5.1 Introduction

The Quaternary (2.58 Ma until present; Gibbard and Cohen, Reference Gibbard and Cohen2008) witnessed a hominin evolutionary tree that became substantially ‘bushier’ than it had been during the preceding Neogene. Here, the term hominin is used to refer to the tribe Hominini, or the clade containing modern humans, all extinct human species, and all immediate ancestors of humans (Wood, Reference Wood2010). This is used in place of the term hominid, which refers to the family Hominidae that includes living apes and their fossil ancestors, as well as members of the tribe Hominini. Globally, climates began cooling and drying from about 3 Ma onwards, while climatic variability increased (Behrensmeyer, Reference Behrensmeyer2006). Local palaeoenvironmental records in the Awash Valley and the Omo–Turkana Basin (Ethiopia and Kenya) indicate a reduction in woody cover and an expansion of more open grasslands between 3.6 and 1.4 Ma (Cerling et al., Reference Cerling, Wynn, Andanje, Bird, Korir, Levin, Mace, Macharia, Quade and Remien2011a). While there are attempts to link global climatic drying trends with shrinking African forests and expanding open grasslands, and in turn implicate these as a causal factor for habitat-specific adaptive radiations in the hominin lineage (e.g. Vrba, Reference Vrba and Grine1988; Stanley, Reference Stanley1992; Reed, Reference Reed1997; Potts, Reference Potts1998; Cerling et al., Reference Cerling, Wynn, Andanje, Bird, Korir, Levin, Mace, Macharia, Quade and Remien2011a), these should be made with caution because regional and global climatic signals may not always be the same (Behrensmeyer, Reference Behrensmeyer2006). What is clear is that, whether driven by climate change and/or additional factors (e.g. material culture technology), the co-existence of multiple hominin taxa became the norm from the start of the Quaternary.

Some noteworthy events marking the Quaternary fossil hominin and archaeological records include the emergences of genus Homo (Kimbel et al., Reference Kimbel, Johanson and Rak1997) and Paranthropus (Walker et al., Reference Walker, Leakey, Harris and Brown1986), a tripling of hominin endocranial volume (Holloway et al., Reference Holloway, Broadfield and Yuan2004), the initial hominin migration out of Africa (Lordkipanidze et al., Reference Lordkipanidze, Jashashvili, Vekua, Ponce de León, Zollikofer, Rightmire, Pontzer, Ferring, Oms, Tappen, Bukhsianidze, Agusti, Kahlke, Kiladze, Martinez-Navarro, Mouskhelishvili, Nioradze and Rook2007), the earliest evidence of stone tools (Semaw et al., Reference Semaw, Renne, Harris, Feibel, Bernor, Fesseha and Mowbray1997), the first widely accepted evidence of cut-marked bone (de Heinzelin et al., Reference 82de Heinzelin, Clark, White, Hart, Renne, WoldeGabriel, Beyene and Vrba1999; Semaw, Reference Semaw2000), and the earliest evidence for controlled use of fire (Brain and Sillen, Reference Brain and Sillen1988; Berna et al., Reference Berna, Goldberg, Horwitz, Brink, Holt, Bamford and Chazan2012). Recent discoveries in Ethiopia, however, suggest that the origin of Homo may extend into the Neogene as early as 2.8 Ma (Villmoare et al., Reference Villmoare, Kimbel, Seyoum, Campisano, DiMaggio, Rowan, Braun, Arrowsmith and Reed2015), and that controversial stone tool cut-marks may have appeared as early as 3.39 Ma (McPherron et al., Reference McPherron, Alemseged, Marean, Wynn, Reed, Geraads, Bobe and Béarat2010; but see Domínguez-Rodrigo et al., Reference Domínguez-Rodrigo, Pickering and Bunn2011).

