Becoming Human

The Question

What had to happen, between an Eocene primate the size of a mouse lemur and the species writing these words, for an animal of our kind to appear on the planet?

The previous entry ended fifty million years ago in the canopy of broadleaved forests, with the early primates already split into the two main lineages they have followed ever since: the strepsirrhines, leading to the modern lemurs and lorises, and the haplorhines, leading to the tarsiers, the monkeys, the apes, and us. From here, the arc narrows further. One branch within the haplorhines will produce the great apes; one branch within the great apes will produce the hominins, the lineage of upright-walking forms more closely related to modern humans than to chimpanzees; one species within the hominins will spread across the planet, occupy every continent except Antarctica, and, as the last ice age ends, begin the long process of building what its descendants now call civilization.

This is the story of that narrowing.

The Split From The Other Apes

The most recent common ancestor of human beings and chimpanzees lived in Africa somewhere between roughly six and eight million years ago. That estimate is anchored by two independent lines of evidence — the fossil record, which contains plausible early hominin candidates from around seven million years ago, and the molecular clock, which uses the rate at which neutral genetic differences accumulate between living species to estimate when their lineages last shared a common ancestor.

The molecular-clock estimate has shifted substantially over the past few decades. Early estimates based on a constant per-year mutation rate placed the chimp–human split around five to six million years ago. The shift came from direct measurement of the human germline mutation rate using parent–offspring whole-genome sequencing, and from more careful calibration against great ape generation times: the rate per year is slower than older work assumed, which pushes the split further back into the past. The 2026 working estimate, drawing on these calibrated rates, places the chimp–human last common ancestor around seven million years ago, with the genuine uncertainty spanning roughly six to nine million years ^[1].

The animal at that node was not a chimpanzee. It was a now-extinct great ape from which both lineages descend. After the split, the chimpanzee lineage continued to occupy the African forest and produced the modern chimpanzees and bonobos. The hominin lineage left a long, fragmentary, often contentious fossil trail across Africa over the next seven million years.

The Earliest Candidates

The fossils from the first two million years after the split are scarce and difficult to interpret. Three named genera carry most of the weight in the current literature.

Sahelanthropus tchadensis is known principally from a remarkably complete cranium, Toumaï, recovered from the Toros-Menalla locality in northern Chad and described by Brunet and colleagues in 2002. Sahelanthropus is dated to approximately seven million years ago, on a combination of biostratigraphy and the more recent radiometric and cosmogenic dating of the host rocks ^[2]. The cranium has a chimpanzee-sized braincase, a small face, and reduced canines, which would not in themselves make it a hominin. The principal argument for placing it on the human side of the split is the position of the foramen magnum, the opening in the base of the skull through which the spinal cord passes. In Sahelanthropus this opening is positioned beneath the skull rather than behind it, the position one would expect in an animal that habitually held its head upright over a vertical body. A femur attributed to the same individual was described in 2022 and was argued to support habitual bipedal locomotion, but the femur attribution and the bipedalism inference have both been contested in subsequent literature, and the question is not yet settled ^[2].

Orrorin tugenensis, from the Lukeino Formation of the Tugen Hills in Kenya, is dated to roughly six million years ago on current stratigraphic calibration of the host rocks and was described by Senut and colleagues in 2001 ^[3]. Orrorin is known from fragmentary remains, including a femur whose internal cortical thickness distribution is consistent with bipedal weight-bearing — a feature that, like Sahelanthropus's foramen magnum, is suggestive without being conclusive.

Ardipithecus kadabba (~5.8 to 5.2 Ma) and the much better-known Ardipithecus ramidus (~4.4 Ma) come from the Middle Awash region of Ethiopia. The 2009 Science monograph on A. ramidus, led by Tim White and a large international team, presented a partial skeleton named Ardi: a small-bodied animal with grasping feet still capable of climbing trees, but with a pelvis and lower limbs reorganised for upright walking on the ground ^[4]. Ardipithecus is widely accepted as a hominin, though the question of whether it sits on the direct human line or on a sister branch is open.

The pattern across these earliest candidates is consistent. Several lineages of African apes appear, between roughly seven and four million years ago, to have begun spending more time on the ground and walking upright when they did so. None of them was an obligate biped. The fossils are fragmentary, the dates are revised regularly, and the placement on or near the hominin trunk is contested in detail. What the record establishes firmly is that the move toward bipedalism began early — close to the chimp–human split itself — and that several lineages were experimenting with it.

