Early hominins — possible and probable

Aside, Card, CardGrid, LinkCard, Steps, Tabs, TabItem, Badge, } from '@astrojs/starlight/components';

Core idea

Between roughly 7 and 4 million years ago, somewhere in Africa, a lineage of African apes split into two: one branch led to modern chimpanzees and bonobos, the other to us. The fossils from that window are the most consequential and the most contested in palaeoanthropology. Four named species — Sahelanthropus tchadensis, Orrorin tugenensis, Ardipithecus kadabba, and Ardipithecus ramidus — have been proposed as the earliest hominins. The total physical evidence for all four would fit, as Wood remarks, inside a single supermarket trolley with room to spare.

The topic draws a sharp line between possible hominins (the morphology is suggestive but the case is fragile) and probable hominins (multiple independent traits support the assignment). It is a topic about how scientists reason under severe data scarcity — and about why the public story of "we found the earliest human ancestor" almost always outruns what the bones can actually support.

Author's framing: The earliest hominins would have differed from the common ancestor by very little. The differences we hunt for at 6 million years ago are not the dramatic contrasts between a living human and a living chimpanzee — they are subtle dental, cranial, and postcranial hints, any of which could have evolved more than once.

Why it matters

If you only remember one thing from the deep-time topics of human evolution, make it this: the further back you go, the harder it is to tell a hominin from any other African ape. That is not a failure of effort — it is a structural consequence of how lineages split. Two species that diverged ten minutes ago are almost identical; the differences that define them have not yet had time to accumulate. The same logic holds across millions of years near the branching point.

This matters because it disciplines how we read every announcement of a new "earliest ancestor." It also matters because the question of when bipedalism evolved — and whether it evolved once or multiple times — depends entirely on whether these four candidates are real hominins or convergent apes. The answer reshapes the timeline of every later development: brain expansion, tool use, the loss of large canines, the origins of Homo.

Key takeaways

Mental model — the 7-to-4 million year window

Copy sourceZoom

Telling a hominin from a panin

The signal we are looking for

When researchers comb 8-to-5 million-year-old sediments, they are not looking for the dramatic gulf between a modern human and a modern chimpanzee. They are looking for the first wisp of divergence. Most of that signal lives in three places: the teeth (smaller canines, larger chewing teeth, thicker enamel, stronger jaw), the base of the skull (a forward-shifted foramen magnum, signalling a vertical trunk), and the lower limb (femur, pelvis, and foot anatomy consistent with bipedal weight-bearing).

The trouble is that any single one of these traits can mislead. Thick enamel, for instance, was long assumed to be a hominin signature — but it appears intermittently across the Miocene apes. A worn canine tip without lateral wear is suggestive, but not decisive. A forward foramen magnum is more diagnostic, but the relevant skull region is fragile and rarely preserved.

Possible versus probable

Wood's "possible / probable" framing is a way of being honest about the strength of inference. A possible hominin is a fossil where at least one feature points the right way, but the assignment depends on a small number of bones, sometimes a single one. A probable hominin is a fossil where multiple independent anatomical systems — teeth and skull and postcranium — point consistently toward the hominin side of the split. The category boundary is not about how old the fossil is; it is about how many independent lines of evidence reinforce one another.

The four contenders

Sahelanthropus tchadensis — the Chad surprise

Found at Toros-Menalla in Chad by Michel Brunet's team from 2001, Sahelanthropus is dated to 7.2–6.8 Mya using biochronology and isotope methods. It matters for three reasons. First, geography: it sits in west-central Africa, far from the Rift Valley sites that had dominated the field, suggesting early hominins occupied a much broader range than previously assumed. Second, habitat: the surrounding sediments preserve lakes, woodland, and gallery forest — a mosaic, not closed canopy. Third, the cranium: distorted in the ground but digitally "straightened" using virtual anthropology, it shows chimp-sized brain volume paired with strong brow ridges and canines worn only at the tip rather than along the sides as in chimps.

That trio of features is suggestive. Whether it is sufficient to call Sahelanthropus a hominin rather than a panin, a member of the common ancestral population, or an extinct sister clade is genuinely unresolved. The associated limb bones, long held back from publication, would help — but until they are openly described, the case stays on the possible side of the line.

Orrorin tugenensis — fragmentary femurs

Discovered in the Tugen Hills of northern Kenya and dated to roughly 6 Mya by potassium/argon, Orrorin is known from about a dozen specimens, mostly teeth and fragments of thigh bones. Brigitte Senut and Martin Pickford argued it is a hominin on two grounds: thicker molar enamel than typical panins, and an asymmetrical reinforcement of the femoral neck consistent with bipedal load-bearing.

Critics push back on all of it: the femur morphology is open to other readings; thick enamel is not a uniquely hominin feature; and the rest of the dentition is broadly ape-like. Wood places Orrorin on the possible side — closer in his judgement to the common ancestor than to an established hominin.

Ardipithecus kadabba and ramidus — the probable cases

The Middle Awash region of Ethiopia has yielded two species in this genus. The older, Ar. kadabba (5.8–5.2 Mya), is known from a mandible, teeth, and a handful of postcranial bones. The younger, Ar. ramidus (4.4 Mya), is represented by the famous partial skeleton informally called "Ardi," along with cranial fragments and many isolated teeth.

Ardipithecus is where the evidence finally becomes thick enough to support the probable label. Multiple independent systems — dental, cranial, and postcranial — point in the same direction. Ar. ramidus in particular shows a pelvis and foot that look adapted for habitual upright walking on the ground while retaining strong grasping capacity for trees: a mixed locomotor profile that lines up with a mosaic habitat.

The bipedalism question

When bipedalism first appeared depends on which contenders you accept. If Sahelanthropus counts, the trait is at least 7 million years old. If you require Ardipithecus-grade evidence, the confident floor is closer to 4.4 Mya. That two-and-a-half-million-year uncertainty is not pedantry — it reshapes the entire chronology of what came next.

Practical application

How to read a "new earliest ancestor" headline

Example — the supermarket-trolley test

Imagine you are asked to reconstruct the early history of, say, modern automobiles, but the entire archive available to you is one engine block, two steering wheels, half a windscreen, and a hubcap — spread across a million years of geological time and recovered from three different continents. You would be able to say something. You could distinguish a wheel from a propeller. You might even argue, from the bolt pattern, that two of the wheels came from related vehicles. But you would not be able to say with confidence which vehicle invented the steering wheel, or whether two manufacturers converged on similar designs independently.

That is roughly the position palaeoanthropology occupies for the 7-to-4 million year window. The supermarket-trolley quip is not just colourful — it is a calibration tool. When you hear a sweeping claim about the earliest hominin, picture the trolley. Most claims look very different at that scale.

Caveats

Related lessons