A new study by Sánchez-Martínez and colleagues, published in Quaternary International,1 works through 19 archaeological levels spanning approximately 39,000 to 13,500 years before present. Their question is precise: did the people occupying this site get their stone from the same places throughout the Upper Palaeolithic, or did procurement strategies shift over time? The answer, drawn from more than 3,000 analysed lithics, is that they shifted considerably — and that the pattern of change is not random.

The raw material in question is chert. The southern slope of the Pyrenees is unusually well-stocked with siliceous resources, and Homo sapiens groups here relied on chert almost exclusively for tool production. This distinguishes them from the Neanderthal populations that preceded them in the same region, who incorporated quartzite into their technological repertoire. The shift to near-exclusive chert use is itself a cultural signature, not simply a response to what was locally available.

The team examined cores and retouched tools from the entire Upper Palaeolithic sequence using an archaeopetrological approach: examining the textural and micropalaeontological characteristics of each piece under binocular microscope, then comparing the results against a geological reference collection of siliceous rocks from the University of Barcelona. This allowed them to identify the sedimentary environment in which each chert type originally formed, and from that to infer potential source areas.

Six chert varieties were distinguished. Two appeared consistently across all levels. The other four showed up only sporadically, in specific assemblages, and those rare appearances turn out to be the most informative.

Read more

That family is now the subject of a genome-wide ancient DNA study published in Nature Communications1 by Jacob Bongers, Jordan Dalton, Lars Fehren-Schmitz, and colleagues. The paper reconstructs kinship, ancestry, and movement in the Chincha Valley between roughly the thirteenth and sixteenth centuries AD, before, during, and just after Inca incorporation of the region. What it finds complicates the standard story of pre-Inca coastal Peru as a patchwork of isolated, static communities.

These people were moving. And they had been doing so long before the Inca gave them any reason to.

The Chincha Valley sits on Peru’s south-central coast, a fertile strip running from the Pacific inland toward the Andean foothills. By the fifteenth century, it was home to a major regional polity called the Chincha Kingdom, with a total population that may have exceeded 100,000. Sixteenth-century colonial sources describe a finely stratified economy: at least 10,000 fisherfolk, 12,000 farmers, and 6,000 artisans and merchants, each occupying distinct sectors of the valley. Chincha merchants traveled by balsa raft along the coast and by llama caravan into the highlands, trading silver, gold, emeralds, and other prestige goods with elites across the Andes. At the pivotal Cajamarca confrontation between Inca forces and the Spanish, the Chincha lord reportedly sat beside the Inca Emperor himself. This was not a marginal valley people.

When the Inca incorporated Chincha in the early AD 1400s, they did so through what current scholarship describes as a negotiated arrangement, rather than straightforward conquest. The Chincha, it seems, agreed to become a seafaring client state in exchange for privileged access to the prized Spondylus trade, which had previously been dominated by the Chimú on the north coast. The relationship was unusual. And new DNA evidence suggests it may have been built on connections that already ran deep.

The research team extracted genome-wide data from 21 individuals buried at two locations in the valley: Las Huacas, a 100-hectare site on the alluvial plain, and a series of cemeteries scattered across the middle valley. The middle valley is what Andean scholars call a chaupiyunga, a transitional ecological zone between the coast and the highland valleys, suitable for growing coca and maize. Over 500 graves cluster into 44 mortuary sites there, falling into two basic types: subterranean cists holding individuals in extended positions, and larger accessible mausolea called chullpas, where the dead were painted with red hematite and cinnabar pigments and had their vertebrae threaded onto reed sticks.

The earliest individuals from the cist graves, two people with median modeled death dates around AD 1290, carry unadmixed north coast ancestry. Their genomes cluster not with southern coastal populations, as geography would suggest, but with individuals from the Chicama Valley and other north coast sites more than 700 kilometers away. The team’s analysis used outgroup-f3 statistics and qpADM admixture modeling to establish this. The result is unambiguous enough that the paper frames it as “strongly supported” by independent historical, ceramic, and textile evidence as well. North coast ceramics and textiles have been documented throughout Chincha, and a sixteenth-century chronicle by Pedro Cieza de León describes the earliest Chincha people as a group led by a captain who came from afar to conquer the valley.