The Australopithecines

By around four million years ago, the hominin record becomes denser, the bipedalism becomes unambiguous, and a recognisable adaptive radiation gets underway. The animals that fill this stretch are usually grouped as the australopithecines, after the genus Australopithecus.

Australopithecus anamensis, from rocks in northern Kenya and northern Ethiopia, dates to roughly 4.2 to 3.8 million years ago. A well-preserved cranium from Woranso-Mille in Ethiopia, described by Haile-Selassie and colleagues in 2019, gives the clearest picture of what the early australopithecines looked like — a chimpanzee-sized animal with a forward-projecting face and a small brain, but with the hip and knee structure of a habitual biped ^[5]. A. anamensis is the immediate predecessor of, and probably ancestral to, the much better-known Australopithecus afarensis.

Australopithecus afarensis lived in eastern Africa from approximately 3.9 to 2.9 million years ago. The species is best known from the partial skeleton AL 288-1, recovered at Hadar in Ethiopia in 1974 and named Lucy after a Beatles song that was playing in the field camp. Lucy is dated, by current 40Ar/39Ar dating of the Hadar volcanic tephras, to approximately 3.18 million years ago ^[6]. She stood about a metre tall, weighed perhaps thirty kilograms, had a brain of around four hundred cubic centimetres (roughly chimpanzee-sized), and walked upright on hips and knees that would not have looked out of place on a modern human child.

The most striking confirmation that afarensis-grade animals walked upright comes not from any skeleton but from a set of footprints. At Laetoli in northern Tanzania, in volcanic ash dated to approximately 3.66 million years ago, three individuals walking across freshly fallen ash left a continuous trackway preserved when the ash was wetted and then re-buried ^[7]. The prints show a striding bipedal gait with a heel strike, an arched foot, and a non-divergent big toe — the gait of a habitual walker, not an opportunist climbing down from the canopy.

After afarensis, the australopithecines diversified. Australopithecus africanus, from southern Africa, lived from roughly 3.3 to 2.1 million years ago. Several robust forms — collectively often given their own genus, Paranthropus, and including P. aethiopicus, P. boisei, and P. robustus — appeared in eastern and southern Africa between roughly 2.7 and 1.0 million years ago. The robust australopithecines had massive jaws, large flat molars, and prominent bony crests on the skull for the attachment of powerful chewing muscles. They are widely interpreted as a specialist herbivorous lineage that ate hard plant foods, persisted alongside the early genus Homo for more than a million years, and then went extinct.

The australopithecines, taken together, established beyond any doubt that habitual bipedal walking on the ground is older than any of the other features readers might consider distinctively human. It predates large brains by more than two million years. It predates stone tools by at least half a million. It predates the genus Homo by more than a million.

The First Stone Tools

The oldest stone tools currently known are from a site called Lomekwi 3, on the western shore of Lake Turkana in Kenya. They are dated to approximately 3.3 million years ago and were described by Harmand and colleagues in 2015 ^[8]. The Lomekwi tools are large, often heavy, struck from cobbles using techniques that include both passive anvil percussion and direct hard-hammer flaking. They are simpler than the later Oldowan industry, but they are unambiguous evidence of intentional flaking and use.

Lomekwi 3 is older than the genus Homo by half a million years. The toolmakers, on present evidence, must have been australopithecines or a close relative. This was a significant adjustment to the older view, in which stone-tool use was treated as a defining feature of Homo. As of 2026 the first appearance of stone tools sits earlier on the timeline than the first appearance of our genus, and the inference is that some australopithecine populations were already shaping stone for use long before the lineage that would become Homo split off from them.

The Genus Homo

The earliest fossil currently assigned to the genus Homo is the lower jaw LD 350-1, recovered from the Ledi-Geraru research area in the Afar region of Ethiopia and dated to approximately 2.8 million years ago. It was described by Villmoare and colleagues in 2015, with the dating of the host sediments published in a companion paper in the same issue of Science ^[9]. The mandible has a mosaic of features: it shows the more derived dental and jaw proportions characteristic of Homo, but on a robustness scale not far from the latest australopithecines. It is exactly the kind of transitional form one would expect, in the right place at the right time.

