The evolution of Homo sapiens, its ancestors, and closely related species from the last common ancestor with chimpanzees onward (~7 million years ago to present).
Early evolutionary biologists answered the question of human origins by searching for the precise location of “man’s place in nature,” in T.H. Huxley’s phrasing, based on comparative anatomy between living species. Research has moved from viewing humanity at the top of the scala naturae to seeing it as “just” a big-brained, bipedal primate, and the focus shifted to explaining how we arrived at “our” place. The post-nineteenth-century focus has been on understanding the evolutionary circumstances that produced Homo sapiens, based on the idea that human-specific traits are the product of the same evolutionary processes that led to all other species. This effort is notably multidisciplinary: human origins fall within the remit of anthropology, biology, genetics, zoology, primatology, geology, and psychology. The (occasionally contentious) synthesis of work within these fields has produced a coherent, albeit pixelated picture, of which this summary is a sketch. It first reviews hominin evolutionary history chronologically and then explores evolutionary patterns, including the evolution of cognition, in more depth.
Hominin Evolutionary History
Hominins in Comparative Perspective
Systematics of the Primates
Outline of Primate Evolution
There is an element of arbitrariness inherent in the idea of a “beginning” of the evolution of a lineage: after all, evolution is a continuous process, and consequently there are no real starting points within it. For this reason, the evolutionary events (like the splitting of lineages) requiring chronological estimates are based on taxonomic frameworks.
The current chronological outline of primate evolution is based on convergences between anatomical, fossil, and molecular approaches, of which the “molecular clock” gives the most frequently used dates. By assuming a constant rate of mutation during evolution, the molecular clock uses similarities and differences between biomolecules to estimate lineage-split times. The divergence between primates and closely related lineages is estimated somewhere between 70 and 80 Ma, in the Cretaceous epoch. The Strepsirhini and Haplorhini diverged from each other around 55–90 Ma, and recent genetic estimates suggest they last shared a common ancestor ~87 Ma, at the very beginning of the primate radiation. Debate continues about the most likely location of their common ancestor – but Asia is a strong contender, because the basal clades of both lineages (the haplorhine tarsiers and strepsirrhine lorises), as well as the closest relatives of primates, Dermoptera and Scandentia, are exclusive to Asia. Fossil evidence of this common ancestor remains elusive. Tarsiers diverged from the rest of the Haplorhini early, around ~81 Ma. Platyrrhines last shared a common ancestor with catarrhines ~43 Ma, but the fossil record is equivocal about where this last common ancestor lived and the platyrrhine route of dispersal to the Americas is therefore also unclear. Platyrrhines remained entirely arboreal, whereas some catarrhines (including humans) adapted to a terrestrial life. The Catarrhini further subdivided into Cercopithecoidea and Hominoidea ~25 Ma. Within the Hominoidea, the consensus order of divergence is an early Hylobatidae (gibbon) split with the rest of the clade at ~21 Ma, followed by Pongo (orangutan) at ~15 Ma and Gorilla at 8–9 Ma, and a final split between the Pan (chimpanzee and bonobo) and Homo lineages between 6 and 7 Ma.
Hominin Relationships to African Apes
Collectively, all modern and extinct great apes, including humans, are currently referred to as hominids. The African contingent of the family Hominidae (comprising Homo, Pan, and Gorilla) is known as the subfamily Homininae. Within the Homininae there are three tribes: Gorillini, Panini, and Hominini. Humans are therefore most closely related to the two Pan species, chimpanzees and bonobos
Implications for Hominin Origins
In the context of the range of living species, hominins share the closest evolutionary history with the African apes and in particular with chimpanzees and bonobos. This evolutionary proximity implies that features we have in common with all Hominidae genera are the consequence of shared ancestry. The list of shared features, which were presumably present in our common ancestors, has expanded considerably over recent years and now includes behavioral traits, potentially including a capacity for “cultural,” or socially transmitted, behavior (Whiten 2011).
The features not shared with African apes are most probably the product of our lineage’s evolution after it diverged from our modern ape cousins’ lineages – in other words, the features that arose specifically during hominin evolution. In this perspective, it is implicit that some features unique to other ape genera are the consequence of their independent evolution, so apes cannot always be used as direct analogues or homologues for human ancestors. This phylogenetic framework is essential in exploring the origins of hominin characteristics, as it is a precursor to discovering why and when these “human” features appeared and what consequences they had.
Environmental Context for Hominin Evolution
The problems to which hominin adaptations provide solutions derive from the physical environment: the ecological context in which hominins evolved. Hominins diverged from the Panini clade at the end of the Miocene (23–5.3 Ma), around 7 Ma. Important climatic changes coincide with their emergence: long-term decline in temperatures (culminating in Pleistocene Ice Ages) and associated fragmentation of habitats (involving the breaking up of Miocene tropical forests and expansion of grassland, woodland, and bushland, producing mosaic habitats). These more open environments played significant roles in hominin evolution. Furthermore, the increased fragmentation probably promoted speciation through population isolation and also produced dynamically changing selective pressures associated with changing resource type, distribution, and seasonality.
The Hominini-Panini split is poorly documented in the fossil record, reflecting an overall paucity of Miocene fossils. Hominin origins consequently remain an elusive paleoanthropological problem. The problem has two key elements: first, the environmental and evolutionary context in which the split and, second, the lineages involved – the last common ancestor (LCA) of humans and chimpanzees, and its daughter species, the earliest member of the hominin clade.
To date, no potential chimpanzee-human LCA is known from the fossil record. This precludes empirical reconstruction of the context in which the ancestral species split into two independent lineages. Even the geographical context of that group is unclear; although hominins belong to the African ape clade, and this clade might have a prolonged African ancestry, an alternative scenario is a Eurasian ancestor that migrated back into Africa in the Late Miocene. Two potential Eurasian ancestors have been proposed: Graecopithecus and Ouranopithecus – but there is insufficient fossil material to conclusively resolve the issue.