The radiocarbon modeling here is itself a significant methodological achievement. Calibrating dates from coastal Peru is notoriously difficult, because the marine reservoir effect, the offset between atmospheric and marine radiocarbon, varies substantially along the Peruvian coast due to strong deepwater upwelling and seasonal El Niño activity. The standard approach of using modern shell data to estimate this offset produces wildly inconsistent results; for this region, the six relevant shells from the early twentieth century yield ΔR estimates ranging from -181 to +219 years, a range so wide it renders precise dating essentially meaningless.

The team addressed this by letting the marine reservoir offset float as a free parameter in a Bayesian model, constrained by stratigraphic relationships and generational intervals derived from the aDNA family tree. The result was a ΔR estimate of -314 ± 52 years, far more negative than the shell-based approaches suggest, and consistent with two earlier estimates from AD 500 contexts in coastal Peru. They also built individual mixed calibration curves for each person based on their specific marine diet proportions, derived from stable isotope analysis, using a custom script they call Mix-Cal-Lot. The practical upshot is a generational-scale chronology that can actually track who lived when, with enough precision to say things like: this grandmother was born around 1280, this grandson around 1320.

Read more

The difference between five hectares and one hectare is not just agricultural. It is the difference between a future and a holding pattern.

This is where archaeologists Per Ditlef Fredriksen (University of Oslo / University of Cape Town) and Foreman Bandama (Field Museum of Natural History / University of Illinois at Chicago) went to do fieldwork, in 2019 and then again in 2022–23, as part of a Norwegian Research Council project called ARCREATE. Their paper, published this year in Cambridge Archaeological Journal,1 reports what they found. It is framed as contemporary archaeology, which means the methods are the same ones archaeologists use on Iron Age ceramics or medieval construction sites. The subject is the present.

What they found was not simply a story of survival. It was something more structurally interesting: a displaced community had developed a coherent set of material strategies, ways of building, making, sourcing raw materials, and moving through a landscape, that simultaneously expressed defiance toward the authorities who uprooted them and maintained active connection to the dead.

The ranch where the third-wave flood victims now lived had already received two earlier waves of settlers. The first arrived in the early 2000s, Tsonga-speakers displaced under Mugabe’s land reform programme, mostly from 40–50 kilometers away. They had occupied vacant land and over time received plots of around eight hectares each. A second wave followed around 2010, predominantly Shona-speakers who relocated voluntarily before the dam finished filling. They received four-hectare plots. Then came the flood victims of 2014, also mostly Shona-speakers but with almost no political leverage, packed into one-hectare plots in a newly cleared section called Chingwizi, physically separate from where the first two waves had settled.

Sections A through D for the first two waves. Section E for the flood victims. Firstcomers and latecomers, with a sharp material gradient between them.

The authorities told the third wave not to build permanent structures. They were designated temporary residents. The infrastructure eventually planned around their tiny plots was the worst possible confirmation of that temporariness: not that they would be moved on, but that they might stay stuck in a barely functional holding space indefinitely. A community member put it plainly by 2022:

“Now they tell us we will get irrigation, so we are not going to be moved. We try to remain hopeful, but it is way overdue. We were much better off where we were.”

They had ignored the directive about permanent structures. They built in stone.

Read more

That is the central finding of a study published in Antiquity1 by Anderson L. Carter, Canan Çakırlar, and colleagues at the University of Groningen and collaborating institutions. The site is Hala Sultan Tekke, on the southeastern coast of Cyprus, occupied during the Late Bronze Age from roughly 1650 to 1150 BCE. The birds are Columba livia — the rock dove, ancestor of every pigeon currently working a city sidewalk.

The isotope signature matters because it is hard to explain away. Stable carbon and nitrogen values in bone collagen reflect what an animal ate over its lifetime. Wild birds forage opportunistically; their isotope values scatter widely. The pigeons at Hala Sultan Tekke did not scatter. Their dietary range was tighter than that of any other species in the regional comparison dataset — tighter even than cattle, which were actively herded and managed. Using a statistical measure of dietary breadth called the corrected standard ellipse area, the team found that pigeons had the most constrained niche of all taxa sampled: a value of 1.26, compared with 1.75 for humans and 1.80 for cattle. In ecological terms, these were animals occupying a narrow, managed dietary slot.