The first species of Homo to be formally named was Homo habilis, described by Louis Leakey, Phillip Tobias, and John Napier in 1964 from material recovered at Olduvai Gorge in northern Tanzania, in sediments dated by current calibration to between roughly 1.85 and 1.65 million years ago ^[10]. Homo habilis — "handy man" — was named for its association with the Oldowan stone tool industry: simple flaked cores and sharp-edged flakes used for cutting flesh, smashing bones to extract marrow, and processing plant material. The Oldowan tradition itself extends back to approximately 2.6 million years ago, on the basis of stone tools recovered from the Gona study area in Ethiopia, predating the habilis type material by several hundred thousand years ^[10].

Two features distinguish the early Homo fossils from their australopithecine predecessors. The brain is somewhat larger — on the order of six hundred to seven hundred cubic centimetres in habilis, compared with around four hundred in afarensis. And the dental and jaw apparatus is reduced — the molars are smaller, the canines are more incisor-like, and the chewing muscles less massive. Both features are consistent with a shift in diet toward more meat and more processed food: a smaller jaw can do less mechanical work, and a larger brain costs more energy than vegetable foods alone can easily supply.

Homo Erectus And The First Departure

By approximately 1.9 million years ago, a more derived species appears in the African record: Homo erectus, with its African form sometimes distinguished as a separate species, Homo ergaster. The taxonomic question of whether erectus and ergaster are best treated as one species or two is unresolved as of 2026. What is not in doubt is that, with these animals, several features converge that we would now recognise as broadly human: a body of modern proportions, with long legs and shorter arms; a brain in the range of eight hundred to nine hundred cubic centimetres; and, above all, a willingness to leave Africa.

The earliest known fossils of Homo outside Africa are from the site of Dmanisi, in the country of Georgia at the foot of the Caucasus. The Dmanisi fossils — including five well-preserved skulls — are dated to approximately 1.8 million years ago ^[11]. They show that within roughly a hundred thousand years of erectus-grade animals appearing in Africa, members of the genus Homo had walked north and east through the Levant and into Eurasia. By a million years ago, Homo erectus fossils are known from sites across Asia, including Java in modern Indonesia. Homo erectus eventually persisted in southeastern Asia until perhaps a hundred thousand years ago — an evolutionary tenure of close to two million years, longer than the entire span of any later human species.

Homo erectus is also associated with a second, more sophisticated stone tool industry: the Acheulean, characterised by large bifacially flaked handaxes, cleavers, and picks. The earliest secure Acheulean assemblages are from the western shore of Lake Turkana in Kenya at approximately 1.76 million years ago ^[12]. Where the Oldowan tools were made by removing a few flakes from a cobble to produce a sharp edge, the Acheulean handaxe is shaped over its entire surface to a roughly symmetrical, almond-shaped form. It represents a sustained mental template applied to stone — a planned object rather than a found edge.

The Control Of Fire

Habitual control of fire is one of the most consequential features of the human lineage. Cooked food yields more useable energy from less chewing time; fire extends the day, deters predators, allows occupation of cold environments, and supports tool production techniques that depend on heat-treated stone. Knowing when our ancestors first controlled fire matters.

The evidence is, however, difficult. Fire is destructive, its archaeological traces are subtle, and natural fires are common enough on the African landscape that any single burned bone or charcoal patch is hard to interpret. The strongest claim for very early fire use is from Wonderwerk Cave in the Northern Cape of South Africa, where Berna and colleagues in 2012 reported microstratigraphic evidence — burned bone and ashed plant material in undisturbed cave sediments — for fire use approximately one million years ago ^[13]. Several other sites in Africa and Eurasia between roughly one million and five hundred thousand years ago show similar but individually weaker traces.

The conservative position, defended in a widely cited 2011 review by Roebroeks and Villa, is that habitual and uncontroversial fire use becomes archaeologically visible only around four hundred to three hundred thousand years ago, in sites in Europe and the Levant where hearths are common, repeated, and embedded in long stratigraphic sequences ^[13]. Between these two thresholds — the strong but isolated claims back to a million years, and the secure record from around four hundred thousand — sits an interval in which fire use was probably present but episodic, opportunistic, or carried out by populations small enough that we have not yet found their hearths. The honest answer in 2026 is that habitual fire control is older than half a million years, plausibly older than a million years, and that the lower bound is still being argued.

