Modern Humans Did Not Evolve in Africa

Modern human origins were once framed around a tidy, linear story. For decades, most paleoanthropologists believed that a widespread middle Pleistocene taxon called Homo heidelbergensis, sometimes subdivided with African representatives labelled Homo rhodesiensis, gradually evolved into modern humans in Africa. The key fossils anchoring that picture were the Kabwe skull from Zambia, and the Bodo cranium from Ethiopia, sometimes called Homo bodoensis. But you may be surprised or even shocked to learn that these African fossils are not the ancestors of Homo sapiens. I used to believe, and some people still do believe, that a species called Homo heidelbergensis was the common ancestor of us and the Neanderthals. But there are problems with that now. So, more work on heidelbergensis suggests that, first of all, it has its own distinctive characteristics, which makes it less likely to be a common ancestor. These fossils, broadly grouped within what some researchers call Chibanian hominins, seem to embody an intermediate condition. Bigger brains than Homo erectus, but still strongly built faces and brow ridges, a mosaic of archaic and more modern features. For a long time, most paleoanthropologists supported a version of this model, arguing that Homo rhodesiensis represented the ancestor modern humans. That tidy picture has unraveled in recent years for two main reasons. First, the chronology has shifted dramatically. Second, the morphology, especially of the face, no longer fits a simple ancestral model. New dating of key fossils, combined with refined analysis of craniofacial growth and the modern human face, has forced a re-evaluation. The result is that many researchers and others now treat the African hominins as a side branch, rather than the direct lineage leading to Homo sapiens. They were widespread and probably interrelated populations, but their faces, their evolutionary trajectory, and their late persistence make them unlikely to be our immediate ancestors. The age problem came into sharp focus with the re-dating of the Kabwe skull. The evidence that our lineage was mostly in Africa is based on an idea, I think, assumption, a kind of inertial idea that our lineage must have always been in Africa, because Africa is the center of human history. But if you look at the archaeological evidence, it's not incredibly clear. And if you look at the genetic evidence, we have many early branches from Eurasia and only one from Africa, and complexity and branching in Eurasia that's sampled in the DNA record. DNA from Denisovans, DNA from unknown archaic lineages that contributed to Denisovans, Neanderthals, and all of those are represented in the Eurasian record, not in the African record. Long thought to be about half a million years old, comfortably old enough to sit at the root of our species, Kabwe was re-analyzed using modern uranium series and electron spin resonance techniques directly on the cranial bone. The results, published by Chris Stringer and colleagues, produced a best age estimate of 299,000 years. This was a shock. The skull turned out to be much younger than assumed. Instead of predating early modern humans, Kabwe is roughly the same age as the oldest Homo sapiens fossils, such as those from Jebel Irhoud in Morocco and Florisbad in South Africa. Yet Kabwe shows none of the derived modern human facial features. It retains a huge continuous brow ridge, supraorbital torus, a projecting mid-face without a canine fossa, a long, low brain case with a receding forehead, and massive zygomatic bones. If this specimen were truly ancestral to Homo sapiens, one would expect at least a hint of the globular brain case and retracted mid-face seen in the earliest modern humans. Its young age means that fully archaic, non-sapiens morphologies persisted in Africa at the same time as early members of our own species. Evidence for modern human evolution outside of Africa, according to Stringer. What's more, Kabwe is not alone. The Bodo cranium from Ethiopia, long cited as a bridge between Homo erectus and Homo sapiens, shares many of the same robust features. Thick brow ridges, wide, flaring cheekbones, zygomatics, a long, receding forehead, and a broad nasal aperture. Collectively, these African specimens have been called Rhodesiensis, but many researchers now prefer to speak of hominins from Africa and Europe, a population or clade with its own derived traits, separate from the Homo sapiens trajectory. Chris Stringer and his collaborators in their study on the human face explicitly noted that the large, non-sapiens-like faces of Bodo and Kabwe represent taxonomic diversity in the African middle Pleistocene, and should not be treated as a simple ancestral morph for Homo sapiens. Instead, the evidence of facial growth and morphology points toward an older ancestor with a more modern-like midface. The facial argument is powerful. Stringer and colleagues in Thiers' study The Evolutionary History of the Human Face carefully compared zygomatic and maxillary shapes across fossil hominins. Modern humans have a short, retracted, and tucked-in face. The zygomatic bone faces laterally, the inferior border is retracted beneath the orbit, there is a canine fossa depression below the cheek, the brow ridge is reduced and segmented, and the forehead rises steeply above the eyes. In contrast, the African hominins have large, projecting midfaces, inflated maxilla, and massive, continuous brow ridges. Their zygomatics, also known as cheekbones, are not retracted but project forward and sweep laterally in a way that supports a long, sloping facial profile. They lack the malar notch and sharp inflection at the alveolar margin that characterizes Homo sapiens. These features indicate a fundamentally different facial growth pattern. Bone deposition and resorption fields studied in juvenile fossils show that modern humans remodeled the midface early, creating a tucked configuration, while middle Pleistocene hominins retained archaic growth zones similar to earlier hominins. Even more striking is the geographic and temporal pattern. Fossils from Jebel Irhoud in Morocco already show a modern human-like face, short, flat, with a canine fossa, and reduced brow ridges. Florisbad and Herto continue this trend. Other East African specimens such as Ndutu and Iliya Springs show varying degrees of this modern facial plan as far back as perhaps 500,000 years. Some middle Pleistocene Chinese fossils—Dali, Jinyushan, Nanjing, and the recently described Harbin cranium— also have surprisingly modern midfaces despite otherwise archaic skulls. In contrast, the big classic Rhodesiensis crania are stubbornly archaic, with no sign of this trend. The simplest explanation, Stringer argues, is that the ancestor of modern humans was not these robust hominins, but an earlier species with a more modern-like midface, possibly akin to Jungtian man of China. For example, Jungtian man has a face whose growth fields and shape are surprisingly modern. Developmental modelling shows that you can derive a modern human face from a Jungtian-like ancestor with only modest change, whereas to go from Bodo or Kabwe to modern humans would require major, unlikely developmental reversals. Genomic evidence reinforces this. Nuclear DNA shows the Neanderthal-modern-human split happened at least 550,000 to 765,000 years ago, long before Kabwe's time. If our line diverged that early, then the 300,000-year-old African cannot be the ancestor. They are too young and already on a separate side path, perhaps closer to the Neanderthal Denisovan clade, or representing a ghost lineage that later went extinct. Stringer's discussion of the Kabwe dating explicitly warns that the skull's young age raises further questions about the mode of evolution of Homo sapiens in Africa and whether Homo rhodesiensis was a direct ancestor of our species. Morphological studies also show that the rhodesiensis clade is internally cohesive, a separate population. Bodo and Kabwe share a set of derived cranial traits not present in earlier Homo erectus, huge, double-arched brow ridges that form a continuous torus, a broad and inflated midface with no canine fossa, thick cranial vault bones, long, low brain cases with only partial forehead elevation, and a strongly buttressed occipital region. They often retain a sagittal keel or parasagittal thickening. Their facial skeletons seem adapted to heavy biting and other behaviours, strong zygomatic arches, deep maxillae, and expanded nasal cavities that may relate to energetic or climatic demands. These traits appear consistently across Africa and Europe from roughly 600,000 years to at least 300,000 years ago, suggesting a single evolving population network rather than unrelated local forms. Thus it is reasonable to call them a single group. But this does not equal ancestry to Homo sapiens. Functionally, their faces likely responded to mechanical and environmental pressures very different from those shaping our own. Some hypotheses propose that the large midface and brow ridges of African hominins were tied to powerful mastication of tough foods, possibly with less cooking and processing than later humans. Others link the broad nasal aperture and paranasal sinuses to cold or dry climates, aiding air conditioning. Still others see social display or structural integration with a large brain and heavy chewing muscles. Whatever the mix of causes, the result is a face that grew forward and remained massive. Modern humans, in contrast, underwent facial gracilization, midface reduction, development of a chin, segmented brows, and a tucked subnasal region. Cultural and dietary shifts, including cooking and tool use, likely relaxed mechanical constraints, allowing natural selection and neutral processes to favor a smaller, more delicate face. The developmental data make this clearer. Modern human midfaces form through early bone resorption beneath the orbits and along the nasal margin, creating a canine fossa and pulling the face under the brain case. In Bodo and Kabwe, growth is dominated by deposition, pushing the midface forward. To evolve a modern face from these robust forms would require re-patterning facial growth in a way that essentially reverses long-established developmental pathways, possible in theory, but phylogenetically unparsimonious compared to deriving both Neanderthals and Homo sapiens from an earlier Jungtian-like base. The timing of the Neanderthal-modern human split also undermines the old model. If Neanderthals were already evolving distinctive faces by 400,000 years ago, and their lineage split from ours perhaps 600,000 to 700,000 years ago, then the true last common ancestor predates the bulk of the Chibanian fossils. Ancestral candidates must be older than 600,000 years ago and show a facial pattern compatible with both Neanderthals and modern humans. Thus, while the clade may share a common ancestor, that ancestor is not the one shared by us and Neanderthals. The young ages of Kabwe and similar fossils also have cultural implications. Archaeologists once assumed that Middle Stone Age toolkits in Africa signalled Homo sapiens, but the 299,000-year age for Kabwe shows that archaic non-sapiens hominins were making, or at least associated with, early Middle Stone Age artefacts. Africa 300,000 years ago was not home to a single evolving lineage, but a patchwork of contemporaneous humans. Early Homo sapiens at sites like Jebel Irhoud and Florisbad, late surviving hominins such as Kabwe, perhaps Homo naledi in the south, and possibly other ghost populations. This mirrors Eurasia, where Neanderthals, Denisovans, late Homo erectus, and small-bodied island forms coexisted. The story of modern human origins is thus reticulate and pan-African, not linear replacement. Chris Stringer's own intellectual journey exemplifies this shift. Once a leading proponent of a simple African ancestry, he now emphasises the complexity and polycentric nature of modern human evolution. The Kabwe redating prompted him to publicly acknowledge that the ancestral link is uncertain. And that this skull is surprisingly young. It shows no significant derived features of our species. His review on the face went further, explicitly rejecting the idea that the robust middle Pleistocene African crania represent the ancestral morph for Homo sapiens. Instead, he and his colleagues argue for a deeper, possibly Jungtian, man-like ancestor, and for recognising the African and European fossils as a separate branch. This re-evaluation also affects how we interpret gene flow and introgression. Genomic studies detect traces of archaic African lineages in modern populations, so-called ghost introgressions. Late surviving hominins, such as Kabwe's population, could have contributed such DNA. Their survival until at least 300,000 years ago means they could have overlapped and interbred with emerging Homo sapiens. And I'm now much less confident about who that common ancestor was. I used to think it was Homo heidelbergensis. Now, I don't know who the common ancestor is and I don't even know where that common ancestor lived. The ancestor may have lived in Africa, but equally it could have lived in Asia, could have even lived in Europe. So I think, and as it says here, this age estimate raises further questions about the mode of evolution of Homo sapiens in Africa and whether Heidelbergensis rhodesiensis was a direct ancestor of our species. And as we say here, the evidence instead suggests that the large and non-sapiens-like faces of Bodo and Broken Hill represent taxonomic diversity in the African middle Pleistocene record, which could exclude the large fossils assigned to Heidelbergensis rhodesiensis as representative of an ancestral morph for Homo sapiens. Thus, even if they are not our direct ancestors, they may have left a genetic legacy. This parallels how Neanderthals and Denisovans, those side branches, contributed small fractions of our genome. But if they didn't come from Africa, then this raises the possibility of an into-Africa migration. Another important aspect is the distinction between morphology and ancestry. The African and European fossils clearly share a common evolutionary root. They are a group that diverged from earlier Homo erectus-grade ancestors, acquired a suite of cranial specializations, and radiated across two continents. But the fact that a group is similar does not mean our species descends directly from it. Phylogenetic trees can have side branches that are internally cohesive, yet end in extinction or in gene-flow mosaics. The African clade seems to be such a branch, a successful widespread population complex that persisted into the late middle Pleistocene, but was not the trunk from which Homo sapiens arose. The facial evidence has become especially decisive because the face integrates many functional and developmental signals. By comparing growth patterns, surface resorption, and bone deposition, researchers can reconstruct ancestral states. The modern human face appears to have deep roots, with some early African and even Eurasian fossils hinting at its beginnings well before 300,000 years ago. This undermines any model that tries to derive our facial plan from late, massively built, Kabwe or Bodo-type skulls. Instead, modern human facial evolution looks like a long-term trend toward reduction and retraction that started early and continued gradually, while the lineage retained or even exaggerated robust features. The Kabwe redating also shows the dangers of assuming age from context. For decades, because the mining site was destroyed and the fossil lacked clear stratigraphy, scientists simply estimated 500,000 years ago, based on faunal correlation and morphological feel. Direct geochronology overturned that. Similar reassessments may affect other key fossils and further scramble the simple ancestor-descendant charts. Taken together, the evidence now points toward this picture. During this time, Africa and Europe were inhabited by a robust hominin clade that shared a suite of derived archaic features and persisted surprisingly late. Parallel to them, and perhaps earlier in origin, were lineages that would give rise to Homo sapiens, already showing facial reduction by 500,000 to 300,000 years ago in specimens such as Jebel Irhoud, Floris Bad, and possibly older African material. Neanderthals diverged even earlier, with their own path to a large mid-face beginning at least 400,000 years ago in Cima de los Huesos. Genetic exchange likely occurred among these groups, but the direct ancestor of modern humans is older and more gracile-faced than the robust hominins. Chris Stringer, once a champion of the Heidelbergensis ancestor model, now cautions that the age and morphology of Kabwe and its kin make them poor candidates for our species' root. In this more complex framework, these hominins are important not as our forebears, but as a window onto the plurality of human forms that shared Africa with our emerging lineage. They remind us that modern human evolution was not a smooth climb, but a braided river of populations, some leaving genes, some leaving only fossils. Their massive faces and late survival challenge the idea of steady, progressive modernization. Instead, they show persistence of robust archaic morphologies alongside early modern humans, underlining the continent's long-term diversity. The shift from a Heidelbergensis ancestry model to a Pan-African reticulate model with deeper roots also influences how we define species. If Kabwe-like hominins were contemporary but morphologically distinct, perhaps reproductively partially isolated, yet potentially capable of interbreeding, then strict cladistic species boundaries blur. Modern human origins may have involved assimilation of multiple African archaic populations, but the core lineage leading to Homo sapiens seems to have branched off before the rise of the robust type. Thus, the combination of new geochronology, facial developmental analysis showing that Bodo and Kabwe faces cannot parsimoniously transform into modern ones, and ancient DNA divergence, estimates pushing the Neanderthal sapiens split back beyond 600,000 years ago, has dismantled the idea that Homo rhodesiensis is our ancestor. Instead, it stands as a side branch. Now click on these other videos to learn more about our human journey. Thank you for watching.