The nitrogen values told a more specific story. The mean nitrogen isotope value for the main pigeon grouping sat at around 8.82 per mille, placing these birds above known scavenging populations like dogs (8 per mille) and foxes (8.26 per mille) and overlapping almost exactly with the human samples drawn from comparable Cypriot Bronze Age sites (9.92 per mille). Whatever these birds were consuming, it was consistent, protein-enriched relative to a purely wild diet, and closely tied to what humans were eating. The team’s interpretation is that the pigeons were living inside the city and eating its waste — grain spillage, scraps, and possibly deliberate supplemental feed from people who were managing them.

“Either way,” writes senior author Çakırlar, “this very likely means that they were domesticated or on their way to being domesticated.”

The presence of juvenile bones at multiple contexts across the site reinforces this picture. Fledgling Columba liviabones in an urban deposit are not what you get from wild birds passing through. They confirm that the pigeons were breeding within the settlement itself.

What Domestication Actually Means Here

The word “domesticated” is doing real work in this study, and the authors handle it with appropriate caution. Bone morphology cannot resolve the question for Columba livia. The species is exceptionally plastic: size varies enormously depending on climate, diet, and management practices. A population raised primarily for fertilizer production in a desert environment, like the Byzantine-period pigeons documented at Shivta and Saadon in the Negev by Marom and colleagues, stayed small even when clearly domesticated and housed in towers. Apply body-size criteria there and you would miss them entirely. The same metric that works reasonably well for cattle or pigs breaks down almost completely for pigeons. The Hala Sultan Tekke team measured 154 long bones and found all fell within the size range for C. livia, but the assemblage is not clearly distinguishable from a wild population on those grounds alone.

What the isotopic and contextual data offer instead is evidence for a commensal relationship that had already tipped toward management. The authors describe this as consistent with the “commensal pathway” to domestication — a process by which animals gradually move into the human niche over many generations, initially attracted by food waste or shelter, and eventually subject to deliberate management. Rock doves naturally nest in rocky outcrops and cliff faces. The ashlar masonry attested at Hala Sultan Tekke suggests the city had multi-story buildings with sheer facades that could have functioned as adequate substitutes.

The relevant comparison for dating purposes is Nea Helos, a Hellenistic site in Greece where the earliest previously identified assemblage of morphologically domesticated pigeon bones was recovered, dated to roughly the fourth through first centuries BCE. The Hala Sultan Tekke material predates that by somewhere between nine hundred and a thousand years. A 2023 genomic study identified modern domestic pigeons as most closely related to wild populations from the Middle East, pointing toward at least one domestication event in that region — consistent with the Cypriot evidence but not specifically pinpointed to it. Carter and colleagues are explicit that Hala Sultan Tekke does not represent the origin of pigeon domestication. It represents the earliest direct biomolecular evidence for a commensal human-pigeon relationship that we currently have.

The Feast

Most of the pigeon bones — 82 percent by minimum number of individuals — came from a single area of the site, City Quarter 1. Within it, a significant concentration was recovered from two interconnected rooms, numbered 70 and 83. These had been flagged during excavation as a cultic space. They contained a large stone structure functioning as a table or altar, burnt deposits of animal remains from multiple species, ceramic vessels representing at least three different production traditions (local Cypriot, Mycenaean, and Canaanite), Egyptian faience beads, gold leaf, a Mycenaean figurine, and carbonized botanical remains including olive, grape, and cereals. The excavators classified these as feasting deposits.

Feasting in Bronze Age Cyprus was not incidental. It was structurally embedded — politically, socially, and in mortuary practice. The finds from rooms 70 and 83 have the material profile of repeated formal events: multiple episodes of burning, tableware from across the Mediterranean, architecturally demarcated space. Pigeon bones appear not only in that cultic core but also in wells that served as refuse pits near tomb areas, and in at least one tomb deposit directly. The birds were showing up across the entire social register of the site’s ritual life.

Just over half of all the C. livia specimens showed contact with fire. The most parsimonious explanation is that the birds were cooked and eaten, then their bones discarded in the ritual space or burned as part of the offering itself. In birds of this body size, butchery marks are not necessary for consumption and are not expected to be preserved even when the animal was processed — so the absence of cut marks does not complicate the picture.

