Energy Affairs, Geological Survey, Republic of South Africa. Malinsky-Buller, A., Hovers, E., & Marder, O. (2011). Making time: ‘living floors’, ‘palimpsests’ and site formation processes–a perspective from the open-air Lower Paleolithic site of Revadim Quarry, Israel. Journal of Anthropological Archaeology, 30(2), 89–101. Marean, C. W. (2010). Pinnacle Point Cave 13B (Western Cape Province, South Africa) in context: the Cape floral kingdom, shellfish, and modern human origins. Journal of Human Evolution, 59(3-4), 425–443. Marean, C. W., Bar-Matthews, M., Fisher, E., Goldberg, P., Herries, A., Karkanas, P., Nilssen, P. P., Thompson, E. (2010). The stratigraphy of the Middle Stone Age sediments at Pinnacle Point Cave 13B (Mossel Bay, Western Cape Province, South Africa). Journal of Human Evolution 59(3-4):234–255. Marean, C. W., & Bertino, L. (1994). Intrasite spatial analysis of bone: subtracting the effect of secondary carnivore consumers. American Antiquity, 59(4), 748–768. Marean, C. W., & Frey, C. J. (1997). Animal bones from caves to cities: reverse utility curves as methodological artifacts. American Antiquity, 62(4), 698–711. 248 Reynard Marean, C. W., Abe, Y., Frey, C. J., & Randall, R. C. (2000a). Zooarchaeological and taphonomic analysis of the Die Kelders Cave 1 Layers 10 and 11 Middle Stone Age larger mammal fauna. Journal of Human Evolution, 38(1), 197–233. Marean, C. W., Goldberg, P., Avery, G., Grine, F. E., & Klein, R. G. (2000b). Middle Stone Age stratigraphy and excavations at Die Kelders Cave 1 (Western Cape Province, South Africa): the 1992, 1993, and 1995 field seasons. Journal of Human Evolution, 38(1), 7–42. Marean, C. W., Bar-Matthews, M., Bernatchez, J., Fisher, E., Goldberg, P., Herries, A. I., Jacobs, Z., Jerardino, A., Karkanas, P., Minichillo, T., & Nilssen, P. J. (2007). Early human use of marine resources and pigment in South Africa during the Middle Pleistocene. Nature, 449(7164), 905–908. Marean, C. W., Cawthra, H. C., Cowling, R. M., Esler, K. J., Fisher, E., Milewski, A., Ptts, A. J., Singels, E., & De Vynck, J. (2014). Stone Age people in a changing South African Greater Cape Floristic Region. In N. Allsopp, J. F. Colville, & G. A. Verboom (Eds.), Fynbos: ecology, evolution and conservation of a megadiverse region (pp. 164–199). London: Oxford University Press. Marean, C. W., Cowling, R. M., & Franklin, J. (2020). The Palaeo-Agulhas Plain: temporal and spatial variation in an extraordinary extinct ecosystem of the Pleistocene of the Cape Floristic Region. Quaternary Science Reviews, 235, 106161. McCall, G. S. (2007). Behavioral ecological models of lithic technological change during the later Middle Stone Age of South Africa. Journal of Archaeological Science, 34(10), 1738–1751. McCall, G. S., & Thomas, J. T. (2012). Still Bay and Howiesons Poort foraging strategies: recent research and models of culture change. African Archaeological Review, 29(1), 7–50. Miller, C. E., Goldberg, P., & Berna, F. (2013). Geoarchaeological investigations at Diepkloof Rock Shelter, Western Cape, South Africa. Journal of Archaeological Science, 40(9), 3432–3452. Mitchell, P. J. (1993). The archaeology of Tloutle rockshelter, Maseru District, Lesotho. Journal of the National Museum Bloemfontein, 9, 77–132. Mourre, V., Villa, P., & Henshilwood, C. S. (2010). Early use of pressure flaking on lithic artifacts at Blombos Cave, South Africa. Science, 330(6004), 659–662. Munro, N. D. (2004). Zooarchaeological measures of hunting pressure and occupation intensity in the Natufian: implications for agricultural origins. Current