There are three African Late Miocene contenders for “early hominin” status: Sahelanthropus tchadensis (7.43–6.38 Ma), Orrorin tugenensis (6.14–5.52 Ma), and Ardipithecus (6.7–4.26 Ma). These three genera are classed as potential basal members of the hominin clade due to probable bipedal adaptations and, where sufficient evidence, dental traits shared with later hominins. Found in Chad, Sahelanthropus has an anteriorly positioned foramen magnum (the opening at the base of the cranium through which the spinal cord passes) compared to quadrupedal species, indicating that its head was held on an upright body consistent with some level of bipedal locomotion. However, the incomplete nature of the fossil makes reconstruction of the position of its foramen magnum, along with the behavioral inferences based upon it, contentious. By contrast, East African Orrorin is only known from fragmentary postcrania. Its femoral morphology suggests it shared distinctive hip biomechanics with the unequivocally bipedal australopithecines (Richmond and Jungers 2008). Finally, Ardipithecus, an Ethiopian genus comprising two species (ramidus and kadabba), shows a mosaic of terrestrial and arboreal features, indicating it was a facultative biped – in other words, that it moved bipedally on the ground, but quadrupedally in trees. Dentally, Ardipithecus showed reduced sexual dimorphism of the upper canine: a hallmark of later hominins, contrasting with marked canine sexual dimorphism in chimpanzees.
Opinions are divided about early hominin taxonomy: the fossils may represent three individual genera, may belong to one basal hominin genus, may not belong to the hominin clade at all, or a combination of these. The real difficulty in assigning Late Miocene apes to a species, genus, or lineage is not just ascertaining the presence or absence of derived traits shared with later hominins, as their evolution may have been paraphyletic. The divergence of later Miocene African hominids may have involved complex speciation with prolonged gene flow among populations and the evolution of extinct lineages with no direct ancestral relationship to later hominids. Further uncertainties about genetic mutation rates and generation times compound these paleontological considerations, because they influence molecular clock split-time calculations. Slow mutation rates produce dates as early as ~13 mya for the Hominini-Panini split, while fast ones estimate it at ~5 mya.
Early Hominin Evolution
Varying levels of terrestriality, with morphological adaptations to ground-dwelling. Mosaics of locomotor features are indicative of a spectrum of bipedal adaptation, with species incorporating arboreal climbing in their locomotor repertoire to greater or lesser degrees. Au. afarensis, a well-represented species (with around 300 individuals found so far) that lived from 3.89–2.9 Ma, has clear derived bipedal adaptations, like a humanlike pelvis and non-grasping big toe, but also primitive features indicative of arboreality, like long, curved fingers and a strong shoulders. A set of footprints found in Laetoli, Tanzania, dated to ~3.7 Ma and likely made by Au. afarensis, are commonly interpreted as the product of a form of bipedalism not unlike that of modern humans (Raichlen et al. 2010). Au. africanus (4.02–1.9 Ma) provides evidence of an increased reliance on arboreality than its older relative, with a grasping big toe and longer arm to leg ratio. Au. sediba (1.98 Ma) moved with a distinct combination of arboreal and terrestrial features, again implying a unique form of bipedalism (Zipfel et al. 2011). So, while early hominins were clearly united in their increased reliance on bipedal locomotion compared to their Miocene ancestors, accumulating evidence suggests at considerable diversity the nature and degree of bipedalism each species exhibited.
Specialization of subsistence behavior with a trend toward postcanine megadonty. The general “australopithecine” trend toward postcanine megadonty reflects increasing specialization in subsistence behavior over ~2 million years. From an ecological perspective, this makes sense: as new environments (such as those resulting from decreasing temperatures in the Pliocene) become saturated with opportunistic species, the evolutionary advantage shifts to specialized species who extract energy from specific resources more efficiently (Foley 1987). Subsistence specialization is especially pronounced in Paranthropus (later “robust” Australopithecines (aethiopicus, boisei, and robustus), ranging from 2.73 to 0.87 Ma), which share adaptations for chewing extremely hard and brittle foods like nuts and seeds. These adaptations include prominent sagittal crests for the attachment of massive chewing muscles, large and robust cheekbones to accommodate these muscles, and thick dental enamel. Some researchers consider their degree of subsistence specialization sufficiently different to that of the “gracile” Australopithecines to suggest they are a separate genus, Paranthropus, in accordance with one definition of a genus as a clade sharing a single adaptive zone (Wood and Collard 1999).
The Origins of Homo
The exact ancestry of Homo is unclear, but the genus tends to be seen as the evolutionary product of the gracile australopithecines. In general, Homo differs from Australopithecus in its trend toward ecological generalism, reflected in efficient ranging behavior, encephalization, and behavioral flexibility. The candidates at the center of the debate about the origin of our genus, which first appears at the start of the Pleistocene (2.5–0.01 Ma), are Homo habilis (2.6–1.65 Ma) and Homo rudolfensis (2.09–1.78 Ma). Their status as members of Homo is contentious. The problems with assigning fossils from around 2 Ma to either Homo or Australopithecus are twofold: first, the process of macroevolution and, second, uncertainty regarding the exact defining characteristics of the Homo genus.
In a now familiar tune, exact taxonomic affinities between species around this time are hotly debated. This derives, in part, from the debated definition of “genus” in the context of macroevolution, where the central question is when differences between species or groups of species are sufficient to merit their distinction at the generic level and of what nature these differences should be. A character-based distinction between Homo and Australopithecus – where Homo is defined on the basis of enlarged absolute brain size (>600 cm2) and “complex” behavior, including tool use and language – long formed the bedrock of investigation into the origins of Homo. This essentially amounts to a “checklist” of features true Homo species should exhibit, but its validity has been questioned in recent years.
On the one hand, the apparent simplicity of a trait list for Homo membership is deceptive, because inferring the presence or absence of “complex” behavior is difficult. Language leaves no direct trace in the fossil record, and no stone tools have so far been found associated directly with Pliocene hominins. In the context of multiple coexisting hominin species around 2 Ma, this precludes assigning makers to lithic assemblages. The synonymity of technological ability and the Homo genus, moreover, is an artifact of the incomplete archaeological record: until 2015 the earliest known stone tool industry was the Oldowan, dated to 2.5 Ma, coincident with the appearance of the first bigger-brained hominin species (H. habilis). The beginning of the archaeological record has now been pushed back to 3.3 Ma with the discovery of the Lomekwian industry (Harmand et al. 2015), implying either that Homo originated much earlier in hominin evolution or, more parsimoniously, that the evolution of technological behavior preceded the emergence of our genus. Simple stone tool use by chimpanzees and capuchins in order to crack nuts supports the latter.