An increasing number of candidates has been put forward as intermediate taxa between either Australopithecus (Australopithecus afarensis or Australopithecus africanus) or Kenyanthropus platyops, and H. erectus during the Quaternary. These taxa, which contribute to the bushiness of the hominin evolutionary tree, include Australopithecus garhi, Australopithecus sediba, Homo rudolfensis, and H. habilis. Each of these exhibit different primitive and derived craniodental and/or postcranial features (Johanson et al., Reference Johanson, Masao, Eck, White, Walter, Kimbel, Asfaw, Manega, Ndessokia and Suwa1987; Wood, Reference Wood1991; Asfaw et al., Reference Asfaw, White, Lovejoy, Latimer, Simpson and Suwa1999; Leakey et al., Reference Leakey, Spoor, Brown, Gathogo, Kiarie, Leakey and McDougall2001; Berger et al., Reference Berger, de Ruiter, Churchill, Schmid, Carlson, Dirks and Kibii2010; Spoor et al., Reference Spoor, Leakey and Leakey2010). H. erectus (sensu lato) has traditionally been considered the first hominin to migrate out of Africa (Klein, Reference Klein2009), possibly facilitated by ecological changes such as grassland expansion (e.g. Antón et al., Reference Antón, Leonard and Robertson2002). However, the 1.77 Ma date for Dmanisi (Georgia) H. erectus material (Lordkipanidze et al., Reference Lordkipanidze, Jashashvili, Vekua, Ponce de León, Zollikofer, Rightmire, Pontzer, Ferring, Oms, Tappen, Bukhsianidze, Agusti, Kahlke, Kiladze, Martinez-Navarro, Mouskhelishvili, Nioradze and Rook2007) implies that at least one of these intermediate taxa may have been the first to leave Africa unless, as some believe (Lordkipanidze et al., Reference Lordkipanidze, Ponce de León, Margvelashvili, Rak, Rightmire, Vekua and Zollikofer2013), the intermediate taxa also all represent H. erectus. Near the end of the Quaternary at least one additional out-of-Africa migration occurred, this time involving modern Homo sapiens (Henn et al., Reference Henn, Gignoux, Jobin, Granka, Macpherson, Kidd, Rodríguez-Botigué, Ramachandran, Hon, Brisbin, Lin, Underhill, Comas, Kidd, Norman, Parham, Bustamante, Mountain and Feldman2011), likely as a result of gradually accumulating behavioural and cognitive changes (McBrearty and Brooks, Reference McBrearty and Brooks2000).

A number of hominin fossils from the Quaternary of southern Africa has contributed to clarifying hominin evolution. Here, evidence from several sites is placed in the broader context of hominin evolution (Fig. 5.1). Some of these fossils have been attributed to H. habilis (e.g. partial crania: SK 27, StW 53; see Fig. 5.2) or H. erectus (e.g. partial mandibles: SK 15, SK 45, StW 80; partial cranium: SK 847; see Fig. 5.3), although these attributions are debated (Grine et al., Reference Grine, Jungers and Schultz1996). Others simply attribute some of these same specimens to Australopithecus or Homo spp. (Kuman and Clarke, Reference Kuman and Clarke2000; Grine, Reference Grine2005; Clarke, Reference Clarke, Reed, Fleagle and Leakey2013; Dusseldorp et al., Reference Dusseldorp, Lombard and Wurz2013). This attribution highlights the differences between southern and eastern African fossils (Grine et al., Reference Grine, Jungers and Schultz1996). Other researchers designate a new species altogether for the southern African collection – H. gautengensis (Curnoe, Reference Curnoe2010). The fossils from southern Africa assigned to early Homo (e.g. StW 53) are predominantly derived from Sterkfontein Cave, either Member 5 (StW 53 Infill: Hughes and Tobias, Reference Hughes and Tobias1977; Partridge, 2000) or Member 4 (Kuman and Clarke, Reference Kuman and Clarke2000), and Swartkrans Cave, Members 1–3 (Susman et al., Reference Susman, de Ruiter and Brain2001; Grine, Reference Grine2005). Recently, dental material from Drimolen has been attributed to Homo (Keyser et al., Reference Keyser, Menter, Moggi-Cecchi, Pickering and Berger2000; Grine et al., Reference Grine, Smith, Heesy, Smith, Grine, Fleagle and Leakey2009; Moggi-Cecchi et al., Reference Moggi-Cecchi, Menter, Boccone and Keyser2010). Comparatively speaking, most Neogene hominin material from southern Africa (predominantly Sterkfontein Member 4: Clarke, Reference Clarke, Reed, Fleagle and Leakey2013) appears to represent A. africanus or another species of Australopithecus altogether, whereas most Quaternary hominin material represents Paranthropus robustus (e.g. Cooper’s Cave: de Ruiter et al., Reference de Ruiter, Pickering, Steininger, Kramers, Hancox, Churchill, Berger and Backwell2009; Drimolen: Keyser et al., Reference Keyser, Menter, Moggi-Cecchi, Pickering and Berger2000; Kromdraai: Thackeray et al., Reference Thackeray, de Ruiter, Berger and van der Merwe2001; Swartkrans Cave, Members 1–4: Sutton et al., Reference Sutton, Pickering, Pickering, Brain, Clarke, Heaton and Kuman2009) or Homo, where A. sediba fossils represent a notable exception (Berger et al., Reference Berger, de Ruiter, Churchill, Schmid, Carlson, Dirks and Kibii2010). Apart from material representing >1 Ma in human evolutionary history, the fossil record of southern Africa is relatively sparse (Dusseldorp et al., Reference Dusseldorp, Lombard and Wurz2013) until the appearance of archaic H. sapiens (e.g. Florisbad) at ~0.259 Ma (Grün et al., Reference Grün, Brink, Spooner, Taylor, Stringer, Franciscus and Murray1996) and eventually modern H. sapiens (e.g. at Border Cave and Klasies River Cave), with a notable exception being the partial hominin cranium from Elandsfontein (Drennen, Reference Drennen1953; Singer, Reference Singer1954) that could be as old as 1 Ma (Klein et al., Reference Klein, Avery, Cruz-Uribe and Steele2006), or considerably younger when based on associated fauna (Klein and Cruz-Uribe, Reference Klein and Cruz-Uribe1991). The more geochronologically recent hominin material from southern Africa continues to play a substantial role in debates concerning the origins of modern humans (Bräuer, Reference Bräuer2008; Pearson, Reference Pearson2008).