Anthropology, 45(S4), S5–S34. Munro, N. (2009). Epipaleolithic subsistence intensification in the southern Levant: the faunal evidence. In J.J. Hublin & M. P. Richards (Eds.), The evolution of human diets: integrating approaches to the study of paleolithic diets (pp. 141–155). New York: Springer. Myers, T. P., Voorhies, M. R., & Corner, R. G. (1980). Spiral fractures and bone pseudotools at paleontological sites. American Antiquity, 45(3), 483–490. Nicholson, R. A. (1993). A morphological investigation of burnt animal bone and an evaluation of its utility in archaeology. Journal of Archaeological Science, 20(4), 411–428. Olsen, S. L., & Shipman, P. (1988). Surface modification on bone: trampling versus butchery. Journal of Archaeological Science, 15(5), 535–553. Oppenheimer, S. (2009). The great arc of dispersal of modern humans: Africa to Australia. Quaternary International, 202(1-2), 2–13. Parkington, J. (2010). Coastal diet, encephalization, and innovative behaviours in the late Middle Stone Age of southern Africa. In S. C. Cunnane & K. M. Steward (Eds.), Human brain evolution: the influence of freshwater and marine food resources (pp. 189–202). Hoboken, New Jersey: Wiley-Blackwell. Perreault, C., Brantingham, P. J., Kuhn, S. L., Wurz, S., & Gao, X. (2013). Measuring the complexity of lithic technology. Current Anthropology, 54(S8), S397–S406. Pickering, T. R., Marean, C. W., & Domınguez-Rodrigo, M. (2003). Importance of limb bone shaft fragments in zooarchaeology: a response to “On in situ attrition and vertebrate body part profiles” (2002), by MC Stiner. Journal of Archaeological Science, 30(11), 1469–1482. Powell, A., Shennan, S., & Thomas, M. G. (2009). Late Pleistocene demography and the appearance of modern human behaviour. Science, 324(5932), 1298–1301. Proches, S., Cowling, R. M., & Du Preez, D. R. (2005). Patterns of geophyte diversity and storage organ size in the winter-rainfall region of southern Africa. Diversity and Distributions, 11(1), 101–109. Rector, A. L., & Reed, K. E. (2010). Middle and late Pleistocene faunas of Pinnacle Point and their paleoecological implications. Journal of Human Evolution, 59(3-4), 340–357. Reynard, J. P. (2014). Trampling in coastal sites: an experimental study on the effects of shell on bone in coastal sediment. Quaternary International, 330, 156–170. Reynard, J. P., & Henshilwood, C. S. (2017). Subsistence strategies during the late Pleistocene in the Southern Cape of South Africa: comparing the Still Bay of Blombos Cave with the Howiesons Poort of Klipdrift shelter. Journal of Human Evolution, 108, 110–130. Tracking Occupational Intensity Using Archaeo-faunal Data: Case... 249 Reynard, J. P., & Henshilwood, C. S. (2018). Using trampling modification to infer occupational intensity during the Still Bay at Blombos Cave, southern Cape, South Africa. African Archaeological Review, 35(1), 1–19. Reynard, J. P., & Henshilwood, C. S. (2019). Environment versus behaviour: zooarchaeological and taphonomic analyses of fauna from the Still Bay layers at Blombos Cave, South Africa. Quaternary International, 500, 159–171. Reynard, J. P., & Wurz, S. (2020). The palaeoecology of Klasies River, South Africa: an analysis of the large mammal remains from the 1984–1995 excavations of Cave 1 and 1A. Quaternary Science Reviews, 237, 106301. Reynard, J. P., Discamps, E., Wurz, S., Badenhorst, S., van Niekerk, K. L., & Henshilwood, C. S. (2016a). Occupation intensity and environmental change at Klipdrift Shelter, southern Cape, South Africa. Palaeogeography, Palaeoclimatology, Palaeoecology, 449, 