On the other hand, the usefulness of a trait checklist is undermined because it oversimplifies distinctions between genera by assuming a definable moment in time at which these distinctions appeared. This might not have been the case, and it has recently been suggested that the abrupt Homo-Australopithecus divide that the checklist approach is based on has been exaggerated and that the emergence of Homo from Australopithecus is better conceived of as an accumulation of small transitions (Foley 2016).
So, theoretical problems abound. What this means in practice for the putative earliest members of Homo, habilis and rudolfensis, is that they likely represent “intermediate” species between Australopithecus and Homo, with features linking them to both genera. They have larger absolute brain sizes than Australopithecines and likely made and used stone tools, but the habilis hand and foot indicate it might have retained some arboreality in its locomotive repertoire. Both species have been suggested to have similar life histories to Australopithecines and certainly have considerably smaller bodies than later Homo. A phylogenetic approach to Homo by Wood and Collard (1999), where they define a genus as a “monophylum [common ancestor and all descendants] whose members occupy a single adaptive zone,” suggested habilis and rudolfensis do not belong in the genus. In summary, the emergence of Homo appears to be characterized by greater complexity than previously assumed.
The Evolution of Homo in the Pleistocene
The Homo genus
While its membership to the Homo genus is uncontested (based on its large brain, tool use, and modern body proportions), the precise phylogenetic affinities of Homo erectus are not. Its ancestry, descendants, and classification are debated, especially in relation to Homo ergaster, which is exclusively found in Africa. Some researchers assign ergaster and erectus to a single evolving lineage, called Homo erectus; others support a variation on this argument, where ergaster is the “African variety” of erectus. From this perspective, Homo erectus sensu lato (s.l.) refers to the lineage as a whole, whereas Homo erectus sensu stricto (s.s.) is used to refer to its Asian representatives (Lordkipanidze et al. 2013). Alternatively, Homo erectus and Homo ergaster can be seen as two distinct species, where ergaster is older and African and H. erectus younger and Asian (Klein 1999). These contrasting perspectives are based on interpretations of morphological variation between fossil groupings (with the unresolved issue being how much variation can be exhibited within a distinct species unit). An unusually large set of contemporaneous hominins found together in Dmanisi, Georgia, illustrates the complexity. The site was inhabited from 1.8 to 1.7 Ma (making these the earliest hominin fossils outside Africa), and its inhabitants exhibit wide morphological variation with features linking them to H. habilis, H. ergaster, and H. erectus (s.s). This mosaic suite of features, combined with the range of variation, it has been suggested, shows a single evolving Homo lineage with some geographically stratified variation around 1.8 Ma (Lordkipanidze et al. 2013). However, across the range of possible Homo paleodemes of the Lower Pleistocene, it is likely that several lineages existed in relative geographical isolation, some becoming established as stable species – for example, Homo erectus.
Regardless of systematics, the Homo lineage is characterized as a whole by fully modern (i.e., humanlike) bipedalism reflected in a rounded pelvis, long legs relative to arms, and arched feet, as well as reduced postcanine dentition and large brains (~900 cm2). They made simple (Mode 1) stone tools and might have used fire.
Whatever their ultimate phylogenetic status, hominins first dispersed out of Africa in the Lower Pleistocene. The earliest evidence possible claim for a dispersal are some disputed stone tools from Pakistan (Dennell et al. 1988), at around 2 Ma; a less contested earliest date outside of Africa are from the hominin remains at Dmanisi (1.8 Ma) and lithic assemblages in Israel (1.9–1.7 Ma). There is considerable uncertainty about the timing, route, and direction of early hominin dispersals, but archaeological evidence implies at least three waves of migrations. The first, around 1.8 Ma, involved the makers of Oldowan core-chopper tools; the second, around 1.4 Ma, makers of Acheulean hand axes; and the third, around 0.8 Ma, is represented by Acheulean assemblages with numerous flake cleavers. It is commonly proposed that these dispersals were facilitated by behavioral flexibility (manifested in greater levels of hunting and cooperation and dependence upon technology) resulting from increased intelligence, as well as changes in body plan reflecting adaptations for long-range walking.
The best-known later descendant of these dispersing hominins is the larger-brained Homo heidelbergensis (0.7–0.1 Ma). As a whole, this species was larger than preceding hominins, stockily built, and had thick cranial bones and massive brow ridges. Its brain size was almost as large as that of modern humans at around 1200 cm3, although its encephalization quotient (the allometric relationship between the brain and body size) is smaller. Technologically, it is associated with the Acheulean industry (Mode 2) and is likely to have hunted and potentially cooked its food. Its relationship to later hominins is controversial. H. heidelbergensis is known from Africa and Eurasia and is sometimes divided into geographical lineages (H. heidelbergensis and H. rhodesiensis). Whether they are a single species or two, H. heidelbergensis represents the core Middle Pleistocene lineage, widespread across the Old World and ancestral in some way to the later hominin lineages in Africa and Eurasia; namely, modern humans and Neanderthals.
Later Homo Diversity
Late Pleistocene species and their evolutionary relationships are relatively well understood, no doubt because they are better well represented in the fossil record, and more significantly because they existed recently enough for ancient DNA to be preserved. There are three lineages of later Pleistocene hominin currently recognized – Neanderthals, Denisovans, and Homo sapiens. The first two are Eurasian, the last African, and all evolved in the last ~300 ky or more.