Several Human Species At Once

For most of the period covered by this entry, more than one species of human-grade animal lived on the planet at the same time. The mid-Pleistocene world, between roughly seven hundred thousand and forty thousand years ago, contained at least five and probably more.

Homo heidelbergensis, known from sites in Africa and Europe, is the species often regarded as the common ancestor of the two best-known later forms. It carried a brain in the range of twelve hundred cubic centimetres — within the range of modern humans — built shelters, hunted large game with wooden spears (the Schöningen spears in Germany, dated to around three hundred thousand years ago, are the oldest preserved wooden hunting weapons in the record), and used fire. From heidelbergensis-like populations, the African branch led toward Homo sapiens; the Eurasian branch led toward the Neanderthals (Homo neanderthalensis) and a sister group called the Denisovans.

The Neanderthals occupied Europe and western Asia from roughly four hundred thousand years ago until they disappear from the fossil record around forty thousand years ago. They were stocky, cold-adapted, with brains slightly larger on average than modern humans, sophisticated stone tool industries (the Mousterian), the use of fire and pigments, evidence of deliberate burial, and care for injured individuals. They were not the brutish caricature of older popular accounts.

The Denisovans were identified from genetics rather than morphology. In 2010, the mitochondrial genome of a small finger bone from Denisova Cave in the Altai Mountains of southern Siberia, sequenced in Leipzig by Krause, Pääbo and colleagues, revealed a previously unsuspected human population — distinct from both Neanderthals and modern humans, but as closely related to Neanderthals as Neanderthals are to us ^[14]. The fuller nuclear genome, published the same year, confirmed the result and showed that the Denisovans are the sister group to the Neanderthals. As of 2026 they are known from a handful of fossils — a few teeth, a partial mandible from the Tibetan Plateau, isolated bone fragments — and from the genomes that have been recovered from them.

Homo floresiensis, from the island of Flores in Indonesia, is a small-bodied human species that stood about a metre tall and persisted until perhaps fifty thousand years ago ^[15]. The original 2004 announcement was contested as a possible pathological modern human, but the accumulation of further skeletal material from Liang Bua and the discovery of a precursor population at Mata Menge has made floresiensis a generally accepted island-dwarfed human lineage.

Homo naledi, from the Rising Star cave system in South Africa, was announced in 2015 from a remarkable assemblage of more than fifteen hundred bones recovered from a chamber accessible only through a series of narrow squeezes. Initial morphological intuition placed naledi close to early Homo. The published U-series and electron spin resonance dating of the fossils, by Dirks and colleagues in 2017, placed them at between approximately 335,000 and 236,000 years ago — much younger than morphology alone had suggested, and overlapping in time with the earliest Homo sapiens ^[16].

The mid-Pleistocene was therefore a world of co-existing human species, distributed across Africa, Europe, mainland Asia, and at least the islands of Flores and the Rising Star cave system. The reduction to a single surviving species is a recent event.

Homo Sapiens

The fossils currently regarded as the earliest of our own species, Homo sapiens, come from a site called Jebel Irhoud in Morocco, on the northwestern shoulder of Africa. The Jebel Irhoud material — partial crania and other remains — was redescribed and redated by Hublin and colleagues in 2017, who used thermoluminescence dating of associated burnt flints to give an age of approximately 315,000 years ago, with confidence limits of plus or minus thirty-four thousand years ^[17]. The Jebel Irhoud fossils combine a face essentially modern in proportions with a braincase still elongated in a more archaic shape, suggesting that the modern face appeared earlier than the modern globular braincase. The 2017 redating moved the working benchmark for the origin of Homo sapiens from the previously canonical figure of around two hundred thousand years ago to approximately three hundred thousand.

Two further sites support the picture. The Omo Kibish fossils from southern Ethiopia, redated in 2005, were placed at around 195,000 years old, though a 2022 redating of overlying volcanic ash by Vidal and colleagues has argued for a substantially older age of at least 233,000 years; the picture is still being argued. The Herto fossils, also from Ethiopia, are around 160,000 years old ^[18]. Taken together, the African record suggests that anatomically modern Homo sapiens appeared somewhere across the continent by around three hundred thousand years ago and was widely distributed within Africa for the next two hundred thousand years before any branch of it left the continent in numbers.

The cultural record of these populations is significant. Fully modern human behaviour — geometric engravings, personal ornamentation, pigment use, long-distance transport of materials, complex projectile weapons, fishing — does not arrive in a single package. It accumulates over the late Middle Stone Age in Africa across roughly the last two hundred thousand years. Pinnacle Point on the southern Cape coast preserves systematic use of red ochre pigment from approximately 165,000 years ago. Blombos Cave, also on the southern Cape coast, has yielded geometric cross-hatched engravings on pieces of red ochre from around 73,000 to 100,000 years ago, perforated marine shells used as beads from a similar period, and the world's oldest known graphic mark drawn with an ochre crayon. Diepkloof Rock Shelter preserves engraved ostrich-eggshell containers from around 60,000 years ago ^[19]. By the time any modern humans left Africa in numbers, this entire cultural toolkit was in place.

Language

Modern human language is plausibly the most consequential biological feature our species has, and it leaves the most indirect archaeological trace. There are no fossils of language. The closest the record offers are anatomical proxies (the shape of the vocal tract, the size of the hypoglossal canal through which the nerve to the tongue passes, the inner-ear bone structures that calibrate hearing for the human speech range) and genetic proxies.

The most discussed genetic proxy has been the FOXP2 gene. A 2002 paper by Enard and colleagues identified two amino-acid changes specific to the human lineage in the FOXP2 protein and proposed that these had been the target of recent positive selection associated with the appearance of speech ^[20]. The picture has been substantially revised since. In 2007, Krause and colleagues sequenced the FOXP2 gene from Neanderthal DNA and found that the Neanderthals carried the same two human-specific amino-acid changes that Enard's team had identified ^[20]. The implication is that these particular FOXP2 substitutions are older than the split between modern humans and Neanderthals, and the ability to speak — if it is even meaningfully encoded by FOXP2 alone, which it almost certainly is not — does not separate us from our closest extinct relatives. Subsequent work has further softened the original signal of selection on FOXP2 in the modern human lineage. Language remains a defining feature of Homo sapiens in the modern sense, but it cannot be reduced to a single gene, and its evolutionary appearance is not pinned down by any single point on the timeline.

Out Of Africa

The standard textbook picture of a single Out-of-Africa migration roughly sixty thousand years ago is now known to be incomplete. There were at least two waves, and probably more.

Earlier excursions out of Africa are well attested in the fossil record. Misliya Cave on Mount Carmel in Israel preserves an upper jaw of Homo sapiens dated to between approximately 177,000 and 194,000 years ago ^[21]. Apidima Cave in southern Greece preserves a partial cranium, Apidima 1, that was redated by Harvati and colleagues in 2019 to approximately 210,000 years ago and described as anatomically modern Homo sapiens; the modern-human affinity has been contested in subsequent reanalyses, but the early date stands ^[21]. These earlier excursions appear to have been demographic dead ends — they left no detectable surviving lineages in any present-day human population — but they show that Homo sapiens was capable of leaving Africa, and did so, several times before the main wave.

The main wave departed Africa between approximately seventy and sixty thousand years ago. From it descend all non-African populations alive today. Within roughly twenty thousand years of leaving Africa, modern human populations had reached Australia (a journey that required watercraft capable of crossing significant stretches of open sea); within another twenty thousand, they were in central and northern Europe and across the steppes of Eurasia; by approximately fifteen thousand years ago, they had crossed the exposed Bering land bridge into the Americas. By twelve thousand years ago, when the last ice age ends and this entry's coverage closes, modern humans were established on every continent except Antarctica.

This main wave did not enter empty land. It moved into territory already occupied by Neanderthals across western Eurasia, by Denisovans across central and eastern Eurasia, and probably by other archaic populations as well. The encounters left a permanent record in the genomes of every reader. Modern humans of non-African ancestry carry approximately one to four per cent of their genome from Neanderthal admixture ^[22]. Modern humans whose ancestors lived in Oceania carry up to a few per cent of their genome from Denisovan admixture, with smaller amounts present in some East Asian populations ^[22]. Several of the Neanderthal-derived alleles now segregating in modern human populations have been associated with adaptation to higher latitudes, including pigmentation and immune function; the Denisovan-derived EPAS1 haplotype carried by modern Tibetan populations is associated with adaptation to high-altitude oxygen levels.

By around forty thousand years ago, the Neanderthals had disappeared from the fossil record. The Denisovans appear to have persisted later in some refugia but were also gone by perhaps thirty thousand years ago. The reduction of human-grade animals to a single surviving species had taken place.

The Most Recent Common Ancestors

A small additional feature of human descent is worth recording, because it tends to be widely misunderstood, and because it concentrates much of what genetics has revealed about our recent past into two memorable images.

Within the modern human population, the most recent common female-line ancestor through whom every living person inherits their mitochondrial DNA — the small, separate genome carried in the cellular organelles called mitochondria — lived in Africa somewhere around 155,000 to 200,000 years ago. She is sometimes called Mitochondrial Eve. The most recent common male-line ancestor through whom every living male inherits his Y chromosome — Y-chromosomal Adam — lived around 200,000 to 300,000 years ago ^[23].

Three things about these two figures need to be said carefully. First, Mitochondrial Eve and Y-chromosomal Adam did not live at the same time, and they were not the only humans alive in their respective generations. Many other women lived alongside Mitochondrial Eve, and many of them have living descendants today; it is only the unbroken female-line inheritance of the mitochondria that traces back through her alone. The same applies to Y-chromosomal Adam. Second, they were not the first humans, and they were not in any meaningful sense Adam and Eve. They were simply the points where two particular inheritance systems happen to coalesce. Third, the dates have moved over time as the genetic data have improved. The Y-chromosomal Adam date is the most striking case: pre-2013 estimates placed him around sixty to ninety thousand years ago. In 2013, Mendez and colleagues described a previously unknown Y-chromosome lineage, haplogroup A00, in some West African populations, that diverged from all other modern Y chromosomes far earlier than any lineage previously known. The discovery of haplogroup A00 pushed the Y-chromosomal coalescence back to roughly 200,000 to 300,000 years ago ^[23]. The earlier figure should no longer be used.

The deeper point is the one that follows from any honest accounting of these dates. Every human being now alive descends from an unbroken chain of ancestors stretching back through every event covered in this entry — through every Out-of-Africa wave, every mid-Pleistocene generation, every erectus parent that fed an erectus child, every australopithecine that walked across an ash field, every early hominin that climbed down from a Miocene tree, every Eocene primate, every Cretaceous-survivor mammal, every Permian-survivor cynodont, every Carboniferous-swamp amniote, every Devonian fish that crawled out of the water, every Cambrian larva, every Ediacaran soft body, every early eukaryote, every prokaryote that successfully divided. None of those ancestors died before reproducing. If a single one of them had, the unbroken chain would have terminated, and the reader of these words — and the writer of these words — would not exist.

That is a strong claim. It is also a true one. Every living person is the surviving end of an unbroken biological inheritance going all the way back to the origin of life on this planet. None of the events in this seven-entry arc was inevitable. Most of them, considered in advance, would have looked unlikely. The lineage that led to us survived all of them.

The Close Of The Arc

The seven-entry How We Came About series ends here. It opened with the first self-replicating chemistry on the early Earth, around four billion years ago. It moved through the rise of cells, the appearance of eukaryotes, the long stable interval of simple multicellular life, the explosion of animal body plans in the Cambrian, the move of life onto land, the rise and fall of the synapsid and dinosaur worlds, and the explosive radiation of the mammals after the dinosaurs were gone. It closes with one species of mammal, recently emerged from the African savannas, distributed across every continent of the planet, and standing on the edge of an event — the agricultural revolution — that will reorganise the entire surface of the Earth within the next ten thousand years.

The next major arc of the project will pick up at that threshold. It will cover the appearance of agriculture, the appearance of cities, the invention of writing, the rise of the first states and empires, the long history of human cooperation and conflict, the slow accumulation of technological and intellectual capability, the recent few centuries of industrial and scientific revolution, and the period in which I am writing — the early decades of the twenty-first century, a moment in which the descendants of those early hominins are reorganising their relationship with their planet, with their machines, and with each other on a scale that has no precedent in the previous seven entries.

For now, though, the Pleistocene closes. The ice sheets are about to retreat. Humans are spread across every continent except Antarctica, hunting and gathering, in small bands, with stone tools and fire and language and song. The rest of what they will do is the subject of what comes next.