349–364. Reynard, J. P., Discamps, E., Badenhorst, S., & Henshilwood, C. S. (2016b). Subsistence strategies in the southern Cape during the Howiesons Poort: taphonomy and zooarchaeology of Klipdrift Shelter, South Africa. Quaternary International, 404(Part B), 2–19. Richerson, P. J., Boyd, R., & Bettinger, R. L. (2009). Cultural innovations and demographic change. Human Biology, 81(2-3), 211–235. Schiffer, M. B. (1987). Formation processes of the archaeological record. Albuquerque: University of New Mexico Press. Schlanger, S. H. (1990). Artifact assemblage composition and site occupation duration. In P. E. Minnis & C. L. Redman (Eds.), Perspectives on southwestern prehistory (pp. 103–121). Boulder: Westview Press. Schlanger, S. (1991). On manos, metates, and the history of site occupations. American Antiquity, 56(3), 460– 474. Schoener, T. W. (1974). The compression hypothesis and temporal resource partitioning. PNAS USA, 71(10), 4169–4172. Sealy, J. (2016). Cultural change, demography, and the archaeology of the last 100 kyr in southern Africa. In S. Jones & B. Steward (Eds.), Africa from MIS 6-2 (pp. 65–75). Dordrecht: Springer. Shennan, S. (2001). Demography and cultural innovation: a model and its implications for the emergence of modern human culture. Cambridge Archaeological Journal, 11(1), 5–16. Shiner, J. I. (2006). Artefact discard and accumulated patterns in stone artefact assemblage composition in surface archaeological deposits from Pine Point and Langwell Stations, western New South Wales. The Rangeland Journal, 28(2), 183–195. Skinner, J. D., & Chimimba, C. T. (2005). The mammals of the Southern African subregion. Cambridge: Cambridge University Press. Steele, T. E., & Klein, R. G. (2009). Late Pleistocene subsistence strategies and resource intensification in Africa. In Hublin JJ., & Richards M.P. (Eds.), The evolution of hominin diets (pp. 113-126). Springer, Dordrecht. Steele, T. E., & Klein, R. G. (2013). The Middle and Later Stone Age faunal remains from Diepkloof Rock Shelter, Western Cape, South Africa. Journal of Archaeological Science, 40(9), 3453–3462. Stein, J. K., Deo, J. N., & Phillips, L. S. (2003). Big sites—short time: accumulation rates in archaeological sites. Journal of Archaeological Science, 30(3), 297–316. Stiner, M. C., Munro, N. D., Surovell, T. A., Tchernov, E., & Bar-Yosef, O. (1999). Paleolithic population growth pulses evidenced by small animal exploitation. Science, 283(5399), 190–194. Stiner, M. C., Munro, N. D., & Surovell, T. A. (2000). The tortoise and the hare: small-game use, the broad spectrum revolution, and Paleolithic demography. Current Anthropology, 41(1), 39–73. Sullivan, M. E. (1984). A shell midden excavation at Pambula Lake on the far south coast of New South Wales. Archaeology in Oceania, 19(1), 1–15. Texier, P. J., Porraz, G., Parkington, J., Rigaud, J. P., Poggenpoel, C., Miller, C., Tribolo, C., Cartwright, C., Coudenneau, A., Klein, R., & Steele, T. (2010). A Howiesons Poort tradition of engraving ostrich eggshell containers dated to 60,000 years ago at Diepkloof Rock Shelter, South Africa. PNAS, 107(14), 6180–6185. Thompson, J. C. (2010). Taphonomic analysis of the Middle Stone Age faunal assemblage from Pinnacle Point Cave 13B, Western Cape, South Africa. Journal of Human Evolution, 59(3-4), 321–339. Thompson, J. C., & Henshilwood, C. S. (2014). Nutritional values of tortoises relative to ungulates from the Middle Stone Age levels at Blombos Cave, South Africa: implications for foraging and social behaviour. Journal of Human Evolution, 67, 33–47. Thompson, E., & Marean, C. W. (2008). The Mossel Bay lithic variant: 120 years of Middle Stone Age research from Cape St Blaize Cave to Pinnacle Point. South African Archaeological Society Goodwin Series, 10, 90–104. 