The Neanderthals are the best-known extinct hominin, and past misconceptions unfairly characterized them in Hobbesian terms as “nasty, brutish, and short.” Relatively short they might have been – they were stockily built, with males averaging 166 cm and females 154 cm – but the archaeological record associated with them suggests they were capable of behaviors that are very similar to those of modern humans. Key questions about Neanderthals concern the degree of behavioral complexity they were capable of and the nature of their distinct morphological features. Anatomically, Neanderthals appear particularly well-adapted to cold Pleistocene European climates: their limb proportions appear to conform to Allen’s Rule (cold-adapted species tend to have short limbs), and their shape and relative robusticity conforms to Bergmann’s Rule (bodies tend to be rounder in colder climates). Neanderthal facial features have also been explained with reference to climate: for example, their relatively large sinuses could have been useful for warming inhaled air. However, this explanation has been challenged, and the distinctive Neanderthal morphology might simply be the consequence of genetic drift in an isolated population.
Neanderthals made standardized stone tools, but this is where consensus about their behavioral repertoire ends. Broadly speaking, there are two major positions: one contends they lacked behavioral flexibility, did not have language, did not bury their dead, did not use fire, and scavenged rather than hunted, and the other suggests the opposite. Current evidence best supports a mild version of the latter. Neanderthal genomes contain the human version of the speech-related FOXP2 gene, and the Kebara Neanderthal hyoid bone shows features consistent with a capacity for speech – although these features do not necessarily imply “modern” language. Archaeological evidence suggests at proficient hunting: the oldest known wooden spears in Europe date to 300 kya, and lithic spear points date as far back as 185 kya. Claims for advanced “cultural” capacities have been based primarily on the French Châtelperronian industry, which included ivory rings, grooved and perforated animal teeth, and other potential ornaments, but this has been contested on the basis of likely stratigraphic mixing (Higham et al. 2010). It appears Neanderthals used bird feathers, perhaps for decorative purposes. Controversial archaeological evidence aside, their large brains and similar encephalization quotient to humans make it unlikely Neanderthals were incapable of at least some “advanced” behavior.
The Neanderthals were not the only Eurasian hominin lineage. The “Denisovans” were first introduced to the paleoanthropological lexicon by Reich and colleagues in 2010, following the successful sequencing of ancient mitochondrial and nuclear DNA extracted from a distal manual phalanx found in southern Siberia. Based on the aDNA evidence, which suggested the individual belonged to a sister group of Neanderthals, the group was named Denisovans after the cave in which their remains were found. So far, they are represented by very fragmented and undiagnostic anatomical elements, and so their phenotype is unknown, but they may have been widespread over large parts of East Asia. When both are compared to modern human genomes, the Denisovan has an unexpected excess of divergent regions in comparison with its Neanderthal sister species, leading Prüfer and others (2014) to propose that these regions could have been contributed by an as yet unidentified species. Alternatively, the similarity between Neanderthals and modern humans might be the result of gene flow from AMHs into Neanderthals, and this is a more parsimonious explanation (Posth et al. 2017).
The Evolution of Modern Humans
It is now almost universally accepted that while Neanderthals and Denisovans evolved in Eurasia, Homo sapiens’ roots lie in Africa. This is known as the “Out of Africa” model, in which AMHs are a distinct African species that replaced existing Eurasian hominins as it dispersed out of Africa around 50 kya. Earlier models proposed a more widespread evolution of modern human traits, with regional continuity from archaic to modern forms, rather than dispersals and replacement. Genetic studies have formed the basis for the Out of Africa model.
The coalescence date of the modern human and Neanderthal lineages gives a lower bound for our species’ independent genetic history. Prüfer et al. (2014) suggest a date of 553–589 kya based on two Neanderthal genomes. All lines of genetic evidence point toward an African origin of AMHs. The highest level of mitochondrial DNA diversity is found in Africa, which can be explained as a result of African populations having a longer period of time to accumulate genetic diversity compared to the populations of AMHs that colonized the rest of the world after their emergence in Africa. Mitochondrial DNA lineages coalesce ~200 kya (implying their LCA lived around this time), in accordance with fossil evidence. The same goes for the Y chromosome, although it coalesces at ~150 kya (Jobling et al. 2014). The relatively low amount of whole-genome genetic variation worldwide compared with, for example, chimpanzees, is suggestive of a relatively small founding population resulting from a genetic bottleneck associated with the AMH migration out of Africa.
The oldest fossils showing significant Homo sapiens traits are found in Africa: in Jebel Irhoud, Morocco (300 kya); Omo, Ethiopia (195 kya); Herto, Ethiopia (160 kya); and Laetoli, Tanzania (120 kya). Some of these are clearly AMH, while others show mixed traits, suggesting a complex population history. Modern humans are characterized by a rounded cranium with a vertical forehead and reduced brow ridges, a small orthognathic face, a chin, and a tall, relatively gracile skeleton. These earliest AMHs are considerably more robust than current forms, and they exhibit a mix of archaic and modern features concurring with an African origin.
There is a general association between the evolution of modern humans and the Middle Stone Age in Africa, although it is the case that both archaic forms in Africa and Neanderthals used Mode 3 technologies (see section “Technology”). The African archaeological record is dramatically understudied compared to its European counterpart, but despite this, more derived traits such as blades, use of ochre, and use of aquatic resources first occur between 280 and 120 kya in Africa, compared to around 50 kya in Eurasia.
A historically contentious concept is “behavioral modernity.” On the one hand, some scholars have argued the archaeological record is indicative of a mismatch between anatomical modernity and behavioral modernity. The most extreme variety of this standpoint is a “human revolution” where modern human culture, linguistic ability, and cognition suddenly appear at 50 kya, 150,000 years after the emergence of morphologically modern humans (e.g., Mithen 1998). On the other hand, Mcbrearty and Brooks (2000) and others suggest there was a slow, sporadic development of modern human cognition in Africa over a period spanning at least 300,000 years; so, behavioral modernity evolved in tandem with its biological correlates (i.e., large brains and the modern human form). Current evidence supports the latter view, with increasing evidence for unambiguously symbolic behavior as early as 100 kya, and modern human cognition generally, as manifested in various technologies and subsistence behaviors, at ~250 kya (for a full discussion (Mcbrearty and Brooks 2000).