Fig. 5.1. Map of relevant hominin-bearing sites in southern Africa. Numbered sites are 1: Diepkloof Rock shelter, 2: Elandsfontein, 3: Sea Harvest, 4: Die Kelders Cave, 5: Blombos, 6: Pinnacle Point, 7: Klasies River Cave, 8: Florisbad, 9: Equus Cave, 10: Witkrans Cave, 11: Swartkrans, 12: Sterkfontein, 13: Drimolen, 14: Kromdraai, 15: Malapa, 16: Cave of Hearths, 17: Border Cave.

Fig. 5.2. The StW 53 cranium from Sterkfontein has been attributed to Australopithecus or Homo by various authors. (A) Frontal view of the original reconstruction by Ron Clarke; (B) frontal view of the subsequent reconstruction by Curnoe and Tobias (Reference Curnoe and Tobias2006), which was reconstructed with a larger cranial capacity; (C) the partial cranium illustrated in its entirety

(photo: Kristian J. Carlson).

Fig. 5.3. The SK 847 cranium from Swartkrans that is often attributed to Homo erectus

(photo reproduced by permission of Jason Heaton).

5.2 The emergence of genus Homo

Traditionally, genus Homo is thought to have emerged from an australopithecine ancestor such as Australopithecus anamensis, A. afarensis, A. africanus, or A. sediba (Klein, Reference Klein2009; Antón et al., Reference Antón, Potts and Aiello2014), or possibly the more poorly known taxon K. platyops (Leakey et al., Reference Leakey, Spoor, Brown, Gathogo, Kiarie, Leakey and McDougall2001). Adaptive transformations that took place (see Fleagle and Grine, Reference Fleagle, Grine, Renfrew and Bahn2014) include (1) enlarged body size; (2) increased encephalisation, probably facilitated by greater exploitation of high value food resources such as meat (e.g. increased emphasis on hunting behaviour; Aiello and Wheeler, Reference Aiello and Wheeler1995); (3) reduced masticatory apparatus (e.g. smaller tooth dimensions, less mandibular reinforcement and lower facial prognathism); and (4) expanded ranging behaviour in the context of drying climates and increasingly open grassland habitats, which selected for elongating lower limbs to improve locomotor efficiency (Bramble and Lieberman, Reference Bramble and Lieberman2004; Pontzer, Reference Pontzer2012). Overall body and limb proportions also changed towards more modern morphologies (Holliday, Reference Holliday2012). The relative order and timing of these changes, however, is less certain and remain important research questions (Wood, Reference Wood2014).

H. habilis was initially put forth as an intermediate taxon between Australopithecus and H. erectus, from Olduvai Gorge, Tanzania (Leakey et al., Reference Leakey, Tobias and Napier1964). Leakey et al. (Reference Leakey, Tobias and Napier1964) defined the new intermediate taxon H. habilis based on a collection of craniodental materials (parietals, mandible and dentition) and hand bones from a juvenile known as Olduvai Hominid (OH) 7, a partial skull with dentition from a second individual (OH 13), craniodental material assigned to a third individual (OH 6), a partial mandible with dentition assigned to a fourth individual (OH 4), and foot bones assigned to fifth individual (OH 8). Others have suggested that the hand and foot material from OH 7 and OH 8, respectively, belong to a single individual (Susman and Stern, Reference Susman and Stern1982). According to Leakey et al. (Reference Leakey, Tobias and Napier1964), the overall morphology of H. habilis did not fit into the range of variation characterising either Australopithecus, predominately represented at the time by southern African fossils (i.e. A. africanus), or the temporally later H. erectus, predominately represented at the time by Asian fossils. Not long before the H. habilis discovery, hominin fossils from Swartkrans in South Africa that had originally been assigned to Telanthropus capensis (Broom and Robinson, Reference Broom and Robinson1949; Robinson, Reference Robinson1953) were reassigned to H. erectus (Robinson, Reference Robinson1961). The fossils from Swartkrans, however, were considered at the time to be too fragmentary for establishing definitive relationships with the new H. habilis fossils from Olduvai, although similarities were noted between them, particularly in the dentition (see Tobias, Reference Tobias1991a). In describing H. habilis, Leakey et al. (Reference Leakey, Tobias and Napier1964) emphasised its maxillary and mandibular reduction and reduced buccolingual breadth of the postcanine dentition to within H. erectus and later Homo levels, but they still noted larger anterior dentition in H. habilis than was exhibited in either Australopithecus or H. erectus. The cranial morphology (e.g. muscular ridge form) and facial prognathism of H. habilis were described as intermediate between Australopithecus and H. erectus. In other anatomical regions, Leakey et al. (Reference Leakey, Tobias and Napier1964) noted some morphological resemblance to modern humans. Specifically, the hand bones of H. habilis exhibited broadened apical tufts of terminal manual phalanges as well as similar forms of metacarpophalangeal joint surfaces, compared to those of modern human hands. However, H. habilis hand bones also exhibited more robusticity and more phalangeal dorsal curvature, with stronger flexor sheath ridges, compared to modern human hand bones. Finally, Leakey et al. (Reference Leakey, Tobias and Napier1964) pointed to similarities between H. habilis and modern human foot bones as evidence of similar bipedal adaptations, including an enlarged and adducted hallux and well-marked longitudinal and transverse arches.