250 Reynard Thompson, J. C., Faith, J. T., Cleghorn, N., & Hodgkins, J. (2017). Identifying the accumulator: making the most of bone surface modification data. Journal of Archaeological Science, 85, 105–113. Van Andel, T. H. (1989). Late Pleistocene Sea Levels and the Human Exploitation of the Shore and Shelf of Southern South Africa. Journal of Field Archaeology, 16(2):133–155. Val, A., Costamagno, S., Discamps, E., Chong, S., Claud, E., Deschamps, M., Mourre, V., Soulier, M. C., & Thiébaut, C. (2017). Testing the influence of stone tool type on microscopic morphology of cut-marks: experimental approach and application to the archaeological record with a case study from the Middle Palaeolithic site of Noisetier Cave (Fréchet-Aure, Hautes-Pyrénées, France). Journal of Archaeological Science: Reports, 11, 17–28. Val, A., Porraz, G., Texier, P. J., Fisher, J. W., & Parkington, J. (2020). Human exploitation of nocturnal felines at Diepkloof Rock Shelter provides further evidence for symbolic behaviours during the Middle Stone Age. Scientific Reports, 10(1), 1–12. Villa, P., & Mahieu, E. (1991). Breakage patterns of human long bones. Journal of Human Evolution, 21(1), 27–48. Villa, P., Soriano, S., Teyssandier, N., & Wurz, S. (2010). The Howiesons Poort and MSA III at Klasies River main site, cave 1A. Journal of Archaeological Science, 37(3), 630–655. Wadley, L. (2010). Were snares and traps used in the Middle Stone Age and does it matter? A review and a case study from Sibudu, South Africa. Journal of Human Evolution, 58(2), 179–192. Wadley, L. (2012). Some combustion features at Sibudu, South Africa, between 65,000 and 58,000 years ago. Quaternary International, 247, 341–349. Wadley, L. (2015). Those marvellous millennia: the Middle Stone Age of southern Africa. Azania: Archaeological Research in Africa, 50(2), 155–226. Wadley, L. (2021). What stimulated rapid, cumulative innovation after 100,000 years ago? Journal of Archaeological Method and Theory, 28(1), 120–141. https://doi.org/10.1007/s10816-020-09499-y. Will, M., Kandel, A. W., Kyriacou, K., & Conard, N. J. (2016). An evolutionary perspective on coastal adaptations by modern humans during the Middle Stone Age of Africa. Quaternary International, 404, 68–86. Winterhalder, B. (2001). The behavioural ecology of hunter–gatherers. In C. Panter-Brick, R. H. Layton, & P. Rowley-Conwy (Eds.), Hunter-gatherers: an interdisciplinary perspective (pp. 12–38). Cambridge: Cambridge University Press. Yeshurun, R., & Bar-Oz, G. (2018). Ungulate skeletal element profiles: a possible marker for territorial contraction and sedentism in the Levantine Epipaleolithic. Quaternary International, 464, 173–186. Yeshurun, R., Malkinson, D., Gershtein, K. M. C., Zaidner, Y., & Weinstein-Evron, M. (2020). Site occupation dynamics of early modern humans at Misliya Cave (Mount Carmel, Israel): evidence from the spatial taphonomy of faunal remains. Journal of Human Evolution, 143, 102797. Zhou, W. X., Sornette, D., Hill, R. A., & Dunbar, R. I. (2005). Discrete hierarchical organization of social group sizes. Proceedings of the Royal Society B: Biological Sciences, 272(1561), 439–444. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Affiliations Jerome P. Reynard 1 1 School of Geography, Archaeology and Environmental Studies, University of the Witwatersrand, Private Bag 3, WITS, Johannesburg 2050, South Africa