Admixture Between Modern Humans and Archaic Hominins
Interbreeding between Late Pleistocene hominins appears to have been a common occurrence. At current count, there is genetic evidence for three to five cases between four distinct populations (Prüfer et al. 2014): between modern humans and Neanderthals, modern humans and Denisovans, Neanderthals and Denisovans, and Denisovans and an unknown hominin. The evidence for interbreeding is debated: it has been shown that ancient population substructure generates highly similar genetic patterns to gene flow, and the two scenarios are difficult to differentiate statistically. If assumptions hold, however, and the genetic patterns are the consequence of gene flow, coding sections of introgressed DNA can provide a clue as to the functional consequences of interbreeding. It looks, for example, like Neanderthals and Denisovans contributed immune system genes to modern Eurasian and Oceanian populations, and altitude adaptations in modern Tibetans have been linked to Denisovan-like DNA (Jobling et al. 2014).
The Evolution of Human Diversity
Homo sapiens geographic dispersals. Data from Jobling et al. (2014)
Date(s) of dispersal(s)
First colonized ~50 kya
First colonized 80–120 kya
At least 3 independent waves of colonization
First colonized ~50 kya
3 (independent) waves
Date of first colonization disputed, but probably between 15 and 20 kya
The Out of Africa model of human evolution emphasizes the genetic homogeneity of humans as a species. Biological differences are thus relatively minor and are the product of differential dispersals, local adaptation, and genetic drift under conditions of isolation. Older models that related major racial divisions in the human population have been abandoned.
The Pattern of Hominin Evolution
Hominin evolution is not just a question of phylogenetic and phenotypic change, but is a pattern of adaptive change in response to changing ecological and selective conditions. The fossil record shows a sequence to these adaptive changes.
Anatomically, obligate bipedalism is one of the most obvious and fundamental differences between our ape relatives and us. The consensus among paleoanthropologists is that bipedalism is a defining feature of the hominin clade and that adaptations for upright locomotion can be used to assign putative hominin fossils from 7 to 5 Ma to the lineage (see section “Hominin Origins”). Accumulating evidence suggests at multiple “versions” of bipedalism in the hominin clade (see section “Early Hominin Evolution”). A key issue in the analysis of locomotor features is interpreting “primitive” arboreal traits shared with Miocene apes: these could be preserved due to functional relevance or be inconsequential vestigial remains that were not selected against. In any case, mosaics of locomotor features in early hominins imply bipedalism need not be monophyletic, meaning it could have evolved several times independently within a rapidly diversifying lineage.
The absence of early chimpanzee fossils complicates reconstruction of the precursor to hominin bipedalism. Suggestions for the ancestral condition have included a knuckle-walking ancestor, an orthograde arboreal clamberer, and pronograde clambering similar to orangutans. It is likely the LCA did not brachiate; the Miocene ape Proconsul, which likely belongs to the common ancestry of African apes, shows no adaptations to brachiation.
The question of why bipedalism evolved remains, and while as many as a dozen explanations have been proposed, they are united in their specific reference to the ecological context within which it did. This context was one of increasing habitat fragmentation. Energetic models suggest bipedalism allows the exploitation of larger areas than the knuckle-walking of modern chimpanzees, with obvious advantages in terms of food acquisition in patchy resource distributions associated with mosaic habitats. Others have suggested a bipedal posture was first habitually adopted in threat displays against predators or as a form of vigilance in increasingly open environments. Thermoregulatory advantages of reducing the surface area exposed to direct sunlight and raising the trunk further from the ground have also been proposed. A further possible advantage of bipedalism was that it freed the hands for carrying, an idea originally proposed by Darwin.These hypotheses are not necessarily mutually exclusive; indeed, it is likely selective pressures associated with changing environments produced multiple challenges (like increased predation, exposure to heat, and more dispersed resources) to which an upright posture provided evolutionary solutions. And while it provided solutions, bipedalism also engendered energetic costs requiring explanation in comprehensive models of its origins.
Encephalization and Life History
Regardless of their evolutionary advantages, larger brains entail clear energetic costs (see section “The Costs of Cognition” for a full discussion) for parents and offspring alike. They require a longer period of postpartum growth to reach adult size because the narrow human birth canal, the product of selection for a rounded bipedal pelvis, limits the maximum brain volume attainable in utero. This period of growth creates increased and prolonged energetic demands, met by secondary altriciality – high and prolonged dependence on others, including conspecifics other than parents. The Homo “solution” to these demands is a changed life history (the timing of key events in an organism’s life cycle). Human life histories differ from those of African apes in two key ways: they include a long period of childhood after weaning, and women live long after reproductive cessation. By contrast, female apes’ physiological systems usually fail together with menopause. These life history shifts appear in tandem with encephalizing brains: analyses of dental and femoral development of fossil hominins have indicated that Australopithecus is more similar to African apes in its rate of development, whereas the development schedules of Homo ergaster and Neanderthals are more similar to that of sapiens (Smith 1994).
A key difference between extant primates and humans is our extreme reliance on technology for survival – while some primates (chimpanzees and capuchins) do use stones to crack nuts, there is no evidence (yet) of nonhumans purposefully manufacturing lithic tools in the wild. Technology is a specific trait within a species’ behavioral repertoire, but it is especially ecologically significant as it provides a foundation for other behavioral traits like hunting.
Hominin Technology. Adapted from Foley (1987)
Early Stone Age
Simple, direct percussion producing flakes and chopping tools
From late Pliocene; especially in Africa, but occurs continuously
Production of large flakes, extensively retouched; hand axes
Early and middle Pleistocene; extensively in Africa, Europe, and parts of Asia
Middle Stone Age
Systematic preparation of cores prior to striking
Early parts of late Pleistocene in Europe and Africa
Reduced striking platforms with blade production
Late Pleistocene onward; in many parts of the world, especially Europe
Later Stone Age
Microlithic flake and blade production with retouch
Global post-Pleistocene distribution, with some late Pleistocene presence
Technology is one element, among many, of human culture. “Culture” is commonly seen as the sine qua non of our ecological niche as it is essentially a storage and transmission vehicle for human behavior. The comparative study of culture in an evolutionary context is an increasingly important field, one example of which is cultural primatology. Paleoanthropologists and cultural primatologists are concerned with, first, when and why “culture” evolved and second, its biological consequences.