Whether any of this had a direct connection to the cult of Aphrodite, with whom pigeons are closely associated in classical Cypriot tradition, remains an open question. Pigeon and dove figurines appear on Cyprus from the Middle Bronze Age onward, and the island was identified in antiquity as Aphrodite’s birthplace. Her affinity for the birds is well-documented in artistic representations from across the Mediterranean. The paper acknowledges this interpretive framework while noting that no temple or sanctuary has been identified at Hala Sultan Tekke specifically. The cultic deposits are real. Their precise theological affiliation is not resolved.

Three individuals in the isotope dataset sat apart from the main grouping, with lower nitrogen values overlapping with wild herbivores rather than with the human range. The authors raise the possibility that these birds came from a different setting — possibly Trypes, a rural site under Hala Sultan Tekke’s administrative control, identified as a probable granary and livestock supplier. Whether they were brought to the city specifically for consumption, or represent pigeons with a more independent relationship to the human environment, the bones do not say.

Hala Sultan Tekke was destroyed twice around 1200 to 1150 BCE and then abandoned, its bay having silted up and severed what had been one of the most active trading connections in the eastern Mediterranean. The destruction coincided with the Bronze Age collapse that ended dozens of cities across the region. The pigeons almost certainly outlasted the city. Columba livia is remarkably good at that.

Further Reading

• Hernández-Alonso, G. et al. (2023). Redefining the evolutionary history of the rock dove, Columba livia, using whole genome sequences. Molecular Biology and Evolution 40. https://doi.org/10.1093/molbev/msad243

• Marom, N. et al. (2018). Pigeons at the edge of the empire: bioarchaeological evidences for extensive management of pigeons in a Byzantine desert settlement in the southern Levant. PLoS ONE 13. https://doi.org/10.1371/journal.pone.0193206

She is now called the Ossick Lass. “Ossick” is the local vernacular for both Great and Little Urswick, a place-name worn soft by dialect. Martin Stables, the self-taught archaeologist from the village who led the excavation starting in 2016, wanted her name rooted in the landscape where she had been buried for eleven millennia. The name fits. She is the earliest known person from northern Britain.

Stables is not a professional archaeologist. He began digging Heaning Wood Bone Cave out of personal fascination with his village’s prehistoric past. The project eventually drew in specialists from the University of Lancashire, the Francis Crick Institute’s Ancient Genomics Laboratory, and the University of Nevada. What started as local curiosity became something with a publication in the Proceedings of the Prehistoric Society1 and a place in the wider story of post-glacial Europe.

The bones themselves are fragmentary. All fragments belonging to Individual F, as she is designated in the research, are cranial. A molar tooth crown, part of the upper jaw, the left cheekbone. The taphonomy is consistent with a whole-body burial placed in the cave shortly after death, the skeleton gradually disarticulating as the sediment shifted over thousands of years. There is no evidence of gnawing, no sign the body had been exposed elsewhere first. She was buried here, in this shaft, by people who treated her death as something requiring a deliberate response.

Radiocarbon dates on her bone collagen place her death between 9290 and 8925 cal BC. A perforated periwinkle shell bead from the same layers dated to 9135-8645 cal BC, overlapping closely enough that it was likely placed with her. Five of these beads in total were recovered from the cave, all of them flat periwinkle (Littorina obtusata) with single perforations, small and fragile. The researchers note that surviving examples are almost certainly a fraction of what was originally deposited. Perforated periwinkle beads appear at other Early Mesolithic burial caves in Britain, including Aveline’s Hole and Gough’s New Cave in Somerset, though the records from those sites are not precise enough to link the beads directly to burials. At Heaning Wood, the radiocarbon dates make that link explicit.

Osteological analysis put her age at 2.5 to 3.5 years. The DNA from her remains was poorly preserved, which is typical of material this old recovered from cave sediment. Three samples were taken from her maxilla. Two produced unreliable sex estimates flagged for possible contamination. The third, while still degraded, yielded a reliable result: XX. She was female. It is the first time a child’s biological sex has been confirmed for remains this ancient from this part of the world.