Importantly, the original naming of H. habilis required lowering the minimum endocranial volume threshold for the genus Homo to 600 cc (Leakey et al., Reference Leakey, Tobias and Napier1964; Tobias, Reference 87Tobias1991b) from a previous minimum of 700–800 cc. Recent discovery of the small-bodied and small-brained (417 cc) Homo floresiensis from the late Quaternary of Flores (Indonesia) (Brown et al., Reference 81Brown, Sutikna, Morwood, Socjono, Jatmiko, Wayhu Saptomo and Awe Due2004; Falk et al., Reference Falk, Hildebolt, Smith, Morwood, Sutikna, Jatmiko, Saptomo and Prior2009) has challenged this threshold, particularly because the postcranial features of H. floresiensis firmly place it within Homo (Jungers et al., Reference Jungers, Larson, Harcourt-Smith, Morwood, Sutikna, Awe Due and Djubiantono2009). Further re-examination is also warranted by the similarly sized A. sediba MH 1 endocast (420 cc) with its accompanying prefrontal reorganisation, which is more in keeping with H. sapiens than earlier A. africanus endocasts (e.g. Sts 5, Sts 60) (Carlson et al., Reference Carlson, Stout, Jashashvili, de Ruiter, Tafforeau, Carlson and Berger2011).

Subsequently, more hominin fossils have been attributed to H. habilis, including craniodental (e.g. OH 24: Leakey et al., Reference Leakey, Clarke and Leakey1971; KNM-ER 1813: Leakey, Reference Leakey1974; StW 53: Hughes and Tobias, Reference Hughes and Tobias1977; SK 27: Curnoe and Tobias, Reference Curnoe and Tobias2006; KNM-ER 42703: Spoor et al., Reference Spoor, Leakey, Gathogo, Brown, Anton, McDougall, Kiarie, Manthi and Leakey2007; OH 65: Clarke, Reference Clarke2012), dental (Suwa et al., Reference Suwa, White and Howell1996), and postcranial specimens (e.g. OH 35: Susman and Stern, Reference Susman and Stern1982; KNM-ER 3735: Leakey and Walker, Reference Leakey and Walker1985; Leakey et al., Reference Leakey, Walker, Ward, Grausz and Giacobini1989; OH 62: Johanson et al., Reference Johanson, Masao, Eck, White, Walter, Kimbel, Asfaw, Manega, Ndessokia and Suwa1987), although these attributions are widely debated, as are those of specimens excluded from this taxon (Blumenschine et al., Reference Blumenschine, Peters, Masao, Clarke, Deino, Hay, Swisher, Stanistreet, Ashley, McHenry, Sikes, van der Merwe, Tactikos, Cushing, Deocampo, Njau and Ebert2003; Curnoe and Tobias, Reference Curnoe and Tobias2006; Collard and Wood, Reference Collard, Wood, Henke and Tattersall2007; Antón et al., Reference Antón, Potts and Aiello2014; Wood, Reference Wood2014). Amongst these excluded specimens, the partial cranium from Sterkfontein, StW 53 (Hughes and Tobias, Reference Hughes and Tobias1977), has substantial significance (see Fig. 5.2). Depending on whether this fossil originates from Member 4 (Kuman and Clarke, Reference Kuman and Clarke2000) or Member 5 (Hughes and Tobias, Reference Hughes and Tobias1977), stone tools found only in the latter Member could be attributed to this taxon. Also, if StW 53, SK 27, or other material from South Africa ultimately represents H. habilis, it would extend the geographic distribution of this taxon to South Africa.

Reconstructing postcranial adaptations of H. habilis has benefited from the recovery of subsequent finds that include associated fossils and partial skeletons. The original material attributed to the taxon (OH 7 and OH 8) exhibited a powerful grasping hand characterised by overall robust and curved middle phalanges with marked flexor sheath ridges (i.e. climbing features), but more human-like distal phalanges (e.g. broadened apical tufts), suggesting human-like manual manipulatory capabilities, a thumb with mixed ape-like and human-like features, and the foot of a dedicated biped (Susman and Stern, Reference Susman and Stern1982). Additional partial skeletons (e.g. OH 62, KNM-ER 3735) also contribute to the picture of postcranial adaptations. While determining the Homo-like or ape-like limb proportions of H. habilis partial skeletons OH 62 and KNM-ER 3735 has been hampered by the incomplete nature of their diaphyses (Richmond et al., Reference Richmond, Aiello and Wood2002; Haeusler and McHenry, Reference Haeusler and McHenry2004), relative limb strength of the OH 62 partial skeleton has shown compelling similarities to chimpanzee-like limb strength proportions (Ruff, Reference 86Ruff2009). Thus, the first appearance in the fossil record of the modern human bauplan – an obligate biped – occurs in H. erectus (e.g. represented by the partial skeleton KNM-WT 15000, as well as the less complete KNM-ER 803 and KNM-ER 1808 skeletons), which is ~0.5 Ma or more after the initial emergence of genus Homo (Bramble and Lieberman, Reference Bramble and Lieberman2004; Klein, Reference Klein2009).

Following some of the additional fossil discoveries (e.g. the partial cranium KNM-ER 1470 from Koobi Fora, Kenya: Leakey, Reference Leakey1973), it was perhaps unsurprising that other taxa challenged the role of H. habilis as the only potential intermediate taxon in the early Quaternary. Overall size and facial configurations of KNM-ER 1813 and KNM-ER 1470 were shown to be too different to encompass both within a single species (Wood, Reference Wood1992, Reference Wood2010; Rightmire, Reference Rightmire1993; Kramer et al., Reference Kramer, Donnelly, Kidder, Ousley and Olah1995; Antón et al., Reference Antón, Potts and Aiello2014). Wood (Reference Wood1992) proposed two coexisting hominin lineages, one being smaller brained (H. habilis) with a less robust mandible and smaller teeth (e.g. KNM-ER 1813), the other larger brained (H. rudolfensis) with a more robust mandible and larger teeth (e.g. KNM-ER 1470). Unfortunately, these comparisons have been constrained by the absence of postcranial material reliably attributed to H. rudolfensis (Wood, Reference Wood2010).

Additional fossil discoveries pre-dating H. habilis (sensu stricto), but still attributed to genus Homo, now extend this genus earlier into the Plio-Pleistocene. For example, a first right lower molar of a juvenile individual (KNM-WT 42718) has been dated to 2.34 Ma (Prat et al., Reference Prat, Brugal, Tiercelin, Barrat, Bohn, Delagnes, Harmand, Kimeu, Kibunjia, Texier and Roche2005). The 2.33 Ma maxilla with partial dentition (A.L. 666-1) from Hadar (Ethiopia) shares several features (e.g. palate shape and thin postcanine enamel) with later Homo fossils, to the exclusion of Australopithecus (Kimbel et al., Reference Kimbel, Johanson and Rak1997). Kimbel et al. (Reference Kimbel, Johanson and Rak1997) assigned the partial maxilla to Homo aff. H. habilis because they argued that these characters are apomorphic in the Homo clade, meaning they would not be useful for distinguishing a different species within that genus. A partial hominin mandible from the Chiwondo Beds in the Malawi Rift (UR 501) was biochronologically dated to 2.4 Ma. The Malawa mandible was attributed to H. rudolfensis based on the similarity of overall corpus dimensions and strength, and morphology in the mental region, to Homo from Koobi Fora (Bromage et al., Reference Bromage, Schrenk and Zonneveld1995). An additional fossil, a partial right temporal (KNM-BC1) from 2.4 Ma deposits in the Chemeron Formation of Kenya, was attributed to Homo based on its medially positioned mandibular fossa and sharp petrous crest (Hill et al., Reference Hill, Ward, Deino, Curtis and Drake1992). Most recently, a partial mandible (LD 350-1) attributed to Homo sp. indet. and dating to 2.8 Ma has been recovered from the Ledi-Geraru research area in the Afar Regional State, Ethiopia (Villmoare et al., Reference Villmoare, Kimbel, Seyoum, Campisano, DiMaggio, Rowan, Braun, Arrowsmith and Reed2015).

Additional fossils from southern Africa that have had important contributions to the debate on genus Homo origins include those from the Malapa site dated to 1.977 Ma and attributed to A. sediba (Berger et al., Reference Berger, de Ruiter, Churchill, Schmid, Carlson, Dirks and Kibii2010) (Fig. 5.1). Comparative studies of A. sediba craniodental material, the MH 1 endocast, and upper limb, pelvis, and lower limb anatomy from two partial skeletons (MH 1 and MH 2) (Berger et al., Reference Berger, de Ruiter, Churchill, Schmid, Carlson, Dirks and Kibii2010; Carlson et al., Reference Carlson, Stout, Jashashvili, de Ruiter, Tafforeau, Carlson and Berger2011; Kibii et al., Reference 84Kibii, Churchill, Schmid, Carlson, Reed, de Ruiter and Berger2011; Churchill et al., Reference Churchill, Holliday, Carlson, Jashashvili, Macias, Mathews, Sparling, Schmid, de Ruiter and Berger2013; DeSilva et al., Reference DeSilva, Holt, Churchill, Carlson, Walker, Zipfel and Berger2013) suggest that A. sediba may have been directly ancestral to H. erectus. Potential incipient prefrontal reorganisation of the MH 1 endocast (Carlson et al., Reference Carlson, Stout, Jashashvili, de Ruiter, Tafforeau, Carlson and Berger2011), specifically in areas of the orbitofrontal cortex (e.g. Broadmann’s areas 10 and 47), that recently have been linked in modern human brains to successful statistical/probabilistic modelling of environmental events (O’Reilly et al., Reference O’Reilly, Jbabdi, Rushworth and Behrens2013), may have signalled improved fitness via better hunting/scavenging skills. Ultimately, if the lineage represented by A. sediba was ancestral to H. erectus, then contemporary taxa such as H. rudolfensis or H. habilis would be precluded from being ancestral to H. erectus, eliminating the possibility that these or other Homo taxa would have made a genetic contribution to H. sapiens.

5.3 Paranthropus and the ‘robust’ australopithecine hypodigm

The initial fossil attributed to the ‘robust’ australopithecine morph (TM 1517) was found by Broom (Reference Broom1938) at Kromdraai, South Africa. Decades later the discovery of a cranium at Olduvai Gorge, Tanzania (OH 5) with tiny incisors, large postcanine teeth, and a prominent sagittal crest, initially called Zinjanthropus boisei at the time by Louis Leakey (Reference Leakey1959), geographically extended the Paranthropus hypodigm to eastern Africa. Paranthropus was pushed back geochronologically to 2.5 Ma when a hyper-robust cranium, KNM-WT 17000, was discovered in the West Turkana area of Kenya (Walker et al., Reference Walker, Leakey, Harris and Brown1986). Both the eastern (P. boisei) and southern African (P. robustus) representatives of Paranthropus are thought to have arisen from P. aethiopicus, but doubt about the monophyly of Paranthropus remains (Constantino and Wood, Reference Collard, Wood, Henke and Tattersall2007). While questions also abound regarding the ancestor of Paranthropus, it is clear that it ultimately disappeared from the fossil record approximately 1.0 Ma in southern Africa, and as early as 1.34 Ma in eastern Africa (Grine, Reference Grine1988; Constantino and Wood, Reference Collard, Wood, Henke and Tattersall2007; Domínguez-Rodrigo et al., Reference Domínguez-Rodrigo, Pickering, Baquedano, Mabulla, Mark, Musiba, Bunn, Uribelarrea, Smith, Diez-Martin, Pérez-González, Sánchez, Santonja, Barboni, Gidna, Ashley, Yravedra, Heaton and Arriaza2013).

The Paranthropus hypodigm is constructed predominantly from craniodental remains, with emphasis placed on megadontia, sagittal crests, and small endocranial volumes (~500 cc), as well as substantial sexual dimorphism (Susman et al., Reference Susman, de Ruiter and Brain2001; Constantino and Wood, Reference Collard, Wood, Henke and Tattersall2007). The genus is well-represented by fossils chiefly from Olduvai, Koobi Fora, Omo, and West Turkana in eastern Africa; and Drimolen, Gondolin, Kromdraai, and Swartkrans in southern Africa. The existence of Paranthropus in both eastern and southern Africa alongside other hominins, including Australopithecus or Homo, is largely due to niche separation. Palaeoclimatic drying trends in Africa over the course of the Quaternary indicate that woodlands diminished as open grasslands returned (Cerling et al., Reference Cerling, Wynn, Andanje, Bird, Korir, Levin, Mace, Macharia, Quade and Remien2011a), which supports the idea that niche specialisation is reflected in the extreme craniodental feeding adaptations of Paranthropus (Rak, Reference Rak1983; Grine, Reference Grine1986). Recent analyses of dental microwear and stable carbon isotope ratios (Ungar et al., Reference Ungar, Grine and Teaford2008; Cerling et al., 2011b, Reference Cerling, Manthi, Mbua, Leakey, Leakey, Leakey, Brown, Grine, Hart, Kaleme, Roche, Uno and Wood2013), however, suggest that Paranthropus emphasised low-quality, fibrous foods in its diet, irrespective of the dietary adaptations suggested by its specialised craniofacial hard tissue anatomy. Thus, there is evidence that Paranthropus likely had a broad enough diet that its eventual extinction should not be attributed solely to dietary overspecialisation (Wood and Strait, Reference Wood and Strait2003).

Few postcranial fossils have been attributed to P. boisei, primarily because of the challenges inherent in finding associated (and diagnostic) craniodental material. Postcranial remains attributed to P. robustus (e.g. partial femora from Swartkrans: SK 82, SK 97, SK 3121, and SKW 19) demonstrate the unambiguous locomotor adaptations of a biped, albeit of a possibly altered form of gait compared to modern humans (Ruff et al., Reference Ruff, McHenry and Thackeray1999; Susman et al., Reference Susman, de Ruiter and Brain2001). Thus far, the only partial skeleton attributed to P. boisei (OH 80) supports such interpretations (Domínguez-Rodrigo et al., Reference Domínguez-Rodrigo, Pickering, Baquedano, Mabulla, Mark, Musiba, Bunn, Uribelarrea, Smith, Diez-Martin, Pérez-González, Sánchez, Santonja, Barboni, Gidna, Ashley, Yravedra, Heaton and Arriaza2013). Paranthropus has not been precluded from the list of tool-makers or tool-users by some workers (e.g. Susman, Reference Susman1988; Semaw, Reference Semaw2000). The capability of P. robustus for performing tool behaviour (e.g. precision gripping) is suggested by its hand skeletal morphology, specifically a well-defined insertion of m. flexor pollicis longus and a broad apical tuft on the pollical distal phalanx (Susman, Reference Susman1988). Moreover, similarities between bone tools from the sites of Swartkrans (Member 3) and Drimolen, and suggestive associations of these tools with P. robustus fossils from both sites (Backwell and d’Errico, Reference Backwell and d’Errico2003, Reference Backwell and d’Errico2008; d’Errico and Backwell, Reference d’Errico and Backwell2009), lend support to this notion. While it is important to note that there remains no firm association between Paranthropus fossils and stone/bone tools, the possibility that Paranthropus was a tool-user or -maker suggests that Homo may not have been the exclusive example at Swartkrans, and that it would not be prudent to attribute the extinction of P. robustus to a lack of tool behaviour (Susman, Reference Susman1988).

5.4 Emergence of modern humans

The oldest fossil evidence attributed to modern H. sapiens comes from Ethiopia: two crania from Member 1 of the Omo–Kibish region (approximately 0.195 Ma; McDougall et al., Reference McDougall, Brown and Fleagle2005), and several crania from Herto (approximately 0.16 Ma) in the Middle Awash region (White et al., Reference White, Asfaw, DeGusta, Gilbert, Richards, Suwa and Howell2003). Subsequent to fossil records of early Homo and Paranthropus in the Cradle of Humankind, the fossil record of southern Africa contributes to this debate with material postdating ~0.2 Ma, but with intermittent gaps still appearing throughout. A notable exception is the partial hominin cranium from Elandsfontein (Drennen, Reference Drennen1953; Singer, Reference Singer1954) that could be as old as 1.0 Ma (Klein et al., Reference Klein, Avery, Cruz-Uribe and Steele2006), or as young as 0.4 Ma based on associated fauna (Klein and Cruz-Uribe, Reference Klein and Cruz-Uribe1991).

Fossils attributed to modern H. sapiens in southern Africa are usually associated with Middle Stone Age (MSA) technologies. Modern H. sapiens material from MSA levels at Klasies River Mouth (Singer and Wymer, Reference Singer and Wymar1982; Rightmire and Deacon, Reference Rightmire and Deacon1991) has played a crucial role in the debate on modern human origins (Stringer, Reference Stringer2002; Rightmire, Reference Rightmire2009; Dusseldorp et al., Reference Dusseldorp, Lombard and Wurz2013). Fossils from the site of Border Cave (de Villiers, Reference de Villiers1973) have also been used to characterise early modern H. sapiens cranial (Smith et al., Reference Smith, Falsetti and Donnelly1989) and postcranial morphologies (Pearson and Grine, Reference Pearson and Grine1996), but these are subject to dating problems (Bird et al., Reference Bird, Fifield, Santos, Beaumont, Zhou, Di Tada and Hausladen2003; Grün et al., Reference Grün, Beaumont, Tobias and Eggins2003). The Hofmeyr skull dated to 36 kyr BP shares affinities with Upper Paleolithic Eurasian rather than southern African populations, suggesting that the former can be traced back to a sub-Saharan Africa ancestor during the Late Pleistocene (Grine et al., Reference Grine, Bailey, Harvati, Nathan, Morris, Henderson, Ribot and Pike2007). Additional sites in southern Africa with MSA levels, typically along the coast, contribute small numbers of modern H. sapiens fossils (mostly teeth). These sites include Blombos (Grine et al., Reference 83Grine, Henshilwood and Sealy2000), Cave of Hearths (Tobias, Reference Tobias1971), Die Kelders Cave 1 (Grine, Reference Grine2000), Diepkloof Rock shelter (Verna et al., Reference Verna, Texier, Rigaud, Poggenpoel and Parkington2013), Equus Cave (Grine and Klein, Reference Grine and Klein1985), Pinnacle Point (Marean et al., Reference 85Marean, Nilssen, Brown, Jerardino and Stynder2004), Sea Harvest (Grine and Klein, Reference Grine and Klein1993), and Witkrans Cave (McCrossin, Reference McCrossin1992) (Fig. 5.1).

Fossils from the MSA of southern Africa that contribute to documenting postcranial differences between modern H. sapiens and their hominin precursors H. erectus, H. antecessor and H. heidelbergensis are relatively sparse, but these fossils reflect improvements in locomotor (bipedalism) efficiency rather than substantial differences in locomotor activities or repertoires (Klein, Reference Klein2009). A reduction in postcranial robusticity, but not to present day levels of gracility, is also evident in modern H. sapiens MSA material compared to earlier hominins, but this trend is mixed with one of retained primitive features (Churchill et al., Reference Churchill, Pearson, Grine, Trinkaus and Holliday1996; Pearson and Grine, Reference Pearson and Grine1996). Of particular note is the massive proximal femur from Berg Aukas (Namibia) of unknown geochronological age. The specimen exhibits external morphology such as a head size relative to proximal shaft breadth recalling that of recent humans, but other features such as neck shaft angle appear to be unique (Grine et al., Reference Grine, Jungers, Tobias and Pearson1995). In terms of its shaft and overall morphology, the enigmatic Berg Aukas specimen most closely resembles archaic H. sapiens (Grine et al., Reference Grine, Jungers, Tobias and Pearson1995). This has led some to suggest that hominin postcranial fossils associated with MSA levels were still behaviourally archaic compared to Later Stone Age (LSA) hominins (Churchill et al., Reference Churchill, Pearson, Grine, Trinkaus and Holliday1996).

Compared to the origins of Homo or Paranthropus, the origins of modern humans benefits from corroboratory or complementary evidence from non-fossil genetics and material culture sources. Studies of genetic diversity amongst worldwide populations have documented the greatest diversity in Africa, suggesting that modern H. sapiens originated there (Conrad et al., Reference Conrad, Jakobsson, Coop, Wen, Wall, Rosenberg and Pritchard2006; Campbell and Tishkoff, Reference Campbell and Tishkoff2009), and particularly in southern Africa where genetic diversity in modern Khoisan-speaking hunter-gatherers is particularly high compared to populations elsewhere in the world (Henn et al., Reference Henn, Gignoux, Jobin, Granka, Macpherson, Kidd, Rodríguez-Botigué, Ramachandran, Hon, Brisbin, Lin, Underhill, Comas, Kidd, Norman, Parham, Bustamante, Mountain and Feldman2011). In terms of material culture, McBrearty and Brooks (Reference McBrearty and Brooks2000) refuted the idea that an abrupt behavioural revolution took place at the origin of modern humans, instead suggesting gradual development of behavioural complexity over the duration of the African MSA, with material culture and symbolic behaviour considered proxies of cognitive complexity. Wurz (this volume) updates the case to be made in the MSA of southern Africa for the emergence and gradual expression of complex behaviours, and thus changing cognition.

Towards the end of the Quaternary, emigration of modern H. sapiens out of Africa, whether modelled as a biological population leaving Africa and replacing existing local populations of hominins around the world by sharing genes or not (i.e. variations of a replacement model), or whether modelled as local biological populations of hominins contemporaneously evolving whilst sharing genes (i.e. a multiregional model), has long been debated (Smith et al., Reference Smith, Falsetti and Donnelly1989). The number of emigration events and their timings is complex (Atkinson et al., Reference Atkinson, Gray and Drummond2008), but the latter is estimated to be ~200–100 kyr BP (Lambert and Tishkoff, Reference Campbell and Tishkoff2009). Genetic diversity is related to both population size and its mobility. A population bottleneck refers to a temporary period of low population size, such as where genetic diversity is constrained due to a massive die-off, often attributed to a global environmental event. A population release refers to the period following the bottleneck where rapid population increase takes place. It appears from genetic evidence that a global population bottleneck occurred at ~50 kyr, after emigration events but before a population release (Atkinson et al., Reference Atkinson, Gray and Drummond2008). Reasons for the bottleneck are not well understood, but the subsequent population release may be linked to technological improvements accompanying the transition to the LSA at ~50 kyr BP (Villa et al., Reference Villa, Soriano, Tsanova, Degano, Higham, d’Errico, Backwell, Lucejko, Colombini and Beaumont2012). An alternative reason may be stabilisation of global climates after the eruption of Mt Toba around 73–71 kyr BP (Ambrose, Reference Ambrose1998). Irrespective of the ultimate causes of the population bottleneck and release, global population growth rates increased rapidly everywhere after ~50 kyr BP, although the timing of these increased rates differed in different regions.

5.5 Summary

The hominin evolutionary tree became substantially more complex during the Quaternary compared to the preceding Neogene, possibly from adaptive radiation driven by a climatic drying trend that took place during much of the Quaternary. Noteworthy events marking its fossil record include the emergence of Homo and its sister taxon Paranthropus in Africa, a tripling of hominin endocranial volume, and the initial hominin migrations of Homo out of Africa. While a number of potential evolutionary links between A. afarensis (or A. africanus) and H. erectus have been proposed, the relationship between these taxa ultimately remains uncertain. Contemporaneous events in the Quaternary archaeological record include arguably the earliest evidence of stone tools, the first evidence of cut-marked bone, and the earliest evidence for controlled use of fire. By the end of the Quaternary, modern H. sapiens had emerged from Africa and spread throughout the globe. At this time, distinct behavioural and cognitive changes in H. sapiens are evident in the archaeological record of southern Africa, more than is evident in its fossil record.