The time depth of “culture,” as well as its evolutionary causes, is entirely subject to definition. Cultural primatologists operationally conceive of “culture” as a process, rather than a product: i.e., as information or behavior acquired from conspecifics through some form of social learning. This definition dramatically deposes human culture from its long-held pedestal of exclusivity in one sense, because in this conception culture can be attributed to other modern taxa such as killer whales and chimpanzees. Despite cultural behaviors in nonhumans, human culture remains unique in its scale and complexity, often summarized as the capacity for cumulative culture. The empirical question that follows on from this perspective is, “when did human-specific (cumulative) culture appear, and under what selective pressures?” Broadly, two models of cultural emergence prevail: one advocating a “revolution”: a relatively late sudden emergence of culture as a complete “package” associated with the development of language and consciousness (e.g., Mithen 1998). The other posits a gradual accretion of elements that comprise modern human culture (e.g., Mcbrearty and Brooks 2000). The expanding archaeological record, especially as more evidence in Africa is discovered, tends to support an accretionary pattern, with some elements, potentially including symbolism, appearing before the emergence of AMHs, and others, like blade technology, appearing after. That said, it is true that human culture has markedly increased in complexity over the last ~10,000 years, and this probably reflects an important transition in cultural evolution, and equally that culture probably played relatively little role in the Pliocene and earliest Pleistocene phases of hominin evolution. If a tendency toward cumulative culture is a process that occurred in the last half million years or so, it raises the question of the “cultural status” of different hominins. The limited evidence of the archaeological record would certainly point to marked contrasts between modern humans and, for example, Acheulean-bearing H. heidelbergensis, but more subtle differences with Neanderthals, who may have shared considerable cultural complexity with AMHs.
The biological consequences of human culture are significant. By any standard, it is unique in the extent to which it has transformed our surroundings and thus our selective environment. This has contrasting evolutionary effects: on the one hand, it removes selective pressures (a culturally constructed environment buffers the genome from “natural” pressures, removing the need to genetically adapt to newly colonized habitats). On the other, it creates them, as the constructed environment exerts new evolutionary stresses culminating in gene-culture coevolution. The classic example of this process is lactase persistence in populations with pastoralist ancestry, where individuals in pastoralist communities who continued producing lactase would have been at a nutritional advantage.
The African apes and earliest hominins in the later Miocene, the products of which are Sahelanthropus, Orrorin, and Ardipithecus.
The bipedal apes (early australopithecines), whose diversification mainly occurred within mosaics of locomotive features.
The megadontic specialists (the Paranthropus genus): it is unclear whether this was a monophyletic (single ancestor) group or the outcome of a general trend toward megadonty within the Australopithecines and, as such, whether it is a true radiation.
The earliest Homo: this is the most problematic of the radiations, because the phylogenetic positions of the species within it are precarious (see section “Hominin Origins”), and it shows relatively little taxonomic diversity or scale of geographical dispersal.
Pleistocene Homo: specifically, Homo erectus (s.l.) and its descendants. Underlying this radiation is a major shift in adaptive complex, reflected in changes in morphology and behavior.
Larger-brained Homo, in the last 0.5 Ma, with the emergence of the Neanderthals, Denisovans, and AMHs.
The radiation of Homo sapiens, which mostly involved social rather than morphological diversification.
These patterns are, in practice, more continuous, but emphasize the role of diversification and geographical spread in the process of becoming human, rather than the classic scala naturae of early reconstructions. The hominin family tree is now speciose, although this is subject to splitting/lumping controversies, where the underlying problem is one of variation – specifically, how much is acceptable between individuals classed as the same species. A splitting perspective would identify about 28 hominin species, and this does not seem unreasonable in the context of modern-day primate diversity (with approximately 60 genera and 200 species) (Fleagle 1995).
Transitions in Hominin Evolution
Ranging and energetics (~5–4 Ma), with the appearance of habitual bipedalism and changes in ranging behavior as habitats became increasingly fragmented
Technology and niche expansion (~3–2 Ma), with the first appearance of lithic technology and animal food processing, as well as features associated with the Homo lineage (large brains, reduced dentition, more humanlike body plan)
Cognition and cultural processes (~0.5–0 Ma), with the first appearances of “advanced” behavior: depending on definition, this is human seen in humans or is common to the Neanderthal and human lineages.
The Evolution of Human Cognition
Evidence for Evolving Cognitive Capacity
Reconstructing the evolution of hominin cognition is fundamentally different to that of other physical features. Thoughts do not fossilize (although their products are occasionally preserved in the archaeological record), but most crucially, the end product – the modern human mind – is more difficult to define than basic anatomy since it is currently not fully understood biochemically nor psychologically. “Cognition” comprises multiple skills and it is possible that these evolved independently, simultaneously, or in a mosaic fashion; but accurate reconstruction of these patterns is difficult. Broader trends in cognitive evolution, however, are clear: as a whole, hominin cognition evolved increasingly complex functions capable of innovation and abstraction.
This is directly evidenced in increasing brain size, particularly within the Homo genus, where both absolute and relative cranial capacities increase exponentially (see Fig. 3). Relative brain size (captured as an allometric relationship between brain and body size in the encephalization quotient [EQ]) is particularly informative about evolving cognition because it suggests “excess” brain presumably used for higher mental processes as opposed to basic functions. Hominin EQs demonstrate a pattern of increasing excess capacity that accelerated during the evolution of Homo. This is classically interpreted as evidence of expanding cognitive abilities or intelligence. Beyond broad patterns, elucidating which cognitive skills evolved when requires finer-grained analysis of where in the brain, specifically, growth occurred. This was long held to be exclusively within the prefrontal cortex, strongly implicating increased social intelligence (Dunbar 1992); but recent reanalyses suggest greater expansion occurred within the cerebellum, implicating fine motor skills (Barton and Venditti 2014).
Evolving cognition is also indirectly captured by its behavioral consequences, which are preserved in the archaeological record. The manufacture of lithic technology is particularly informative because it provides a long and consistent record of change, although other markers of advancing cognition could be used. Two variables – duration of technological tradition and complexity of technology – reflect cognitive states well, but it is important to note that these data are proxies and therefore provide only a scale of the relative cognitive abilities of stone tool-producing hominins. Faster turnover of technological traditions and increased complexity within them, manifested in numbers of stone tool “types,” are generally taken as evidence for more advanced cognition; but there are caveats. Stability of tradition is taken as evidence of socially transmitted “culture” in nonhumans, with associated inferences about advanced cognition; and classifying “complexity” within technocomplexes is subjective. In any case, both variables evidence trends toward advancing cognition over time: technologies change from simple flaked chopping tools to assemblages comprising highly diverse tool types (see section “Technology”), and there is much faster turnover of later technocomplexes, on average, than earlier ones. Within-species change suggests change was not always in the direction of greater complexity, with later Asian Homo erectus (s.l.) displaying fewer technological “types” than earlier African forms – although this might reflect the use of alternative local materials like bamboo. The clearest shift toward complexity in both variables occurs with the emergence of Homo sapiens, which produced dramatically more variable and rapidly changing technology than all other hominins.
Crucially, when direct and indirect evidence are analyzed together, it becomes clear that markers of cognitive evolution do not change simultaneously. Encephalization is a universal temporal trend in the Homo lineage, but this is not directly translated into equal rates of change in technology. This suggests a complex interaction between encephalization, changes in behavior, and evolutionary processes such as speciation. These patterns demonstrate that elements comprising cognition may have evolved independently. The idea of a sudden emergence of “modern” human cognition as a single, unified package comprising culture, language, and consciousness relatively recently is currently not supported by the evidence, and a more cumulative model is more likely.
Social and Ecological Models for the Evolution of Hominin Cognition
What selective pressures and advantages underlay the observed pattern of hominin cognitive evolution? Various ecological and social answers have been proposed. Clutton-Brock and Harvey (1980) examined the relationship between primate brain sizes and relative home ranges, concluding that larger brains are linked to frugivorous diets and larger home ranges. For hominins, whose diets were likely omnivorous and who became increasingly specialized for long-distance ranging in fragmented habitats, this link strongly implies a selective role for sophisticated mental processing of large amounts of spatial data. In this view, encephalization was thus primarily ecologically functional.
An alternative model is the “social brain hypothesis.” Nicholas Humphrey, Alison Jolly, and Robin Dunbar suggested the problems of living in social environments prompt selection for larger brains. Dunbar (1992) observed a correlation between the relative size of the neocortex and the size of primate social groups. Larger group sizes and associated social complexity thus created selective pressures for bigger, more cognitively advanced brains. This model fits well with the very high levels of social complexity and activity found among humans, with a particularly high degree of social cooperation.
Not all encephalization, however, is found in the neocortex. Recently, Barton and Venditti (2014) demonstrated that the cerebellum, which contains four times more neurons per unit of space than the neocortex, underwent rapid increase during hominin evolution. Consequently, humans have larger cerebella relative to neocortices than other extant primates do. In this “sensory-motor hypothesis,” selection for technical intelligence played a more significant role in cognitive evolution than did social or ecological factors.
These factors, of course, need not be mutually exclusive, and there are clearly interactions between social and ecological factors. Group size, for example, is largely determined by availability and distribution of resources, and equally the size and structure of a group can affect how resources are acquired – for example, hunting among chimpanzees and almost certainly also among hominins.
The Costs of Cognition
The Evolution of Language
The evolution of language is closely related to that of human cognition. Language both facilitates cognitive functions (as internal language, or the language of thought) and their externalization (speech). Models of language evolution range from an extremely recent, non-Darwinian emergence in Homo sapiens alone to extreme time depth with incremental evolution of its component parts throughout hominin evolution. These contrasting paradigms reflect both differences in definitions of language and the paucity of direct evidence.
Elucidating the evolution of human internal language falls far out of the traditional reach of paleoanthropology (although ingenious attempts have been made), but genetic and anatomical proxies for externalized language suggest at least some form of language in pre-sapiens hominins. Neanderthals possessed the derived human version of the speech-related FOXP2 gene (Krause et al. 2007), but its exact role has been debated and its genetic presence alone is insufficient evidence for completely humanlike vocal language. The single Neanderthal hyoid bone from Kebara, Israel, was essentially modern, suggesting a capacity for articulated speech. However, the hyoid provides tentative evidence at best: a modern form is not equivalent to a modern position in the vocal tract. Other attempts to reconstruct speech production by extinct hominins, like analyzing endocast shapes to infer relative sizes of Broca’s and Wernicke’s areas and linking the size of the thoracic vertebral canal to voluntary control of breathing required for speech, have provided highly ambiguous or statistically unsupported conclusions. “Modern” behavior is linked to both speech and internal language. To the extent that human language is interdependent with symbolic thought, we can infer its existence from “symbolic” artifacts of which the first evidence might extend as far back as shell engravings to 430 kya or a more recent package of evidence for symbolism around the last interglacial (130–70 kya), particularly in Africa. Overall, the evidence suggests language is not the prerogative of modern humans alone; linguistic capacity was likely shared with Neanderthals, also implying a form of language in the LCA they shared. This would place some level of language competence to closer to half a million years ago.
“Our” place in nature is specifically characterized by striding bipedalism; slow life history strategies; large brains capable of complex cognition; human-specific culture based on abstract, symbolic, and internalized thought with a metacognition; articulated language; and an extreme reliance on technology. These traits are related and interdependent, evolving both in a serial sequence and in parallel, depending on the trait. Fossil, archaeological, and genetic data suggest clear patterns, with bipedalism evolving earlier in human evolution as part of a general shift in ranging and energetics, followed by a technology-based niche expansion. The final period of human evolution was marked by the development of advanced cognition and its behavioral corollaries. However, within each of these phases, there would have been associated coevolution of traits; for example, the coevolution of brain size and changes in life history or diet and technology.