Her stable isotopes are slightly unusual. The ratio of carbon-13 to carbon-12 and the elevated nitrogen-15 in her bone collagen sit higher than the seven other individuals from the cave. The research team considers two explanations: one is that she was still being breastfed or recently weaned, which can elevate isotopic values in young children; the other is that her diet included a marine component. If the marine explanation is correct, her radiocarbon date would actually be slightly younger than the raw figure suggests, because marine foods introduce older carbon. That would bring her date into closer alignment with the shell bead. The question cannot be resolved with current data.

Read more

The men buried at Likneset died during the seventeenth and eighteenth centuries, during the period when European fleets had moved their whale processing offshore, pelagic, and were using sheltered Svalbard fjords as seasonal anchorages. They died far from wherever they had come from. Their ships went back without them. Someone buried them where they fell.

A new study by Lise Loktu of the Norwegian Institute for Cultural Heritage Research and forensic specialist Elin Therese Brødholt of Oslo University Hospital, published this week in PLOS One,1 draws on three decades of excavations at Likneset to address two questions at once: what is the warming Arctic doing to the site, and what do the remains of those men tell us about what it meant to live and die in early modern Arctic whaling?

The answers to both questions are uncomfortable in different ways.

Read more

A new study in Nature Ecology & Evolution1 by Bourgon and colleagues has analyzed zinc isotope ratios from the enamel of 24 human specimens and 57 faunal samples from three well-documented Sri Lankan cave sites: Fa-Hien Lena, Batadomba-lena, and Balangoda Kuragala. The combined record spans roughly 20,000 to 3,000 years ago, taking in the tail end of the Last Glacial Maximum, the Pleistocene-Holocene transition, and the period immediately before domesticated crops appear in the island’s lowland zones. What that record shows is a slow, measurable drift in the human diet toward plants, long before anyone was farming.

The method deserves some explanation, because it’s doing something that wasn’t possible until recently. Standard dietary reconstruction in archaeology relies on stable nitrogen isotopes in bone collagen, which track trophic level reliably. The problem is that collagen degrades. In tropical environments, it’s often gone entirely. Sri Lanka’s cave sites preserve some of the earliest Homo sapiens fossils found anywhere in South Asia, but those fossils are old enough, and wet enough in their burial context, that collagen is frequently absent or unreliable.

Zinc isotopes in enamel are different. Enamel is dense and highly mineralized, and it preserves the biogenic zinc signature even under conditions that destroy organic tissue. The key property is directional: herbivores show the highest δ⁶⁶Zn values, carnivores the lowest, and omnivores fall in between. As trophic level increases, δ⁶⁶Zn decreases. The mechanism involves fractionation across tissues and the fact that animal foods contain more zinc and promote greater absorption than plant foods. This last point matters for interpretation. Because animal protein has an outsized effect on zinc isotope ratios relative to its caloric contribution, even moderate meat intake can pull values toward the carnivore end of the scale. Which means that when humans sit consistently in the intermediate range, the contribution from plants must have been substantial.

The faunal dataset the team assembled provides the interpretive scaffold. Deer, sambar, barking deer, rhinoceros, elephant, wild boar, flying squirrel, porcupine, bat, civet, and various primate species are all represented. This spread gives a local trophic structure against which the human values can be plotted. The human specimens consistently sit between the herbivores and the macaques.

The macaques are interesting. Macaca sinica, the toque macaque endemic to Sri Lanka, shows very low δ⁶⁶Zn values, overlapping with what you’d expect from obligate carnivores, despite being an omnivore that eats a substantial proportion of plant food. The low values probably reflect its habit of consuming eggs, insect larvae, small reptiles, birds, and mammals alongside fruit and leaves. Because obligate carnivores are absent from Sri Lanka’s cave fauna, M. sinica ends up occupying the low end of the isotopic range and functions as a carnivore analogue in the system. The team validated this by comparing the spacing between macaque values and herbivore values against equivalent data from mainland Southeast Asian sites where carnivores are present, and the match is close enough to support the interpretive framework.

Against that baseline, humans sit in the middle. That alone is not surprising, given what the archaeological record at these sites already suggests: microliths, bone tools, and what is currently the earliest confirmed bow-and-arrow technology outside Africa all point to skilled, organized hunting of arboreal and semi-arboreal mammals. Mollusc shells, breadfruit, and candlenut show up too. Mixed subsistence seems obvious in hindsight.

The direction of change over time is less obvious. That’s the study’s more striking finding.

Read more

Over the following four centuries, as Mohenjo-daro continued to grow and fill in, housing inequality fell. Not marginally. By around 2100 BC, the Gini coefficient of residential disparity in DK-G South had dropped to 0.23, a figure typical of Neolithic farming villages, not major urban centers. The gap between the largest and smallest homes had narrowed to something archaeologists usually associate with societies that don’t yet have cities.

Mohenjo-daro had tens of thousands of residents. It had standardized brick construction, a citywide drainage network, a grid of streets, a sophisticated system of weights and measures, and a vigorous long-distance trade network that extended to Mesopotamia. It was, by any reasonable measure, one of the most functionally complex urban environments on earth in the third millennium BC. And it appears to have been getting more equal as it matured.

That finding is the core of a new study by Adam Green, Iqtedar Alam, and Cameron Petrie, published in Antiquity1 in 2026. Their analysis draws on excavation data from the earliest twentieth-century digs at the site, converting house footprints into Gini coefficients through a GIS-based approach. The Gini coefficient is a standard economic measure of distributional inequality: zero means total equality, one means total concentration. Their results put the city’s overall residential Gini at 0.44, well below contemporaneous Mesopotamian cities like Ur and Ugarit, which both score above 0.6. The ancient Greek site of Knossos sits at 0.86. The Classic Maya city of Palenque at 0.75.

Mohenjo-daro was already an outlier by those comparisons. But the diachronic data from DK-G South is what makes the study genuinely provocative.

What the houses say

The standard archaeological assumption is that cities generate inequality. The logic is intuitive: larger populations mean more specialization, more specialization means more occupational hierarchy, and hierarchy means wealth concentrates at the top. V. Gordon Childe codified this in his 1950 “urban revolution” framework, which treated social stratification as a defining feature of cities. Neoevolutionary theory went further, treating the state as essentially defined by stratification. If you had a city, you had elites; if you couldn’t find the elites, you weren’t looking hard enough.

Indus archaeologists spent much of the later twentieth century fighting this prior. Mohenjo-daro and the other major Indus sites lack the usual material signatures of hierarchical power: no palaces, no royal tombs, no temples accessible only to a priestly class, no statues of rulers. The arguments went back and forth. Some scholars concluded the Indus must have been a complex chiefdom or non-state. Others countered that a civilization that built cities at this scale must have had elites, even if the evidence was missing. Absence of evidence wasn’t evidence of absence, or so the argument ran.

What Green and colleagues have added is a quantitative dimension to a debate that had stayed largely qualitative. And the quantitative data doesn’t just confirm the impression that Mohenjo-daro was relatively equal. It shows the city moving in a direction opposite to what the theory predicts.

The method has known limitations, which the authors are explicit about. Gini coefficients of house size are not a comprehensive picture of social inequality. They don’t capture differences in movable wealth, in gender or kinship status, or in access to political power that doesn’t manifest architecturally. A wealthy person might not invest in a larger house. A small structure might house a powerful family. The measure also struggles with people who have no residence at all, the poorest members of any ancient city. These caveats matter.

But the measure does capture one real economic dimension: access to labor and resources at the household level. And it allows direct comparison across sites and time periods, which is why the Global Dynamics of Inequality (GINI) project has assembled this kind of data from hundreds of ancient settlements worldwide. Mohenjo-daro now sits in that database, and what it shows is distinctive.

The governance question

The DK-G South sequence tracks five periods from roughly 2500 to 2100 BC. The earliest period shows a Gini of 0.39 and a median house size of 161 square meters. By the latest period, the Gini is 0.23 and the median has dropped to 141 square meters, but with much less variance. The log standard deviation of house areas falls from 0.73 to 0.44 over the same span. Houses weren’t just smaller on average; they were converging toward a common size.

This convergence happened alongside something else: the city’s residences became more uniformly aligned with Mohenjo-daro’s street grid. In the earlier levels, structures are arranged more loosely. In the later levels, they lock into the urban plan. Green and colleagues read this correlation as suggestive of governance, that the same collective decisions that built and maintained public infrastructure also shaped residential space in ways that narrowed the gap between households.

What that governance looked like in practice is uncertain, and the study doesn’t overclaim. There are no deciphered texts from the Indus Civilization, no royal inscriptions, no administrative archives of the kind that survive from Mesopotamia. But the material record offers indirect evidence for distributed authority. Indus seals, small stone stamps that likely facilitated exchange and credit, were found primarily in private residences at Mohenjo-daro rather than in temples or central administrative buildings. In Mesopotamia, equivalent objects concentrated in institutional contexts, where a palace or temple controlled trade. At Mohenjo-daro, the tools of economic governance were spread across the city’s households. Green has previously characterized this contrast in terms of where the power to govern exchange sat: distributed rather than monopolized, and for that reason probably harder to weaponize against ordinary residents.

The same logic extends to the standardized weights and measures used across the entire Indus region. This wasn’t a system enforced from a single administrative center. It was a shared protocol adopted by communities across a vast area, a form of coordination without centralized control.

Green and colleagues point to the decline in residential disparity coinciding with more structured urban development, and suggest that the same governance producing civic amenities may also have constrained how unevenly residential space was distributed. Green has described Mohenjo-daro as having had “deliberative spaces” where collective decisions about the city were made. The new Gini data, tentatively, supports something like that interpretation. The declining inequality wasn’t random drift. It tracked with urban development in a way consistent with intentional decisions about how the city was organized.

One further pattern in the data is worth noting carefully. As inequality declined, productivity appears to have risen. The GINI project uses mean residence area as a rough proxy for household wealth and productivity. In DK-G South, median residence area bottomed out at about 98 square meters in the Late III Period, around 2200 BC, as the Gini was still falling. Then, in the latest period, when the Gini reaches its low of 0.23, median residence area rebounds to 141 square meters. Gregory Possehl had already noted, separately, that there is more evidence for craft production in the city’s later levels. The authors are cautious here, and rightly so: the sample sizes are small and the relationship isn’t statistically robust. But the pattern raises a pointed question about causality.

Development economics since Kuznets has generally assumed that growth generates inequality, and that only after growth reaches a certain threshold does redistribution become viable. Thomas Piketty’s analysis of capitalist economies complicates this picture but doesn’t overturn the basic direction of the arrow. The Mohenjo-daro data, if interpreted at face value, runs that arrow the other way: collective governance constrained inequality, and declining inequality may have contributed to a subsequent rise in productivity. The city’s most equal period appears to be its most productive.

That is a speculative reading, and the paper says so. But it’s the kind of speculation that makes a dataset useful beyond its immediate empirical scope. Mohenjo-daro isn’t a policy prescription. It is a natural experiment, running across four centuries, that complicates the idea that inequality is an unavoidable cost of urban complexity.

The site still holds a great many unanswered questions. The chronological resolution outside DK-G South is poor; most other excavated areas collapse multiple building phases into a single unit, which almost certainly inflates the overall Gini by mixing large early structures with smaller later ones. Future work revisiting structural chronologies from the rest of the site could sharpen the picture considerably. Projects like M-LAB at the Pratt Institute are already working to reinterpret Mohenjo-daro’s legacy datasets with more rigorous methods.

For now, the numbers from DK-G South say something that’s hard to dismiss: as this Bronze Age city built itself into maturity, its wealthiest households and its poorest ones were moving toward each other. That doesn’t happen by accident.

Further Reading

• Childe, V.G. 1950. The urban revolution. The Town Planning Review 21: 3–17. https://doi.org/10.3828/tpr.21.1.k853061t614q42qh

• Green, A.S. 2020. Debt and inequality: comparing the ‘means of specification’ in the early cities of Mesopotamia and the Indus civilization. Journal of Anthropological Archaeology 60. https://doi.org/10.1016/j.jaa.2020.101232

• Green, A.S. 2022. Of revenue without rulers: public goods in the egalitarian cities of the Indus civilization. Frontiers in Political Science 4. https://doi.org/10.3389/fpos.2022.823071

• Green, A.S. et al. 2025. Kuznets’ tides: an archaeological perspective on the long-term dynamics of sustainable development. Proceedings of the National Academy of Sciences USA 122. https://doi.org/10.1073/pnas.2400603121

• Kohler, T.A. et al. 2025. Introducing the special feature on housing differences and inequality over the very long term. Proceedings of the National Academy of Sciences USA 122. https://doi.org/10.1073/pnas.2401989122

• