Last year around this time, my students asked me, “How threatened do you feel the current political climate is regarding human evolution, medical genetics, and your own research?” I gave them my honest answer and shared it here with readers.

Now it’s time for an update. My outlook on the science has gotten stronger. We are seeing advances like never before. Meanwhile, the federal funding outlook for human origins research is worse than any time since the Second World War.

It’s understandable that many U.S. scientists are dispirited by the challenges coming from the current administration. This is not, as I’ve seen some say, the end of the science—far from it. Our scientific work is racing forward faster than ever. But this is a turning point. The future depends on building better relationships for people and science.

For my regular readers, I first want to thank you. Your support for this site is making an enormous difference. One of the reasons why I’m able to keep my research at the highest level is because of you, and I have some really exciting work coming up.

I plan to continue to keep all my writing here free for anyone to access. Many people all over the world for whom a Substack subscription would be a big part of their monthly budget are able to read here for free. Teachers from all over the U.S. and internationally are relying on the site in their work. So again I want to especially thank all the supporters who are making this possible with your founding memberships and paid subscriptions.

The big news on federal science funding in recent weeks is that the National Science Foundation (NSF) has made its fiscal year 2027 budget request, including the elimination of its Directorate for Social, Behavioral, and Economic Sciences (SBE). That is the part of NSF that includes anthropology and archaeology. Biological anthropology, cultural anthropology, and archaeology programs are among the many that will be eliminated at NSF by this proposal.

I’ve been expecting this move for a long time. For two decades or more, federal funding for social sciences has received more and more criticism from legislators, especially—but not only—Republicans. Once, the biggest threats to federal funding in archaeology and anthropology were members of Congress and lobbyists who saw evolutionary biology as a threat to creationist teachings. That’s changed over the last two decades. Public acceptance of evolution has reached new highs. At the same time, more and more people now oppose public spending on scientific work related to social, racial, or political issues. The deep history of humankind is not a partisan subject. Yet understanding the social, racial, and political aspects of humans and the history of those ideas matters a lot to the study of human origins—always has, and always will.

Of course the final decision on budget is not with the Executive Branch but with Congress. Last year, Congress eventually allocated much more money to NSF than the Executive Branch requested. Many saw that as a positive step for support of science, and it might happen again during the budget negotiations for next year.

But what we’ve seen in 2026 is that NSF is not spending the funds that Congress approved. A number of graphs have been going around, showing the drop-off of grants in 2025 from earlier years, and the incredible reduction so far in 2026. The Grant Witness site is compiling this information from public data, with declines in funded awards across all areas of NSF. For the SBE directorate, the situation is especially dire. The agency’s change in priorities has already taken effect:

Again, nearly all anthropology and archaeology funding comes under the SBE budget line. As of this writing, zero grants have been approved in these areas of study in fiscal year 2026 so far.

How big will be the effect on these research fields? Taking biological anthropology as an example, the number of grants to PhD scientists in biological anthropology has averaged twenty or so per year for the last decade. Fourteen researchers were awarded grants with start dates in 2025; twenty-five with start dates in 2023 and 2024, nineteen with start dates in 2022. In terms of the number of active researchers in the field, that’s not so large—but it will have a strongly negative effect on some kinds of research by some of the most active researchers. NSF has higher impact on dissertation research by graduate students. In biological anthropology, around 25 to 30 dissertation proposals have been funded each year for the last decade. This program was “archived” by NSF last year, effectively ending it.

The federal funding scene for human origins is wider than NSF. Some biological anthropology and evolution research is funded by the National Institutes of Health (NIH), especially those areas that impact lifespan, genetic variation, and pathogens. These are tiny parts of the overall NIH funding landscape, but because NIH is so much larger than NSF it plays an important role, especially in the area of human genetics. Like NSF, NIH has also seen massive shifts in priorities and a strong reduction in new programs under the Trump administration. It’s hard to assess at present how much the changed priorities will reduce support for work that has importance in understanding things like Neanderthal genetics and population history—but it is certain to be a substantial reduction, not an increase.

Fulbright has also been an important source of federal funding for field research in human origins, through international exchange. My research was greatly advanced by a Fulbright as I began working in South Africa, and these grants enable many students to complete their dissertation research. The Fulbright program is continuing, but I know people whose proposals have been rejected during the last year due to their inclusion of language related to climate and environment. Again, changing priorities are excluding some research in biology, anthropology, and archaeology.

I also have to say that human origins research is a broad topic that draws on results and methods in geology, genetics, biology, and other fields. A lab doing geochronology may be supported through a geosciences funding stream. Comparative DNA research may be supported by a funding stream for human genome research. In the U.S., these other areas are experiencing their own funding disruptions, but they are not ending. They were always supported more reliably than archaeology or biological anthropology. What suffers when redirecting funding into those other areas is fieldwork and original data collection on ancient people and their relatives.

The changes under the current administration are abrupt and severe. But they are not a sudden reversal; they represent an acceleration of the long decline in federal support for this kind of science. To be sure, there was never a golden age of federal support for human origins. Still, twenty years ago and longer, there was relatively more support for field research from NSF, including programs such as the HOMINID grants that were important to a small number of larger-scale projects. In recent years, NSF has prioritized fields like physics, computer science, and artificial intelligence. Grants related to human origins have gotten smaller in direct costs (correcting for inflation), more of the funding has gone to indirect costs, and fewer have been funded. In fact, if you look at the graph above, every year of the Biden administration saw a decline in SBE awards.

I strongly support increased federal funding for human origins research. It’s a smart investment. Even though I’ve never received funding from NSF, and even though I don’t always agree with the direction that funds have gone, this is an area where a federal role just makes sense. The research in this field does not merely benefit institutions or corporations, it benefits everyone. It’s an area where public leadership can be transformative: helping scientists share their data more broadly, work across disciplinary boundaries, and build research connections across international borders.

Other nations, including China, Canada, many individual European nations as well as the European Union, have increased their investment in this area of scientific research. (For readers in those nations, thank you for your support!)

The field regularly generates “top ten science stories” every year. Last year it was the “Dragon Man” DNA and proteomic discoveries, and a spectacular partial skeleton of Homo habilis described already this year with more to come. What it takes to maintain substantial field research at a significant archaeological or fossil site including staffing is generally on par with the budget of one single laboratory in biomedical research. Very few opportunities for discovery come at such low cost relative to the scientific results. Moreover, almost every project in this area of research involves people from many disciplines solving problems together—a model for transdisciplinary work that is increasingly important in bigger, more expensive fields.

Consider the enormous achievements in just the last decade. Ancient DNA and other molecular evidence have transformed our understanding of human variation and its history. Research at known fossil sites has generated abundant new information about human ancestors and relatives, while exploration has turned up many new sites—some in unexpected places and times. Technology is opening windows into how sites formed, how ancient hominins behaved, how they grew and developed, and how they overcame or succumbed to disease. The leverage right now on new problems is extraordinary.

All that being said, I have a broader call to action. Most research in human origins is funded by foundations, universities, and private donors. My work has been almost entirely funded by university, foundation, and small individual contributions. Connecting people with research in this field is more important than it’s ever been to its future.

I know a lot of researchers who are waiting to hear from NSF, but I don’t know any who are standing still. They’re finding other ways to support their work and building new innovative collaborations.

I’m going to keep doing as much as I can to bring attention to new work, especially the work that is making me think. I hope that you’ll join me. If you’re an active supporter, thank you once again for helping keep this going. And if you’re just reading along, please spread the word about the discoveries that are changing our view of deep human history. Whether it’s a lecture, or an outreach event at your local institution, or a shared email, there’s a place for you as we uncover our shared past.

Earlier this month, Ran Barkai and Muataz Shalata published a new article describing fascinating Acheulean artifacts from the Sakhnin Valley of northern Israel. They describe handaxes made from geodes, handaxes with fossils on the cortex, handaxes with weird holes that look like tiny caves.

A handaxe is the ultimate stereotypical tool. It’s easy to imagine them all the same, again and again, minor variations on a global pattern. But the definition of a handaxe is really quite loose: almost any large tool flaked on both sides toward a teardrop-like shape might qualify. They became more stereotyped in some contexts, especially after around 400,000 years ago.

It was in that time that occasionally archaeologists find an anomaly. A handaxe with a fossil. Or with exceptionally fine flaking around the edge, with a bit of a deviation. What seems like intentional symmetry, or intentional violation of symmetry.

Archaeologists have spent a lot of thought on such anomalies. Who doesn’t like sparkly things? Still, beyond the bling, the question remains: What did the toolmakers have in mind when they produced these objects? It’s an old question with a lot of importance for the role of technology in our cognitive evolution.

Stones from the olive groves

Sakhnin sits between the Sea of Galilee and the Mediterranean coast, in a low valley that makes a logical travel route for the 60-kilometer trek between them. Muataz Shalata is a resident of Sakhnin with an avocation of archaeological exploration who saw the promise of the valley with its many flint tools scattered across various locations. He reached out to Ran Barkai, and together they launched an archaeological survey.

They found what seems to be a rich Acheulean landscape. As their team worked, they recovered over 200 handaxes, alongside other typical Lower Paleolithic artifacts. Most are made from local flint nodules, well suited for flaking large cutting tools.

Most of the handaxes had a standard shape, and with the degree of standardization and intensity of flaking, Barkai and Shalata suspect that they are products of the Late Acheulean, likely between around 500,000 and 200,000 years ago. They did not excavate profiles with methods that could date in situ artifacts.

Among the handaxes were some artifacts that stand out. The ancient knappers worked with some flint nodules that contained marine fossil imprints or even entire embedded geodes. On some of these nodules, they flaked the stone so that the anomalous features sit within the face of the tool.

A few of the artifacts are produced on natural geodes, with the geode exposed on the face of the handaxe. Two that I’ve pictured at the top of the post have the geodes centered in the artifact. Another has it sligthly offset, but highly striking in a way that indicates the knapper worked around it.

Another geode was used for its naturally spherical shape like other spheroid pounding tools.

One curious piece is a large, thick, pointed handaxe on which the knapper left a weathered fossil-bearing surface. To my eye, the shape of the handaxe is similar to those I’ve seen made from elephant bone, where the original shape of the material dictates a limited degree of flake removal. In the Sakhnin case, the remaining weathered surface with its tiny fossils stands out with a striking texture.

An aesthetic sense

Over the years, archaeologists have often noted Acheulean handaxes or other objects with fossils centered on one side, or other natural features that seem to have been enhanced by the knapper. These are some of the most iconic artifacts from our evolutionary past.

What do they mean? Many archaeologists over the years have argued that such artifacts are evidence of an aesthetic sense. The

It is easy to forget how many cultures have organized—and still organize—technology around notions of spirit and power. People notice natural variations in stone and other materials. The unusual draws their attention. They interact with it. Those interactions sometimes attract the attention of others. In any hominin species that transmits culture, which was all of them, shared attention may initiate a shared perception of interest, spirit, power.

A simplistic economic analysis of stone toolmaking may omit the role of cultural preference, status, and power in the analysis of Acheulean artifacts. But have humans ever invented a tradition of handcrafting that is purely utilitarian? As they say in AI, attention is all you need.

When is a shell just a shell?

Still, archaeological finds are a very sparse record of ancient behavior. The “Acheulean” as understood by archaeologists spans more than a million years across parts of three continents. A few of the sites with handaxes and other large bifaces also have hominin fossils, and those belong to a diverse set of groups as recognized by biologists. With this kind of evidence, it may not be possible to show that a particular artifact, however special-looking, really had anything special about it.

The best-known example of possible aesthetic intent in a handaxe came from West Tofts, near Norfolk, England, discovered in 1911. This flint handaxe that features a beautifully preserved shell right in the middle of one face. For over a century, archaeologists and art historians have argued about what this handaxe might signify about the minds of ancient hominins.

The shape of the handaxe seems like a perfect frame highlighting the fossil. This is no chance occurrence; at the very least, the knapper noticed the shell and worked around it. But was it more than this? Many archaeologists have considered the object symbolic, intended to communicate something to other individuals who saw it.

The problem of a singular object of this kind is the unanswered hypothetical. How likely are such objects with no aesthetic intent at all?

In 2023, Emily Flanders and Alastair Key tried to answer this question. They looked at the West Tofts handaxe from the perspective of knapping. They put the artifact through a micro-CT scanner, mapping the internal structure of the flint. The nodule of flint is relatively thin, so the kind of flake removals needed to shape this handaxe was very minimal. The fact that cortex—the outer surface of the natural nodule—still remains on both faces of the artifact shows that leaving the fossil did not impede the knapper from making a functional cutting edge.

The scan also showed the internal air pockets and other inclusions within the nodule, possibly including additional fossilized shells. The heterogeneity of the material, with natural internal fracturing, would have made it challenging to remove more flakes. Since the shell itself is located in a slightly concave area of the cortex, the flakes around the edge of the handaxe naturally ended where they did.

Flanders and Key concluded that the West Tofts handaxe is no more than a “remarkably average, structurally flawed, utilitarian biface”. Maybe sometimes a shell really is just a shell.

Bottom line

As Flanders and Key put it, there is no disagreement that today’s people see aesthetic qualities in the West Tofts handaxe. Nor is there much disagreement that aesthetics may have been important to hominins who made Acheulean artifacts. They considered a much more limited question:

The problem in my view is that archaeological samples from this time depth are never sufficient to answer that question. It’s equifinality—the principle that different sets of events can lead to the same outcome. It would take a lot of handaxes to test the hypothesis that some fossil-bearing ones were intentionally shaped.

That’s one reason that Sakhnin valley collection may be very interesting. A localized cluster of handaxes with fossils, geodes, and other strange inclusions might help understand how much intention there was to the creation of these fascinating objects.

What I keep in mind is that random chance is not an option. The most utilitarian scenario for the production of the West Tofts handaxe nonetheless involves many judgments of symmetry and consistency across the object. The recognition of various symmetries—not only reflective but radial and others—is part of the basic perceptual equipment of all hominins. There could be no handaxe production without it. Indeed, Oldowan core reduction involves various symmetries that set the production sequence apart from those known from nonhuman primates. Because of this cognitive architecture, the aesthetics of stone artifacts are inseparable from their production.

References

Barkai, R. (2021). The Elephant in the Handaxe: Lower Palaeolithic Ontologies and Representations. Cambridge Archaeological Journal, 31(2), 349–361. https://doi.org/10.1017/S0959774320000360

Barkai, R., & Shalata, M. (2026). Lower Palaeolithic Tools of Potency: Handaxes Shaped around Fossils and Other Extraordinary Features at Sakhnin Valley, Israel. Tel Aviv: Journal of the Institute of Archaeology of Tel Aviv University, 1–31. https://doi.org/10.1080/03344355.2026.2637187

Flanders, E., & Key, A. (2023). The West Tofts handaxe: A remarkably average, structurally flawed, utilitarian biface. Journal of Archaeological Science, 160, 105888. https://doi.org/10.1016/j.jas.2023.105888

McNabb, J., Cole, J., & Hoggard, C. S. (2018). From side to side: Symmetry in handaxes in the British Lower and Middle Palaeolithic. Journal of Archaeological Science: Reports, 17, 293–310. https://doi.org/10.1016/j.jasrep.2017.11.008

A few weeks ago I looked at the remarkable underground structures in Bruniquel Cave, created by Neanderthals more than 170,000 years ago and sealed until their recent discovery by cavers. Another cave with a similar history is La Roche-Cotard, around ten kilometers west of Tours, France.

Within an area known as the Pillar Chamber, around 15 to 20 meters from the current cave entrance, Neanderthals marked several parts of the cave wall. They made these marks with their fingers, tracing them firmly into a soft surface layer of the cave wall that still covers the harder bedrock beneath.

The entrance of the cave was uncovered in 1846 when railway workers were quarrying sand and gravel to build an embankment. Before that, the cave had been sealed under deep layers of sediment.

In a recent paper, Jean-Claude Marquet and coworkers described the panels of finger markings, each including dozens of parallel lines often converging into shapes or filling areas of the wall. The researchers also applied OSL dating to remaining sediments outside the cave entrance, to better understand the time that these sediments closed up the cave entrance. They determined that the cave closed around 57,000 years ago, with a minimum (95%) age of 51,000 years. This date is only a minimum for the engraved markings inside the cave, which may have been made much earlier. Whatever their true age, they were made before any evidence of modern humans in this region of France.

Finger-flutings

The cave of La Roche-Cotard formed within a Cretaceous limestone that has a crumbly, sandy texture. The cave formed by gradual dissolution of the rock, which has left a relatively soft, thin clay surface still adhering to the hard cave walls. It is this surface that the Neanderthals could shape with their fingers, plowing little furrows.

The finger-flutings at La Roche-Cotard are the only ones known so far from Neanderthal times. Finger flutings and similar trace markings within soft sediment occur very widely in the ancient rock art record.

• Hand and footprints in travertine at Quesang, China date to between 220,000 and 170,000 years ago, probably made by Denisovans.

• On the southern coast of South Africa, geometric tracings made in ancient sand surfaces date to more than 100,000 years ago.

• Finger flutings occur at many sites in Australia, with some of the oldest noted at Koonalda Cave, dating to more than 20,000 years.

• Some Upper Paleolithic contexts in Europe have finger flutings, with notable examples including some of the earliest-known painted caves, El Castillo, Spain, and Chauvet Cave, France.

Why were these markings made? In some caves and rockshelters, the finger flutings are only a part of a more extensive corpus of painted or engraved markings. They were made in many different contexts, extended across different periods of time and different geographic regions.

This gives little reason to think there was a single impetus for these marks wherever they occur. They are just one of many ways that people have marked their environments. For the Neanderthals, geometric or repeated linear marks have been recovered from engraved pieces of bone from many sites, as well as from the floor of Gorham’s Cave, Gibraltar.

Over the years some researchers have suggested that finger-flutings were often made by children. The idea initially came from the unusual places where markings occur, including the high ceilings of cave chambers. Keryn Walshe, April Nowell, and Bruce Floyd evaluated this idea in a 2024 article, testing whether finger sizes could reliably be assessed from fluted markings. They found that the errors from measurement of such markings outweigh the small differences separating the sizes of adult and child fingers, and fingers of men and women, giving no basis for estimating the age or sex of people who made Upper Paleolithic fluted markings.

In La Roche-Cotard, Marquet and coworkers used an experimental approach to better understand the way the fluted markings were made. They were especially concerned to ensure that the marks could not be attributed to nonhuman species such as cave bears or hyenas—important because some claw marks consistent with cave bears do occur in the cave. But in their experiments, Marquet and coworkers found the smooth, repeated parallel nature of the finger-flutings were unmistakably human in their manufacture.

Red ochre markings

Evidence for pigments is quite abundant at many Neanderthal sites, including red ochre and black manganese dioxide. In some places it is clear that the pigments were reduced to powder and mixed with fluid. In others, small blocks or “crayons” of pigment show traces of wear upon them.

Pigment marks on perishable items do not survive long in archaeological settings. Rock walls inside of caves may sometimes retain such markings, but archaeologists have attempted to find minimum ages for only a few such marks in European cave sites.

Some of those date to Neanderthal times. Cueva de Ardales, Spain, has pigment marks with estimated minimum ages of 68,000 years ago. At other sites, marks on walls or red ochre stains on natural objects like shells show that the Neanderthals were applying pigment.

La Roche-Cotard also has a few red ochre marks, described briefly by Marquet and coworkers in a 2014 article. These small pigment spots occur within the area just beyond the Pillar Chamber and its fluted markings, in the passage toward the Hyena Chamber. Whether these were deliberately placed marks or incidental results of a painted hand or body surface placed on the rock surfaces is not clear.

Marks like these occur in many cave sites, some much better known for figurative paintings that date to the Magdalenian or earlier Upper Paleolithic periods. In such contexts it has been common for rock art specialists to assume that all marks were made by artists from the same time period. But increasingly, researchers are recognizing that caves are palimpsests with marks made in very different eras by different populations of humans. Sometimes modern people walked into caves where Neanderthals had made their mark.

In La Roche-Cotard, they didn’t get the chance. The Loire River changed its course away from the north side of the valley in front of the cave. Sediments built up, no longer swept clear by the river, and covered the cave entrance, sealing the work of the Neanderthals within.

The “Mask” of La Roche-Cotard

The most enigmatic object comes not from inside the cave but from the sedimentary layers outside its entrance. In 1975, archaeologists put a trench into the terrace outside the cave to understand what lay beneath. Within layer 7 they found an array of Mousterian artifacts, butchered animal bones, and an apparent hearth. Among them was a curiously-shaped flint core.

It has a roughly trapezoidal shape, flaked around the edges so that the shape and size were chosen by the maker or makers. There is a natural hole passing through the rock in the shape of a tube. Within this tube is wedged a long bone splinter, 74 millimeters in length, held in place by small pieces of flint.

The shape gives it the striking appearance of a face with eyes. After its discovery, Michel Lorblanchet called it a “mask”, and that is how it has been known ever since: the mask of La Roche-Cotard.

In 2016, Marquet and coworkers undertook a redating of the site using OSL, which reaches into time periods earlier than the 1970s-era radiocarbon results allowed. This is the same work that ultimately showed the cave closed before 51,000 years ago. They found that the layer 7 Mousterian assemblage outside the cave, including the mask, dates to 75.6 ± 5.8 thousand years ago.

As I started investigating this object, the thing that surprised me is how big it is. It’s roughly four inches in diameter and weighs just under a pound, 299 grams. This is no incidental product; somebody wanted it to look the way it does.

Does the object represent a human (or animal) face? Anyone who reads this site will not be surprised to hear that archaeologists have diverse opinions about this question.

Marquet and coworkers have proposed that the mask is similar to other objects from the deep archaeological record that resemble human forms, slightly shaped to enhance the resemblance. These pierres-figures, or figure stones, include such varied artifacts as the so-called “face” from Makapan dating to more than 2 million years ago, and the Berekhat Ram figure from Israel, more than 230,000 years old.

Bottom line

What do these markings say about the Neanderthals? As I’ve written many times, it is abundantly clear that these ancient people had an aesthetic sense. Markings like these cannot speak for themselves; they were part of a broader system that is invisible to us today.

There were hundreds, more likely thousands, of different Neanderthal cultures across their long existence. People within each culture came to know each other, and many of them found markings a way to reinforce their similarities or signal their differences. Some left signs for each other. Others marked places as a way of expressing their connection, ownership, or spiritual bond to their homes.

This cave site, with its south-looking entrance on the slopes above the great Loire River was no doubt the kind of place that the Neanderthals noticed and valued. When they built a fire outside it, eating their kill, who can say what a curious stone with a hole perforating it may have meant to them? Did it strike their fancy, inspiring a bit of play that one of them shared with the others?

I think once sitting by a fire with friends I may well have done the same.

Notes: We owe a great deal to Jean-Claude Marquet and coworkers, who have published most of their work in open access journals under Creative Commons licenses, including Paléo and PLOS ONE.

References

Hoffmann, D. L., Standish, C. D., García-Diez, M., Pettitt, P. B., Milton, J. A., Zilhão, J., Alcolea-González, J. J., Cantalejo-Duarte, P., Collado, H., de Balbín, R., Lorblanchet, M., Ramos-Muñoz, J., Weniger, G.-Ch., & Pike, A. W. G. (2018). U-Th dating of carbonate crusts reveals Neandertal origin of Iberian cave art. Science, 359(6378), 912–915. https://doi.org/10.1126/science.aap7778

Kelly, M., David, B., Rivero Vilá, O., Garate Maidagan, D., Delannoy, J.-J., Mullett, R., Birkett-Rees, J., Petchey, F., Barker, A., Arnold, L. J., Green, H., Fresløv, J., & GunaiKurnai Land and Waters Aboriginal Corporation. (2025). Finger flutings at New Guinea II Cave, lower Snowy River valley (Victoria), GunaiKurnai country. Australian Archaeology, 91(2), 133–163. https://doi.org/10.1080/03122417.2025.2529627

Marquet, J.-C., Freiesleben, T. H., Thomsen, K. J., Murray, A. S., Calligaro, M., Macaire, J.-J., Robert, E., Lorblanchet, M., Aubry, T., Bayle, G., Bréhéret, J.-G., Camus, H., Chareille, P., Egels, Y., Guillaud, É., Guérin, G., Gautret, P., Liard, M., O’Farrell, M., … Jaubert, J. (2023). The earliest unambiguous Neanderthal engravings on cave walls: La Roche-Cotard, Loire Valley, France. PLOS ONE, 18(6), e0286568. https://doi.org/10.1371/journal.pone.0286568

Marquet, J.-C., Lorblanchet, M., Oberlin, C., Thamo-Bozso, E., & Aubry, T. (2016). Nouvelle datation du « masque » de La Roche-Cotard (Langeais, Indre-et-Loire, France). Paléo, 27, 253–263. https://doi.org/10.4000/paleo.3144

Marquet, J.-C., & Lorblanchet, M. (2003). A Neanderthal face? The proto-figurine from La Roche-Cotard, Langeais (Indreet-Loire, France). Antiquity, 77(298), 661–670. https://doi.org/10.1017/S0003598X00061627

Pettitt, P. (2003). Is this the infancy of art? Or the art of an infant? A possible Neanderthal face from La Roche-Cotard, France. Before Farming, 2003(4), 1–3. https://doi.org/10.3828/bfarm.2003.4.11

Walshe, K., Nowell, A., & Floyd, B. (2024). Finger Fluting in Prehistoric Caves: A Critical Analysis of the Evidence for Children, Sexing and Tracing of Individuals. Journal of Archaeological Method and Theory, 31(3), 1522–1542. https://doi.org/10.1007/s10816-024-09646-9

A new preprint from the geneticist David Reich focuses on the interactions of Neanderthal and African ancestral humans some 250,000 years ago. The research connects with several ideas about Neanderthal DNA that I’ve discussed here and elsewhere, including the matrilineal population dynamics I outlined recently. Reich adds a hypothesis that hasn’t been discussed much by geneticists: Maybe the range expansion that delivered African genes also brought cultural innovations to early Neanderthals.

Archaeologists have long noted technical commonalities between stone tool industries associated with Neanderthals and Middle Stone Age (MSA) tools made around the same time in Africa. Reich discusses the Levallois flaking strategies that underlay Mousterian toolkits, greater use of fire, and improved hafting technology. I would add the use of ochre and other pigments, greater resharpening and recycling of artifacts, and more organized structuring of campsites. Could such similarities in material culture have followed one or more waves of migration from Africa into the Neanderthal range?

Ancient peoples who could trade genes could surely also trade ideas. Understanding such multiregional connections has long been important to my own work. Episodes of gene flow, idea exchanges, and coevolution of genes and cultural adaptations are a generalized version of this model that I share.

Still, Reich’s specific idea that an African range expansion may have “formed” the Neanderthal population and its cultural repertoire is quite a bold statement. In it, I see something very much like the “Mode 3 hypothesis” proposed by Rob Foley and Marta Lahr in 1997. In this strong form, Reich’s idea is disproved by the archaeological record of Asia, where Denisovans shared the Levallois and other cultural innovations of Neanderthals and MSA Africans.

This broader pattern of cultural exchanges is not surprising or unexpected. It may be consistent with a role for African genes in the evolution of Neanderthal behavior.

Sex and dispersal

In the article, Reich examines the African gene flow into Neanderthal groups sometime around 250,000 years ago. (In my diagram of genetic mixture among ancient genomes above, this episode is designated with the number 8.) The most profound effect of this gene flow was the complete replacement of earlier Neanderthal mitochondrial DNA (mtDNA) and Y chromosome branches by African lineages. In addition to these two uniparental genetic systems, the African gene flow makes up around 6% of the whole genomes of later Neanderthals. Reich notes that this combination of mtDNA and Y chromosome replacement seems unusual when coupled with more modest nuclear gene flow.

A couple of weeks ago, I recounted new research by Alexander Platt and coworkers, who were the first to characterize the X chromosome bias of that African gene flow into Neanderthals. One possible explanation for their findings is a matrilineal social organization in an expanding African population as they encountered and mixed with Neanderthals.

Reich shows that sex-biased dispersal might also explain the mitochondrial and Y chromosome results. The article’s main innovation, established in simulations, is that sex bias in either a paternal or maternal direction would increase genetic drift with disproportionate effects on both uniparental systems. Reich notes the results from Platt and coworkers that suggest a maternal bias, but suggests that the X chromosome results are much weaker than either uniparental system and suffer from the general challenge of accurately identifying introgressed segments. So he leaves open the possibility of an excess of gene flow from men rather than from women.

The article further suggests that a similar process may have unfolded within Africa. The evidence that suggests today’s Africans derive a fraction of their ancestry from an “archaic ghost population” might be explained if a single early modern group began to grow in numbers and geographic range within Africa, mixing with other long-diverged groups.

The current idea of the timing of these events within Africa is that ancestors of today’s Khoesan, central African hunter-gatherers, and other groups started to differentiate after 300,000 years ago. These early ancestors may have absorbed genes from archaic African groups. One or more groups may have pushed into Eurasia at this time, spreading similar genes to the Neanderthals.

My take on the genetics

There is a lot in Reich’s article that I agree with. The core of the hypothesis is that the gene flow brought African genes to Neanderthal populations and may have transformed them in functionally important ways. I’ve been saying this for a long time. Indeed, a central theme in one of my earliest scientific papers, now twenty-five years old, is that Africa-to-Europe gene flow played a role in the transition from early Neanderthals at Sima de los Huesos and other sites to later Neanderthals.

Further, I see great potential in connecting the Middle Pleistocene dynamics within Africa with the dynamics outside of Africa. Over the last decade, the paleoanthropological trend has been to promote the early evolution of Homo sapiens as a “Pan-African” phenomenon, an idea that has been called “African multiregional evolution”. This idea, in my view, may be both too little and too much.

Too little: The Pan-African model ignores the genetic interactions of early African groups with Neanderthals and possibly other Eurasian peoples. We know the human network included these populations, but the Pan-African label leaves them out.

Too much: If early modern humans were networked across the entire continent of Africa, what room was there for archaic forms of humans and Homo naledi? These groups make up the majority of the later Middle Pleistocene fossil record, and their degree of niche overlap with ancestral modern humans is unclear.

In reality the patterns of movements, fissions, and mixtures of groups must have been heterogeneous and sporadic enough to enable persistence of many divergent groups, even while some became more and more tightly connected. By 40,000 years ago the dispersal and networking of modern human groups was certainly dominated by their strengths, but during their early evolution 250,000 years ago their limitations were still very evident. One reason I like Reich’s model is that it can unfold within this looser network of populations.

Still, there are aspects of the genetics that I think are unlikely. One is Reich’s treatment of the mitochondrial DNA and Y chromosome. Uniparental inheritance systems are highly susceptible to genetic hitchhiking—part of my work since my 1999 dissertation and discussed at length in 2006. Today the low variation of mitochondrial DNA and Y chromosomes compared to autosomal variation stands out, particularly so when we look at variation within Africa. An easy explanation for this mismatch is that one or more favorable variants arose in Middle Pleistocene Africans.

For mitochondrial DNA, selection likely coincides with the genetic load of slightly deleterious mutations that can become common or fixed in small populations. Connectivity between larger and smaller populations tends to help purge deleterious variants, and increased connectivity might favor the dispersal of mtDNA and Y chromosome variants from larger African populations into Neanderthals even without new adaptive mutations.

All this is to say that a strong sex bias in dispersal is not needed to explain the uniparental systems. All you need is a slight increase in connectivity. The same might be true within Africa, if the structure of populations before 300,000 years ago gave rise to enough drift.

A Levallois connection

When it comes to the archaeology, Reich’s hypothesis is very close to the “Mode 3 hypothesis” proposed by Rob Foley and Marta Lahr in 1997. Foley and Lahr noted a coincidence of timing of early Levallois assemblages in Europe and Africa starting around 300,000 years ago. Levallois and related methods of stone reduction are known as “prepared core” techniques, because they require the toolmaker to shape the stone core in anticipation of striking one desired flake. The archaeologist Grahame Clark had termed these methods as “Mode 3” technology. The planning required for these techniques seemed to represent an advance over the handaxe-dominated Acheulean (Mode 2) and chopper-based Oldowan (Mode 1) tools that preceded them.

Like Reich, Foley and Lahr thought that Levallois and other prepared core techniques may be markers of a biological dispersal.

Who was this dispersing population? Foley and Lahr proposed that Neanderthals and modern humans both evolved from a Levallois-using common ancestor. Like Reich, they accepted that Neanderthals were aligned with modern humans and different from the “true archaics” in Europe that preceded them.

The Mode 3 hypothesis was a departure from the way most archaeologists thought about material culture and modern human origins. Many archaeologists had focused upon the European Upper Paleolithic as the most important phenomenon, and these toolkits were centered on long, thin flakes called blades—“Mode 4” in Clark’s terminology. But Foley and Lahr considered Mode 4 techniques as a red herring. Blades were not universal across modern human societies, and instead Mode 3 seemed to be a technological common denominator. The interesting twist was that this seemed to include the Neanderthals. These authors provided several different ideas about how to classify the Neanderthal-modern common ancestor. The most memorable (illustrated in their diagram) attributed the first Mode 3 population to Homo helmei.

This wasn’t the only way to sort the archaeological, fossil, and genetic record of the time. I looked at this problem differently in my own work. The trajectory of increasing brain size evolution in Neanderthal populations across the last 400,000 years I interpreted as evidence for genetic exchange with African populations. In this context, cultural exchanges that might be marked by Levallois core reduction and other innovations were not surprising at all. Reich suggests that it’s a “conundrum” that African and Neanderthal populations share Levallois technology, but I don’t think that way. Such exchanges are a normal consequence of past human interactions.

Can a Mode 3 hypothesis work today?

The Mode 3 hypothesis seemed to work in the 1990s when evidence from much of Asia was scarce and poorly dated. Foley and Lahr noted a near lack of archaeological sites between 250,000 and 50,000 years ago in China, and noted nothing between Acheulean (Mode 2) and microliths (Mode 5) in South Asia. In that era, the old idea of a “Movius Line” separating the artifact traditions of eastern and western Asia was still a commonly held assumption.

Today’s record does not support that old way of thinking.

In China, an array of sites present evidence of prepared core techniques or hafting in the time range between 160,000 and 70,000 years ago. Hafting is evidenced at Xigou in central China, Lingjing provides evidence of ochre use and discoidal reduction with some Levallois cores, at a site with archaic human fossil material. Somewhat earlier, as early as 180,000 years ago, Guanyindong has many kinds of Middle Paleolithic core reduction techniques including core-on-flake, discoidal, and Levallois. The Panxian Dadong assemblage, roughly the same age, represents some of these same methods.

These Chinese assemblages are not identical to those from western Europe. They combine polar core reduction, some large cutting tools, and a smaller proportion of prepared core methods. They represent related solutions in different proportions. Almost certainly they were produced by Denisovans.

In South Asia, Levallois-based Middle Paleolithic assemblages at Attirampakkam, India, began by around 385,000 years ago, with representation in several levels before 170,000 years ago. Another site older than 245,000 years with Levallois and discoidal reduction is Hanumanthunipadu in Andhra Pradesh. The hominins of India during the later Middle Pleistocene are represented only by the Narmada partial skull, which is ambiguous. The fraction of Denisovan ancestry in today’s peoples of India suggests that one or more groups of Denisovans may be part of the hominin prehistory of this region.

The strongest evidence that Denisovans used Levallois and other prepared core techniques comes from Denisova Cave itself. The earliest stone assemblages from Denisova include Levallois cores as well as sediment DNA attributable to Denisovans, dating to between around 250,000 and 200,000 years. Throughout the Denisova sequence, these Mousterian-like elements are common, and in a number of layers elongated flakes and true blades. There is no categorical technical distinction between layers with Denisovan fossils and DNA compared to Neanderthals.

All this is to say that the Neanderthals were not the only population with technical similarities and other cultural overlaps with MSA Africans. Denisovans, too, used prepared core techniques, hafted points on spears, used ochre, and made markings. The record in East and South Asia remains underrepresented, but as it stands these techniques are not later or offset in time compared to Europe or Africa.

With this extent of cultural overlap, it is very interesting that Denisovans seem to have had less gene flow from African sources that Neanderthals did. I don’t think it’s likely that gene flow was absent—instead, it’s just an indication that cultural pathways for exchange were not one-to-one matches for gene flow.

Bottom line

When it comes to Reich’s hypothesis, I strongly agree with the parts that are established by other evidence—especially the importance of African mixture into Neanderthals and the possible sex bias of those exchanges. I also fully concur with the general idea that cultural exchanges accompanied these genetic exchanges.

But the part of the hypothesis that sets up a relationship between African range expansion and Levallois technology is a throwback to the 1990s that is out of step with today’s data. Denisovans, too, had Levallois and other material culture similarities with Neanderthals and contemporary African peoples. The Denisovans did not, as far as we can tell, have the same extent of gene flow from African sources. Still, future data might show that we are underestimating African-to-Denisovan genetic exchanges.

The specific idea that the sex bias was the cause of the mitochondrial and Y chromosome replacement in Neanderthals is questionable in my view. Natural selection on these uniparental systems is a better explanation.

When it comes to cultural exchanges, the important questions are how much such technologies may have been coadapted with neuroanatomy and development. Those cultures were the environments to which our species adapted. It does not take much gene flow to move adaptive genes, and understanding the pattern of Neanderthal adaptive evolution with African gene flow is the direction we should go.

Finally, maybe the most valuable part of Reich’s paper is that it reveals the weakness of current search strategies for “modern” genetic adaptations. Most have asserted that the important evolutionary changes in modern humans will be genetic variants that modern people have and Neanderthals lack. But our Neanderthal and African ancestors were part of the same genetic network, overlapping in a fraction of their genomes. To find the evolutionary changes that affected us, we’ll have to look more closely at the genes that found value in Neanderthals.

Notes: The technical modes derive from the classification by Grahame Clark in his 1969 book, World Prehistory: A New Synthesis. These have gone into and out of fashion over the years, and I tend to think they cause more confusion. A more functional technical classification was presented by John Shea in his 2016 book, Stone Tools in Human Evolution. In Shea’s perspective, Levallois and other prepared core techniques are classified as “Bifacial Hierarchical Core” reduction.

References

Akhilesh, K., Pappu, S., Rajapara, H. M., Gunnell, Y., Shukla, A. D., & Singhvi, A. K. (2018). Early Middle Palaeolithic culture in India around 385–172 ka reframes Out of Africa models. Nature, 554(7690), 97–101. https://doi.org/10.1038/nature25444

Anil, D., Chauhan, N., Ajithprasad, P., Devi, M., Mahesh, V., & Khan, Z. (2022). An Early Presence of Modern Human or Convergent Evolution? A 247 ka Middle Palaeolithic Assemblage from Andhra Pradesh, India. Journal of Archaeological Science: Reports, 45, 103565. https://doi.org/10.1016/j.jasrep.2022.103565

Bar-Yosef, O., & Wang, Y. (2012). Paleolithic Archaeology in China. Annual Review of Anthropology, 41(1), 319–335. https://doi.org/10.1146/annurev-anthro-092611-145832

Clark, G. (1969). World Prehistory: A New Synthesis. Cambridge University Press.

Foley, R., & Lahr, M. M. (1997). Mode 3 Technologies and the Evolution of Modern Humans. Cambridge Archaeological Journal, 7(1), 3–36. https://doi.org/10.1017/S0959774300001451

Hawks, J. (2006). Selection on mitochondrial DNA and the Neanderthal problem. In J.-J. Hublin, K. Harvati, & T. Harrison (Eds.), Neanderthals Revisited: New Approaches and Perspectives (pp. 221–238). Springer Netherlands. https://doi.org/10.1007/978-1-4020-5121-0_12

Hawks, J. D., & Wolpoff, M. H. (2001). The Accretion model of Neandertal evolution. Evolution, 55(7), 1474–1485. https://doi.org/10.1111/j.0014-3820.2001.tb00667.x

Ma, D.-D., Pei, S.-W., Xie, F., Ye, Z., Wang, F.-G., Xu, J.-Y., Deng, C.-L., & De La Torre, I. (2024). Earliest Prepared core technology in Eurasia from Nihewan (China): Implications for early human abilities and dispersals in East Asia. Proceedings of the National Academy of Sciences, 121(11), e2313123121. https://doi.org/10.1073/pnas.2313123121

Reich, D. (2026). Hypothesis: A modern human range expansion ~300,000 years ago explains Neandertal origins. Preprint. bioRxiv. https://doi.org/10.64898/2026.03.11.711219

Shea, J. J. (2016). Stone tools in human evolution: behavioral differences among technological primates. Cambridge University Press.

Shunkov, M. V., & Kozlikin, M. B. (2023). The Earliest Paleolithic Assemblages from Denisova Cave in the Altai. Archaeology, Ethnology & Anthropology of Eurasia, 51(1), 18–32. https://doi.org/10.17746/1563-0110.2023.51.1.018-032

Soriano, S., & Villa, P. (2017). Early Levallois and the beginning of the Middle Paleolithic in central Italy. PLOS ONE, 12(10), e0186082. https://doi.org/10.1371/journal.pone.0186082

White, M., & Ashton, N. (2003). Lower Palaeolithic Core Technology and the Origins of the Levallois Method in North‐Western Europe. Current Anthropology, 44(4), 598–609. https://doi.org/10.1086/377653

Yue, J.-P., Song, G.-D., Yang, S.-X., Kang, S.-G., Li, J.-Y., Marwick, B., Ollé, A., Fernández-Marchena, J. L., Shu, P.-X., Liu, H.-Y., Zhang, Y.-X., Huan, F.-X., Zhao, Q.-P., Qiao, B.-T., Shen, Z.-S., Deng, C.-L., & Petraglia, M. (2026). Technological innovations and hafted technology in central China ~160,000–72,000 years ago. Nature Communications, 17(1), 615. https://doi.org/10.1038/s41467-025-67601-y

In 2016, the archaeologist Jacques Jaubert and collaborators described a fascinating discovery deep within the Bruniquel Cave, France. This roughly linear cave is nearly a half-kilometer in length, situated within the steep-sided Aveyron valley. Most of this long system had been hidden by a scree slope inside a smaller cave until cavers dug through in 1990, opening passage much deeper into the hillside.

Three hundred thirty-six meters from the entrance, explorers noticed that the floor of the cave holds a series of circular structures made from broken stalagmites. More than 400 stalagmites were broken, moved, and piled into rings on the cave floor. Calcite had formed upon many parts of these structures after they were built, which enabled Jaubert and coworkers to use uranium-thorium to see how long ago they were built. The result was a minimum of 176.5 ± 2.1 ka. This was a time period when early Neanderthals inhabited western Europe, long before the arrival of any modern humans.

Upon some of those rings—sitting in a few cases right on top of stalagmites—are reddened or blackened “speleofacts”, objects interpreted as evidence of fire. Some fragments of burned bone and an area of “char” in one of the structures suggest that the makers of these structures had built and manipulated fires here, deep in the cave.

Last month, research led by Kim Génuite provided some new evidence about the use of Bruniquel Cave. By investigating the geology of the cave entrance, they have established a timeline for the opening and closure of the cave system and its use by early Neanderthals. Their work shows that the cave became closed sometime before 142.9 ± 1.3 ka, and likely was closed by gradual rockfall corresponding to the cold period of Marine Isotope Stage 6.

This new information corroborates the earlier findings. Early Neanderthals walked into this cave, went three football fields into the earth, created 15-foot-wide bubbles of rock, lit and tended small fires upon them. Then they left. Tens of thousands of years later, sliding and falling rock closed the cave, sealing in the evidence for more than 140 millennia.

Without a trace

I wrote about Bruniquel in 2016 when Jaubert and coworkers first established the age of the structures. At the time the site stood out as an outlier in many ways.

Little evidence was known in 2016 of any hominin-made structures anywhere as early as 170,000 years ago. Hominin access into deep caves at this geological age was not well known outside of the Sima de los Huesos site and the Rising Star cave system of South Africa. There were in 2016 no small number of archaeologists who doubted that Neanderthals ever had reliable use of fire.

Bruniquel combines all these things: fires upon structures built in a deep cave, all long before any direct evidence of modern humans in the region.

The structures were an enigma. The floor of the cave looks like someone tossed stones into a pool that suddenly froze, but the rings weigh hundreds of kilograms. No other hominin-made markings exist within the cave system. There is no evidence of a living floor or occupation area, and no stone tools or artifacts whatsoever. There are no hominin bones or physical remains. Whoever made these rings on the floor, they came in, broke and moved tons of speleothems, tended small fires in ways that heated and discolored many of the stones, and left no other trace.

When I last wrote about Bruniquel, I didn’t yet know much about the backstory of the discovery. That’s a sign of how the site had become a footnote to 1990s-era archaeology. The cavers and scientists who first entered the cave noticed the stalagmite rings and obtained an AMS radiocarbon age on burnt bone of >47.6 ka BP. As noted in the subsequent work by Jaubert and collaborators, any date above 45,000 years ago suggested that the site was earlier than the Upper Paleolithic, but at the very edge of the capabilities of the method it could not be taken as real. In the 1990s, archaeologists were primed to doubt any cultural capabilities of the Neanderthals.

Little could anyone have suspected that these structures are actually more than 170,000 years old. If not for Jaubert and coworkers who revisited the site starting in 2013, the site might remain a footnote today. Looking back, it’s striking to me how many other footnotes from the 1990s are left, waiting for someone to re-examine them with today’s approaches.

Fire

In earlier eras, archaeologists often noted discolored bone, blackened or reddened stone, and dark chunks of organic material as possible charcoal. That’s exactly the kind of evidence that Bruniquel presented, localized into small areas on the structures themselves.

But by the early 2000s Paleolithic archaeologists were experiencing a wave of skepticism about the firemaking capabilities of Neanderthals, and by extension all early humans. Too often blackened bone had turned out to be not burned, instead permeated by black or reddish chemicals like manganese and iron oxides. Without dense concentrations of burned material and ash, many archaeologists were willing to dismiss even genuine burned bone as the product of accidental or natural fires, carried into caves by erosional processes or chance.

Tiny blackened or reddened areas on speleothems, traces of blackened bones and unidentifiable “char”, of the kind found at Bruniquel, these kinds of traces were too easily dismissed.

The confirmation of the fire evidence at Bruniquel came from a method called “Raman spectrometry”, which was applied to three blackened fragments. Two of those fragments were embedded within calcite speleothems that showed ages consistent with the stalagmite structures. The results of the Raman spectrometry method showed bone fragments that had been heated to temperatures consistent with controlled fires. These gave the broader pattern of reddened and blackened speleothem evidence meaning as trace evidence of the manipulation of small fires. The situation is more akin to the votive offerings in a temple than the long-tended hearths that might have gotten more archaeological attention.

Over the last decade or more, I’ve felt a sea-change in the way that archaeologists examine ancient fire evidence. Forensic methods have become more standardized, as reflected by the Bruniquel results.

Later Neanderthals were masters of fire technology. Already by 2016 researchers had observed that the birch pitch used by Neanderthals at some sites would have required prolonged heating and sophisticated control of oxygen supply. Recently Juan Ochando and collaborators identified a specialized fire structure at Vanguard Cave, Gibraltar, that seems well-suited to exactly such tar production from resinous plants, in this case rockroses.

In past excavations, such occurrences might be noted as small hearths or possible cooking fires, but the structure and import of the evidence was certainly missed.

At the same time, the processual perspective on how fire fit into Neanderthal ways of living has also become richer and more capable of testing hypotheses. For example, in northern Europe the Neanderthals of the last interglacial, around 125,000 years ago, used fires to clear landscapes and generate favorable habitats that attracted grazing animals for hunting, as well as shifting plant foraging opportunities. Neanderthals were engineering the ecosystem at the Neumark-Nord site for more than 2000 years during this warm climate period, setting and using fire as part of their strategy.

Was it a stretch for the Neanderthals of Bruniquel to transport coals and create small fires for lighting, almost a quarter-mile into the rock? For people who lived by firelight any night they wished, it was surely routine.

Structures

All human societies all around the world have at least basic architecture. Shelters built from wood, brush, grass, or reeds are nearly ubiquitous. Many societies add other kinds of structures. Some are utilitarian: workbenches for crafting large wooden items, windbreaks and fire rings, hunting blinds, thorn bomas for keeping out predators. Others are ritual in purpose: mortuary stands for exposing corpses to the elements, astronomical markers or henges,

Great apes, too, all build structures. Every great ape sleeps in a nest of their own construction, generally built each night, and around 1–2% of their waking time is spent in construction.

It’s not too surprising that evidence for structures at early archaeological sites has been hard to find. Big rocks are not a primary component of most of the structures made by foraging peoples. But considering the evolutionary context, it is inevitable that many ancient groups would have built structures.

This is one reason why the preserved evidence of a large, carved log structure from Kalambo Falls, Zambia, was so striking when revealed in 2023. More than 470,000 years ago, hominins notched logs to interlock with each other and built some kind of low structure out of them. It’s not clear how large or extensive the finished structure may have been, nor is its purpose evident. In that way the Kalambo Falls evidence resembles the evidence from Bruniquel: Each is enigmatic in purpose, but each was clearly part of a tradition that was much broader than the single piece of evidence that survived.

Few ancient structures as evocative as the eight-meter circle built from mammoth bones at Molodova 1, Ukraine, more than 44,000 years ago. This evidence is old, first excavated in the 1950s and subject to repeated analysis and reanalysis by Soviet archaeologists and outsiders. Most recently, Laëtitia Demay and coworkers presented a review and summary of the evidence in 2012.

The stone tools associated with mammoth butchery at this site are Mousterian, and combined with the date cause archaeologists to attribute this site to Neanderthals. The broad circle contains several ancient fireplaces, cutmarked bones of red deer and bison, a cache of red ochre, and stone tools.

Was this the base of a hut made from skins and wooden poles? Was it the base of a windbreak or hunting blind? Or did the Neanderthals who used this site merely kick and shove mammoth bones out of the way of their campsite, making a circle as an accidental side effect? Aside from the first hypothesis, these ideas were suggested by outside researchers reading the work, without direct familiarity with the site itself. Demay and coworkers emphasized that the site was used repeatedly, with remains of at least fifteen mammoths represented there. Hearths were built several times in the southern part of the bone circle, and there is much evidence for human activity within it. They interpret this as a domestic area of a long-term campsite, repeatedly used by the mammoth hunters.

Resetting expectations

One of my recurring areas of interest is the way that archaeologists dismiss or minimize the cultures of earlier human relatives. I’ve written that we need to replace a “Culture Last” way of looking at the evidence with a “Culture First” model. I keep pushing this concept because I’ve seen so many cases where evidence is published not at the time of its discovery, but only after someone else says the evidence should exist somewhere.

Bruniquel has a unique status in being used only for a very short time and closed to later access. In this limited sense, the site is the best possible case of preservation of ancient behavior.

A decade after the work establishing the age of the Bruniquel evidence, discussion of fire evidence has shifted quite strongly. Scientific discussion of building of shelters and other structures has shifted much less. The Bruniquel evidence of structures is often mentioned in other recent research, and it seems that other researchers struggle to apply the evidence of building outside the context of shelter or windbreaks. Clearly much more work remains to integrate an understanding of built environments into the early archaeological record.

The new work on the closure of Bruniquel’s entrance has some fascinating elements. At the time the site was used, the rockshelter entrance stretched fairly widely, with large speleothem pillars interrupting the view. Génuite and coworkers present a view of the cave mouth from inside, as it would have been for the Neanderthals 170,000 years ago in the site.

I see it as a metaphor. How little we know of the world of these Neanderthals, outside the cave where they left those circles of stone.

References

Barham, L., Duller, G. a. T., Candy, I., Scott, C., Cartwright, C. R., Peterson, J. R., Kabukcu, C., Chapot, M. S., Melia, F., Rots, V., George, N., Taipale, N., Gethin, P., & Nkombwe, P. (2023). Evidence for the earliest structural use of wood at least 476,000 years ago. Nature, 622(7981), 107–111. https://doi.org/10.1038/s41586-023-06557-9

Demay, L., Péan, S., & Patou-Mathis, M. (2012). Mammoths used as food and building resources by Neanderthals: Zooarchaeological study applied to layer 4, Molodova I (Ukraine). Quaternary International, 276–277, 212–226. https://doi.org/10.1016/j.quaint.2011.11.019

Gao, X., Zhang, S., Zhang, Y., & Chen, F. (2017). Evidence of Hominin Use and Maintenance of Fire at Zhoukoudian. Current Anthropology, 58(S16), S267–S277. https://doi.org/10.1086/692501

Génuite, K., Verheyden, S., Camus, H., Burlet, C., Boudadi-Maligne, M., Bruxelles, L., Cheng, H., Cochard, D., Delaby, S., Fabre, E., Ferrier, C., Lézin, C., Ledoux, L., Medina-Alcaide, M.-A., Mora, P., Muth, X., Pélissié, T., Soulier, D., Xue, J., & Jaubert, J. (2026). The evolution of the cave’s entrance of Bruniquel and consequences for its accessibility by early Homo neanderthalensis. Quaternary Science Reviews, 377, 109866. https://doi.org/10.1016/j.quascirev.2026.109866

Huang, C., Li, J., & Gao, X. (2022). Evidence of Fire Use by Homo erectus pekinensis: An XRD Study of Archaeological Bones From Zhoukoudian Locality 1, China. Frontiers in Earth Science, 9, 811319. https://doi.org/10.3389/feart.2021.811319

Jaubert, J., Verheyden, S., Genty, D., Soulier, M., Cheng, H., Blamart, D., Burlet, C., Camus, H., Delaby, S., Deldicque, D., Edwards, R. L., Ferrier, C., Lacrampe-Cuyaubère, F., Lévêque, F., Maksud, F., Mora, P., Muth, X., Régnier, É., Rouzaud, J.-N., & Santos, F. (2016). Early Neanderthal constructions deep in Bruniquel Cave in southwestern France. Nature, 534(7605), 111–114. https://doi.org/10.1038/nature18291

Ochando, J., Jiménez-Espejo, F. J., Giles-Guzmán, F., Neto de Carvalho, C., Carrión, J. S., Muñiz, F., Rubiales, J. M., Cura, P., Belo, J., Finlayson, S., Martrat, B., van Drooge, B. L., Jiménez-Moreno, G., García-Alix, A., Lozano Rodríguez, J. A., Albert, R. M., Ohkouchi, N., Ogawa, N., Suga, H., … Finlayson, C. (2024). A Neanderthal’s specialised burning structure compatible with tar obtention. Quaternary Science Reviews, 346, 109025. https://doi.org/10.1016/j.quascirev.2024.109025

Roebroeks, W., MacDonald, K., Scherjon, F., Bakels, C., Kindler, L., Nikulina, A., Pop, E., & Gaudzinski-Windheuser, S. (2021). Landscape modification by Last Interglacial Neanderthals. Science Advances, 7(51), eabj5567. https://doi.org/10.1126/sciadv.abj5567

Rouzaud, F., Soulier, M., & Lignereux, Y. (1995). La grotte de Bruniquel. Spelunca. Bulletin, (60), 27-34.

Soressi, M. (2016). Neanderthals built underground. Nature, 534(7605), Article 7605. https://doi.org/10.1038/nature18440

Stepka, Z., Azuri, I., Horwitz, L. K., Chazan, M., & Natalio, F. (2022). Hidden signatures of early fire at Evron Quarry (1.0 to 0.8 Mya). Proceedings of the National Academy of Sciences, 119(25), e2123439119. https://doi.org/10.1073/pnas.2123439119

My PhD student, Ayken Askapuli, is a specialist on the human genetics of Kazakhstan and nearby areas of Central Asia. Over the last several years he has been engaged in an ancient DNA project, concentrating on the genomes of bodies in medieval tombs attributed to the Golden Horde period. Last week his paper describing some of this work was published by the Proceedings of the National Academy of Sciences.

It’s great to be able to share this work, and I’m so pleased to see Ayken’s success publishing it. My own part in the paper was small. What I’ll do here is try to relate part of this fascinating story, which hasn’t yet come to a conclusion.

The Golden Horde

The “Golden Horde” is the common name for the northwestern extension of the Mongol Empire during the period from 1240 to 1480 CE. This dynasty founded by Joshi, the eldest son of Genghis Khan, ruled over the Kipchak steppe, western Siberia, and the Pontic-Caspian region. It controlled the trade routes of the Silk Road and was important in the formation of modern ethno-linguistic groups like the Kazaks, Tatars, and Özbeks.

On or near the Kenggir River in central Kazakhstan are several mausoleums that are attributed to the Golden Horde period. Some of these were subject of archaeological work during the Soviet era, and some have been investigated by Kazakh archaeologists more recently, including Ayken’s collaborators in the present study.

Local traditions attribute some of these mausoleums to particular Mongol leaders—one is popularly attributed to Joshi Khan himself, another, Bolghan Ana, is popularly attributed to Joshi’s daughter. However, these local traditions may not be literal historic truths. For example, archaeological work by Irina Panyushkina and coworkers has shown that timbers used in the construction of the Joshi Mausoleum post-date his death by 75 to 100 years. Interestingly, one of the coffins in this mausoleum was made of wood that is roughly the right timing, although no record of reburial or movement of a body exists.

All this is to say that the identity of the skeletal remains in these tombs is not something that can be taken at face value.

Y chromosomes and Genghis Khan

One reason for interest in the Golden Horde mausoleums is their possible connection to the paternal lineage of Genghis Khan, subject of quite a lot of interest in human genetics over more than 20 years. Back in 2003, Tatiana Zerjal and collaborators from Chris Tyler-Smith’s research group published a groundbreaking study on the Y chromosome variation across Asia. Their work drew attention to a common Y chromsome haplogroup in many populations today that originated a fairly short time ago. Due to the pattern of rapid diversification of this lineage, they called it the “star cluster”.

In the 2003 work, the suggestion was that the paternal ancestor of this star cluster lineage had lived within the last 1000 years. From this estimate of age and its current geographic distribution in eastern and central Asia, Zerjal and coworkers suggested it might be the paternal lineage of Genghis Khan.

The logic was based in history. Few men within the last 1000 years have had the long-term offspring number of the Mongol elite, of which Genghis was the founder. Also, Zerjal and coworkers found that the Hazara ethnic group of present-day Afghanistan were especially marked in sharing star cluster Y chromosomes, and according to them, this group maintains an oral tradition of paternal descent from the Mongol elite.

In 2018 that story was revised substantially by Lan-Hai Wei and collaborators. These researchers greatly expanded the sample of Y chromosomes to more than 18,000 individuals, including many whole-chromosome sequences. With so much more data, they could add better resolution of the timing and branching pattern within the “star cluster”, redesignated as C2a1a3-F1918, or C3*. The main innovation of this work was to show that the C3* did not start 1000 years ago but instead closer to 2500 years ago. Wei and collaborators suggested that the present widespread geographic distribution of this C3* is indeed a legacy of the Mongols. But that isn’t because Genghis and his heirs specifically had so many offspring, but instead because ordinary Mongol peoples sharing the same Y chromosome heritage shared more generally in the conquests and dispersal led by Genghis and his heirs.

All this is background to our new study. The Golden Horde presence in Kazakhstan began with the immediate heirs of Genghis Khan. Ayken worked with Japanese collaborators led by Naruya Saitou, who led the DNA extraction from the skeletal material. They obtained data from four ancient individuals, three male and one female individual, excavated from four different mausoleums.

Results of the study

The analysis of the data led to several interesting insights. To begin, the traditional stories that related tombs to particular ancient individuals like Joshi were not supported. One of the individuals, from the Mausoleum of Alisha Khan, yielded radiocarbon results consistent with the 1700s CE, more than 200 years after the Golden Horde period. The other three all were consistent with the early to mid 1300s, so they were the right age to be Golden Horde elite, but not the specific individuals. For example, the radiocarbon results on the individual in the Joshi tomb suggest that this individual lived in the years after 1300 CE, long after Joshi’s historical death in 1227, and close to the era when the mausoleum was built. The woman in the Bolghan Ana mausoleum did not share a first- or second-degree relationship with the man in the Mausoleum of Joshi meaning she was not this individual’s daughter.

Although the local traditions about the occupants of the mausoleums were not upheld by the DNA results, the results did confirm some other aspects of the broader story of the Golden Horde and the Mongol empire. The Kenggir individuals are genetically closer to Mongolian and East Asian populations living today than they are to Kazakh or other local ethnic groups. Ayken undertook an analysis of relatedness using identity-by-descent (IBD) of these genomes, and a broader sample of ancient genomes from Asia. He found several genealogical relatives for these people in ancient genomes from Mongolia, and one from eastern Kazakhstan.

He also found kin relationships among three of the Kenggir individuals, two of which—from the Ayakkamir Mausoleum and the Alisha Khan Mausoleum—seem likely to be ancestor and descendant across around 5 to 6 generations. All of the Kenggir individuals show fairly high numbers of runs of homozygosity (ROH) which are indicative of cousin marriages within their genealogical backgrounds.

All three of the men sampled in the study share the C3* Y chromosome haplogroup. While the data do not connect this haplogroup directly to the genealogical heirs of Genghis Khan, the results do help to establish that the Golden Horde elite did have this haplogroup.

For me, another interesting result of the study concerns the cultural change in the Golden Horde elite. One source of differences in burial customs is religion. In many Mongol burials, bodies are oriented in a north or northeast orientation, and burials of the true elite were all conducted in secret locations. These four mausoleums represent a change in the custom. Bodies were placed in a west or northwest orientation, which is more consistent with Islamic practice. Even so, the few grave goods including a camel head in one burial are not typical of Islamic burials, and may represent a blending of traditions among people who had nominally adopted Islam.

Ayken’s work is continuing on the social and cultural legacies of these genetic connections from the Mongol empire. Already he has a preprint that uses SNP data combined with Y chromosome haplotypes to look at the ways that clan structure and genealogies are reflected in genetics and in oral histories. For me it has been great to learn about the rich histories of human connections across Central Asia. These dynamics in Medieval groups may be different from the prehistoric record, but they nonetheless help inform about the deep evolutionary history of this region.

Notes: The terminology for Y chromosome haplogroups has changed over time, and there is still some inconsistent use of terms in different studies. In our work we discuss the C3* haplogroup, and this corresponds to C2*-ST in Wei and coworkers (2018).

References

Askapuli, A., Kanzawa-Kiriyama, H., Kakuda, T., Kassenali, A., Yessen, S., Schamiloglu, U., Schrodi, S. J., Hawks, J., & Saitou, N. (2026). Genomes of the Golden Horde elites and their implications for the rulers of the Mongol Empire. Proceedings of the National Academy of Sciences, 123(8), e2531003123. https://doi.org/10.1073/pnas.2531003123

Askapuli, A., Vilar, M. G., Zhabagin, M., Sabitov, Z., Zhumadilov, Z., Ragsdale, A., Schurr, T. G., Hawks, J., & Saitou, N. (2026). Genetic Basis of Social Structure in the Pastoral Nomads of Central Eurasia. Preprint. https://doi.org/10.64898/2026.01.27.701587

Panyushkina, I. P., Usmanova, E. R., Uskenbay, K. Z., Kozha, M. B., Dzhumabekov, D. A., Akhatov, G. A., & Jull, A. J. T. (2022). Chronology of the Golden Horde in Kazakhstan: 14C Dating of Jochi Khan Mausoleum. Radiocarbon, 64(2), 323–331. https://doi.org/10.1017/RDC.2022.24

Wei, L.-H., Yan, S., Lu, Y., Wen, S.-Q., Huang, Y.-Z., Wang, L.-X., Li, S.-L., Yang, Y.-J., Wang, X.-F., Zhang, C., Xu, S.-H., Yao, D.-L., Jin, L., & Li, H. (2018). Whole-sequence analysis indicates that the Y chromosome C2*-Star Cluster traces back to ordinary Mongols, rather than Genghis Khan. European Journal of Human Genetics, 26(2), 230–237. https://doi.org/10.1038/s41431-017-0012-3

Zerjal, T., Xue, Y., Bertorelle, G., Wells, R. S., Bao, W., Zhu, S., Qamar, R., Ayub, Q., Mohyuddin, A., Fu, S., Li, P., Yuldasheva, N., Ruzibakiev, R., Xu, J., Shu, Q., Du, R., Yang, H., Hurles, M. E., Robinson, E., … Tyler-Smith, C. (2003). The Genetic Legacy of the Mongols. The American Journal of Human Genetics, 72(3), 717–721. https://doi.org/10.1086/367774

News sources this week tell us that “modern human women preferred Neanderthal men”. One story added, “the results do not indicate whether mating was consensual”. AI-painted images of star-crossed Pleistocene lovers are making eyes at each other up and down my social feeds.

As usual headline writers have tried to make the research sexy. The actual result is less titillating but even more interesting. In the center of these stories is the strongest genetic test yet showing that Neanderthals and the African ancestors of modern people were the same biological species.

The results may also give one of our best opportunities to understand the social factors that helped modern human populations expand and disperse throughout the world.

Introgression one way

The new research comes from Alexander Platt, Daniel Harris, and Sarah Tishkoff, published this week in Science. Building on earlier work by their team, these authors looked closely at the Neanderthal ancestry on the X chromosome compared to the autosomes.

This is well-worn ground. In 2014, Sriram Sankararaman and coworkers showed that X chromosomes of today’s people have a lot less Neanderthal ancestry than the 22 pairs of autosomes. The X chromosome is also home to more large genomic segments that are almost devoid of Neanderthal ancestry—called “introgression deserts”.

One hypothesis proposed at the time was partial hybrid incompatibility between the expanding population of modern people and the Neanderthals they met. This could fall into a broader pattern known as Haldane’s Rule.

Haldane’s Rule explained: When two populations divide and are isolated for a long time, they may start to evolve reproductive incompatibilities—as these evolve, they are the basis of speciation. Hybrids of the two populations may exhibit sterility or reduced fertility. When such problems emerge, they are more likely to manifest in the sex with two different sex chromosomes—in humans, this is males with the XY genotype—than in the sex with two of the same sex chromosome—in humans, females with the XX genotype. This happens because males have the full phenotypic effect of any X chromosome allele—even when it causes sterility—while females have a second copy of the X that may mask or moderate some phenotypic effects. More generally, deleterious X chromosome alleles are more exposed to selection due to effects in XY individuals than XX individuals. This works in the opposite direction in birds, where males have a ZZ genotype and females a WZ genotype, and it was J.B.S. Haldane who generalized based on homogametic and heterogametic sex chromosome genotypes.

When Sankararaman and coworkers noticed the lower Neanderthal ancestry of the X chromosome, they suggested that this might be the result of reduced fertility of hybrid males. Such a reduction in fertility would be a strong sign that the process of speciation in modern humans and Neanderthals had come to the point where they should be considered distinct biological species.

This is the kind of result that I take very seriously. Speciation in sexual organisms is about reproduction. If hybrid offspring of two populations have reduced fertility or sterility, those populations are different species.

But as I noted at the time, there are other possible explanations for the reduced Neanderthal ancestry on the X chromosome. Neanderthals had a long-term smaller population size than African ancestors of modern people, resulting in a genetic load of weakly deleterious Neanderthal alleles. From today’s vantage point—more than 1500 generations after the populations mixed—slow, weak selection against many slightly deleterious X chromosome alleles is hard to tell apart from hybrid male reduced fertility across a single generation.

There’s also the positive selection on the X chromosome. Just as the X chromosome is more exposed to purifying selection against deleterious alleles, it can also more quickly register positive selection on beneficial alleles. This effect has transformed human X chromosomes, resulting in lower diversity than the autosomes and greater differentiation across living human populations. Work by Laurits Skov and collaborators in 2020 showed that the early modern dispersal into Eurasia was accompanied by strong selection on advantageous X chromosome alleles. That effect may have wiped out a lot of Neanderthal variation.

So, I’ve been a Haldane holdout. My feeling has been reinforced lately by the evidence that mixture with Neanderthals took place over a large geographic range as modern people dispersed. It’s hard to maintain the notion of hybrid incompatibility when the groups met and mixed continuously.

Introgression the other way

The new research by Platt, Harris, and Tishkoff adds a much older period of intermixture to the story. The paper builds on earlier work by this team, led in 2023 by Harris, that looked at the mixture of African people into the Neanderthal population around 250,000 years ago.

This 250,000-year-old period of African dispersal has been coming into focus for more than a decade. The first clue was a greater degree of genomic similarity of the “Altai Neanderthal” (Denisova 5) genome and African peoples, compared to the “Denisovan” (Denisovan 3) genome. Later, geneticists understood that when it comes to Neanderthal-modern differences, there is a large genomic discrepancy between the autosomes on the one hand and the mitochondrial DNA and Y chromosome on the other. Both these genetic systems were adopted wholesale by Neanderthals from African ancestors sometime around 250,000 years ago. Along with them came around 6% of the genome as a whole. Every Neanderthal after that time—roughly, the age of the Petralona skull and 200,000 years after Sima de los Huesos—was part African.

As data have grown, it has become possible to study this early episode of introgression in more detail. The new study asks quite a simple question: Did this early introgression of African DNA into Neanderthals have the same consequences for the X chromosome as the later introgression of Neanderthal DNA into modern humans?

The answer is no. In fact, the pattern was exactly the opposite. There was no reduction of X chromosome introgression into Neanderthals, there was a big increase.

It’s like the mirror image of the later contacts of Neanderthals and modern humans. Haldane’s Rule would predict a similar fate for the X chromosome in both cases of introgression. Instead, it seems that the X chromosomes of African ancestral groups had a big advantage in both these periods of population mixture, regardless of whether they expanded with the modern population or were absorbed by the Neanderthal population.

Platt and coworkers note that this is not what would be predicted by reduced fertility of male hybrids, nor is it what the selection hypotheses predict. I underline those conclusions: The early introgression pattern is absolutely not what you’d expect from two populations that had started the process of speciation.

The authors suggest that a long-term sex bias in mating between the populations might produce the effect they found. Introgression into modern groups was male-biased, and introgression into Neanderthal groups was female-biased. A single long-term mating bias, with pairings of Neanderthal men and modern women more likely than the opposite, would generate these observations.

An unusual pattern

In both these periods of time, it was the African newcomer population that was expanding, and Neanderthals who had long resided in the regions. In both cases the women of the expanding group were more likely to mate with men of the resident group. That pattern cuts against the grain of most historic and prehistoric cases of human population contact.

In recent cases, geneticists have not generally traced the X chromosome but instead the mtDNA and Y chromosome due to the greater resolution of haplotype frequency data compared to X-single nucleotide polymorphisms (SNPs). Y chromosome haplotypes trace male-mediated migration and relationships; mtDNA traces female-mediated migration and relationships.

In historic and prehistoric population dispersals and contacts, the strongest pattern is often the arrival of new Y chromosome haplotypes. When Bronze Age populations from the Eurasian steppe entered central and western Europe, the Y chromosomes of some resident populations were almost entirely lost, while Y chromosome lineages of steppe origin became very common. The mitochondrial DNA variation was much less affected by newcomer lineages, reflecting a disproportionate impact of the incoming men and greater incorporation of resident women. The same pattern happened when Vikings arrived in Ireland, when European populations established colonies in the Americas, and when Bantu-speaking peoples dispersed into southern Africa and mixed with Khoesan-speaking hunter-gatherers and herders.

Still, the opposite pattern with female-mediated migration and gene flow has sometimes transpired. This was the pathway of early emergence of Polynesian peoples. People speaking Austronesian languages with Asian mitochondrial lineages expanded into near Oceania, meeting and combining with Melanesian peoples, whose Y chromosome haplotypes became common in the ancestral Polynesians. A similar case may be the growth and dispersal of Muskogean-speaking populations of the southeastern United States.

In cases like these, the political and economic organization of the expanding group is based in matrilineal clan or descent networks. The members of the culture reckon their kinship, loyalty, and affinity by their mothers and more distant maternal kin.

In anthropological theory, these different modes of organization and patterns of migration and dispersal are connected to subsistence and technology. Patrilineal migration and dispersal are often tied to mobile measures of wealth, like cattle and horses, as well as situations where technologies provide a warrior elite with mobility and the means of conquest. Expansion and dispersal of matrilineal networks is possible when fertility itself is of higher value, when the linguistic and cultural ties around community can be maintained by strong political consensus even as men from differing groups are integrated into the cultural group.

A matrilineal dispersal hypothesis

The X chromosome results are not an invitation to think in absolutes. Human societies—even small-scale human societies that hunt and gather wild foods—have always been varied in their residence patterns, kinship, and political structures. Neanderthal societies could not all have been cut from the same cloth, certainly not across thousands of miles and up to 200,000 years. As groups of these peoples came into contact with each other, some would have been culturally aligned, some ready for exchanges of communication and marriages, others less compatible or much more reticent for such contacts.

All this makes it unlikely that the same mating bias would carry across any very large geographic area, or that the same phenomena would unfold in a similar way 200,000 years apart. Yet, the data suggest that broadly similar patterns did unfold.

This is not a matter of cultural preference. But it may be a function of the kind of social organization that sometimes made growth and dispersal of early modern populations possible.

• Around 50,000 years ago, a branch of the African ancestral population underwent a significant founder effect before dispersing north, west, and east from southwest Asia. This movement was a demic expansion characterized by continuous population growth along its entire trajectory. Marriage networks during this period occasionally incorporated Neanderthals, with unions between male Neanderthals and female modern humans being significantly more common.

• An earlier period of contact, occurring more than 250,000 years ago, had a different demographic outcome. During this phase, the Neanderthal population did not suffer a global decline, and the expansion of modern human groups eventually stalled without population replacement. However, this contact was sufficient to cause a complete replacement of Neanderthal mitochondrial DNA and Y-chromosome variation with modern human lineages. The female-centered nature of this interaction is evidenced by a relative excess of African ancestry on the X chromosomes of all later Neanderthals compared to their autosomes.

• In neither case was the pattern of gene flow exclusively female-mediated. If for no other reason, we know this because of what happened to their Y chromosomes: If either period of contact was exclusively female-mediated, then the African Y chromosome would not have replaced the indigenous Neanderthal Y chromosome variation.

These expansions could not have been exactly like recent expansions of matrilineal groups. But the circumstances of such recent expansions may provide some very useful insights. Consider:

• Expansion of groups into the ranges of existing resident populations relies on cultural solidarity and high-fidelity cooperation along kinship networks. Matrilineal organization typically keeps cultural and genetic means of reckoning connections aligned across more generations than patrilineal systems.

• Extra-group marriages likely enabled land-use rights and strategic alliances. A beneficial side-effect of these pairings is the introduction of locally adaptive genes, such as those related to immune response, environmental factors, or local foods.

• Collective foraging strategies facilitate exploitation of a somewhat broader range of local resources, making it possible to grow into spaces that the resident population may find subsistence more marginal.

• Disease and immunity—especially maintaining fertility rates in the presence of disease—may be an initial block to expansion and dispersal, but once local genes or cultural strategies are taken onboard, growth itself provides a snowballing advantage over other resident populations.

• The newcomer and resident populations are never incompatible, but as the expanding population grows, their mixture has less and less impact on the subsequent gene pool. The genetic patterns most strongly reflect the earliest contacts.

• The matrilineal expansion hypothesis may help explain why Neanderthal-modern contacts did not follow the same patterns as the male-mediated dispersals of later agricultural or pastoral peoples, which are typically associated with paternal networks and “elite dominance” models.

Many anthropologists over the years have thought that matrilineal organization may have been the ancestral mode for early humans. Right now there’s not much hard data to test this idea. But the increasing resolution of kinship studies from ancient DNA have shown some advances, including the identification of kin at Chagyrskaya Cave, which shows some evidence of female-biased migration. It will be a major advance if genomic data from African and Near Eastern sites such as Klasies River Mouth or Qafzeh may add information about early African ancestral humans.

Note: Thanks to a reader for pointing out a typo with bird genotypes, this is now corrected.

References

Harris, D. N., Platt, A., Hansen, M. E. B., Fan, S., McQuillan, M. A., Nyambo, T., Mpoloka, S. W., Mokone, G. G., Belay, G., Fokunang, C., Njamnshi, A. K., & Tishkoff, S. A. (2023). Diverse African genomes reveal selection on ancient modern human introgressions in Neanderthals. Current Biology. https://doi.org/10.1016/j.cub.2023.09.066

Harris, K., & Nielsen, R. (2016). The Genetic Cost of Neanderthal Introgression. Genetics, 203(2), 881–891. https://doi.org/10.1534/genetics.116.186890

Platt, A., Harris, D. N., & Tishkoff, S. A. (2026). Interbreeding between Neanderthals and modern humans was strongly sex biased. Science, 391(6788), 922–925. https://doi.org/10.1126/science.aea6774

Sankararaman, S., Patterson, N., Li, H., Pääbo, S., & Reich, D. (2012). The Date of Interbreeding between Neandertals and Modern Humans. PLoS Genetics, 8(10), e1002947. https://doi.org/10.1371/journal.pgen.1002947

I’ve been trying to understand the fossils of Sahelanthropus tchadensis since Michel Brunet led a team in describing its skull—nicknamed Toumaï—back in 2002.

This fossil skull and all other known fossils of S. tchadensis come from the Djurab desert of Chad, where they lived an estimated 7 million years ago. Many accept S. tchadensis as the earliest hominin. But unanswered questions have left some scientists skeptical.

I’ve been one of the skeptics.

One of the most contentious issues has been locomotion. Today, we’re coming closer to an understanding of the early adaptations to upright walking in our relatives. After being hidden for almost two decades, the partial femur and two ulnae of S. tchadensis have now started to clarify how this species compares with known hominins like Australopithecus, as well as with living and extinct apes.

This story has changed a lot over the years. With today’s much better genetic information about human and chimpanzee relationships, the stakes of the question have also changed. Sahelanthropus tchadensis may be just one window into a network of populations that exchanged genes as the human lineage slowly differentiated from our closest relatives.

History of the problem

Did S. tchadensis walk upright? For almost twenty years, the Toumaï skull seemed to be the only evidence. The cranial base and back of the cranium can be informative about posture and movement, but both areas were badly distorted in the course of fossilization. Researchers led by Christoph Zollikofer who reconstructed the skull from CT scans in 2005 suggested it was configured like early hominins, including Australopithecus.

Working with several collaborators, I challenged this interpretation—first in a 2002 comment and then in a longer 2006 research article. One area that impressed us was Toumaï’s elongated attachment area for muscles of the neck, much more like great apes than hominins. To be sure, S. tchadensis shares traits with hominins like Australopithecus—for example, canines worn on the tip instead of the blade-like wear seen in many living apes. But traits like this can also be found in other long-extinct apes. We observed that across the decades many fossils had a similar story: at first interpreted as human relatives, later shown to be extinct apes.

More bones were needed. If none had ever turned up, I’m not sure I’d have much more to say about S. tchadensis today.

But new evidence did turn up, from an unexpected quarter. Just at the time of discovery of the Toumaï skull, members of the same field team also collected a femur shaft and a partial ulna of S. tchadensis at the same locality. Six months later, they found a second ulna fossil.

These could have been gamechangers—if anybody could have seen them. But for eighteen years, they remained undescribed.

The journalist Scott Sayare related the weird story of these fossils last year in a fascinating article for The Guardian. Accounts differ on why Brunet and collaborators omitted these bones from their 2002 description of S. tchadensis and their subsequent 2005 article in which they described two additional mandibles and several teeth recovered from nearby sites. Since posture and locomotion were so important to understanding S. tchadensis, leaving out these bones was a bad scientific decision.

From the first time I saw photos of the femur, I knew it would not be easy to interpret how it functioned. I wrote as much back in 2009 when I was able to share some photos with my readers, courtesy of Aude Bergeret-Medina, who had examined the fossil. I hardly guessed at the time that another decade would pass before I learned any more about it.

Things changed in 2020, when Guillaume Daver and coworkers described the femur and two ulnae in a preprint. Their work, later published in Nature, concluded: “The Toros-Ménalla femur exhibits several hallmarks of selection for bipedalism as a regular behaviour”.

At around the same time, the Journal of Human Evolution released a peer-reviewed article by Roberto Macchiarelli and coworkers, presenting their own interpretation of the femur’s anatomy: “there is no compelling evidence that it belongs to a habitual biped”.

Pretty stark difference.

Whenever fossils are interpreted differently by different groups of smart people, I go looking for an explanation. Often the problem is that one or both groups do not have access to all the data. So I’ve been following the research closely to try to puzzle out the truth.

A recent study led by Scott Williams in Science Advances has been helpful. Williams’ team was able to scan and analyze high resolution replicas of the bones, held at Harvard University. Looking at them in person yielded insight into some features that earlier studies had overlooked. With their scans, they also built a better idea of the relative sizes of the bones and their comparisons with other species of apes and hominins.

The most valuable thing about their paper is that Williams and coworkers have provided clear illustrations of the fossils, together with close comparisons with other species. While I’ve read earlier papers closely, the illustrations left much to be desired, and to be quite honest in most cases I don’t think they have shown the fossils in enough detail for me to assess whether they’ve described them accurately. The open access license of Williams’ and collaborators’ work lets me configure those here for readers.

The femur

You wouldn’t think it would be that hard to look at a femur and tell whether an ape was a biped. But most of the things that work for human legs also have to work for ape legs. All living apes can and do stand and walk bipedally—they just aren’t tuned for it to the exclusion of other forms of movement, as humans are. Of the features that set human femora apart from the living great apes, some relate to our taller stature and longer legs. Many hominins like Australopithecus were pretty different from living people in leg length and stature, so their patterns of femoral anatomy were not quite the same as today’s people.

Because of the diversity among bipedal hominins, it takes more than comparing humans and apes to understand how the femoral anatomy of an extinct species related to their pattern of movement. It takes careful attention to muscle and joint function.

Apes tend to keep knees slightly bent when they stand upright. When they walk bipedally, it is with less hip extension, less knee extension, and without the distinctive toe-off of a human stride. They rarely run bipedally, although they may rear up on two legs when charging forward at another individual.

Bipedal walking and running in hominins like Australopithecus involved functional innovations in the hip joint and knee joint. Walking with a more extended hip required massive alterations to the pelvis. The changes to the femur were more subtle. The knee is most strongly involved: Australopithecus had slightly flattened femoral condyles sitting at a distinct angle from the shaft. The femur neck of Australopithecus is longer and angles more forward—called femoral neck anteversion—than in apes. The femur neck is flatter from front to back, orienting the cortical bone and internal trabecular structure to resist vertical force. Several ligament and tendon attachments of the proximal femur important to bipeds also have bony traces, especially involving the gluteus muscles and the iliofemoral ligament.

These are the features I would point out to any introductory student in my courses. As important as they are, neither the distal end nor femoral head and neck are well-preserved in the S. tchadensis femur TM 266-01-063. This is why I had very little to say about the shape from the photos of the specimen. The dark coloration of the specimen also makes photos hard to interpret. What was needed was for someone to make the data available.

Daver and coworkers pointed out that the base of the femoral neck of the TM 266-01-063 femur seems to have a flatter, or anteroposteriorly compressed nature, similar to humans but also to Miocene apes like Danuvius guggenmosi. Just at the edge where the bone is broken, Williams and coworkers observe a small bump, which they identify as the femoral tubercle. That’s an attachment site for the iliofemoral ligament, which stabilizes the hip joint when the leg is extended. It’s seen in humans and other bipedal hominins, but not in great apes.

The shaft of the femur also matters to bipedal walking. Near its proximal end, humans and chimpanzees differ in muscle attachment configuration. Humans and Australopithecus have a gluteal tuberosity or roughened ridge for attachment of the gluteus maximus, while chimpanzees and other great apes have a more anterior insertion for this muscle with a smooth, more rounded ridge descending from it, called a spiral pilaster. All the researchers who have examined TM 266-01-063 have paid attention to this area, and they express diverging opinions: the papers led by Scott Williams and Guillaume Daver say it’s hominin-like, the papers led by Roberto Macchiarelli and Marine Cazenave say it’s ape-like.

Well, from Williams’ illustrations, I can see what both sides have been seeing. In the image above the red arrows point to gluteal tuberosity, including the bump that Williams and coworkers interpret as this feature in TM 266-01-063. The spiral pilaster curving down the side of the chimpanzee femur is pretty clear in this view. The complication is that the TM 266-01-063 femur appears to have a angled fracture or stress line that almost parallels the spiral pilaster of the chimpanzee. Looking at a photograph and not a 3D replica or scan, I can see how someone might misinterpret the shape here.

I believe I’ve mentioned many times that secrecy and closed access to scans does nothing but hold back scientific work. I have found that the easiest way to get other researchers to see what you are describing is to simply give everyone the data.

Another thing that is hard to see in a photograph is how twisted the shaft is. The angle between the femoral neck and femoral condyles is positively twisted—neck more forward—in humans, and neutral or negatively twisted—neck to the side—in great apes. That twist is manifested at the base of the femoral neck and the proximal femoral shaft. In S. tchadensis, Williams and coworkers find that the twist is more human than most humans.

Together with the muscle and ligament attachments, that’s a fairly good argument for weight support with an extended hip. The earlier study by Daver and coworkers included data from CT scanning, enabling them to observe the bone tissue inside the femur. A key observation in that study was an alignment of the trabecular bone plates known as the calcar femorale, often found in humans and rarely observed with a similar form in living great apes.

Still, extended hip posture doesn’t necessarily mean Australopithecus-like bipedality. The mention of Danuvius above is a hint that many of these features do overlap either with extinct apes or with the range of variation within living great apes.

The mechanics coming into focus from the TM 266-01-063 femur are generally compatible with the way that Owen Lovejoy, Tim White, and collaborators have described the femoral anatomy of Ardipithecus ramidus. It’s a pity that there isn’t more overlap between the preserved portions of the femur in Ar. ramidus and S. tchadensis. Lovejoy and coworkers do picture the femoral fragment designated ARA-VP-1/701, an isolated find that has a similar gluteal tuberosity morphology, continuing toward what Lovejoy termed a “proto-linea aspera”. With a fairly flat separation of the attachments of the vastus group from the adductor group, it’s different from the pronounced, rough line in humans.

Summarizing their observations, Williams and coworkers describe the Sahelanthropus femoral morphology as reflecting “a non-obligatory terrestrial biped that retained powerful hip extension for flexed hip locomotor behaviors such as vertical climbing”.

What I would say is that the S. tchadensis femoral form would work well in an Ardipithecus-like leg. It was not specialized in the same way as the femora of chimpanzees, bonobos, and gorillas; and that may simply mean that S. tchadensis was not a terrestrial quadruped as those living African apes are. Even so, Williams and coworkers find that the shape dimensions of the TM 266-01-063 femur fit between chimpanzees and gorillas when compared with other apes, hominins, and humans.

The ulnae

The S. tchadensis arm bones are a left ulna shaft, TM 266-01-050, and a right ulna partial shaft, TM 266-01-358. An obvious first question is whether any of these bones collected at the TM 266 locality come from the same individual—either as each other, the femur, the Toumaï skull and mandible, or all of the above. The proportion of length and joint diameters of the ulna compared to the femur shaft can be very informative about locomotion.

Details about the fossils’ context are scanty. The fossils were surface finds and, as told by Daver and coworkers, the TM 266 locality covered an estimated 1.5 square kilometers in 2001 when these fossils were collected. The only published photos from the field show fossils after being moved. The hominin teeth from the locality represent at least three different individuals.

Alain Beauvilain and Jean-Pierre Watté told the story of the collection of the Toumaï skull, including field photos showing the femur and skull. They suggested that the skull and other bones had been moved from their original contexts by desert peoples sometime before the scientific expedition found them. Whether that is the case or not, there is no evidence of anatomical association of the fossils at the time of discovery.

Probably for this reason, Daver and collaborators did not address the relative sizes or lengths of the ulnae and femur. Williams and coworkers observe that even though the ulnae may represent different individuals, they seem to come from individuals that are not very different in mass. With that in mind, they attempt a length comparison, finding that the more complete ulna isn’t long in the way that a chimpanzee or gorilla would be.

Furthermore, the coronoid notch that articulates with the humerus at the elbow joint faces forward. Knuckle-walking apes have this notch oriented a bit upward, helping with weight support across the elbow joint. Forward-facing ulnae are found in humans and Australopithecus, but also in orangutans and gibbons.

Otherwise, however, the ulnar shafts are a lot like chimpanzee and gorilla ulnae. They’re curved along their shaft. Marc Meyer and coworkers, including Williams, have observed that this aspect of the S. tchadensis ulnae is similar to knuckle-walkers, related to the demands of propulsion across the elbow joint. An analysis of ulna shape puts the S. tchadensis form right in the middle of the chimpanzee range of variation, far from any hominins. There’s little in this anatomy to suggest that upright posture had any influence upon the S. tchadensis upper limb.

Still, a handful of early hominin ulnae do share a chimpanzee-like level of curvature, especially two fairly complete ulnae attributed to Paranthropus boisei. With the recent description of the highly robust hand bones of P. boisei by Carrie Mongle and collaborators, it may be that the curved ulna is part of an adaptation for foraging on the ground, maybe uprooting tubers or digging. That might account for a similarity of form even with different locomotor demands.

Where my thoughts are trending

When Brunet and collaborators published their description of S. tchadensis in 2002, I was skeptical of the idea that this species was a member of the our branch of primates. In addition to the poor evidence for a pattern of upright posture or locomotion from the Sahelanthropus skull, there was something else shaping my thinking.

Mainstream interpretations of DNA evidence at the time suggested that the human-chimpanzee last common ancestor lived only around 5 million years ago. But the fossils of Sahelanthropus tchadensis at the time of their discovery were placed between 7 and 6 million years ago, which was later refined to a range between 7.2 and 6.8 million years. That range of ages was seemingly too old by 2 million years for this primate to be a member of the human branch.

A lot has changed in twenty years. New discoveries of Australopithecus have shown that these early hominins were not all alike in their pattern of movement. The skeleton of another relative, Ardipithecus ramidus, was first described in 2009. This species had its own distinctive anatomical pattern: Upright trunk, neck, and head posture combined with an ability to stand with extended hip, yet retaining long arms, fingers, and grasping feet for climbing.

Meanwhile, the genomic record has grown to encompass much larger samples of all living apes and humans. DNA now suggests that human ancestors may have started to diverge from chimpanzee and bonobo ancestors by the time of Toumaï, but with continued interbreeding for a long time.

The reason why we notice the differences between hominins after 4 million years ago and living great apes is the extinction of the intermediate forms. But in the period between 8 million and 5 million years ago, the genetic evidence suggests the ancestral populations were mixing with each other, occasionally exchanging DNA. Perhaps the early hominin lineage and early chimpanzee-bonobo lineage had different habitat preferences and subtly different locomotor strategies during this early period of differentiation. But they were not too different to mix and mate.

I tend to agree with Williams and coworkers’ new analysis. Sahelanthropus tchadensis may have walked and climbed in a similar way to Ardipithecus ramidus. This species is accepted as early hominins by many researchers, and S. tchadensis is not very different in the parts both preserve. If that’s the case, almost certainly early chimpanzee-bonobo ancestors had this general body form also. Maybe the most important question about S. tchadensis is not whether it was yet a biped, but whether its early chimpanzee-bonobo contemporaries were yet knuckle-walkers.

It is hard to write about comparisons of two groups without making it sound like a dichotomy between the two groups. But evolution defies dichotomies. Each species evolves from ancestors through a continuum of genetic, anatomical, and behavioral variation.

Our ancestry is no different. Was S. tchadensis a hominin? To really answer that, we’ll need much more information about the many extinct populations that existed as part of this network of early apes.

Notes: I highly recommend reading the Guardian article by Scott Sayare, “The curse of Toumaï: an ancient skull, a disputed femur and a bitter feud over humanity’s origins”. The piece includes perspectives from many scientists who were involved in study of the Sahelanthropus remains. Even though I’ve followed this story for years, I learned a lot from Sayare’s account that I had not previously known.

What the opposing teams have written about the fossils has helped me learn a lot. This post would not have been possible without Scott Williams and colleagues’ paper that provides clear illustrations. Still, without open access to the data from these fossils, it has been hard for me to judge what any of the researchers may not have noticed. I’d love to see these become available for anyone to inspect and print for their own research and teaching.

References

Beauvilain, A., & Watté, J.-P. (2009). Was Toumaï (Sahelanthropus tchadensis) buried? Anthropologie, 47(1/2), 1–6.

Brunet, M., Guy, F., Pilbeam, D., Lieberman, D. E., Likius, A., Mackaye, H. T., Ponce de León, M. S., Zollikofer, C. P. E., & Vignaud, P. (2005). New material of the earliest hominid from the Upper Miocene of Chad. Nature, 434(7034), Article 7034. https://doi.org/10.1038/nature03392

Brunet, M., Guy, F., Pilbeam, D., Mackaye, H. T., Likius, A., Ahounta, D., Beauvilain, A., Blondel, C., Bocherens, H., Boisserie, J.-R., De Bonis, L., Coppens, Y., Dejax, J., Denys, C., Duringer, P., Eisenmann, V., Fanone, G., Fronty, P., Geraads, D., … Zollikofer, C. (2002). A new hominid from the Upper Miocene of Chad, Central Africa. Nature, 418(6894), Article 6894. https://doi.org/10.1038/nature00879

Cazenave, M., Pina, M., Hammond, A. S., Böhme, M., Begun, D. R., Spassov, N., Gazabón, A. V., Zanolli, C., Bergeret-Medina, A., Marchi, D., Macchiarelli, R., & Wood, B. (2025). Postcranial evidence does not support habitual bipedalism in Sahelanthropus tchadensis: A reply to Daver et al. (2022). Journal of Human Evolution, 198, 103557. https://doi.org/10.1016/j.jhevol.2024.103557

Daver, G., Guy, F., Mackaye, H. T., Likius, A., Boisserie, J.-R., Moussa, A., Pallas, L., Vignaud, P., & Clarisse, N. D. (2022). Postcranial evidence of late Miocene hominin bipedalism in Chad. Nature, 609(7925), Article 7925. https://doi.org/10.1038/s41586-022-04901-z

Lebatard, A.-E., Bourlès, D. L., Duringer, P., Jolivet, M., Braucher, R., Carcaillet, J., Schuster, M., Arnaud, N., Monié, P., Lihoreau, F., Likius, A., Mackaye, H. T., Vignaud, P., & Brunet, M. (2008). Cosmogenic nuclide dating of Sahelanthropus tchadensis and Australopithecus bahrelghazali: Mio-Pliocene hominids from Chad. Proceedings of the National Academy of Sciences, 105(9), 3226–3231. https://doi.org/10.1073/pnas.0708015105

Lovejoy, C. O., Suwa, G., Spurlock, L., Asfaw, B., & White, T. D. (2009). The Pelvis and Femur of Ardipithecus ramidus: The Emergence of Upright Walking. Science, 326(5949), 71. https://doi.org/10.1126/science.1175831

Macchiarelli, R., Bergeret-Medina, A., Marchi, D., & Wood, B. (2020). Nature and relationships of Sahelanthropus tchadensis. Journal of Human Evolution, 149, 102898. https://doi.org/10.1016/j.jhevol.2020.102898

Meyer, M. R., Jung, J. P., Spear, J. K., Araiza, I. Fx., Galway-Witham, J., & Williams, S. A. (2023). Knuckle-walking in Sahelanthropus? Locomotor inferences from the ulnae of fossil hominins and other hominoids. Journal of Human Evolution, 179, 103355. https://doi.org/10.1016/j.jhevol.2023.103355

Mongle, C. S., Orr, C. M., Tocheri, M. W., Prang, T. C., Grine, F. E., Feibel, C., Patel, B. A., Laureijs, O., Hobbs, T. E., Maiolino, S., Rossie, J., Mbogo, W., Du Plessis, A., Lonyericho, J., Woto Huka, W., Sale, H., Umuro, A., Yirgudo, P., Dalacha, I., … Leakey, L. N. (2025). New fossils reveal the hand of Paranthropus boisei. Nature, 647(8091), 944–951. https://doi.org/10.1038/s41586-025-09594-8

Sayare, S. (2025, May 27). The curse of Toumaï: An ancient skull, a disputed femur and a bitter feud over humanity’s origins. The Guardian. https://www.theguardian.com/science/2025/may/27/the-curse-of-toumai-ancient-skull-disputed-femur-feud-humanity-origins

Williams, S. A., Wang, X., Araiza, I., Guerra, J. S., Meyer, M. R., & Spear, J. K. (2026). Earliest evidence of hominin bipedalism in Sahelanthropus tchadensis. Science Advances, 12(1), eadv0130. https://doi.org/10.1126/sciadv.adv0130

Wolpoff, M. H., Hawks, J., Senut, B., Pickford, M., & Ahern, J. (2006). An Ape or the Ape: Is the Toumaï Cranium TM 266 a Hominid? PaleoAnthropology, 2006, 36–50.

Wolpoff, M. H., Senut, B., Pickford, M., & Hawks, J. (2002). Sahelanthropus or “Sahelpithecus”? Nature, 419(6907), 581–582. https://doi.org/10.1038/419581a

Zollikofer, C. P. E., Ponce De León, M. S., Lieberman, D. E., Guy, F., Pilbeam, D., Likius, A., Mackaye, H. T., Vignaud, P., & Brunet, M. (2005). Virtual cranial reconstruction of Sahelanthropus tchadensis. Nature, 434(7034), 755–759. https://doi.org/10.1038/nature03397

Ancient peoples crafted technologies from wood, fiber, bark, leather, sinew, feathers, and countless other organic materials. With these they made all manner of implements, clothing, watercraft, and structures. Some such items were simple, made from only a single piece like a digging stick. But most were complicated, made from many pieces, each piece shaped in ways necessary to the overall function.

Nearly all these things have passed into dust.

Still, a few archaeological sites provide exceptional records of ancient organic technologies. They come only from particular times and places: Desert caves, anoxic peat bogs, or fine silty sediments that preserve impressions of objects long decayed.

I’m fascinated by a new article that reports on some of the preserved organic technologies from caves of the Great Basin of North America. The first peoples to leave traces in these caves lived more than 13,000 years ago. The article, by Richard Rosencrance and collaborators, discusses the ways of life that were supported by tightly sewn clothing, cordage, carved wooden traps, and other perishable technologies. Most of the artifacts described in their work come from Paisley Caves and Cougar Mountain Cave, both in eastern Oregon.

Both Paisley Caves and Cougar Mountain Cave were first investigated by archaeologists more than 60 years ago—Paisley Caves by Luther Crossman in the 1930s, Cougar Mountain Cave by John Cowles in 1958 and later by Thomas Layton in 1966. Recording of the positions and context of material in these early excavations was not as precise as might be wished, and some aspects that might be highly interesting today were not systematically collected and ended up in backfill.

At Cougar Mountain Cave, Rosencrance and coworkers relate Cowles’s reporting that many coprolites were in the deposits, and these were not collected. That’s a pity because at Paisley Caves, more recent excavation led by Dennis Jenkins brought to light many human coprolites, which have provided an abundance of biological information about the ancient people who left them.

The jewel of the earlier excavations were the recognition of artifacts of fiber, leather, bark, and wood. The collection by Cowles and Layton from Cougar Mountain Cave includes a wide range of exceptional material, now housed at the Nevada State Museum. The new work by Rosencrance and coworkers includes a lot of new analysis of this collection, with radiocarbon dating on many of the artifacts.

Both sites cover a range of dates but much of the most interesting material comes from the period of time known as the Younger Dryas, roughly 12,900 to 11,600 years ago. In a world that was generally warming from the Last Glacial Maximum 19,000 years ago, the Younger Dryas represents a resurgence of relatively cold conditions.

Rosencrance and coworkers focus on the pattern of technical complexity represented by these objects. Many involve the application of multiple techniques or tools in their production.

One kind of example is tailored clothing. For example, a fragment of elk hide from Cougar Mountain Cave of Younger Dryas age has a cord sewn into its edge—the oldest example of sewn hide in the world. This is too small a fragment to know whether it was part of a garment or moccasin. But the utility of needles and sewing in producing fitted garments and footwear is widely recognized. Tailored clothing was valuable in cold environments as a means of reducing energy expenditure and avoiding frostbite, and the article presents evidence of bone and wood needles from many sites in addition to the two sites with most of the fiber and hide preservation.

Another example is weaving rushes or plant fibers into mats or baskets. Fragments of woven plant material are evidenced in both the Paisley Caves and Cougar Mountain Cave collections.

A different type of complexity is the production of special parts that must function together within a compound implement or machine. Some of the wooden pieces in the collections are interpreted as parts of deadfall traps, in which a large stone is supported by an intricate set of sticks that trigger the fall of the rock when disturbed by a small animal.

The article discusses a concept rooted in the ecology of northern hunter-gatherers that has a broad utility when thinking about such complex organic repertoire: the idea of “serial specialists”. This term was used by Lewis Binford as a way to think about the ecological flexibility and shifting foraging strategies that are needed in places where resources may have specific seasonal patterns.

The idea is that people move during the year—residential mobility—and pursue specialized foraging or extraction strategies for each of the resources they rely upon. One key requirement is being in the right place at the right time. That means learning the timing of the resource and the affordances of the landscape around it. Another key requirement is knowing how to make the specialized implements or follow the specific procedures that enable exploitation of the resource.

Binford emphasized that this was different from a generalist strategy; it’s a strategy of becoming specialists in multiple distinct survival activities.

The ethnographic record of Great Basin groups like the Northern Paiute provides a lot of evidence for this kind of sequential application of different specialized foraging strategies across an annual cycle. These groups moved to exploit specific resources with seasonal abundance, such as the hatching of brine fly larvae in Mono Lake, the harvest of piñon pine nuts, and the communal driving and capture of jackrabbits.

It’s very hard for any single site to provide evidence about residential mobility. Evidence of different activities at a single site doesn’t quite do this: A forager who can craft a well-balanced spear point for bison and a deadfall trap for a rabbit is mastering two very different technologies, yet may employ them in the same place and time. True, the archaeological record sometimes comes through in surprising ways: at some sites archaeologists can establish that human activity was seasonal, often through the preservation of botanical remains or animals that were killed at a single time of year. Still, piecing together an entire cycle would take an impressive correlation from many archaeological sites across a region, all representing the same range of time.

Sites in the northern Great Basin that share Younger Dryas-age deposits might provide such correlations. They fall within a limited timespan and share a Western Stemmed stone tradition. At Paisley Caves, the coprolites have produced abundant biological evidence including dietary remains and mitochondrial DNA.

These sites have received a lot of attention as part of the long debate over whether they provide evidence of human activity prior to the widespread Clovis tradition. The Clovis phenomenon is named for the distinctive fluted spear tips found at a mammoth kill site at Blackwater Draw, near Clovis, New Mexico, and similar points and associated finds stretch from Canada to Central America, with a strong concentration in the southeastern United States. Sites with Clovis points are concentrated in a narrow band of time just before the Younger Dryas, from just over 13,000 to around 12,750 years BP.

Some of the human coprolites from Paisley Caves are indeed older than Clovis, ranging as much as 14,800 years BP, and are authenticated with mitochondrial DNA and fecal sterol chemistry. Only one of the fiber items described by Rosencrance and coworkers comes from sediments predating the Younger Dryas; the rest are contemporary with Clovis or more recent.

The initial habitation of this part of North America is indeed an important topic. I’m just as interested in the connection between Paisley Caves, Cooper’s Ferry, Idaho, and other sites that predate the Clovis phenomenon as anyone.

Even so, the work from these sites has me much more often thinking beyond the question of who was earliest.

What is most valuable in understanding the lives of the ancient people—including their adaptability upon their arrival in new circumstances—is the richness of the record of behavior. In that regard, these sites from the northern Great Basin have few parallels. I can’t wait to see more examination of the biological evidence from these sites, their interrelationships across sites, and their connection to the surrounding environment.

Notes: All dates in this post are in calibrated years BP, and I’ve generally limited to round numbers. In the debate over the timing of sites relative to Clovis more significant digits are sometimes needed, and changing calibration scales sometimes argue for reporting radiocarbon years rather than calibrated years.

References

Binford, L. R. (1980). Willow Smoke and Dogs’ Tails: Hunter-Gatherer Settlement Systems and Archaeological Site Formation. American Antiquity, 45(1), 4–20. https://doi.org/10.2307/279653

Jenkins, D. L., Davis, L. G., Stafford, T. W., Campos, P. F., Hockett, B., Jones, G. T., Cummings, L. S., Yost, C., Connolly, T. J., Yohe, R. M., Gibbons, S. C., Raghavan, M., Rasmussen, M., Paijmans, J. L. A., Hofreiter, M., Kemp, B. M., Barta, J. L., Monroe, C., Gilbert, M. T. P., & Willerslev, E. (2012). Clovis Age Western Stemmed Projectile Points and Human Coprolites at the Paisley Caves. Science, 337(6091), 223–228. https://doi.org/10.1126/science.1218443

Rosencrance, R. L., Smith, G. M., Jenkins, D. L., Connolly, T. J., & Layton, T. N. (2019). Reinvestigating Cougar Mountain Cave: New Perspectives on Stratigraphy, Chronology, and a Younger Dryas Occupation in the Northern Great Basin. American Antiquity, 84(3), 559–573. https://doi.org/10.1017/aaq.2019.22

Rosencrance, R. L., Smith, G. M., McDonough, K. N., Jazwa, C. S., Antonosyan, M., Kallenbach, E. A., Connolly, T. J., Culleton, B. J., Puseman, K., McGuinness, M., Jenkins, D. L., Stueber, D. O., Endzweig, P. E., & Roberts, P. (2026). Complex perishable technologies from the North American Great Basin reveal specialized Late Pleistocene adaptations. Science Advances, 12(6), eaec2916. https://doi.org/10.1126/sciadv.aec2916

Shillito, L.-M., Whelton, H. L., Blong, J. C., Jenkins, D. L., Connolly, T. J., & Bull, I. D. (2020). Pre-Clovis occupation of the Americas identified by human fecal biomarkers in coprolites from Paisley Caves, Oregon. Science Advances, 6(29), eaba6404. https://doi.org/10.1126/sciadv.aba6404

On February 24, 1939, Antonio Borrani was digging to level the ground and extract stone for the expansion of the Villa Guattari. Alessandro Guattari was expanding his family’s hotel at its favorable beachside site at the edge of San Felice Circeo. To the west stretched the massif of Monte Circeo, its rocky flank dotted with caves where it juts into the Tyrrhenian Sea. The picks had already turned up bones. On this day, the digging uncovered a hole.

Guattari climbed through the hole into a narrow winding tunnel, its floor littered with bones. In a small chamber was a skull, upside-down and surrounded by a circle of stones.

The next day brought the paleontologist Alberto Carlo Blanc. He recognized bones of ancient deer and aurochsen. It was clear the cave had been sealed for a very long time. The skull he recognized also. It belonged to a Neanderthal.

Blanc thought the cave was a ritual site. The skull and stones he interpreted as intentionally placed. The Neanderthal skull and jaw were studied by the anthropologist Sergio Sergi in Rome, who interpreted the damaged cranial base as a result of intentional brain removal. After the war, Blanc worked with the archaeologist Luigi Cardini to excavate parts of the site. In the hardened breccia deposits just outside the cave’s entrance their work uncovered stone tools belonging to a local Mousterian tradition, as well as a second Neanderthal jaw.

Yet aside from the skull there was little evidence of Neanderthals inside the cave. On the other hand, there was a lot of evidence of hyenas, from the marks on the bones to the dozens of coprolites. Later research concluded that the Neanderthal individual and other animals within the cave had been hyena prey.

Ever since the discovery the cave has been a heritage site and tourist attraction, with Blanc and Cardini’s excavations expanded into small walkway. In 2018, the local archaeological authority started a small project to renovate access, cleaning the old trenches. When this work uncovered intact archaeological deposits, a new project led by Mario Federico Rolfo aimed to re-evaluate the cave.

Over the last year, several papers describing new finds from Grotta Guattari have come out. These include two articles that present more than a dozen new fossils of Neanderthals.

Part of the new work unfolded within a small chamber where water collects into a pool during seasons of higher rainfall. This “Pond Chamber”, or Antro del Laghetto, was beyond the area with foot traffic and its deposits were sealed by a crust of calcite. Through this crust, they could see some bones.

As the archaeologists worked in this chamber, from 2019 to 2023, they uncovered an intact paleosurface, likely the same that had formed the floor of chamber where the skull was found in 1939, Antro dell’Uomo. At a low point in the chamber, they found a dense array of animal and Neanderthal bones intermingled.

The surfaces of many of the bones have carnivore damage. An article led by Ivana Fiore last year described the overall condition of the assemblage with close analysis of carnivore damage and possible cutmarks. The bones from the Antro del Laghetto show how hyenas gnawed bone ends, and selectively transported limbs, skulls, and ribcages into the cave. They found many small bone fragments that had been intensively chewed by juvenile hyenas. These so-called “nibbling sticks” were basically hyena teething toys.

The Neanderthal bones in this area include parts of at least three skulls, a partial femur, partial radius, left and right hip bones, and several teeth.

The team also undertook excavations in the rockshelter area outside the cave entrance, to better understand the layers found there during Blanc’s work in the 1950s. This area was long assumed to be connected with the deposits within the cave itself. But the system of excavation and dating has shown that these areas are very different in time. The hyena den inside the cave is somewhere around 65,000 years old, but the external deposits are much older, estimated roughly 125,000 to 110,000 years ago.

Unlike the internal bone assemblage, the external deposit includes abundant cultural material. Animal bones in this area have cutmarks and intentional impact fractures, which are diagnostic of human processing for meat and marrow. Many of these bones are burnt. Fiore and coworkers describe a piece of red deer tibia from this area, with markings similar to those found on bone retouchers used by knappers making stone tools. The

Where do these fossils fit into Neanderthal evolution?

The fossils from the external shelter area are too few to consider in much detail. Previous work had shown some differences in form between the Guattari 2 mandible from within the cave, and the Guattari 3 partial mandible from outside it. The new teeth from outside the cave may add some information when full analyses of their internal structure are carried out.

The age of these fossils at or just after the last interglacial makes them a very interesting comparison for other fossils from this time. These include the large sample of Neanderthals from Krapina, Croatia, as well as the fossil child from Scladina, Belgium, and the Neanderthals from Moula-Guercy, France. The last interglacial was a time of very mild climate across Europe, with Neanderthals expanding their presence into the northern parts of Germany, France, and Great Britain. Two thousand kilometers to the east, Neanderthals of this era encountered early populations with African ancestry. These people were culturally similar in many ways. Every fossil from this interesting time may provide valuable clues about their interactions.

The fossils from inside Grotta Guattari represent a much later time, when long periods of cold spurred the growth of great ice sheets in the northernmost parts of Europe. The sea level was 70 to 80 meters lower than today, and a broad coastal plain stretched from Monte Circeo away from the site, rich grazing for aurochsen and other herbivores. This time period, part of Marine Isotope Stage 4 (MIS4), was a time of contraction of Neanderthal populations into the parts of southern Europe that remained temperate, including the Italian peninsula.

The Guattari 1 skull found in 1939 has long been considered a part of the “classic” Neanderthal population. Its shape and many details resemble fossils from southern France like the La Ferrassie 1 skull, or the “old man” of La Chapelle-aux-Saints.

The shape of the new Guattari 17 skull, most complete of the new finds, fits generally into this pattern also with its rounded, en bombe shape when seen from behind, and its flattened profile where the occipital and parietal bones meet. This skull as well as the Guattari 11 occipital bone share a small depression above the area where the neck muscles join the skull, known as a suprainiac fossa. A hallmark of Neanderthal skulls, this is thought to be a side effect of development.

Still, across many of the finds are features that are a bit out of the ordinary for the Neanderthal population of this time. Some of those are detailed in a new article by Mauro Rubini and coworkers, who propose broader networks of gene flow—even as far as Southeast Asia—might explain the variability of some of the Guattari fossils. On the other hand, the migration of populations from north to south, and the contracting population size, may have caused different patterns of evolutionary change in Neanderthal populations in Italy from France, Spain, and the Levant.

Testing those possibilities may require DNA data. Sperduti and coworkers end their brief description of the fossils by noting the many kinds of studies that are underway. They mention ancient DNA and also proteomics, histological, and isotopic studies, as well as the broader comparison of the morphology of the fossils with other hominins. What they describe as a “multi-methodological approach” to the site is increasingly the pattern of today’s paleoanthropological research.

For readers who would like to see the cave and excavation context, the Italian Ministry of Culture produced a video in 2022 showing the excavations that were still ongoing at that time. The voiceover and interviews are in Italian, but even for non-Italian speakers it is a great video for its overview of the work.

References

Di Stefano, G., Ceruleo, P., Ferracci, A., Fiorillo, A., Fiore, I., Gatta, M., Rolfo, M. F., Salari, L., & Petronio, C. (2025). The Late Pleistocene Cervus elaphus from Grotta Guattari (San Felice Circeo, Central Italy). Geobios, S0016699525000841. https://doi.org/10.1016/j.geobios.2025.09.002

Fiore, I., Ceruleo, P., Di Mario, F., Ferracci, A., Gatta, M., Petronio, C., Salari, L., & Rolfo, M. F. (2025). New excavations at Grotta Guattari. Humans and carnivores: Prey and bone surface modifications. Archaeofauna, 34(1). https://doi.org/10.15366/archaeofauna2025.34.1.038

Piperno, M., & Giacobini, G. (1991). A taphonomic study of the paleosurface of Guattari Cave (Monte Circeo, Latina, Italy). Quaternaria Nova, 1990–91(1), 143–161.

Rubini, M., Zaio, P., Spanó, F., Cognigni, F., Rossi, M., Gozzi, A., & Di Mario, F. (2026). Hominin Variability and Evolutionary Relationships at Guattari Cave During the Middle and Late Pleistocene (San Felice Circeo, Latina, Italy). Genes, 17(2), 132. https://doi.org/10.3390/genes17020132

Salari, L., Gatta, M., Fiorillo, A., Fiore, I., Ceruleo, P., Di Stefano, G., Ferracci, A., Rolfo, M. F., & Petronio, C. (2025). The Late Pleistocene cave hyena from Grotta Guattari (San Felice Circeo, central Italy). Historical Biology, 1–21. https://doi.org/10.1080/08912963.2025.2526019

Sperduti, A., Alhaique, F., Borrani, A., Candilio, F., Bondioli, L., Di Vincenzo, F., Ferracci, A., Gatta, M., Piccirilli, E., Benazzi, S., Caramelli, D., Nava, A., Rolfo, M. F., & Manzi, G. (2026). The New Neanderthal Fossil Sample from Grotta Guattari, Monte Circeo (Italy): A Preliminary Synopsis. PaleoAnthropology, 2026(1), 99–121. https://doi.org/10.48738/2026.iss1.3951

Stiner, M. C. (1991). The Faunal Remains From Grotta Guattari: A Taphonomic Perspective. Current Anthropology, 32(2), 103–117. https://doi.org/10.1086/203930

White, T. D., Toth, N., Chase, P. G., Clark, G. A., Conrad, N. J., Cook, J., d’Errico, F., Donahue, R. E., Gargett, R. H., Giacobini, G., Pike-Tay, A., & Turner, A. (1991). The Question of Ritual Cannibalism at Grotta Guattari [and Comments and Replies]. Current Anthropology, 32(2), 118–138. https://doi.org/10.1086/203931

The human chin is a quirk of evolution. In everyday language, a chin is just the front of the jaw below the mouth. But to bone specialists, a chin is a very particular piece of mandibular real estate. The anatomical chin is a thickened projection of the anterior surface of the mandible, usually forming a triangle or “T” shape, below the sockets that hold the incisor and canine roots.

From at least the eighteenth century—before anthropologists even existed—writers including Linnaeus and Blumenbach accepted that a chin is a unique feature of the human jaw, not seen in other kinds of mammals. The anatomist W. D. Wallis, who wrote about the human chin back in 1917, took the idea back to Pliny the Elder who mentioned the unique nature of the human chin in his Natural History during the first century CE.

By the early twentieth century, the growing fossil record showed that the chin was a fairly recent innovation in human evolution. Extinct human relatives tended to lack a projecting chin of any kind. The Mauer mandible attributed to Homo heidelbergensis was a prominent example, as was the La Naulette, La Chapelle-aux-Saints, and the Krapina series of Neanderthal mandibles. The mandibular symphysis of these fossils—where the two halves of the mandible meet—is as thick or thicker than in most living people. But the symphysis angles from the incisors back, sometimes more strongly buttressed on the internal surface instead of an external chin.

But why? Where every other species of primate has a mandibular symphysis running down and backward from the teeth, why does the human symphysis bulge forward?

Over the years, anatomists and biologists proposed a half-dozen ideas. A few were wacky from the start, “just-so stories” that seem hardly testable.

Some imagined that the evolution of human language required a larger oral cavity in humans than other primates. With natural selection trying to give the tongue more room to work, the bony reinforcement of the mandibular symphysis had to move to the outside.

This chin-wagging notion led the twenty-sixth U.S. President to further a stereotype about Homo heidelbergensis:

Today, anthropologists often mention the idea that language might somehow be related to the chin’s evolution. The most common version of this idea is that speech constrains the placement of the tongue and musculature connected to the hyoid bone and the base of the mandible. The teeth and face became smaller in recent human evolution, but the mandibular base could not follow.

The idea is appealing on its surface, but struggles with a basic fact of human development: Children learn to speak with oral cavities vastly smaller than most adults.

Another idea was that the chin might provide support for human facial expressions by anchoring the lip and lower face musculature. Because the larger muscles of facial expression do not have attachments on the chin itself, this idea leans on smaller muscles—especially mentalis, which originates just above the chin and tends to protrude the lower lip and wrinkle the skin.

An interesting angle on the chin’s relation to facial expressions is attributable to Charles Darwin. In The Expression of the Emotions in Man and Animals, he noted that human facial expressions related to sadness look uniquely different in humans compared to other primates. The reason is simply the shape of the human chin, which shifts the soft tissue’s movement. The structure of the face not only anchors the muscles of expression, but is also the stage upon which they play.

A chin is an important part of what makes a face memorable. This underlies a third hypothesis for the chin’s function: sexual selection. The most common version of this hypothesis is that female adults tend to choose mates that have broader and more prominent chins.

Male and female adult humans all have chins, of course. Most sexual signals evolve to be unambiguous. The male pattern of facial hair, including a beard on the chin, makes for a pretty clear signal. The bone underlying all of that is hardly visible.

Still, chins of male individuals tend to be broader with well-developed lateral tubercles, and they protrude forward more. Chins of female individuals are generally narrower and less protruding. It’s this sexual dimorphism that some researchers have highlighted as support for the idea of sexual selection. Sometimes a structure that first appeared for some other reason gets co-opted as a signal. Some have suggested that the broad male chin is a signal correlated with testosterone, others that the narrow female chin is a signal of estrogen. Neither has been confirmed with physiological data.

Zaneta Thayer and Seth Dobson in a 2013 study observed that populations of humans from different parts of the world have different chin morphology. If mate preference is part of the explanation for their mandibular form, cultures seem to have different tastes. This variability detracts from the idea that a mating preference universal to Homo sapiens might explain why human jaws are differently shaped from our evolutionary ancestors.

The variation among recent human populations in chin form suggests humans are less distinctive than Pliny imagined. Some fossil jaws usually attributed to archaic groups, especially among the Neanderthals, also project slightly at the symphysis, or have raised relief overlapping with modern people.

My first encounters with fossil chins in the 1990s were at a time when researchers hotly debated the meaning of this overlap. Many researchers maintained that any resemblances between modern and archaic chins were purely superficial. The most well-known breakdown was written by Jeffrey Schwartz and Ian Tattersall. They emphasized that the details of chin shape matter. A central, raised keel along the midline, known in anatomy as the tuber symphyseos, and a pair of broadly spaced bumps on the mandibular base, tubercula lateralia, combine to make a triangular or T-shaped structure. Find the inverted T and you find a modern human.

My view was that if chins blurred the line between Neanderthal and modern, probably the line is illusory. With today’s genetic data it’s clear that ancient populations belonged to a network of genetic exchanges. Mandibular form is just one of many areas of overlap.

A similar point of view can be found in a 2002 study by Seth Dobson and Erik Trinkaus. They recognized that some earlier humans had faint traces anticipating parts of the modern human chin. On a receding mandibular symphysis, such as the massive million-year-old Tighènif 3 jaw, a trace of a tuber symphyseos may be unexceptional. But in a more vertically-oriented symphysis like the Amud 1 Neanderthal, a slight bulging at the mandibular base becomes clearer. If such discrete elements of the chin were already present in archaic humans, then expanding these elements and reducing alveolar portion of the mandible brings them into a modern configuration. Maybe all that was needed was a push.

But was that push functional, or was it evolutionary happenstance?

Out of all the ideas about the evolution of the chin, two have occupied the most attention over the last few decades. One idea holds that the chin is an external buttress, helping human jaws to minimize bending strain at the mandibular symphysis. The other holds that the chin is a spandrel: a trait that evolved as a by-product of selection for some other trait.

The functional perspective is tied to the deeper evolutionary history of hominins. Many early human relatives had truly massive jaws and teeth. Paranthropus boisei, with molar teeth four times the size of today’s people, is the most extreme of these fossil species, but even species like Australopithecus afarensis outclassed chimpanzees and humans in jaw thickness, mandibular height, and molar and premolar sizes.

The branch that led first to Homo and later to modern humans tended to reduce tooth and jaw sizes at every step. With temporalis and masseter muscles pulling from the larger-brained human skull, and with shorter tooth rows relative to the palate’s width, the geometry of the human bite tends to create a wishbone effect at the middle of the jaw.

Dobson and Trinkaus looked at the mandibles of modern humans and Neanderthals to understand what difference a chin might make to mandibular function. For mandibles of similar size, they found that a chin is irrelevant to the mandible’s resistance to bending strain. In a later study, James Pampush and David Daegling pointed out that extending the mandible further forward tends to increase the wishboning effect, not decrease it. A chin doesn’t reduce strain; it makes matters worse.

The spandrel idea reaches deep into evolutionary biology. Louis Bolk was a Dutch anatomist of the early twentieth century, who observed that human adults retain some of the form seen in babies of other primate species, such as high, curved foreheads and small faces tucked beneath the front of the cranial vault. He considered such neoteny—retention of juvenile traits—as a result of an evolutionary process he termed fetalization. Bolk saw this as an intrinsic organizing principle of human evolution.

The chin, Bolk thought, came from a developmental slowdown of the part of the jaw that holds the teeth. The basal part of the mandible was left protruding beyond it.

A similar view had been proposed by Franz Weidenreich, who much later played an important role in analysis of the Homo erectus “Peking Man” fossils from Zhoukoudian, China.

Stephen Jay Gould drew attention to the legacy of this idea in his 1977 book, Ontogeny and Phylogeny. One of Gould’s aims was to revive the old idea of neoteny in human evolution. But in the chin he perceived a challenge. Neoteny is an arrestation of growth. The human face fails to grow forward into the larger snout of other apes, holding the pattern of youth. The chin protrudes more than in any ape. It is not an arrestation, it’s an addition.

Gould squared the chin’s circle by taking a page from Weidenreich and Bolk. The chin did slow down, Gould wrote. It just slowed down less than the part of the mandible that holds the teeth.

But clever as it is, this idea begs the real question. If the chin is explained by a differential slowdown of mandibular growth, why did the two parts of the mandible slow down differently?

During the last ten years, research on the mandible has focused on this question by trying to understand the modularity of jaw development. As the body grows, different parts are affected by regulatory programs that are to some extent integrated across parts of the body. When parts must function in tandem with each other, natural selection tends to maintain an integration of their development.

The alveolar portion of the mandible is a prime example. The lower teeth must work in tandem with the upper teeth. If they developed at different rates or became very different sizes, normal chewing might become difficult or impossible. The strength of developmental integration of the maxilla and alveolar portion of the mandible keeps such mismatches from happening.

Weaker is the integration between the basal portion and alveolar portion of the mandible. The differences in pacing of these portions suggest that they are affected by distinct patterns of gene regulation.

A recent study by Noreen van Cramon-Taubadel and coworkers worked toward an understanding of the pattern of selection on the various parts of the mandible. They examined mandibular variation and evolution across great apes and humans, finding that aspects related to the chin in living humans show varying patterns of selection across the apes. Their work aligns with earlier studies led by James Pampush, which followed the evolution of a more vertical symphysis and shorter mandible across hominins. A centerpiece of these studies is that the features that make up the modern human chin were subject to selection, but not fully integrated with each other. Nor are they integrated with the longer-term reduction of human jaw length and tooth size.

The story emerging from this work is that the chin is indeed a by-product of competing demands on mandibular shape and size. One part of the story was clearly the developmental requirements of smaller teeth and the resulting evolution of a shorter face.

The other part of the story traces far before birth. Just as the brain expands rapidly before birth, the chin adopts the inverted T shape as early as the 11th week of gestation. Studies led by Michael Coquerelle may show why this is a structural necessity. As a fetus develops, it begins swallowing and breathing amniotic fluid—vital exercises for building the lungs and digestive tract. But human anatomy poses a challenge: our upright posture places the spine and neck in a way that constricts the throat. To prevent the airway from collapsing and to give the tongue and throat room to work, the bottom of the jaw must grow faster than the top.

The adult shape of the chin may have been misleading anatomists all along. This helps make sense of the diversity in the shape and size of components of the chin across human populations. Even the trends in many populations toward loss or malocclusion of wisdom teeth—strongly affecting the alveolar portion of the jaw—seem not to have been correlated with changes to the chin. After the first years of life, the structure has been free to vary independently of changes in tooth size and function. In extinct populations of humans, like the Neanderthals, larger teeth and more projecting faces prevented the fetal mismatch from occurring.

The bottom line for evolution is not how we divide the body into parts, but instead how structure matters to survival or reproduction. From Pliny’s time, naturalists have had no shortage of ideas for why a chin matters. Studies of development have helped advance our understanding of how traits are correlated with each other and with survival across the lifespan, even in the earliest phases of development.

Notes: Any subject with as long a history as the study of mandibular evolution is hard to summarize in a reasonable length. The developmental biology of this feature has occupied anatomists since well before Bolk’s time, and continued on along several avenues. Another interesting sideline that matters for studies of fossils is the challenge of defining what is meant by a chin. I mentioned the work of Schwartz and Tattersall in the post, and for readers who would like to follow up more on that aspect, I recommend the recent article by Andra Meneganzin and coworkers.

References

Bolk, L. (1924). The chin problem. Verslagen van de Akademie Wetenschapen Amsterdam, 27, 329-344.

Coquerelle, M., Prados-Frutos, J. C., Rojo, R., Drake, A. G., Murillo-Gonzalez, J. A., & Mitteroecker, P. (2017). The Fetal Origin of the Human Chin. Evolutionary Biology, 44(3), 295–311. https://doi.org/10.1007/s11692-017-9408-9

Dobson, S. D., & Trinkaus, E. (2002). Cross-sectional geometry and morphology of the mandibular symphysis in Middle and Late Pleistocene Homo. Journal of Human Evolution, 43(1), 67–87. https://doi.org/10.1006/jhev.2002.0563

Meneganzin, A., Ramsey, G., & DiFrisco, J. (2024). What is a trait? Lessons from the human chin. Journal of Experimental Zoology Part B: Molecular and Developmental Evolution, 342(2), 65–75. https://doi.org/10.1002/jez.b.23249

Pampush, J. D. (2015). Selection played a role in the evolution of the human chin. Journal of Human Evolution, 82, 127–136. https://doi.org/10.1016/j.jhevol.2015.02.005

Pampush, J. D., & Daegling, D. J. (2016). The enduring puzzle of the human chin. Evolutionary Anthropology: Issues, News, and Reviews, 25(1), 20–35. https://doi.org/10.1002/evan.21471

Pampush, J. D., Scott, J. E., Robinson, C. A., & Delezene, L. K. (2018). Oblique human symphyseal angle is associated with an evolutionary rate-shift early in the hominin clade. Journal of Human Evolution, 123, 84–95. https://doi.org/10.1016/j.jhevol.2018.06.006

Roosevelt, T. (1916). How old is man? National Geographic 29(2), 111-116.

Schwartz, J. H., & Tattersall, I. (2000). The human chin revisited: What is it and who has it? Journal of Human Evolution, 38(3), 367–409. https://doi.org/10.1006/jhev.1999.0339

Thayer, Z. M., & Dobson, S. D. (2010). Sexual dimorphism in chin shape: Implications for adaptive hypotheses. American Journal of Physical Anthropology, 143(3), 417–425. https://doi.org/10.1002/ajpa.21330

Thayer, Z. M., & Dobson, S. D. (2013). Geographic Variation in Chin Shape Challenges the Universal Facial Attractiveness Hypothesis. PLoS ONE, 8(4), e60681. https://doi.org/10.1371/journal.pone.0060681

Verhulst, J. (1994). Speech and the retardation of the human mandible: A Bolkian view. Journal of Social and Evolutionary Systems, 17(3), 307–337. https://doi.org/10.1016/1061-7361(94)90014-0

Von Cramon-Taubadel, N., Scott, J. E., Robinson, C. A., & Schroeder, L. (2026). Is the human chin a spandrel? Insights from an evolutionary analysis of ape craniomandibular form. PLOS One, 21(1), e0340278. https://doi.org/10.1371/journal.pone.0340278

Wallis, W. D. (1917). The development of the human chin. The Anatomical Record, 12(2), 315–328. https://doi.org/10.1002/ar.1090120212

White, T. D. (1977). The anterior mandibular corpus of early African Hominidae: Functional significance of shape and size [PhD, University of Michigan]. https://www.proquest.com/docview/302839985

Today’s people have some Neanderthal ancestors. It’s remarkable as I think about it, that we’ve known about this ancient connection for the last 16 years.

Narrowing down exactly which Neanderthals those ancestors were is a bigger challenge. Where did they live? Does everyone living today have genetic ancestry from the same Neanderthal group? Or from many different ones? How many were there?

A series of papers from the research group led by Mathias Currat has put a new spin on these questions. These papers use the current dataset of ancient genomes from across Eurasia, numbering more than 4000 individuals in total across the time from 40,000 years ago up to 600 years ago. With this sample it is possible to look at variation in Neanderthal ancestry over both time and space.

The developing picture suggests that when the early founders of today’s Eurasian populations were mixing with Neanderthals, that mixture was not limited to one small region or group. Instead, the wave of founders spread into the Neanderthals’ geographic range, mixing with the locals across thousands of kilometers. Ultimately that process extended across most of the Neanderthals’ geographic range, from Afghanistan to France.

The first story

When the first ancient genomes were recovered in 2010, researchers first hypothesized that Neanderthal-modern genetic mixture happened only within a very narrow geographic region.

Early analyses showed that the fraction of Neanderthal genetic ancestry was more or less the same in living people across Europe, Asia, Oceania, and the Americas. All these people had an estimated 1% and 4% Neanderthal-derived DNA.

Only in Africa were people clearly much lower in Neanderthal heritage, small enough to be undetectable with the first statistical approaches to the question. Today we know that across Africa south of the Sahara Desert, living people have only a trace of Neanderthal genetic ancestry. This pattern is a legacy of thousands of years of migration and mixing between Africa and Eurasia. The level is not very high: a fraction of a percent. Still, the sheer distances involved—some 8,000 km across the continent north-to-south—are a vivid reminder of how gene flow has diffused across groups.

To a lot of people, the Neanderthal inheritance was a surprise. Not me, I expected it. What did surprise me was the seeming uniformity of Neanderthal ancestry in so many parts of the world. After all, the Neanderthals lived in Europe, not China or Australia. Like many people, I thought Neanderthal genes should be most common in Europe.

Altogether, these observations led to a hypothesis: An early founder population got its start in Africa, met Neanderthals somewhere in the Near East, and then dispersed throughout the world. Their descendants carried the same Neanderthal legacy with them everywhere they went. The timing of that mixture coincided with the dispersal, sometime between 70,000 and 50,000 years ago.

In East Asia and Southeast Asia, the expanding wave of people never met any more Neanderthals. They did meet others. Varied branches of modern descendants mixed with several deep lineages that we group together as Denisovans, to different degrees, carrying that DNA forward into Oceania and the Americas.

In Europe, on the other hand, the expanding wave of modern people met Neanderthals at every step. Yet somehow they did not mix any further. Maybe, some geneticists thought, some kind of genetic incompatibility stopped further Neanderthal genes from being integrated into later people.

An extra dash of Neanderthal

In 2013, Jeff Wall and coworkers shifted the story by showing that East Asian populations have more Neanderthal ancestry than Europeans. That was really the reverse of what people had expected. Again, Neanderthals lived in Europe, not China.

The difference is small but may be consequential. As data have grown, the numbers have solidified to around 0.2 to 0.5% of the genome, which adds up to around 31 million nucleotides. Still, it’s a good deal less than people within each region often differ from each other.

How did that extra Neanderthal ancestry get into East Asian populations? According to one idea, the founder population of East Asian groups was smaller in numbers than the west. With higher genetic drift, purifying selection was less effective in weeding out many slightly deleterious Neanderthal gene variants. They persisted, where in other parts of Eurasia they disappeared. Another scenario proposes that when the original out-of-Africa founders split into eastern and western branches, the eastern branch met another group of Neanderthals and picked up a small extra burst of DNA.

Other researchers suspected that the dash of extra Neanderthal might be a statistical illusion. Researchers might conceivably have been misrecognizing genes from Denisovans. Or maybe the longer average haplotype lengths across genomes in East Asia—a legacy of the founder effects early in modern human arrival to the region—meant that geneticists were slightly underestimating Neanderthal haplotypes everywhere else.

One thing is for certain: the small differences are hard to measure accurately. Making things harder is that subsequent movements and mixtures of people tended to blur the patterns that once existed.

The impact of Basal Eurasians

I’ve tended to favor the idea that parts of western Eurasia received an extra burst of African ancestry during the last 20,000 years. My thinking traces back to the partial sequencing of the genome from the Oase 1 individual by Qiaomei Fu and collaborators in 2015. This individual had more than double the Neanderthal ancestry of anyone living today, and the length of haplotypes suggested that the individual’s most recent Neanderthal ancestor had lived just a handful of generations earlier. Later work uncovered other early Europeans with similar recent Neanderthal ancestry, from Bacho Kiro, Bulgaria, and Zlatý kůň, Czechia. All of these early Europeans had ancestry from the same founder population that ultimately gave rise to most of today’s Eurasian gene pool. But these people who lived at the time of the earliest Upper Paleolithic in Europe made little dent on the gene pool of later Europeans.

It’s probably too much to say that the first waves of dispersal into Europe “failed”, or became extinct. Genetic samples of subsequent Upper Paleolithic groups are too small to test strong claims about replacement of the first peoples. Still, it is clear that living people haven’t got much ancestry from the earliest European groups.

No matter what happened later, the wave of Neanderthal mixture in the first modern Europeans was mostly erased.

No single event established today’s pattern across western Eurasia. Every wave of newcomers had its own effect. One important influence was the spread of early farmers from Southwest Asia. A series of papers on ancient genomes by Iosif Lazaridis and collaborators, starting in 2014, hypothesized that ancestry from a very early out-of-Africa group—which geneticists call the Basal Eurasians—was important to the populations of Southwest Asia. The Basal Eurasians, possibly stemming from the very first out-of-Africa founders, had a lot less Neanderthal ancestry than other peoples of the continent. That lower Neanderthal ancestry is observed within their descendants in the Levant and Arabia today. The Basal Eurasians made up nearly half the background of Neolithic peoples who brought farming into Europe. With their lower Neanderthal ancestry, the coming of the farmers could have covered up some of the evidence of earlier mixture.

Ancient DNA from early modern Europeans has added enormously to this picture. In a 2023 paper, Claudio Quilodrán and collaborators took the known sample of ancient genomes from across Eurasia, a dataset numbering more than 4000 individuals, and looked at the pattern of Neanderthal genetic ancestry over space and time.

They found many clear relationships. In both Europe and Asia, Neanderthal ancestry forms a slight gradient with higher values at higher latitudes. Going further north, you encounter slightly more Neanderthal ancestry.

In population genetics, this kind of gradient of variation across space is known as a cline. Clines result from processes that mix genes across space. Ultimately genes from different places mix because people move and mate. But whether movement was a massive wave or a slow percolation between two or more populations can be hard to work out.

Ancient genomes help by adding the time dimension. Using the sample of 4000+ ancient genomes, Quilodrán and coworkers were able to show how the cline in Europe changed with the arrival of early farmers. The spread of this group happened around 6000 years ago. By that time Neanderthals were long gone—for more than 30,000 years except possibly at the very extreme in Iberia where they may have persisted a touch longer. Farmers brought along their Basal Eurasian ancestry. That didn’t change the direction of the genetic gradient, but it did cause a reduction of around half a percent Neanderthal ancestry all across Europe.

Today’s European component of Neanderthal ancestry is actually higher than that found in the early European farmers. After their initial arrival in Europe, early farmers mixed with local hunter-gatherers. The genetic makeup of later populations was also affected by the arrival of steppe peoples during the Bronze Age. In the end, today’s complement of Neanderthal genes in Europe was affected by the early farmers only moderately.

A Neanderthal ancestry cline

One of the most valuable figures in the work by Quilodrán and coworkers is the plot of Neanderthal ancestry according to latitude and longitude. The plot is separated into the two continental regions, Europe and Asia, and that separation helps a great deal in recognizing the overall pattern.

Higher latitudes have higher Neanderthal ancestry, as mentioned above. The pattern with longitude has an inflection point. Across Europe, Neanderthal ancestry increases slightly as you go west; across Asia Neanderthal ancestry increases slightly as you go east.

Slight means very slight. In both eastern and western directions, the cline of increasing Neanderthal ancestry adds up to much less than one percent of the genome. The differences between the ends of the cline are much smaller than the variation among individuals at any one geographic location.

Still, a difference of a few tenths of a percent makes up a fair fraction compared to the 1 to 3 percent Neanderthal genetic ancestry that Eurasian people have today and in the past.

A new study from the same research team, led in this preprint by Lionel Di Santo, works to explain how these clines got their start. The team simulated a growing modern population moving step by step across a zone where they mixed with Neanderthals. That stepwise population growth gives rise to a cline as long as interbreeding continues. The further into the zone of mixture, the more Neanderthal genetic input there should be.

In this way, both the latitude and longitude clines can be explained by a wave of advance of modern people, moving into territories occupied by Neanderthals and interbreeding with them along the way.

But there is an important twist. A cline established by an initial wave of advance with mixture should only extend as far as the mixture happened.

To investigate the extent of the clines more precisely, Di Santo and coworkers applied a nonlinear regression model. They found that the European cline keeps increasing almost all the way to the western extreme of Europe, but not quite—it stops in France, beyond which there is no further increase in Neanderthal ancestry. Likewise for the cline’s northern extent, which stops in Germany, well short of Scotland and Norway.

In Asia the cline’s limits are more striking. Neanderthal ancestry increases among the ancient genomes from the Levant to Kazakhstan and northern Pakistan. But further to the north and east there is no further increase in Neanderthal ancestry. Across the continent there is indeed a correlation of latitude and longitude with Neanderthal ancestry, but that correlation is almost entirely the result of the western half of Asian samples.

From this, Di Santo and coworkers draw a map of the extent of interbreeding between the expanding modern human population and the Neanderthals they encountered. That map shows the range of mixture extending from the Levant to Kashmir, northward to the Volga, and westward encompassing most of Europe to the Marne. It’s not much short of the full known geographic range of Neanderthals.

Bottom line

These studies from Mathias Currat’s research group help to put a more decentralized Neanderthal ancestry back on the table. There was not one small, special place where incoming African people and Neanderthals mixed. The mixture happened as a wave.

The clinal pattern of Neanderthal mixture is real, but visible only once you start looking at very large numbers of genomes. With small numbers, the effect is overwhelmed by individual variation.

Equally real is the abbreviated timeline during which mixture seems to have happened. In 2025, two research studies suggested that the Neanderthal ancestry of living people had mostly come in one extended pulse, estimated from 50,000 to 43,000 years ago. These studies have placed an upper limit on the time that some 95% of the genetics of today’s Eurasian people started from a small founder population.

I think these converging lines of research align with a single scenario.

• Sometime after 70,000 years ago, a small founder population left its close relatives in North Africa and entered the Levant.

• The Basal Eurasians are simply the descendants of these founders nearest the starting line, the lowest end of Neanderthal mixture in the cline.

• The uptake of Neanderthal ancestry initially was high, owing to the small size of the founder groups. Some of the Neanderthal gene regions proved to be deleterious in the growing population. But other variants were adaptive, adding to the population’s growth potential.

• Once a wavefront was established, the groups grew at an intrinsic growth rate of between 2% and 4% per generation. That’s a growth rate comparable to many prehistoric human groups, an order of magnitude slower than many post-industrial human populations.

• With a range size and birth-to-marriage distance comparable to human hunter-gatherers, the wave advanced at a velocity of around 20 km per generation. In 7000 years, the wave traveled between 4000 and 6000 km.

• At this point, Neanderthals remained in small pockets at the ends of their former range, and nearly all remaining descendants of the Neanderthals were modern people with between 1% and 4% Neanderthal ancestry.

A fast wave of advance is a good match to the skeletal record. The people from sites like Bacho Kiro, Bulgaria, and Ranis, Germany, arrived in Europe within a few thousand years after the founder group started to expand.

The earliest Neanderthals to mix into this expanding population were from Southwest Asia, making up a disproportionately high fraction of the ultimate Neanderthal genomic legacy. The Neanderthals that lived further toward Europe and Central Asia may have been fewer in number and lower in density. But the main reason they had less genetic impact was simply the uninterrupted growth of the newcomers.

Still, the cline of Neanderthal ancestry shows that their integration at the frontier of modern human expansion was a common occurrence, having a persistent effect on the variation of their descendant populations.

Looking at today’s people, more than 1500 generations have passed since any Neanderthal ancestors lived, fragmenting great introgressed blocks of chromosomes into much shorter shards of sequence. Geneticists can amplify the weak signal from living people’s genomes by building up samples of many thousands of participants. The remarkable thing about ancient genomes is that they now number in the thousands, too. That is changing the way we can look at these critical events in our evolutionary history.

References

Di Santo, L. N., Quilodrán, C. S., Cerrito, P., & Currat, M. (2026). Tracing the Neanderthal–Modern Human hybrid zone using paleogenomic data. bioRxiv. https://doi.org/10.64898/2026.01.06.697899

Fu, Q., Hajdinjak, M., Moldovan, O. T., Constantin, S., Mallick, S., Skoglund, P., Patterson, N., Rohland, N., Lazaridis, I., Nickel, B., Viola, B., Prüfer, K., Meyer, M., Kelso, J., Reich, D., & Pääbo, S. (2015). An early modern human from Romania with a recent Neanderthal ancestor. Nature, 524(7564), 216–219. https://doi.org/10.1038/nature14558

Iasi, L. N. M., Chintalapati, M., Skov, L., Mesa, A. B., Hajdinjak, M., Peter, B. M., & Moorjani, P. (2024). Neanderthal ancestry through time: Insights from genomes of ancient and present-day humans. Science, 386(6727), eadq3010. https://doi.org/10.1126/science.adq3010

Lazaridis, I., Patterson, N., Mittnik, A., Renaud, G., Mallick, S., Kirsanow, K., Sudmant, P. H., Schraiber, J. G., Castellano, S., Lipson, M., Berger, B., Economou, C., Bollongino, R., Fu, Q., Bos, K. I., Nordenfelt, S., Li, H., de Filippo, C., Prüfer, K., … Krause, J. (2014). Ancient human genomes suggest three ancestral populations for present-day Europeans. Nature, 513(7518), 409–413. https://doi.org/10.1038/nature13673

Quilodrán, C. S., Rio, J., Tsoupas, A., & Currat, M. (2023). Past human expansions shaped the spatial pattern of Neanderthal ancestry. Science Advances, 9(42), eadg9817. https://doi.org/10.1126/sciadv.adg9817

Quilodrán, C. S., Tsoupas, A., & Currat, M. (2020). The Spatial Signature of Introgression After a Biological Invasion With Hybridization. Frontiers in Ecology and Evolution, 8, 569620. https://doi.org/10.3389/fevo.2020.569620

Sankararaman, S., Patterson, N., Li, H., Pääbo, S., & Reich, D. (2012). The Date of Interbreeding between Neandertals and Modern Humans. PLoS Genetics, 8(10), e1002947. https://doi.org/10.1371/journal.pgen.1002947

Sümer, A. P., Rougier, H., Villalba-Mouco, V., Huang, Y., Iasi, L. N. M., Essel, E., Mesa, A. B., Furtwaengler, A., Peyrégne, S., de Filippo, C., Rohrlach, A. B., Pierini, F., Mafessoni, F., Fewlass, H., Zavala, E. I., Mylopotamitaki, D., Bianco, R. A., Schmidt, A., Zorn, J., … Krause, J. (2024). Earliest modern human genomes constrain timing of Neanderthal admixture. Nature, 1–3. https://doi.org/10.1038/s41586-024-08420-x

Wall, J. D., Yang, M. A., Jay, F., Kim, S. K., Durand, E. Y., Stevison, L. S., Gignoux, C., Woerner, A., Hammer, M. F., & Slatkin, M. (2013). Higher Levels of Neanderthal Ancestry in East Asians than in Europeans. Genetics, 194(1), 199–209. https://doi.org/10.1534/genetics.112.148213

Yuan, K., Ni, X., Liu, C., Pan, Y., Deng, L., Zhang, R., Gao, Y., Ge, X., Liu, J., Ma, X., Lou, H., Wu, T., & Xu, S. (2021). Refining models of archaic admixture in Eurasia with ArchaicSeeker 2.0. Nature Communications, 12(1), 6232. https://doi.org/10.1038/s41467-021-26503-5

Yesterday I was talking with the Jakarta Post about how exciting I find the ongoing archaeological discoveries from Indonesia. The occasion of the interview was the new research on rock art from southeastern Sulawesi, the oldest dating back to more than 68,000 years ago.

The new work is the latest revelation from a program of rock art dating that has been underway for more than a decade. The paintings themselves, within caves and on rock walls in several parts of Indonesia, are not new discoveries. Many are already well-trafficked tourist sites, known and studied by Indonesian scholars for decades. What is new is the chemical analysis of small calcite formations that formed on top of some of the painted surfaces. These samples enable archaeologists to determine a minimum age for the art, integrating the pictorial record within the chronology of human arrival and activity in this region.

I wrote about the archaeological record of Sulawesi just last week. Work in Leang Bulu Bettue in southwestern Sulawesi shows an unchanged pattern of toolmaking extending from more than 130,000 up to 40,000 years ago, when a new tradition arrived. That age seems out of phase with the earliest-known rock art in the area, estimated to be more than 51,000 years old. The mismatch, I wrote, needs an explanation.

The new work by Adhi Agus Oktaviana pushes the earliest rock art back by an additional 17,000 years. The oldest panel is from Liang Metanduno, on the small island of Muna. Among a vast array of artworks dating to the last few thousand years, Oktaviana noticed two faint handprints, seemingly behind other more recent brightly-painted figures. Those two handprints generated dates averaging 71,000 years ago—the error bar of one going down to 67,800 years.

That’s 28,000 years or more before any known change in local stone tools.

On the face of things, I’d say the data suggest that the early hominin occupants of the island, possibly Denisovans, were marking their landscapes.

Some authors working on rock art suggest that the art itself is evidence for modern human presence. According to this line of thinking, rock art could not have been made by any other population of hominin. It is uniquely modern.

I don’t agree. Traditions of marking are widespread among Pleistocene hominins. In a network of interactions across islands and between hominin groups, I expect that the transfer of knowledge and traditions may have taken some surprising paths.

How rock art dating works

To carry out this new study, Adhi Agus Oktaviana and a large team of collaborators surveyed rock art sites and took small samples of calcite that they found overlaying some parts of the paintings. In the lab, they prepared thin sections of the calcite and sampled thin layers, for each one developing a series of measurements of uranium and thorium, one of the decay products of uranium. The age of the layer closest to the painting is a minimum age for the painting itself.

Nature has a nice short commentary that accompanies the research article, which includes a couple of paragraphs from the lead author telling the story of the find.

A common criticism of this kind of work is that the uranium within a layer of calcite may be affected by later events. Groundwater carries uranium and leaves it within the calcite formation, which is how it gets there in the first place. When uranium atoms decay, the resulting thorium remains in the calcite. But later dripping or atmospheric water can leach the uranium out. If the uranium content is removed while the thorium remains, the resulting dates will be overestimates of the true age of the calcite.

This doesn’t necessarily overstate the age of the underlying painting. Even in the ideal case the calcite formed after the pigments were applied. But such leaching removes the logic for the calcite being a minimum age.

Throughout several publications using this methodology, the lead investigator Maxime Aubert and team have argued that the leaching of uranium can be ruled out. The main approach to detect such leaching is the dating of thin calcite layers in a sequence. Leaching from the outside would deplete the uranium in the layers furthest from the painting, resulting in an inversion of the ages. That isn’t seen in the Liang Metanduno hand stencils or in any of the other early art.

I would add that the broader picture of dating of Pleistocene art in the region, including examples in Borneo and Sulawesi, shows that there are many, many examples older than 30,000 years. A date of 68,000 years may seem like an outlier at the moment, but beyond island Southeast Asia that age is comparable to estimates for pigment markings in caves in Spain such as Maltravieso, Ardales, and La Pasiega, where marks were likely created by Neanderthals.

Who made the marks?

For me the null hypothesis is that the marks were made by the archaic hominins known to have been on the island at the time. The archaeological record of Sulawesi thus far offers no evidence of cultural changes that would point to an influx of newcomers until around 40,000 years ago.

In their paper, Oktaviana and coworkers suggest that modern humans made the handprints. Their logic is that the art itself is evidence of newcomers. They further suggest a possible stylistic link between the handprints in Liang Metanduno and some sites with later ages.

Could it be that these marks and other later painted figures, such as the 51,000+ year-old paintings from Leang Karampuang, were made by modern island-hoppers on their way to Australia?

I think it’s just as likely that the handprints were made by Denisovan coastline-adapted people whose relatives were already making their homes in coastal Sahul by this time. The notion that there was uniquely one dispersal of humans across eastern Indonesia requires a bunch of assumptions that I wouldn’t make. Instead, I think the genetic and archaeological data may be best explained by a network of interactions with multiple pathways for introgression and mixture.

For readers in Europe, my understanding is that the Arte television channel will be broadcasting a documentary on the Sulawesi rock art finds in conjunction with the paper’s publication today. The channel has posted a few clips on YouTube that are available to viewers outside Europe. I’ve embedded one here that shows the research team in Liang Metanduno, with Adhi Oktaviana pointing to the faint handprints.

It’s a truly remarkable finding, and with such a regional tradition of art, I am sure there will be many more dates to flesh out the true antiquity of this tradition and its makers.

Notes: The first habitation of Australia is a tough topic to write about because of the scope of evidence that is part of recent debates. The newsroom version of this topic doesn’t get into the kinds of details that are really at issue, and different research groups have built very different concepts of how the archaeological findings relate to movements of modern humans and others. As I write more about this topic, I’ll try to flag some of the points of agreement and disagreement that have featured in the published research.

Meanwhile, here are links to some of the key recent posts that cover research from this part of the world.

References

Aubert, M., Brumm, A., Oktaviana, A., & Joannes-Boyau, R. (2026). Humanity’s oldest known cave art has been discovered in Sulawesi. The Conversation. https://doi.org/10.64628/AA.j4xujax6w

Aubert, M., Brumm, A., Ramli, M., Sutikna, T., Saptomo, E. W., Hakim, B., Morwood, M. J., Van Den Bergh, G. D., Kinsley, L., & Dosseto, A. (2014). Pleistocene cave art from Sulawesi, Indonesia. Nature, 514(7521), 223–227. https://doi.org/10.1038/nature13422

Brumm, A., Langley, M. C., Moore, M. W., Hakim, B., Ramli, M., Sumantri, I., Burhan, B., Saiful, A. M., Siagian, L., Suryatman, Sardi, R., Jusdi, A., Abdullah, Mubarak, A. P., Hasliana, Hasrianti, Oktaviana, A. A., Adhityatama, S., Van Den Bergh, G. D., … Grün, R. (2017). Early human symbolic behavior in the Late Pleistocene of Wallacea. Proceedings of the National Academy of Sciences, 114(16), 4105–4110. https://doi.org/10.1073/pnas.1619013114

Brumm, A., Oktaviana, A. A., Burhan, B., Hakim, B., Lebe, R., Zhao, J., Sulistyarto, P. H., Ririmasse, M., Adhityatama, S., Sumantri, I., & Aubert, M. (2021). Oldest cave art found in Sulawesi. Science Advances, 7(3), eabd4648. https://doi.org/10.1126/sciadv.abd4648

Oktaviana, A. A., Joannes-Boyau, R., Hakim, B., Burhan, B., Sardi, R., Adhityatama, S., Hamrullah, Sumantri, I., Tang, M., Lebe, R., Ilyas, I., Abbas, A., Jusdi, A., Mahardian, D. E., Noerwidi, S., Ririmasse, M. N. R., Mahmud, I., Duli, A., Aksa, L. M., … Aubert, M. (2024). Narrative cave art in Indonesia by 51,200 years ago. Nature, 631(8022), 814–818. https://doi.org/10.1038/s41586-024-07541-7

Westaway, K. E., Louys, J., Awe, R. D., Morwood, M. J., Price, G. J., Zhao, J. -x, Aubert, M., Joannes-Boyau, R., Smith, T. M., Skinner, M. M., Compton, T., Bailey, R. M., van den Bergh, G. D., de Vos, J., Pike, A. W. G., Stringer, C., Saptomo, E. W., Rizal, Y., Zaim, J., … Sulistyanto, B. (2017). An early modern human presence in Sumatra 73,000–63,000 years ago. Nature, 548(7667), 322–325. https://doi.org/10.1038/nature23452

White, R., Bosinski, G., Bourrillon, R., Clottes, J., Conkey, M. W., Rodriguez, S. C., Cortés-Sánchez, M., de la Rasilla Vives, M., Delluc, B., Delluc, G., Feruglio, V., Floss, H., Foucher, P., Fritz, C., Fuentes, O., Garate, D., González Gómez, J., González-Morales, M. R., González-Pumariega Solis, M., … Willis, M. D. (2020). Still no archaeological evidence that Neanderthals created Iberian cave art. Journal of Human Evolution, 144, 102640. https://doi.org/10.1016/j.jhevol.2019.102640

Sulawesi has become one of the most exciting areas for discoveries in human evolution. Last year, archaeologists led by Budianto Hakim showed that hominins have been on the island for at least a million years—although whether continuously or intermittently is not clear.

Unknown also is who the first inhabitants were. Maybe they were descendants of the Homo erectus population that inhabited Sundaland from 1.5 million years ago or earlier. Without fossils we cannot be certain.

The earliest skeletal evidence of hominins from the island is vastly younger, a mere 25,000 to 16,000 years old. A fragmentary upper jaw of a modern person of this date comes from a cave called Leang Bulu Bettue, near Maros in south Sulawesi.

This cave has been heating up for the last decade as excavations reach deeper into the its sediment floor.

In an open access paper from late 2025, Basran Burhan and collaborators give the first report of much earlier finds from the deepest-known layers of Leang Bulu Bettue. The oldest artifacts they report come from between 208,000 and 132,000 years ago. They make Leang Bulu Bettue the first site on Sulawesi with a record spanning from the Middle Pleistocene across the arrival of early modern people.

Such a record has high value. It may reveal what happened to the earlier inhabitants. Comparable sites in the region are Callao Cave, on Luzon, and Liang Bua on Flores—both showing evidence of an early island hominin species followed by modern humans.

Leang Bulu Bettue

The Leang Bulu Bettue cave has a long tunnel leading from a rockshelter entrance at the base of a cliff. The researchers note that the cave’s name in the Bugis language means roughly “cave of the tunnel through the hill”. It sits within the highly karstic landscape of the Maros regency, where steep, narrow valleys cut between tall, isolated towers of limestone.

The cave is near some of the earliest figurative rock art in the world, dated to more than 50,000 years ago, at Leang Karampuang and Leang Bulu’ Sipong 4.

Many of the finds from later, higher levels in the cave have been previously reported in other papers. Those include:

• Rock art on the cave’s ceiling, including hand stencils, apparently dating to a similar period as art from nearby sites like Leang Karampuang. These are overlain by black pigment drawings from the era when Austronesian language-speaking people lived on the island, a few thousand years ago.

• A modern human bone fragment, Maros-LBB-1a: the right half of a maxilla with three molars, from layer 4a of the site with an age estimated between 25,000 and 16,000 years ago.

• Layers dating to between 30,000 and 22,000 years have ochre blocks, animal bone with traces of ochre, and a few artifacts that were personal ornaments or part of the production chain for ornaments. These may show some of the portable components of the art tradition reflected by cave paintings.

In the new study, Burhan and coworkers probe more deeply. The rich and dense record of the last 40,000 years becomes spotty and intermittent going further back.

In the paper is a great figure showing the spatial distribution of artifacts, bone, and other objects plotted throughout their two excavation areas. The finds plotted in yellow come from Austronesian culture and Neolithic periods, all within the last few thousand years. The orange dots represent finds from the period between 40,000 and 16,000 years ago, which Burhan and coworkers denote as “Phase II”, while the darker red dots represent “Phase I” finds in layer 5 of the site and below, all older than 40,000 years.

The finds reaching back to lowest parts of the excavation are older than 132,000 years. Clearly a key time in the site’s history is the intensification of site use that started around 40,000 years ago.

A tale of two industries

In an earlier paper led by Yinika Perston in 2022, working from the higher layers, the team could already see that something different was happening before 40,000 years ago.

An “Upper Industry” after 40,000 years ago includes many large, elongated flakes and many microblades. Some of the artifacts are coated with patches of ochre residue, and some are shaped in ways that suggest specialized processing of ochre. The toolmakers of this period used several different flaking methods. Their tools were mostly made from chert, often brought to the site from outside the immediate local area.

The “Lower Industry”—at the time known mainly from layer 5—includes large cores, some flaked from limestone and some made from volcanic stream-rolled cobbles. Nearly all the artifacts are locally available materials, and chert is rare.

The recent study from Burhan and coworkers extends that record substantially deeper into the site. Even so, the Lower Industry in layers 5 and below have only 40 artifacts, compared to more than 25,000 artifacts excavated from layer 4 alone. The current evidence shows that Lower Industry artifacts from Leang Bulu Bettue go back to layer 10, with an estimated age between 208,000 and 132,000 years.

The difference between Lower and Upper industries looks like a cultural break. The tools from the Lower Industry are not only limited in their flaking processes and raw material sources, they reflect a vastly less intensive use of the cave than the Upper Industry.

Perston and coworkers observed that 40,000 years ago is close to the first arrival of modern humans in the region, who were likely the makers of the Upper Industry. The Lower Industry, on the other hand, was likely produced by earlier hominins.

Who were the earlier inhabitants?

Burhan and coworkers note that stone artifacts from Talepu, also in south Sulawesi, are in the same time range and similar in their manufacture. A decade ago, Gerrit van den Bergh and coworkers sunk a deep trench there, finding artifacts more than two meters down, in sediments estimated between 194,000 and 118,000 years old. The artifacts belong to a flake and core industry made from river cobbles and targeted toward production of sharp edges without much shaping or retouching. They resemble the Lower Industry tools from Leang Bulu Bettue.

Who made these tools? There are several possibilities.

• Maybe Denisovans reached the island, accounting for their strong legacy within the genomes of people in the Philippines, Papua, and the surrounding area.

• Maybe Homo erectus was present on the island. This species certainly existed on Java around a million years ago, the time the first evidence of habitation of Sulawesi is recorded. Sulawesi is a large enough island that a population of H. erectus might have persisted there without necessarily changing much from the earliest inhabitants.

• When the island was inhabited a million years ago, possibly by H. erectus, the population may have adapted to local circumstances. If that adaptation went far enough, it may have become an endemic island species along the lines of Homo luzonensis or Homo floresiensis. On a larger island than either Flores or Luzon, its characteristics might look very different from those other hominins.

I wish the stone tools gave more of a clue. But nothing in the cultural evidence gives me a reason to prefer one of these hypotheses over the others.

Cutmarks

One thing I want to highlight briefly is that Leang Bulu Bettue has evidence for cutmarks on animal bones in a lower level than any stone artifacts have yet been identified. In layer 13 of the Cave Mouth Trench portion of their excavation, Burhan and collaborators found four bones with marks from butchery by sharp artifacts: two macaque bones, a bone from an anoa (dwarf buffalo), and an unidentifiable bone fragment. These are older than any stone artifact in the site.

This isn’t strange. The overall intensity of cave use during the Phase I period was very light. Flaking debitage is mostly absent, which means tools of the Lower Industry were mostly not made in the cave. Hominins that carried tools in would usually have carried them back out. Meanwhile, they may have processed animal carcasses outside the cave, only carrying parts inside occasionally. Or in some instances, animal scavengers might move parts of butchered carcasses into a cave opening.

What makes me point to the early cutmarks is that they are part of a larger statistical pattern. The first evidence of hominin presence in many sites—and in some entire regions of the planet—is cutmarked bone.

Of course the Leang Bulu Bettue evidence is not the earliest occupation evidence on Sulawesi. The stone tools from Calio, in the Walanae valley, show that toolmakers got their start on the island by 1.26 ± 0.22  million years ago. This was one of the biggest stories in human evolution that I reviewed in 2025.

In a region where individual hominins may have made thousands of artifacts in their lives, and butchered thousands of animals, both kinds of traces may vastly outnumber the skeletal and dental remains of hominins themselves. But which comes first? The answer may depend what kind of sites archaeologists happen to investigate.

Probing a mismatch

Readers will probably notice what stood out to me in these papers. If rock art at nearby sites is now dated to more than 50,000 years ago, why is the threshold dividing occupation phases reflected in the sediments at Leang Bulu Bettue only 40,000 years old?

Maybe some disconnect is to be expected. Rock art may have been such a part of the lives of the early coastal travelers that people left marks long before actually settling on Sulawesi. It’s clear from the early archaeological record of Australia that pigments and marking were important part of cultures from their arrival. Possibly across the region, paintings tied together the places travelers had been long before new populations grew to densities where a cave like Leang Bulu Bettue would get much use.

Still, the persistence of the Lower Industry into the period after 50,000 years is curious. These artifacts are so few, so differently made than later stone tools. Is it possible that the earlier hominin inhabitants of Sulawesi persisted for up to 10,000 years after the first arrival of modern people? The cave record seems to suggest not only did those hominins survive, they were dominant in their presence in Leang Bulu Bettue, only a stone’s throw from the Leang Karampuang paintings.

My sense is that cultural interchange should have been much more likely than isolation. Earlier publications on Leang Bulu Bettue have emphasized the connections between the wider regional rock art and the evidence for ochre and portable art in the cave. If something impeded such interchange for 10,000 years, I want to understand how.

References

Aubert, M., Brumm, A., Ramli, M., Sutikna, T., Saptomo, E. W., Hakim, B., Morwood, M. J., Van Den Bergh, G. D., Kinsley, L., & Dosseto, A. (2014). Pleistocene cave art from Sulawesi, Indonesia. Nature, 514(7521), 223–227. https://doi.org/10.1038/nature13422

Brumm, A., Bulbeck, D., Hakim, B., Burhan, B., Oktaviana, A. A., Sumantri, I., Zhao, J., Aubert, M., Sardi, R., McGahan, D., Saiful, A. M., Adhityatama, S., & Kaifu, Y. (2021). Skeletal remains of a Pleistocene modern human (Homo sapiens) from Sulawesi. PLOS ONE, 16(9), e0257273. https://doi.org/10.1371/journal.pone.0257273

Brumm, A., Hakim, B., Ramli, M., Aubert, M., Van Den Bergh, G. D., Li, B., Burhan, B., Saiful, A. M., Siagian, L., Sardi, R., Jusdi, A., Abdullah, Mubarak, A. P., Moore, M. W., Roberts, R. G., Zhao, J., McGahan, D., Jones, B. G., Perston, Y., … Morwood, M. J. (2018). A reassessment of the early archaeological record at Leang Burung 2, a Late Pleistocene rock-shelter site on the Indonesian island of Sulawesi. PLOS ONE, 13(4), e0193025. https://doi.org/10.1371/journal.pone.0193025

Brumm, A., Langley, M. C., Moore, M. W., Hakim, B., Ramli, M., Sumantri, I., Burhan, B., Saiful, A. M., Siagian, L., Suryatman, Sardi, R., Jusdi, A., Abdullah, Mubarak, A. P., Hasliana, Hasrianti, Oktaviana, A. A., Adhityatama, S., Van Den Bergh, G. D., … Grün, R. (2017). Early human symbolic behavior in the Late Pleistocene of Wallacea. Proceedings of the National Academy of Sciences, 114(16), 4105–4110. https://doi.org/10.1073/pnas.1619013114

Burhan, B., Hakim, B., Sumantri, I., Suryatman, Saiful, A. M., Oktaviana, A. A., Sardi, R., Hasliana, Ramli, M., Siagian, L., Jusdi, A., Abdullah, Syahdar, F. A., Hamrullah, Ilyas, I., Muhammad, P. H., Budi, S. S., Djindar, N. I., Adhityatama, S., … Brumm, A. (2025). A near-continuous archaeological record of Pleistocene human occupation at Leang Bulu Bettue, Sulawesi, Indonesia. PLOS One, 20(12), e0337993. https://doi.org/10.1371/journal.pone.0337993

Hakim, B., Wibowo, U. P., van den Bergh, G. D., Yurnaldi, D., Joannes-Boyau, R., Duli, A., Suryatman, Sardi, R., Nurani, I. A., Puspaningrum, M. R., Mahmud, I., Haris, A., Anshari, K. A., Saiful, A. M., Arman Bungaran, P., Adhityatama, S., Muhammad, P. H., Akib, A., Somba, N., … Brumm, A. (2025). Hominins on Sulawesi during the Early Pleistocene. Nature, 1–6. https://doi.org/10.1038/s41586-025-09348-6

Perston, Y. L., Moore, M. W., Suryatman, N. F. N., Burhan, B., Hakim, B., Hasliana, N. F. N., Agus Oktaviana, A., Lebe, R., Mahmud, I., & Brumm, A. (2022). Stone‐flaking technology at Leang Bulu Bettue, South Sulawesi, Indonesia. Archaeology in Oceania, 57(3), 249–272. https://doi.org/10.1002/arco.5272

van den Bergh, G. D., Li, B., Brumm, A., Grün, R., Yurnaldi, D., Moore, M. W., Kurniawan, I., Setiawan, R., Aziz, F., Roberts, R. G., Suyono, Storey, M., Setiabudi, E., & Morwood, M. J. (2016). Earliest hominin occupation of Sulawesi, Indonesia. Nature, 529(7585), 208–211. https://doi.org/10.1038/nature16448

Neanderthals, Denisovans, and African ancestors of modern people are a deep trident of populations whose tendrils spanned Africa, Europe, and Asia. The Neanderthals and Denisovans began to differentiate around 600,000 years ago from a small ancestral group. That group, the “Neandersovans”, began an estimated 825,000 to 700,000 years ago as one small thread from a tangle of populations. Some larger strands from this tangle lead to later African populations, including ancestors of modern people.

This much we know from genomes. Maybe paleoanthropologists have already found fossils of those ancestral populations. But which fossils could they be?

In a new study, Jean-Jacques Hublin and collaborators highlight the fossil hominins from Grotte à Hominidés, Morocco. Their new work places the fossils’ geological age in the time shortly before 773,000 years ago. The Grotte à Hominidés fossil teeth and jaws share some patterns of shape with humans and Neanderthals that are not seen in Homo erectus. Hublin and coworkers describe the fossils as a “basal lineage” to our species, Homo sapiens.

That place in the hominin tree is the big tangle.

Could the Grotte à Hominidés fossils be a thread that led to modern people or Neandersovans? You can be sure I have an opinion.

Context of the fossils

The Thomas Quarry I (ThI) site in general and the Grotte à Hominidés (ThI-GH) are not new discoveries. This and other sediment deposits are exposed on the wall of a large quarry, which archaeologists have investigated for more than 60 years.

The quarry is within the outskirts of the city of Casablanca, near the present-day Atlantic coastline. The sediments are partially cemented together by calcite, some deposited by wave or water action and some carried by wind. Today the site is around 28 meters above sea level, but sometime in the later Early Pleistocene, when the sea was at a higher level than today, wave action carved a cave into the existing sediment deposits and cliff face. This cavity is the Grotte à Hominidés.

Half a hominin mandible was discovered in 1969, which would draw archaeologists’ attention to this part of the quarry, where they found the cave. By 2011, teams had uncovered many stone tools and ancient animal remains from the ThI-GH sediments. Under the leadership of Jean-Paul Raynal they reported some isolated hominin teeth in 2012, and in 2016, they described a human femur shaft fragment that had been chewed by a large carnivore, likely a hyena.

Other hominin remains recovered in the excavations from before 2011 are reported in the new study by Hublin and coworkers. They include a complete mandible excavated in 2008, ThI-GH-10717, and a series of vertebrae found near it that may represent the same individual. They also include a small section of juvenile mandible and teeth, ThI-GH-10978.

The stone tools within the GH deposit include cores made from quartzite or silicite beach cobbles, flaked bifacially. The flaking techniques and a few handaxes mark the assemblage as Acheulean. There’s little evidence that anyone made stone tools within the cave.

The large assemblage of animal bones appears mostly to have been accumulated by carnivores, with jackal-sized tooth marks especially common. None have cutmarks from hominins.

How did the hominins arrive? There is no occupation floor or significant activity that could be attributed to one group’s lifetimes. At least one individual was ravaged by carnivores, like most of the nonhuman remains. The occasional tool may have been discarded in the cave when people sheltered near the entrance of the cave or used it for shade when foraging closer to the shoreline.

Geological age

Hublin and coworkers report that the fossils come from individuals who lived close to the major Matuyama-Brunhes paleomagnetic reversal, around 773,000 years ago. Their conclusion is based on 119 new samples of magnetic orientation across the sediment layers before, during, and after the Grotte à Hominidés infilling.

An age as early as 773,000 years ago seems to contradict a lot of other work that the geochronologists have done in the site. Electron spin resonance (ESR) combined with uranium-series (US) dating led to estimates for animal teeth from the hominin-containing layers as old as 700,000 and as recent as 400,000 years ago, and ESR-US reported on one of the hominin teeth gave an age between 595,000 and 426,000 years. Those approaches guided earlier interpretation of the site, including the 2016 study that reported the hyena-chewed GH femur, which has “500,000-year-old” in the title. The ESR-US timeline is seemingly reinforced by optically-stimulated luminescence (OSL) dating of sediment samples, which suggested they were between 450,000 and 350,000 years old.

Hublin and colleagues resolve the site’s chronological discrepancies by leaning heavily on paleomagnetism. Because the hominin layers show reversed magnetic polarity, they conclude the site must be older than 773,000 years. They treat the ESR-US dates as minimum ages—citing issues with uranium uptake—and discard the OSL ages altogether.

While this narrows the timeline, the fossils still likely represent populations scattered across thousands of years. Without clear signs of a specific ‘moment’ in time, like a brief cultural occupation, precision is less vital to me than accuracy. I prefer an accurate broad range over a precise range that is potentially incorrect.

Where do these fossils fit?

The new study persuades me that the ThI-GH fossils don’t fit easily into the variation of Homo erectus. Nor do they share most of the traits of recent humans or Neanderthals. If I were trying to imagine a population connecting these other groups, I might imagine something like these.

The teeth carry the most information, just because there are more teeth than other parts, and they show an anatomical mosaic. Each graph with one of the ThI-GH teeth is a different picture. Some show ThI-GH teeth are within the range of recent human form, some are like Homo erectus, in a few cases like Neanderthals, and often off by themselves.

Hublin and coworkers present principal components plots to compare the shapes of the two adult ThI-GH mandibles with other groups. In this particular kind of shape comparison, modern humans and Neanderthals are fairly consistently different—I suspect because their samples are the largest—and the Th1-GH mandibles are near the small area of overlap between these groups. The European Middle Pleistocene mandibles, including early Neanderthals from Sima de los Huesos—also are within that area of shape overlap.

Beyond the multivariate methods, the complete ThI-GH-10717 mandible is a very small one. It’s smaller in size than any H. erectus mandible. Its third molars are particularly small compared both to H. erectus and H. antecessor. Its mandibular corpus height and breadth are smaller than any H. erectus or H. antecessor mandibles. Its small size does fall within the range of modern humans and Neanderthals.

The vertebrae are more erectus-like than modern or Neanderthal, but the aspects they share with Homo erectus are all ancestral traits, not derived similarities that would be evidence of a close relationship.

The hypothesis that connects these results is that the ThI-GH fossils come from a group close to the node that connects all these groups on the tree.

It’s never easy to sort out the details of fossil relationships close to ancestral nodes. Without long branches separating them, such groups did not develop much derived anatomical variation of their own. Instead, genetic drift and selection tend to reapportion variation from ancestral populations, leading to a mosaic.

Bottom line

In my way of thinking, these fossils are Homo sapiens. They fall somewhere within that tangle of groups that ultimately gave rise to Neanderthals, Denisovans, and modern people.

Until now, the hominin fossils that are closest to the same place in the tree of relationships are the fossils from Gran Dolina, Spain, attributed to Homo antecessor. These hominins lived sometime between 949,000 and 773,000 years ago, overlapping with the time Hublin and coworkers propose for the ThI-GH fossils. Both the anatomy and the protein data from Homo antecessor fossils show them to be closer to the Neanderthal-Denisovan-modern branch than to Homo erectus.

The obvious question is whether these fossils from Grotte à Hominidés and Homo antecessor are simply the same thing. I thought the study would answer this question. It really doesn’t. The ThI-GH dental sample differs only in a handful of traits from the Gran Dolina sample. I don’t think the study rejects the hypothesis that these belong to similar or closely related populations.

More likely than not, both the Grotte à Hominidés fossils and Gran Dolina fossils shared ancestors within the few hundreds of thousands of years before they lived. They may well both be part of the tangle.

We’re now a decade or more into a simmering scientific debate about whether the Neanderthal-Denisovan-modern (N-D-M) ancestral population lived in Africa, or whether instead those ancestors lived in Eurasia. It may seem intuitive that Africa is the ancestral location, since African groups already had begun to differentiate at the time the Neandersovan founders lived. What has prompted the debate is Homo antecessor—the closest known outgroup to the N-D-M branch, but ensconced at the extreme western end of Eurasia. The debate heated up a bit last year with the suggestion that a fossil skull from Yunxian, China, might be a close connection of this ancestral group.

Focusing on one or two fossils at a time is the wrong approach. The last part of the Early Pleistocene has dozens of hominin localities, many of them considered “Homo erectus”, which really just means they lack some derived features found in later humans and Neanderthals. Very few have been well dated.

Some of them really were different lineages. Homo naledi, with its known sample much later in the Middle Pleistocene, must have had ancestors, possibly among the African fossils previously attributed to Homo erectus. This thread had already spun its own way before the N-D-M ancestors lived. Some other fossils from a million years ago may be near the ends of their own long threads, stories we don’t know yet.

For the survivors, who became Neandersovans or early African Homo sapiens, the important points of uncertainty are about connections. How much were groups in far-flung places genetically isolated? How frequent was long-distance dispersal of populations? How much did gene flow connect groups over time?

Populations can travel across thousands of kilometers in space more easily than surviving across thousands of years of time. Crossroads of ancient peoples may have existed almost anywhere. Especially Casablanca.

Notes: I praise the authors of this paper and previous work for their open access publications, which have enabled me to share many images of the fossils, stone tools, and contextual data with readers.

References

Daujeard, C., Geraads, D., Gallotti, R., Lefèvre, D., Mohib, A., Raynal, J.-P., & Hublin, J.-J. (2016). Pleistocene Hominins as a Resource for Carnivores: A c. 500,000-Year-Old Human Femur Bearing Tooth-Marks in North Africa (Thomas Quarry I, Morocco). PLOS ONE, 11(4), e0152284. https://doi.org/10.1371/journal.pone.0152284

Hublin, J.-J., Lefèvre, D., Perini, S., Muttoni, G., Skinner, M. M., Bailey, S. E., Freidline, S., Gunz, P., Rué, M., El Graoui, M., Geraads, D., Daujeard, C., Davies, T. W., Kupczik, K., Imbrasas, M. D., Ortiz, A., Falguères, C., Shao, Q., Bahain, J.-J., … Mohib, A. (2026). Early hominins from Morocco basal to the Homo sapiens lineage. Nature. https://doi.org/10.1038/s41586-025-09914-y

Raynal, J.-P., Sbihi-Alaoui, F.-Z., Mohib, A., El Graoui, M., Lefèvre, D., Texier, J.-P., Geraads, D., Hublin, J.-J., Smith, T., Tafforeau, P., Zouak, M., Grün, R., Rhodes, E. J., Eggins, S., Daujeard, C., Fernandes, P., Gallotti, R., Hossini, S., & Queffelec, A. (2010). Hominid Cave at Thomas Quarry I (Casablanca, Morocco): Recent findings and their context. Quaternary International, 223–224, 369–382. https://doi.org/10.1016/j.quaint.2010.03.011

Raynal, J.-P., Sbihi-Alaoui, F.-Z., Mohib, A., El Graoui, M., Lefèvre, D., Texier, J.-P., Geraads, D., Hublin, J.-J., Smith, T., Tafforeau, P., Zouak, M., Grün, R., Rhodes, E. J., Eggins, S., Daujeard, C., Fernandes, P., Gallotti, R., Hossini, S., Schwarcz, H. P., & Queffelec, A. (2011). Contextes et âge des nouveaux restes dentaires humains du Pléistocène moyen de la carrière Thomas I à Casablanca (Maroc). Bulletin de La Société Préhistorique Française, 108(4), 645–669.

Niah National Park is a striking area within the state of Sarawak, part of the nation of Malaysia that is located on the island of Borneo. The park is a UNESCO World Heritage Site, known for its massive limestone cave system—particularly the Great Cave, Gua Niah, home to innumerable small birds known as swiftlets. The birds build their nests on the cave walls from hardened saliva, which people harvest and sell to traders, eventually to become the famous “bird’s nest soup”.

People have used these caves for many thousands of years. Work by archaeologists in the west mouth of the cave starting in the 1950s tells some of the stories of the ancient people. Early in the Holocene, rainforest people buried their dead in Gua Niah, placing them in varied positions, flexed, seated, sometimes mutilating the bodies, and often coupling fire with the burials. Later, between 3500 and 2000 years ago were more than 200 burials in the cave, pottery, stone tools, and metal showing connections with broader cultures across the region.

In another cave, Gua Kain Hitam or the “Painted Cave”, a long mural painted from hematite shows boat-shaped coffins on a journey carrying the souls of the deceased. Burials in that cave include remains of coffins like those pictured, dating to within the last 2000 years.

My interests go quite a bit older.

I’m writing about the Niah Caves today because of new work from Darren Curnoe and collaborators, who may have recovered the first fossil evidence of archaic people from Borneo. This comes from Gua Danang, “Trader’s Cave”, named for the bird’s nest traders who used the cave in recent times.

Gua Danang

Like Gua Niah, this smaller cave was also investigated starting in the 1950s. Tom and Barbara Harrisson, who uncovered so much in the larger cave, found little here. The newer round of excavation work took place from 2017 to 2019, with teams under the leadership of the Australian archaeologist Darren Curnoe. They found more.

The first report of their work is a new article in PLoS ONE, covering the discovery of a single hominin incisor. I don’t often focus on publications that describe a single tooth. But sometimes discoveries of this kind can shed new light on old questions. This is one of them.

Borneo is one of the biggest question marks on the map of human occupation for most of the Pleistocene. The island was connected to the Asian mainland during periods of low sea level, just like its smaller neighbors Java and Sumatra. Java was home to populations of Homo erectus from at least 1.5 million years ago right through the Middle Pleistocene. By contrast, Borneo and Sumatra have hardly any evidence of human occupation across this time period.

That makes each find very interesting.

In the case of the Gua Danang tooth, AA210, its size would be quite big for a modern human and fits into the range of Neanderthals and other archaic people of mainland Eurasia. Only a few upper central incisors are known from Pleistocene sites on Java, and the AA210 tooth measurements fit within their range of variation. All are much smaller than orangutan central incisors, so Curnoe and coworkers rule out that possibility.

The incisor crown is rounded on its front surface—known as “labial convexity”—and it has ridges on both its mesial and distal edges. This shoveled shape, combined with the labial convexity, is more or less what a Neanderthal incisor looks like. Its shape a bit more convex than most of the archaic human teeth from mainland East Asia, although I wouldn’t consider the convexity to be a very strong indication of relationships.

Curnoe and coworkers report OSL ages for the stratigraphic layers in Gua Danang, with the tooth coming from a layer that has an age estimate of 54 ± 5 ka.

The range of 59,000 to 49,000 years is very interesting in light of the finds from nearby Gua Niah. Beneath the early Holocene burials in that cave are much older finds, going back to as early as 46,000 years ago. Until now, these were the earliest hominin finds from Borneo, leading up to the skeletal remains that include the so-called “Deep Skull”.

The Deep Skull

Barbara Harrisson directed excavations in 1957 and 1958 that uncovered a fragmented human cranium. This became known as the Deep Skull from its location nearly 3 meters beneath the present cave floor. A few postcranial bones, including a femur, were also recovered from these excavations. It’s not clear whether these all represent a single individual, although some may, and the femur in particular was recovered close to the Deep Skull.

At the time of their discovery the Deep Skull was thought to be around 40,000 years old. By the beginning of the twenty-first century, this context was revisited in new excavations led by Graeme Barker. A new regime of dating results suggested that the skull comes from between 40,000 and 34,000 years ago. Barker’s work suggests that cultural finds from Niah go back another 5000 or more years earlier than the Deep Skull’s lifetime, back to around 46,000 years ago.

Anthropologists have given this material a great deal of attention for clues about the dispersal of modern humans into the region. The best current idea of early habitation of Australia goes back to 50,000 years or earlier. Modern humans are in evidence at Tam Pa Ling, Laos, by a date of 65,000 years ago. In that context the Niah dates are not so early. But as I’ve written over the last year, the pattern of Neanderthal genetic introgression shared by today’s peoples of the region suggests that their common ancestors came to the area only after 46,000 years ago. Some sites with earlier dates may represent different peoples; some may potentially be in error.

The overwhelming fact is that there are very few human fossils pinning down the pattern of dispersal. Every additional fossil will help. Still, being able to test ideas about these movements will require much more data.

Naturally today everyone loves to see DNA evidence from skeletal remains, me included. Investigation of the Deep Skull for potential radiocarbon dating found no collagen preservation, and that would suggest that DNA preservation is also unlikely.

The morphology of the skull and femur provide some hints about their connections to other populations. These come from a small-bodied person or people, similar in body size and build to some of the island peoples of historic times, such as the Aeta from Luzon. Populations with small body size emerged very quickly in both island and mainland regions of Southeast Asia, from the Andaman Islands to the Philippines.

In their work, Barker and coworkers observe that the Niah people of this time foraged a wide range of rainforest plants, trapped small mammals, fished, and likely practiced deliberate forest burns. These all speak to the kind of effective cultural adaptation to forest life that has been observed in early contexts in mainland Southeast Asia and Sri Lanka.

Bottom line

In the new article on the Gua Danang work, Curnoe and collaborators limit themselves to the description of the hominin incisor and its context. But they mention that the excavations uncovered many stone artifacts that are currently under study. These may help provide a better understanding of the lifeways of the archaic people.

I’ll be very interested to learn whether the Gua Danang evidence shows a similar adaptation to rainforest resources as that suggested by Barker and coworkers for the Deep Skull and its population.

The natural questions that readers will have: Was this Gua Danang tooth Denisovan? Was it the last-surviving Homo erectus? Was it something we haven’t seen yet?

When stone tools are found in a new and unexpected context, we might never make more than an educated guess about who made them. The publication earlier this year of stone tools from Sulawesi more than one million years ago present that challenge. An educated guess is that Sulawesi was likely first inhabited by descendants of Homo erectus similar to those from Java. But until archaeologists find their fossil remains, we won’t know whether these people stayed connected to the larger Sundaland populations, whether they had evolved into something new, or whether they might indeed have come from an unexpected source.

With a hominin fossil, there is at least the chance of something more than a guess. True, the skeletal remains with the Deep Skull may not have preserved DNA. But protein data are another story. Researchers have begun to succeed in protein recovery from teeth of this age and older from East Asia, including the Penghu mandible from the Taiwan Strait. Protein data are much more limited than DNA yet may provide a clue about the relationship of this tooth to other populations.

If I had to bet, I’d guess Denisovan. This site may become one of the keys to understanding the interactions of early modern people with this earlier group.

References

Barker, G., Barton, H., Bird, M., Daly, P., Datan, I., Dykes, A., Farr, L., Gilbertson, D., Harrisson, B., Hunt, C., Higham, T., Kealhofer, L., Krigbaum, J., Lewis, H., McLaren, S., Paz, V., Pike, A., Piper, P., Pyatt, B., … Turney, C. (2007). The ‘human revolution’ in lowland tropical Southeast Asia: The antiquity and behavior of anatomically modern humans at Niah Cave (Sarawak, Borneo). Journal of Human Evolution, 52(3), 243–261. https://doi.org/10.1016/j.jhevol.2006.08.011

Curnoe, D., Datan, I., Goh, H. M., Bin Sauffi, Moh. S., & Ruff, C. B. (2021). Further analyses of the Deep Skull femur from Niah Caves, Malaysia. Journal of Human Evolution, 161, 103089. https://doi.org/10.1016/j.jhevol.2021.103089

Curnoe, D., Datan, I., Taçon, P. S. C., Leh Moi Ung, C., & Sauffi, M. S. (2016). Deep Skull from Niah Cave and the Pleistocene Peopling of Southeast Asia. Frontiers in Ecology and Evolution, 4. https://doi.org/10.3389/fevo.2016.00075

Curnoe, D., Sauffi, M. S., Goh, H. M., Sun, X., & Peiris, R. (2025). A Late Pleistocene archaic human tooth from Gua Dagang (Trader’s Cave), Niah national park, Sarawak (Malaysia). PLOS One, 20(12), e0338786. https://doi.org/10.1371/journal.pone.0338786

Krigbaum, J., & Datan, I. (2005). The Deep Skull and associated human remains from Niah Cave. In J. Majid (Ed.), The Perak Man and Other Prehistoric Skeletons of Malaysia (pp. 131–154). Universitisains Malaysia.

Lloyd-Smith, L. (2014). Early Holocene burial practice at Niah Cave, Sarawak. Bulletin of the Indo-Pacific Prehistory Association, 32(0), 54–69. https://doi.org/10.7152/jipa.v32i0.12844

It was the year of the Denisovan. We learned at last what they looked like, detected one that was dredged from the seafloor, and saw a new tooth become the oldest-ever high-coverage genome.

In this field of study, what seems outside the realm of technology today can come surprisingly fast into focus. That’s why it’s always fun to look back at the year’s research and point to what’s new.

One trope of ancient DNA research is the “previously-unknown lineage”, and as usual there were some new ones this year. On the list I’ve included two at opposite ends of the great dispersal of humans from Africa.

Meanwhile, new ways of looking at data from genomes often help illuminate new questions. I’ve thrown in a couple of stories about function, one focused on the expression of genes that may differ in 3D organization across populations, another on minor components of diet that may have major effects.

Finally, the ethical landscape of genetic research has continued to mature. As I reflected in last year’s review, the best practices for research on human remains include leadership and consultation from descendants of the groups being studied. A valuable example of that kind of research from the American Southwest has a place on this year’s list.

Recognizing the Longren

I don’t see how anybody is supposed to compete for top ancient DNA stories this year with Qiaomei Fu.

The biggest mystery in paleoanthropology was set up 15 years ago with the discovery that an 80,000-year-old finger bone from Denisova Cave, Russia wasn’t who anyone thought it would be. In June, Fu’s research team finally put a face to the enigmatic group known as Denisovans, connecting the mitochondrial genome and proteome of the 146,000+ year old Harbin skull to the group.

The Harbin skull was named Homo longi by Qiang Ji and coworkers after its discovery, and has commonly been called Longren, or “Dragon Man” by the press and other researchers. The skull looks a lot like some other Middle Pleistocene fossils from China, including the Dali skull, which means one Denisovan mystery is solved.

These were not the only interesting Denisova-related stories this year. In April Takumi Tsutaya and colleagues reported on proteomic data from a jaw recovered from the seafloor of the Penghu Strait, showing it to have been a male individual within the broader Denisovan population.

Oldest high-coverage genome

Tripling down on Denisovans has not usually been my approach in these year-end lists. But in October, Stéphane Peyrégne and coworkers released a high-coverage genome from the molar tooth designated as Denisova 25, roughly 200,000 years old. It’s the oldest high-coverage genome yet reported.

Adding even one early data point like this makes a surprisingly big difference to some findings. For example, a model including the new genome suggests that African ancestors of modern people diverged from Neanderthal-Denisovan ancestors between 825,000 and 694,000 years ago—maybe 50,000 to 100,000 years earlier than the previous estimate.

Meanwhile, Denisova 3 and the hybrid Denisova 16 both already showed that mixture between these groups and Neanderthals was happening a lot. Denisova 25 adds a deeper time dimension to this mixture, from even earlier Neanderthal groups. The genome also confirms the ancient passage of “superarchaic” ancestry into the early Denisovans.

There’s a lot in this analysis to add detail to the mixture of Denisovans and modern people, including the observation that estimated admixture fraction goes up when the second Denisovan genome is added to analyses. There is a lot on the functional variation of Denisovan genomes, also. This is a study that I’ll be returning to over the next year.

Debunking Pompeii fantasies

Pompeii is possibly the world’s most famous archaeological site. It is certainly one of the oldest to be visited by the public. In 1863 Giuseppe Fiorelli, who was directing the excavations, recognized that hollow cavities in the hardened volcanic ash were negative molds of human bodies. He poured in plaster of Paris, waited for it to harden, and continued the excavation, exposing the plaster casts of the ancient people.

Enabling visitors to see the victims of Vesuvius at the times of their deaths was an immediate sensation. Today the plaster casts number more than 100 across the site. Over the years, stories arose about some of the figures. For instance, in the House of the Golden Bracelet, an adult wearing a golden bracelet with a child on their lap was long suspected to be a mother and her child. Other individuals found in the same house were assumed to be part of the family group.

Last year, Elena Pilli and coworkers carried out genetic analysis of bone from 14 out of 86 plaster casts that were undergoing restoration. Five preserved DNA sufficient for examining ancestry and relatedness.

Even with just this handful of individuals, the results showed that some longstanding stories were fiction. The “mother” with the golden bracelet was actually male; the child was unrelated. None of four people in the house were genetically related.

Two individuals found in the House of the Cryptoporticus ended their lives one touching the other, long interpreted as sisters in an embrace. But DNA from one of the individuals shows that he was male.

Pilli and coworkers end their study noting that narratives from such limited data are unreliable. One problem has been projecting modern assumptions about gendered behaviors and kinship onto ancient groups, where different cultural expectations were at play. Another is the temptation by previous generations of researchers to manipulate poses or otherwise emphasize some aspects of the evidence.

Marriages of the Avars

One of the most valuable uses of ancient DNA is to help understand how culture and genealogy are related. When people move en masse from one place to another, carrying their culture with them, the appearance of new culture and new genealogical lineages may be easily apparent. But it has been more challenging to see the genealogical effects of trade and interchange between established population centers. That kind of analysis has become possible with evidence of genealogical relatives from distant sites.

A well-designed study in January by Ke Wang and collaborators looked at two cemeteries from the late Avar period near Vienna, dating to the seventh century or eighth century CE. The Avars originated as steppe peoples before settling in central Europe north of East Roman provinces in the Balkans. Wang and coworkers sampled almost every skeleton from the Late Avar period from both cemeteries: Leobersdorf with 155 individuals and Mödling-An der Goldenen Stiege with 485 individuals. They added 83 from the nearby Csokorgasse cemetery, working to construct genealogies connecting individuals with each other.

All of these cemeteries lie within 20 km of each other and represent culturally similar peoples who lived at the same time, with the same burial practices and artifacts reflecting social status. So it is surprising that there is little genetic similarity between the Leobersdorf people, who derive most of their ancestry from eastern Eurasia, and the Mödling burials, which look like other nearby central Europeans.

Especially striking are the many instances of close genealogical relatives that Wang and coworkers could identify between Leobersdorf people and other cemetery sites hundreds of kilometers to the east, on the Hungarian plain. For Mödling, there were hardly any—instead, there were quite a few genealogical relatives in the nearby Csokorgasse cemetery.

As the authors note, the Leobersdorf people of this period seem to have sought out specific groups for marriages, focusing on the Avar heartland. For the Mödling people, the story was different. Two groups with such different patterns of intermarriage, yet within a common cultural setting: that’s something that we see today in many places, but can be almost invisible in the past. This is a remarkable case study.

Green Sahara

The Sahara Desert today covers more than seven million square kilometers, almost the land area of Australia. Much of the desert is inhabited very sparsely by groups of herders such as the Bedouin and Tuareg, with a few large population centers at oases where water can be found.

Between 15,000 and 5,000 years ago things were different. The central Sahara had shallow lakes and streams, well-watered grassland and brushland. This time is known as the African Humid Period.

This year, research from Nada Salem and collaborators drew attention to the people who flourished in this green Sahara. They obtained genomes from two individuals excavated from the Takarkori rock shelter in extreme southwest Libya, with radiocarbon ages between roughly 7000 and 6000 years ago. There, in the in the Tadrart Acacus Mountains, people herded cattle, sheep and goats. The running water in the nearby Wadi Takarkori enabled the people to continue a mixed economy with hunting and fishing. Abundant rock art depicts a vibrant society.

Salem and coworkers found that a deep, previously-unrecognized lineage makes up most of the ancestry of the Takarkori individuals. This deep ancestry shares a common stem with the founder group that dispersed from Africa into Eurasia more than 50,000 years ago. Identifying this deep lineage helped the researchers reinterpret findings from the 15,000-year-old genomes of people from Taforalt, Morocco. These people, representing the Iberomaurusian culture of hunter-gatherers, shared a connection with the Takarkori genomes, suggesting how widespread this lineage was through the later Pleistocene.

The research helps to provide context for the introduction of pastoral ways of life into North Africa. The idea of herding animals may have come from southwest Asia with the domesticated breeds themselves, but gene flow from earlier Levantine peoples did not make a major impact at Takarkori.

Finding unexpected genetic lineages has been a common occurrence in the ancient DNA era. For me, this one is especially interesting because it helps ground the out-of-Africa founders in Northeast Africa. Some archaeologists have long held that cycles of population growth and dispersal in a green Sahara were important to the eventual dispersals of modern humans into Eurasia. The Takarkori findings show that the greening of the desert did not pull much gene flow either from the south nor back into Africa from the Levant. The staging area for the dispersal may have been the Sahara itself.

Selection for micronutrients

One of the most important factors in recent human evolution has been diet. Agricultural processes have shifted nutrient availability, so that calories come much more from high-carbohydrate grains, alcohol, and milk, for example. Less obvious are shifts in minerals like selenium and zinc, ultimately determined by their abundance in soils and local plants. But such micronutrients have impacts on health, and some human genes make a difference to how people’s bodies take in and process them.

In October, Jasmin Rees, Sergi Castellano, and Aida Andrés published a synthesis of how micronutrients have mattered to natural selection around the world. They identify several cases of local adaptation, some occurring in parallel in different groups.

For example, both Maya people and Mbuti hunter-gatherers live in iodine-deficient environments, and both have candidate gene sets that may reflect adaptation to low iodine. In East Asia, populations like the Yakut and Japanese show significant evidence of adaptation to selenium-deficient soils.

Many of these signals are made up of variants from several genes, a pattern of change called oligogenic adaptation—meaning that anywhere from three to ten genes may act together with fairly large effects on the trait. This evolutionary pattern may be especially relevant for humans moving into new regions and encountering new environments—seen in other contexts with pigmentation, adaptation to high altitude, and immunity.

Ancient genomes in 3D

People living today have some clear differences in skeletal and dental form from Neanderthals and Denisovans. Tracking down the genetic correlates of those phenotypic differences has been a challenge. An early focus of research was amino-acid-coding differences, but there are surprisingly few of these between living people and the archaic genomes. Many researchers have pointed to the idea that it’s not protein-coding variation that drove the evolution of modern people, it’s how genes are regulated—focusing on regulatory sites and methylation as ways of understanding gene expression.

A growing area of research is the three-dimensional configuration of the genome. Genes are regulated in part by the physical proximity of sequence motifs after DNA-containing chromatin has been folded and looped into 3D shapes. In a preprint updated in September, Evonne McArthur and coworkers report on advances in understanding 3D genome organization in Neanderthal and Denisovan genomes.

What they find is fascinating. They identify between 60 and 75 parts of the genome in archaic human data where the archaic sequence exhibits high 3D genome organization divergence from all modern samples. That’s not a lot. Still, a pass across these areas in gene function datasets provides some interesting hits: Eye function, brain, skin and hair, and some skeletal traits.

The 3D genome organization may be a key to understanding some of the introgression of Neanderthal genes into recent people. McArthur and colleagues show that areas of the genome with high 3D variability in today’s people are also more likely to have introgressed variants, while areas with large differences in 3D organization between Neanderthal and modern genomes are less likely to have introgression. They also show some places where introgression has brought the pattern of 3D genome organization of some Eurasian individuals into a Neanderthal-like configuration.

Filling in the southern cone

The last decade has seen some large-scale syntheses of the dispersal of humans into the Americas. At a smaller scale, researchers have been investigating DNA records of particular regions. This year, Javier Maravell-López and collaborators reported in more than 200 ancient genomes from the Paraná River to the Andes Mountains, and across twenty degrees of latitude, from northern Paraguay to Bahía Blanca.

The people of this region developed a distinctive genetic lineage very soon after first arriving. An ancient genome from around 10,000 years ago already has a pattern of genetic drift distinguishing the individual from other groups around South America, and shared with later people from the same region.

That identity remained strong over time as groups of this area mixed with nearby regions. At the northwest extreme of the study area, ancient people mixed with populations of the central Andes, including Tiwanaku and related groups. In the south, they encountered the descendants of another fast-diverging group, expanding into their range in the Pampas region.

It’s easy to underestimate the vast land spaces of South America. Headlines this year flashed that an unknown lineage remained isolated for 8000 years. That’s not actually what the study says. What happened was fast growth followed by diffusion of people into and out of the region. Understanding that pattern may take us a long way toward the ancestral population structure across both Eurasia and the Americas.

Ancestral Puebloans

Many Indigenous people of the U.S. are deeply interested in what genetics can reveal about genealogy and history, but are deservedly wary. Archaeologists built up a bad track record across almost two centuries, often ignoring Indigenous histories, digging into sacred sites, and hauling ancestors away to museums. But across the last twenty years, many researchers who focus on ancient DNA have built stronger relationships with tribes. These relationships have led to better science and a stronger trust for research into the past.

This year provides a great example of research co-developed with Indigenous people. Eske Willerslev’s Copenhagen-based research team worked together with people from Picuris Pueblo, New Mexico, to focus on the ancestry of today’s community members and their connections to Ancestral Puebloans. A crucial aspect of the research involved a commitment to ensure data sovereignty for Picuris Pueblo throughout the collaboration.

By sampling both living members and ancient individuals from the last millennium, the study provides a strong picture of a large, diverse group prior to European presence in the region. The data also show a strong connection between Picuris people and Ancestral Puebloan people who lived in Chaco Canyon, around 250 km to the west. This part of the study reinforces oral histories that have been held by Picuris people and by other groups.

The work may also show how community-led research can help repair relationships that were broken by earlier researchers, including both archaeologists and geneticists. Work on DNA from Chaco Canyon published in 2017 became one of several high-profile instances where researchers pushed into ancient DNA sequencing without any consultation with descendant communities.

It is widely recognized within anthropology and human genetics that working with potential descendants leads to better, more accurate scientific results. That outcome can be seen in the new Picuris study, where the participation of the community has added greatly to understanding of the genetic results.

The other leprosy

One of my favorite stories of the year concerns an infectious agent of Hansen’s disease, also known as leprosy. Until 2008, scientists thought that this global malady was always caused by the bacterium Mycobacterium leprae. Then a second pathogen was identified: Mycobacterium lepromatosis. It was first seen in the Americas, and later a few human cases were noted in Eurasia. Then, researchers found it in some of the lesions affecting red squirrels in Britain.

Was it an emerging pathogen, or a neglected one? In May, a massive study led by Maria Lopopolo reported on large-scale screening of bacterial genomes from more than 400 living patients, and 389 ancient genomes across the Americas.

They found M. lepromatosis in 34 living cases, mostly in the U.S. and Mexico. They also found three ancient genomes with the bacterium: one from Canada and two from Argentina, all from the period before European contact. This shows the presence of M. lepromatosis infection was widespread in the ancient peoples of the Americas.

To me the most fascinating element of the study is the estimation of the ancestral divergence between M. lepromatosis and M. leprae, with two models pointing to different dates: around 795,000 to 660,000 years ago, or alternatively between 2.5 and 1.5 million years ago. Either of these sets of dates would imply a passage of the ancestors of one clade from Eurasia to the Americas, or vice-versa.

Red squirrel cases of M. lepromatosis, possibly acquired after gray squirrels were introduced from North America into Britain, may point to the possibility that a nonhuman mammal host carried the pathogen across Beringia, where ancient Americans encountered it. But the obvious species to have dispersed M. lepromatosis in ancient times is humanity itself. It’s interesting to consider whether ancient humans may have contracted Hansen’s disease from earlier contacts with Denisovans or other hominins.

A look forward

Large-scale research teams have filled out many regions of the world with ancient genomes. Africa and southern Asia remain obvious outliers, where studies of a single genome can still make a big difference. Among the stories I didn’t include from this year: a study based on a single genome from Old Kingdom Egypt, as well as a study of a couple dozen ancient genomes from South Africa. I expect in the next few years that advancing technical methods will unleash new datasets from these regions.

Meanwhile, in Europe, China, and parts of the Americas, the state of the art has moved toward investigating cultural and social practices with genealogical genomics. What matters to these questions is not only regional ancestry but the pattern of genealogical connections between people. The studies of the Avar and Pompeii remains are examples of this kind of research from 2025. Investigation of more broad-scale aspects of kinship were also important in the Argentina ancient DNA study

Last year I pointed to the advancing work on ancient pathogens, which has continued to accelerate. This was a year of consolidation of some old results—like the many studies of bubonic plague from ancient sites. Researchers have also published a number of meta-analyses and reviews, starting to bring together the big epidemiological picture. The area is ripe for a new synthesis.

It’s paleoenvironmental analysis with DNA where I’m expecting some big innovation in the next year. Sediment DNA has extended the knowledge that can be brought from animal, plant, and microbial remains. A massive study of subsistence at Denisova Cave this year, a study of human-carnivore interactions from El Mirón Cave in Spain, and a look at ancient wolf, bison, and human DNA from Satsurblia Cave, Georgia, are all recent examples of this kind of work.

What will be next? We’ll see in 2026!

Notes: Several articles have looked closely at the genetic research on Pueblo Bonito individuals and their implications for tribal consultation and repatriation. I can suggest the works I’ve cited below led by Katrina Claw, and by Amanda Daniela Cortez.

There were some amazing DNA studies that I wrote about during the year and didn’t repeat for this post. Think of them as a bonus! A standout was the study of the MUC19 gene by Fernando Villanea and coworkers. Another excellent study was the work on Papuan adaptations with a focus on immunity. And there was the interesting study of lactase persistence genetics in China, with a Neanderthal haplotype implicated.

References

Fu, Q., Bai, F., Rao, H., Chen, S., Ji, Y., Liu, A., Bennett, E. A., Liu, F., & Ji, Q. (2025a). The proteome of the late Middle Pleistocene Harbin individual. Science, 0(0), eadu9677. https://doi.org/10.1126/science.adu9677

Fu, Q., Cao, P., Dai, Q., Bennett, E. A., Feng, X., Yang, M. A., Ping, W., Pääbo, S., & Ji, Q. (2025b). Denisovan mitochondrial DNA from dental calculus of the >146,000-year-old Harbin cranium. Cell, 0(0). https://doi.org/10.1016/j.cell.2025.05.040

Claw, K. G., Lippert, D., Bardill, J., Cordova, A., Fox, K., Yracheta, J. M., Bader, A. C., Bolnick, D. A., Malhi, R. S., TallBear, K., & Garrison, N. A. (2017). Chaco Canyon Dig Unearths Ethical Concerns. Human Biology, 89(3), 177. https://doi.org/10.13110/humanbiology.89.3.01

Cortez, A. D., Bolnick, D. A., Nicholas, G., Bardill, J., & Colwell, C. (2021). An ethical crisis in ancient DNA research: Insights from the Chaco Canyon controversy as a case study. Journal of Social Archaeology, 21(2), 157–178. https://doi.org/10.1177/1469605321991600

Lopopolo, M., Avanzi, C., Duchene, S., Luisi, P., De Flamingh, A., Ponce-Soto, G. Y., Tressieres, G., Neumeyer, S., Lemoine, F., Nelson, E. A., Iraeta-Orbegozo, M., Cybulski, J. S., Mitchell, J., Marks, V. T., Adams, L. B., Lindo, J., DeGiorgio, M., Ortiz, N., Wiens, C., … Rascovan, N. (2025). Pre-European contact leprosy in the Americas and its current persistence. Science, 389(6758), eadu7144. https://doi.org/10.1126/science.adu7144

McArthur, E., Rinker, D. C., Cheng, Y., Wang, Q., Wang, J., Gilbertson, E. N., Fudenberg, G., Pittman, M., Keough, K., Yue, F., Pollard, K. S., & Capra, J. A. (2025). Reconstructing the 3D genome organization of Neanderthals reveals that chromatin folding shaped phenotypic and sequence divergence. bioRxiv. https://doi.org/10.1101/2022.02.07.479462

Peyrégne, S., Massilani, D., Swiel, Y., Boyle, M. J., Iasi, L. N. M., Sümer, A. P., Mesa, A. B., De Filippo, C., Viola, B., Essel, E., Nagel, S., Richter, J., Weihmann, A., Schellbach, B., Zeberg, H., Visagie, J., Kozlikin, M. B., Shunkov, M. V., Derevianko, A. P., … Kelso, J. (2025). A high-coverage genome from a 200,000-year-old Denisovan. bioRxiv. https://doi.org/10.1101/2025.10.20.683404

Pilli, E., Vai, S., Moses, V. C., Morelli, S., Lari, M., Modi, A., Diroma, M. A., Amoretti, V., Zuchtriegel, G., Osanna, M., Kennett, D. J., George, R. J., Krigbaum, J., Rohland, N., Mallick, S., Caramelli, D., Reich, D., & Mittnik, A. (2024). Ancient DNA challenges prevailing interpretations of the Pompeii plaster casts. Current Biology, 34(22), 5307-5318.e7. https://doi.org/10.1016/j.cub.2024.10.007

Pinotti, T., Adler, M. A., Mermejo, R., Bitz-Thorsen, J., McColl, H., Scorrano, G., Feizabadifarahani, M., Gandy, D., Boulanger, M., Gaunitz, C., Stenderup, J., Ramsøe, A., Korneliussen, T., Demeter, F., Santos, F. R., Vinner, L., Sikora, M., Meltzer, D. J., Moreno-Mayar, J. V., … Willerslev, E. (2025). Picuris Pueblo oral history and genomics reveal continuity in US Southwest. Nature, 642(8066), 125–132. https://doi.org/10.1038/s41586-025-08791-9

Ramirez, D. A., Sitter, T. L., Översti, S., Herrera-Soto, M. J., Pastor, N., Fontana-Silva, O. E., Kirkpatrick, C. L., Castelleti-Dellepiane, J., Nores, R., & Bos, K. I. (2025). 4,000-year-old Mycobacterium lepromatosis genomes from Chile reveal long establishment of Hansen’s disease in the Americas. Nature Ecology & Evolution, 9(9), 1685–1693. https://doi.org/10.1038/s41559-025-02771-y

Rees, J., Castellano, S., & Andrés, A. M. (2025). Global impact of micronutrients in modern human evolution. The American Journal of Human Genetics, 112(10), 2538–2561. https://doi.org/10.1016/j.ajhg.2025.08.005

Salem, N., van de Loosdrecht, M. S., Sümer, A. P., Vai, S., Hübner, A., Peter, B., Bianco, R. A., Lari, M., Modi, A., Al-Faloos, M. F. M., Turjman, M., Bouzouggar, A., Tafuri, M. A., Manzi, G., Rotunno, R., Prüfer, K., Ringbauer, H., Caramelli, D., di Lernia, S., & Krause, J. (2025). Ancient DNA from the Green Sahara reveals ancestral North African lineage. Nature, 641(8061), 144–150. https://doi.org/10.1038/s41586-025-08793-7

Tsutaya, T., Sawafuji, R., Taurozzi, A. J., Fagernäs, Z., Patramanis, I., Troché, G., Mackie, M., Gakuhari, T., Oota, H., Tsai, C.-H., Olsen, J. V., Kaifu, Y., Chang, C.-H., Cappellini, E., & Welker, F. (2025). A male Denisovan mandible from Pleistocene Taiwan. Science, 388(6743), 176–180. https://doi.org/10.1126/science.ads3888

Wang, K., Tobias, B., Pany-Kucera, D., Berner, M., Eggers, S., Gnecchi-Ruscone, G. A., Zlámalová, D., Gretzinger, J., Ingrová, P., Rohrlach, A. B., Tuke, J., Traverso, L., Klostermann, P., Koger, R., Friedrich, R., Wiltschke-Schrotta, K., Kirchengast, S., Liccardo, S., Wabnitz, S., … Hofmanová, Z. (2025). Ancient DNA reveals reproductive barrier despite shared Avar-period culture. Nature, 638(8052), 1007–1014. https://doi.org/10.1038/s41586-024-08418-5

When I was a kid, I spent an unusual amount of time making fires with flint and steel. We had races to see who could ignite a fire the fastest, building it up to boil water first, or to burn high enough to sever a string.

A lot of people have misconceptions about firemaking. Some of those ideas come from watching competition reality shows like Naked and Afraid or Survivor. Producers often supply contestants with a “flint” that works magically well. This is actually a commercial firelighting device for survivalists, using magnesium shavings that ignite readily at a spark and burn at 3000°C. In other words, they cheat.

Neanderthals didn’t cheat. They did sometimes strike fires from sparks. They didn’t have steel but they did have iron pyrite for sparking.

I’m prompted to write about firemaking this week because Nature has published some new research from the Barnham site, northeast of Cambridge, England. Rob Davis and collaborators report on evidence of pyrite firemaking in a 400,000-year-old site. There are no hominin skeletal remains at the site, but based on the date the evidence likely was left by early Neanderthals.

This research is getting some attention as the “earliest evidence of ignition”. Some headlines say that the study “pushes back” the evidence for firemaking by a quarter million years. I think the findings are indeed very cool, but I don’t interpret them in this way at all.

Smoking guns are rare at murder scenes

Archaeologists have found extensive evidence for hominin management of fire from many earlier sites, some a million years and more earlier than the Barnham site. In the earliest phase, before around 800,000 years ago, there are a half-dozen sites in Africa and the Levant with fire evidence. After 800,000 years ago, sites become widespread in Africa, Europe, the Levant, and East Asia.

Just before the Early-Middle Pleistocene transition is the site of Gesher Benot Ya’aqov, Israel, around 790,000 years old. This site on the bank of the Jordan River preserves an intricate record of repeated hominin-controlled firebuilding, with charred remains of food plants and fish.

The Gesher Benot Ya’aqov site provides an unparalleled record of fire as a cultural reality for the hominins. As at the Barnham site, the lack of hominin skeletal remains leaves us unsure which hominin group or population made the fires. But their lives revolved around fires in many ways: they used botanical knowledge to choose wood, they worked to gather and transport wood, they brought foodstuffs together for cooking, and their world of smell and taste was one of cooked foods and smoke. They structured their activities around the fires, very much like recent foraging peoples.

Despite the very strong evidence for control and manipulation of fire at Gesher Benot Ya’aqov, there is no evidence to suggest how the fires were first started. No smoking gun.

Some archaeologists have emphasized ignition of fire as a major advance that required greater cognitive planning than controlling fires. But this ignores the logistics of fire as a functional part of a society’s subsistence economy.

Consider what is needed to have fire function for cooking, light, or heat. You have to manage the logistics of fuel for the fire. That means transporting wood to the exact location where you want to burn it, at least enough to keep the fire going long enough to go out for more. You have to have familiarity of which kinds of wood are suitable for burning, distinguishing fuel from both green and rotted wood, usually focusing on tree species that burn well and without noxious smoke. You have to manage the gathering of dry tinder and kindling, and you need an understanding of how to use these materials in sequence to allow the fire to start small—from a single coal—and grow.

Not all these have to happen perfectly, of course. The process tolerates much variation. But if we have evidence that an ancient society was shaping its activities around fire, then fire was certainly reliable. It could not be reliable if the logistics went wrong very often.

In the context of human fire use, being able to ignite a fire from sparks or friction is a failsafe. Human societies that rely daily on fire for heating and cooking mostly build them from hot coals that are retained from earlier fires. When people want to build a fire somewhere new, they can transport smoldering coals safely in pouches or containers that are lined with insulating materials.

People who rely on fire don’t strike a new fire every time they want one. Most fire is conserved—curated. Rekindling a fire from a smoldering coal is a bit easier than starting from friction or from a spark, but the steps involved are very much alike.

So why hasn’t there been more evidence for how fires were started? In paleoanthropology, the “first” evidence of a behavior often is more indicative of archaeological visibility than the behavior’s evolutionary breadth and origin. Evidence for fire is unlikely to preserve, evidence for friction or sparking as a way of starting fires is even less likely to preserve.

Even minerals are perishable

The new study by Davis and coworkers is a valuable example of just how unlikely it is for archaeologists to find the “smoking gun” for firestarting. At the Barnham site, the evidence for knowledge of firestarting is two small chunks of pyrite. One was found in the same level as the fire-reddened sediment interpreted as a fire feature. The other was a surface find.

There are some important details that most reporting has missed. First, the ancient pyrite at the Barnham site is not literally pyrite. Instead, these are “pyrite pseudomorphs”—originally chunks of pyrite (iron sulfide, FeS₂) that were oxidized into goethite (FeOOH) and hematite (Fe₂O₃), presumably after they were buried. The reason Davis and coworkers surmise that these small chunks were once pyrite has to do with their current shape, which retains the textural appearance of crystalline pyrite.

The pyrite itself is gone. In other burial contexts, the retention of a pyrite crystalline appearance after oxidation would be unusual, or would be unrecognized in smaller fragments. In archaeology, it’s common to recognize that organic materials undergo decay and loss over time. But even rocks are affected by diagenetic processes and may be transformed and disappear.

I’ll add that even presence of pyrite fragments does not show that this particular fire was started by hominins. Instead, this study assumes that because the pyrite pieces were near the fire, that the hominins were likely familiar with the flint-pyrite method of fire starting.

Their interpretation draws upon earlier work, especially by Andrew Sornesen, who over the last decade has advanced the scientific understanding of the flint-pyrite method of firestarting. Writing with Marie Soressi, Sorensen has documented intense scratches in small patches on the site of flaked flint artifacts, interpreted as the result of sparking pyrite off the flint.

The later sites with this kind of evidence have mostly not preserved any pieces of pyrite; the Barnham site has preserved no flint artifacts with such traces. Only by putting the pattern together across these hundreds of thousands of years does the story seem to appear.

Fire evidence is growing

I don’t want to leave the impression that I have a negative attitude toward this new work. I think it’s ingenious and I’m especially gratified that the archaeologists were careful enough to notice and preserve these small pieces of pseudopyrite.

What I find a little frustrating is the relentless emphasis on the “first” evidence of starting fires, as if that had been in doubt.

Now to be clear, there remain some archaeologists who take a highly skeptical attitude toward hominin firemaking capabilities. When I was a student, there were some who argued—vociferously—that Neanderthals had no ability to start fires. Those days are thankfully past. Still, even today there are some who suggest that hominins may have controlled fire for hundreds of thousands of years but only by curating fires that were started by lightning.

Poppycock. That “quest for fire” nonsense is a fictional way of looking at the past.

The record of ancient fire that archaeologists gathered across the last century is weak in some important ways, and many specialists have been working to improve the record. Where color changes in sediment were once accepted as compelling evidence for fires, today researchers can apply FTIR, environmental magnetism, and microscopic analysis of sediment to add precision to the data. These approaches have indeed revealed a few cases where the original interpretation of fire is no longer supported. More often, they have helped solidify the evidence or brought attention to cases that were missed. A fascinating paper from last year by Ángela Herrejón-Lagunilla and coworkers uses magnetic evidence to show the timeline of a series of fires made in the same site, giving an unparalleled evidence about the duration of traditions of fire within a Neanderthal society.

Fire can be hard to find, and evidence for starting fires even harder. But archaeologists are better and better at meeting this challenge. As they have done so, they have uncovered a depth of relationship between fire and human ancestors that mattered to our evolutionary pathway. Ancient hominins cooked foods to release energy, to strip toxins from otherwise-inedible plants, or simply to taste better. They systematically burned some landscapes to attract game. Those behaviors started much earlier and were adopted much more broadly than we once thought.

I’ll end with a fascinating anecdote from Andrew Sorensen’s work. Chunks of a black mineral called manganese oxide have been found in many Neanderthal sites, some with clear friction or rub marks on their surfaces. They have usually been interpreted as pigment, some with evidence of rubbing on skin or hide.

But Sorensen realized that manganese oxide has interesting properties when burned. Sparking pyrite into a small pile of this powder would create a very hot fire—not so different from the survivalist tools on reality TV.

In other words, maybe even Neanderthals cheated sometimes.

References

Davis, R., Hatch, M., Hoare, S., Lewis, S. G., Lucas, C., Parfitt, S. A., Bello, S. M., Lewis, M., Mansfield, J., Najorka, J., O’Connor, S., Peglar, S., Sorensen, A., Stringer, C., & Ashton, N. (2025). Earliest evidence of making fire. Nature, 1–7. https://doi.org/10.1038/s41586-025-09855-6

Herrejón-Lagunilla, Á., Villalaín, J. J., Pavón-Carrasco, F. J., Serrano Sánchez-Bravo, M., Sossa-Ríos, S., Mayor, A., Galván, B., Hernández, C. M., Mallol, C., & Carrancho, Á. (2024). The time between Palaeolithic hearths. Nature, 630(8017), 666–670. https://doi.org/10.1038/s41586-024-07467-0

Sorensen, A. C. (2024). Lucky strike: Testing the utility of manganese dioxide powder in Neandertal percussive fire making. Archaeological and Anthropological Sciences, 16(8), 134. https://doi.org/10.1007/s12520-024-02047-9

Sorensen, A. C., Claud, E., & Soressi, M. (2018). Neandertal fire-making technology inferred from microwear analysis. Scientific Reports, 8(1), 10065. https://doi.org/10.1038/s41598-018-28342-9

I am really pleased to see that my first article from 2026 has been released in the Journal of Archaeological Science. The article is a synthesis bringing together ideas from cognitive science and social behavior, and applying those to better understand the Pleistocene archaeological record. The article was led by Agustín Fuentes, together with Jennifer French, Marc Kissel, and Penny Spikins—all amazing collaborators whose work is helping to enable a clearer view of hominin behavior in the deep past.

The article is open access, and I hope everyone feels free to check it out. Here I’ll share my thoughts on some of the main ideas. What I’ll write isn’t necessarily the view of all my coauthors, but I think we’re in pretty broad agreement on these aspects of the work.

Emotional is rational

First, a quick setting of the scene. Natural selection is a calculus of survival and reproduction. Any individual’s time and energy are limited, and following one course of behavior may foreclose the possibility of others.

Behavior thus often involves trade-offs: Choosing to forage in more open habitats may improve food quality but increase the risk of predation. By delaying weaning, a mother may increase her infant’s chance of survival but may prolong the time until her next child can be born.

Humans today have faster spacing between births than any of our close living relatives. Chimpanzee and bonobo mothers tend to have an interval of five years between births, gorillas more than four, orangutans up to eight. For human mothers in natural-fertility societies the spacing between births is closer to three years.

This is possible because human mothers are supported in childrearing by other people, a practice known as allocare. Many anthropologists and behavioral ecologists have thought about allocare and what it means for the behavior of living people and our ancient relatives:

• Sarah Blaffer Hrdy has long been a thought leader in this area, first with her analysis of maternal support networks, and in her more recent book Father Time focusing on fathers and paternal care.

• Kristen Hawkes has written about the role of grandmothers in allocare, with a focus on the trade-off for older mothers between investing in another—possibly risky—pregnancy and investing in the growing families of their existing daughters.

• The late Frank Marlowe wrote about provisioning as an important aim of hunting in social groups of hunter-gatherers, as well as the role of men in protection.

For ants, bees, and other social insects, allocare is a matter of chemical signaling, including chemical controls on gene regulation that drive the development of workers. Humans are built different. We rely on the signaling and cognitive systems used by other primates for regulation of social relationships, and these center around emotion: anger, fear, surprise, sadness, joy, trust, disgust. A wider participation by other people in allocare required changes in emotional cognition: a greater ability to control or regulate emotions in some contexts, as well as changes to the interpretation or meanings of facial, vocal, and other signals of emotions.

Emotional changes overlap with some human-specific cognitive innovations that have been noted by Michael Tomasello, Richard Wrangham, and many others: Humans tolerate the proximity of other individuals, including strangers, vastly more than any other primates do, and humans can direct other individuals’ attention in ways that other primates rarely do. Such abilities rely upon emotional regulation, and they underlie most human collaboration and cultural learning.

This is all mainstream social psychology and behavioral ecology. What we observe in our new paper is that this cognitive background matters to the archaeological evidence left by Pleistocene hominins.

This idea bucks some long-held conceptions of “rationality” in Paleolithic archaeology. Actualistic research in archaeology—the kind that examines living hunter-gatherers as a model for past ways of life—has for decades focused upon measuring of the costs and yields of decisions made by foraging peoples: Measurements of caloric cost of locomotion, measurements of caloric returns of hunted animals and gathered plant foods, measurements of the distances and energy expenditure of raw material transport, measurements of the energetic and nutritional content of cooked food versus raw, of food supplementation for infants versus breast milk, of the energy yield of high-ranked and low-risk prey animals versus lower-ranked, higher-risk prey animals. Researchers may assume that ancient people strived to minimize costs and maximize returns in their decision-making.

Such analyses may be valuable and important but it is a mistake to limit consideration of “rational” behavior to those things that can be weighed or quantified with an oxygen mask on a treadmill. Human behavior engages social networks, and relies upon our mechanisms for social collaboration and cooperation. These involve emotional cognition. Consequential decisions entail the management of emotions, often of many people. People don’t avoid or minimize emotions in their behavior. They manage them.

Archaeological traces of emotional and social cognition

Signs of emotional management are not hard to see in the archaeological record once you start looking. The manufacture of stone tools to specific shapes, such as the iconic teardrop-shaped handaxe, would be impossible without the ability to share joint attention. Macaques and capuchin monkeys learn to crack nuts or shells by watching from a distance, and Lomekwian hammering might conceivably require little more. But young chimpanzees learn to crack nuts and making many other kinds of tools by close observation of their mothers, and it’s likely that the selection of appropriate raw material and rotation of cores by Oldowan toolmakers would have required a higher tolerance of close observation by older children or adults.

In our article, we point to healthcare as strong evidence of emotional cognition. Humans often help the sick and injured, sometimes for prolonged periods of time and at great cost. Healthcare in this sense has been present across societies worldwide, although there are cultural differences that affect who may be eligible for such help, and how it is delivered. In these aspects healthcare is like allocare of infants and children—it matters who they are, how they are related to others in the group, and what forms of care may be provided.

There is growing evidence of healthcare in the archaeological and skeletal record of hominins. In our article we point to some of the famous cases, like the KNM-ER 1808 individual, who lived around 1.6 million years ago and survived for some time with debilitating and painful periosteal reaction and cortical bone thickening. In the Middle Pleistocene, the case of craniosynostosis of the Cranium 14 individual from Sima de los Huesos is notable; in the Late Pleistocene record, Neanderthals with evidence of extensive osteoarthritis, losses of limbs, and near-complete loss of teeth are also very well known.

The point of these examples is not to suggest something extraordinary or unique about the specific individuals involved. After all, sick and injured individuals sometimes survive or overcome disabilities in societies of nonhuman primates. Instead, what is interesting about the pattern in Pleistocene hominins is how they reveal the umbrella of caregiving extending beyond allocare of infants and young children across a broader segment of society, at least in some contexts.

Why do we care for the sick? I don’t know how many times I’ve read previous writers suggesting that ancient people were sneakily rational: Sure, they cared for aging individuals, but their real motive was that the cultural knowledge of older people had great value. But I think it is short-sighted to see Pleistocene healthcare as a quid pro quo. Empathy is a building block of social cognition in hominins. I doubt that it’s possible to build a system of social collaboration without that empathy sometimes manifesting as care.

Likewise, mortuary behavior is universal across human societies today. Many—although far from all—kinds of mortuary behavior leave traces that can be archaeologically visible. One form of mortuary behavior is burial, and the burials of many Neanderthals in Late Pleistocene contexts have gotten a great deal of attention, as has the evidence of burial that I’ve helped publish on the Middle Pleistocene species Homo naledi.

There are many other ways that mortuary behavior may be manifested in archaeological sites. From the evidence for curation and handling of the skulls from Herto, Ethiopia, to the cutmarks on the face of the Bodo 1 skull, also from Ethiopia, to the occurrences of cannibalism at several Neanderthal sites, all these are doorways into the practices surrounding death in their respective hominin groups.

As Emma Pomeroy and others have reviewed, mortuary behaviors are manifested in many other living species of mammals. From our close relatives like chimpanzees and gorillas, to more distant social mammals in many orders including orcas, elephants, and horses, individuals alter their behavior when encountering dead bodies of their own species. They may exhibit profound grief or sadness, they may show fear, and in some heartbreaking cases they continue to manifest care—intervening to try to revive or sustain a juvenile or adult group member—days or weeks after death has occurred.

Humans and our close hominin relatives have often elaborated these natural reactions to death into shared rituals that integrate emotion and memory within our social groups. Culture gives us a shared setting for grief, and emotional cognition enables us to empathize more widely. Such funerary rituals are a part of the collaborative construction of new futures, providing some structure in the absence of the valued deceased.

No behavioral modernity

In our article we don’t get much into pigments, ornaments, cave art, or other signs of “symbolic” culture. But these, too, reflect the growing importance of emotional regulation and social collaboration in our evolutionary history.

Steven Kuhn and Mary Stiner have emphasized that ornaments have a primarily social function. Wearing a distinctive bauble, having a visible tattoo, or marking one’s garments or skin pigment all have the effect of making strangers aware of the wearer’s status and rank within society. Archaeologists are increasingly focusing upon the demographic aspect of these kinds of behavior: As the population grows there are not only more people, there are more social groups, more need for communication among strangers, and more encounters between people who may not share a common language.

There is also a greater chance in a larger population that a new skill or idea may be transmitted forward to other people. The half-life of information depends on how many people there are. It also depends on how willing people are to share space with each other. Emotional cognition and social collaboration are at the core of these abilities. Markers of social identity, rank, and status are some of the best indicators of this system in the archaeological record.

Back in the 1980s, some archaeologists imagined that recent humans were products of a “cognitive revolution” that set them apart from earlier people like Neanderthals. This “great leap forward” was marked by blades, ornaments, and above all, cave art. It was inevitable that some researchers hypothesized that recent people had emerged as a new, anatomically modern set of populations that carried advanced cultural and cognitive abilities.

There was a problem with this idea even at the time. There were archaeological sites like Skhūl, Israel, where people with so-called “modern” anatomy seemed to have had indistinguishable cultures from those at other sites with burials of Neanderthals. Some researchers suggested that anatomical modernity was not the important dimension. What mattered was behavioral modernity: the real revolution was in the brain, not the body.

By 2000, Sally McBrearty and Alison Brooks could detail the gradual appearance in Africa of many innovations that had once been assumed to have emerged suddenly in Europe. There was no “human revolution”, they wrote. There was a gradual evolution of material culture piecemeal across Africa through much of the later Middle Pleistocene. Even so, many archaeologists took as their working assumption that the African archaeological record documents the gradual transition to behavioral modernity, which includes symbolic culture shared by all living people.

We think that the idea of behavioral modernity has obscured the deeper causes of evolution of behavior in humans and other hominins. Again and again, archaeologists have fallen back on the notion of a “transition”—whether fast or slow—with the implicit idea that the transition was singular and progressive.

Biologists study evolutionary change by understanding phylogeny. The phylogeny of Pleistocene Homo involves the divergence of some groups over many hundreds of thousands of years, and networks of contacts and interbreeding connected many of those groups.

All of them were social species. As we write, individuals did not “live outside social groups, at least not for long or typically”. As children they met their ecologies through the interface of other individuals; they rarely survived without others. Their niche was learned socially and propagated by construction of rules, conventions, and habits.

Even if archaeologists could recognize “symbolic” culture from artifacts—which is far from clear—the fact is that symbolism is not the primary basis of collaboration and cooperation in living humans. I would go a bit further than we do in the article. My personal view is that much that has been called “symbolic” in the archaeological record should instead be understood to reflect shared emotional regulation. It is fascinating that some ancient people and other hominins ventured far into caves to make marks on their walls. But the important bit is not the marks—which are, after all, found fairly widely across the Pleistocene world. What’s actually fascinating is the shared journey.

We are not our ancestors. Humans today are often innovative, creative, and strategic in ways that we do not see in deep time. These aspects of human behavior do draw upon emotional cognition but they require integration with other kinds of cognition also. Besides this, we are aided by communication systems that never existed in any Pleistocene society, extending the reach of our attention and forethought.

How can we get at the evolution of these abilities, which we continue to invent? My thought is that what matters today is not well correlated with the kinds of archaeological traces that have been grouped as “behavioral modernity”. The data tell us that those very traces are shared by many Neanderthals, Denisovans, and other ancient groups. However, better understanding the basis of social collaboration among Pleistocene groups is certainly a valuable start.

References

Mayer, D. E. B.-Y., Groman-Yaroslavski, I., Bar-Yosef, O., Hershkovitz, I., Kampen-Hasday, A., Vandermeersch, B., Zaidner, Y., & Weinstein-Evron, M. (2020). On holes and strings: Earliest displays of human adornment in the Middle Palaeolithic. PLOS ONE, 15(7), e0234924. https://doi.org/10.1371/journal.pone.0234924

Fuentes, A., French, J. C., Hawks, J., Kissel, M., & Spikins, P. (2026). Social and emotional cognition in Pleistocene hominin evolution: The role of biocultural processes. Journal of Archaeological Science, 185, 106441. https://doi.org/10.1016/j.jas.2025.106441

Hawkes, K. (2003). Grandmothers and the evolution of human longevity. American Journal of Human Biology, 15(3), 380–400. https://doi.org/10.1002/ajhb.10156

Hrdy, S. B. (2005). Comes the Child before Man: How Cooperative Breeding and Prolonged Postweaning Dependence Shaped Human Potential. In Hunter-Gatherer Childhoods. Routledge.

Hrdy, S. B. (2026). Father Time: How Nurturing Is Natural for Men. Princeton University Press.

Hrdy, S. B., & Burkart, J. M. (2020). The emergence of emotionally modern humans: Implications for language and learning. Philosophical Transactions of the Royal Society B: Biological Sciences, 375(1803), 20190499. https://doi.org/10.1098/rstb.2019.0499

Kuhn, S. L., & Stiner, M. C. (2007). Paleolithic Ornaments: Implications for Cognition, Demography and Identity. Diogenes, 54(2), 40–48. https://doi.org/10.1177/0392192107076870

Marlowe, F. (1999). Showoffs or Providers? The Parenting Effort of Hadza Men. Evolution and Human Behavior, 20(6), 391–404. https://doi.org/10.1016/S1090-5138(99)00021-5

Mcbrearty, S., & Brooks, A. S. (2000). The revolution that wasn’t: A new interpretation of the origin of modern human behavior. Journal of Human Evolution, 39(5), 453–563. https://doi.org/10.1006/jhev.2000.0435

Pomeroy, E., Hunt, C. O., Reynolds, T., Abdulmutalb, D., Asouti, E., Bennett, P., Bosch, M., Burke, A., Farr, L., Foley, R., French, C., Frumkin, A., Goldberg, P., Hill, E., Kabukcu, C., Lahr, M. M., Lane, R., Marean, C., Maureille, B., … Barker, G. (2020). Issues of theory and method in the analysis of Paleolithic mortuary behavior: A view from Shanidar Cave. Evolutionary Anthropology: Issues, News, and Reviews, 29(5), 263–279. https://doi.org/10.1002/evan.21854

I’ve been fascinated after reading new research from David Braun and collaborators focusing on the early Oldowan tool assemblages from Namorotukunan, north of Ileret, Kenya. The report describes three sites, all located within a radius of around 200 meters from each other, but separated in different geological levels stretching from around 2.75 million to 2.44 million years ago.

Across these 300,000 years, the technology and strategies of the toolmakers appear to have been much the same.

What changed was their environment. Starting from a well-watered floodplain with many trees, the local habitat became drier with less woody cover and more natural fires. Later the spot was inundated by a newly resurgent paleolake. It seems that the toolmakers’ lifestyles tolerated the shifting local resources without needing to change the way they made and used tools.

The limits of archaeology

Now, when it comes to causal connections of climate change and hominin behavior, I am especially alert to the weakness of most data from Paleolithic archaeology. I don’t doubt that changes in climate and local environments were made a difference to our evolution, probably at many times and places. But when it comes to testing changes in behavior over time, even a very rich archaeological site ultimately distills down to a single data point.

In this study, the archaeological data represent three brief intervals out of a 300,000-year span. The study presents strong data that the local environment was not constant across this interval. It is the apparent lack of correlation between archaeology and environmental change that makes the story. Environment changes, hominin behavior didn’t.

I have my doubts about the strength of this conclusion, based as it is upon a lack of evidence of change between small artifact samples.

Even so, I really like the study for many reasons. Every one of the samples of early stone artifacts are small. The Oldowan archaeological sites of this range of ages are incredibly rare: They include a few localities in the Gona field region of Ethiopia, one assemblage described from the Ledi-Geraru area, and the site recently described from Nyayanga, Kenya. So a report like this one adds substantially to the knowledge base.

Namorotukunan

When it comes to archaeology, the work described by Braun and coworkers is a long term study. The researchers spent the period from 2013 to 2022 excavating, collecting information, and analyzing finds from the three distinct sites, designated as NMT1, NMT2, and NMT3—and their surroundings.

What blows me away about the Namorotukunan work is the extent of geological and environmental evidence that the team has recovered from the area. Their article includes a remarkable three-dimensional rendering of the present-day topography, a reconstruction of the paleoriver network that created many of the sedimentary deposits, and the sequence of changes to sedimentation over time.

It’s really lovely work. These kinds of reconstructions are important not only for the immediate region but for understanding the changes in hydrology across the entire Turkana Basin. If you’re someone who could use a visual understanding of how this kind of ancient site is formed, this is a great one to study.

The team recovered 1290 artifacts from the three sites. The richest of the three was the NMT2 site in the middle of the time span at around 2.6 million years ago, which produced a bit more than half of the artifacts. There were no hominin bones in the area, and little preservation of faunal bones in the excavated sites, with only 32 found. Further, only the NMT2 site preserved a few bone surfaces well enough to examine for cutmarks or other traces of hominin activity. The team identified cutmarks on one bone.

The toolmakers chose river-worn cobbles of rock that were abundant within the streambeds of the immediate area. There is no evidence that they transported any pieces of rock any distance further than the nearest stream, within dozens of meters of the sites. Among these, the toolmakers mostly chose chunks of a kind of crypto-crystalline silica known as chalcedony, preferring it over other types of rock for its fracture qualities that give rise easily to sharp flakes. The three sites all show very similar selectivity of this type of rock, and are similarly flake-dominated. And when Braun and collaborators looked at how the flakes were made, they could show similar sequences of rotation of the small cobble cores to remove flakes from them across all three of the sites.

The three sites do present a slight difference in their assemblages. Percussion—pounding rocks, bones, or plant material—is one of the major uses of Oldowan artifacts at most sites. But out of the Namorotukunan sites only NMT2 has artifacts that show evidence of percussion, and only very few at that. Braun and coworkers note that the local cobbles are not rock types that stand up to much pounding, so it’s maybe not surprising they didn’t use them much in this way. But an Oldowan site where nobody pounded anything is an unusual Oldowan site, really suggesting that the hominins were very constrained in their activities by the locally available materials.

Oldowan stability

Why was the Oldowan apparently stable for so long? I don’t think I’m saying anything very profound when I say that what looks like stability is partly an illusion of sampling.

The basis of technical behavior that archaeologists have categorized as “Oldowan” is production of sharp flakes and hammering or pounding with roughly fist-sized rocks. Hominins used the sharp flakes for disarticulating and slicing meat from animal carcasses, and probably for cutting into parts of plants and shaping wood. They pounded animal bones with hammers to get into the marrow-rich medullary cavities, and pounded plant foods including nuts, large seeds, and edible underground parts such as tubers.

Their use of wood is mostly invisible to us today. Wood was essential for digging for underground plant foods and invertebrates, for probing for honey, and for spearing small animals. Chimpanzees do all these things. Our only trace of such activities from archaeology has been close analysis of the surfaces of some stone flakes that suggest they had been used to shape wood.

The Namorotukunan sites tell us that hominins in this place shaped their activities based on the available rock types. That’s very similar to the patterns of behavior shown by stone use among chimpanzees today. Chimpanzees will transport rocks that they use for pounding and cracking nuts over short distances of dozens of meters, and these short trips can add up over time. They have not been observed to intentionally flake stone, and they do not crack nuts everywhere that nuts and suitable rocks both exist.

What would an archaeological perspective look like on any single location over hundreds of thousands of years of chimpanzee presence? Over many generations, chimpanzees that sustain a tradition of nutcracking might disappear from any given area. There might be long absences of chimpanzees altogether, with periods of local environmental unsuitability for them. New chimpanzee groups would sometimes pass through, lacking nutcracking altogether. Their wood and leaf-based technologies would leave no trace. Again later, local chimpanzees might invent nutcracking, or newcomers might bring the technique from elsewhere. Using local rocks, they would leave much the same record.

Archaeologists looking specifically for stone artifacts would find occasional assemblages. They would observe how similar the assemblages look.

Still, there are some differences across space and time that tell us something important about Oldowan toolmakers’ behavioral variation. At some places, Oldowan toolmakers went out of their way to retrieve and transport the stone they preferred, across many kilometers.

Earlier this year, Emma Finestone and collaborators reported on the transport of stone to the site of Nyayanga, in western Kenya near Lake Victoria. This site dates to between 3.03 and 2.58 million years, making it possibly the earliest Oldowan site, depending on its exact position in that range of ages. The transport of stone over such distances is something chimpanzees do not do; it shows that these Oldowan toolmakers had larger geographic ranges and knowledge of resources across space.

The Namorotukunan evidence fits together with most other early Oldowan sites, including Hadar, Gona, and Ledi-Geraru. Long-distance transport of stone does not appear to have been typical for early Oldowan toolmakers. Such a site may show the potential for invention. But demonstrating regional and temporal variability will require many more data points. With more sites, archaeologists will begin to have the kind of samples that might test correlations with environmental or climatic data.

Reference

Braun, D. R., Aldeias, V., Archer, W., Arrowsmith, J. R., Baraki, N., Campisano, C. J., Deino, A. L., DiMaggio, E. N., Dupont-Nivet, G., Engda, B., Feary, D. A., Garello, D. I., Kerfelew, Z., McPherron, S. P., Patterson, D. B., Reeves, J. S., Thompson, J. C., & Reed, K. E. (2019). Earliest known Oldowan artifacts at >2.58 Ma from Ledi-Geraru, Ethiopia, highlight early technological diversity. Proceedings of the National Academy of Sciences, 116(24), 11712–11717. https://doi.org/10.1073/pnas.1820177116

Braun, D. R., Palcu Rolier, D. V., Advokaat, E. L., Archer, W., Baraki, N. G., Biernat, M. D., Beaudoin, E., Behrensmeyer, A. K., Bobe, R., Elmes, K., Forrest, F., Hammond, A. S., Jovane, L., Kinyanjui, R. N., de Martini, A. P., Mason, P. R. D., McGrosky, A., Munga, J., Ndiema, E. K., … Carvalho, S. (2025). Early Oldowan technology thrived during Pliocene environmental change in the Turkana Basin, Kenya. Nature Communications, 16(1), 9401. https://doi.org/10.1038/s41467-025-64244-x

Finestone, E. M., Plummer, T. W., Ditchfield, P. W., Reeves, J. S., Braun, D. R., Bartilol, S. K., Rotich, N. K., Bishop, L. C., Oliver, J. S., Kinyanjui, R. N., Petraglia, M. D., Breeze, P. S., Lemorini, C., Caricola, I., Obondo, P. O., & Potts, R. (2025). Selective use of distant stone resources by the earliest Oldowan toolmakers. Science Advances, 11(33), eadu5838. https://doi.org/10.1126/sciadv.adu5838

Plummer, T. W., Oliver, J. S., Finestone, E. M., Ditchfield, P. W., Bishop, L. C., Blumenthal, S. A., Lemorini, C., Caricola, I., Bailey, S. E., Herries, A. I. R., Parkinson, J. A., Whitfield, E., Hertel, F., Kinyanjui, R. N., Vincent, T. H., Li, Y., Louys, J., Frost, S. R., Braun, D. R., … Potts, R. (2023). Expanded geographic distribution and dietary strategies of the earliest Oldowan hominins and Paranthropus. Science, 379(6632), 561–566. https://doi.org/10.1126/science.abo7452