One hundred and 50 years of research into human evolution has revealed much about the tempo and mode of human evolution, showing a complex pattern of novel traits and diversification of lineages, set against a dynamic ecological background. Recent research has been increasingly interdisciplinary, and will continue to be so, with evidence coming from both direct studies of the past – the paleosciences – and inferences from contemporary populations at every level, from the genomic to the cultural.
- Boyle, E. K., & Wood, B. (2017). Evolution of nervous systems. In J. Kaas (Ed.), (2nd ed., pp. 19–36). Elsevier. Retrieved from http://www.sciencedirect.com/science/referenceworks/9780128040966#ancv04.
- Clarke, D. L. (1968). Analytical archaeology. Routledge. Schulkin: Front. Evol. Neurosci. Methuen & Co Limited (11 New Fetter Lane London EC4P 4EE). Retrieved from https://books.google.nl/books?hl=en&lr=&id=8BQcBQAAQBAJ&oi=fnd&pg=PP1&dq=clarke+1968+analytical+archaeology&ots=h0ioEfALf8&sig=XbXE1HNdlEXa_rBxHuaY0zjq7oI#v=onepage&q=clarke 1968 analytical archaeology&f=false.
- Clutton-Brock, T. H., & Harvey, P. H. (1980). Primates, brains and ecology. Journal of Zoology, 190(3), 309–323. https://doi.org/10.1111/j.1469-7998.1980.tb01430.x.CrossRefGoogle Scholar
- Foley, R. (1987). Another unique species: Patterns in human evolutionary ecology. Longman Scientific & Technical. London: England Mithen Google Scholar
- Foley, R. A. (2016). Mosaic evolution and the pattern of transitions in the hominin lineage. Philosophical Transactions of the Royal Society of London B: Biological Sciences, 371(1698). pp 1–14. http://rstb.royalsocietypublishing.org/content/371/1698/20150244.long.
- Foley, R. A., Lee, P. C., Widdowson, E. M., Knight, C. D., & Jonxis, J. H. P. (1991). Ecology and energetics of encephalization in hominid evolution [and discussion]. Philosophical Transactions of the Royal Society of London B: Biological Sciences, 334(1270), pp 223–232. Retrieved from http://rstb.royalsocietypublishing.org/content/334/1270/223.
- Harmand, S., Lewis, J. E., Feibel, C. S., Lepre, C. J., Prat, S., Lenoble, A., … Roche, H. (2015). 3.3-million-year-old stone tools from Lomekwi 3, West Turkana, Kenya. Nature, 521(7552), 310–315. 10.1038/nature14464.
- Hawkes, K., O’Connell, J. F., Jones, N. G. B., Alvarez, H., & Charnov, E. L. (1998). Grandmothering, menopause, and the evolution of human life histories. Proceedings of the National Academy of Sciences, 95(3), 1336–1339. Retrieved from http://www.pnas.org/content/95/3/1336.abstract.CrossRefGoogle Scholar
- Higham, T., Jacobi, R., Julien, M., David, F., Basell, L., Wood, R., … Ramsey, C. B. (2010). Chronology of the Grotte du Renne (France) and implications for the context of ornaments and human remains within the Châtelperronian. Proceedings of the National Academy of Sciences of the United States of America, 107(47), 20234–9. https://doi.org/10.1073/pnas.1007963107.
- Jobling, M., Hollox, E., Hurles, M., Kivisild, T., & Tyler-Smith, C. (2014). Human evolutionary genetics (2nd ed.). New York: Garland Science.Google Scholar
- Klein, R. G. (1999). The human career: Human biological and cultural origins (3rd ed.). Chicago: University of Chicago Press.Google Scholar
- Krause, J., Lalueza-Fox, C., Orlando, L., Enard, W., Green, R. E., Burbano, H. A., … Pääbo, S. (2007). The derived FOXP2 variant of modern humans was shared with Neandertals. Current Biology, 17(21), 1908–1912. https://doi.org/10.1016/j.cub.2007.10.008.
- Lordkipanidze, D., Ponce de León, M. S., Margvelashvili, A., Rak, Y., Rightmire, G. P., Vekua, A., & Zollikofer, C. P. E. (2013). A complete skull from Dmanisi, Georgia, and the evolutionary biology of early & homo. Science, 342(6156), 326–331. Retrieved from http://science.sciencemag.org/content/342/6156/326.abstract.CrossRefPubMedGoogle Scholar
- Mithen, S. J. (1998). The prehistory of the mind: A search for the origins of art, religion, and science. Phoenix.Google Scholar
- Posth, C., Wißing, C., Kitagawa, K., Pagani, L., van Holstein, L., Racimo, F., … Krause, J. (2017). Deeply divergent archaic mitochondrial genome provides lower time boundary for African gene flow into Neanderthals. Nature Communications, 8, 16046. https://doi.org/10.1038/ncomms16046.
- Prüfer, K., Racimo, F., Patterson, N., Jay, F., Sankararaman, S., Sawyer, S., … Pääbo, S. (2014). The complete genome sequence of a Neanderthal from the Altai Mountains. Nature, 505(7481), 43–49. https://doi.org/10.1038/nature12886.
- Reich, D., Green, R. E., Kircher, M., Krause, J., Patterson, N., Durand, E. Y., … Pääbo, S. (2010). Genetic history of an archaic hominin group from Denisova Cave in Siberia. Nature, 468(7327), 1053–1060. https://doi.org/10.1038/nature09710.
- Schulkin, J. (2011). Social allostasis: Anticipatory regulation of the internal milieu. Frontiers in Evolutionary Neuroscience. Retrieved from http://journal.frontiersin.org/article/10.3389/fnevo.2010.00111.
- Zipfel, B., DeSilva, J. M., Kidd, R. S., Carlson, K. J., Churchill, S. E., & Berger, L. R. (2011). The foot and ankle of Australopithecus sediba. Science, 333(6048), 1417–1420. Retrieved from http://science.sciencemag.org/content/333/6048/1417.abstract.CrossRefPubMedGoogle Scholar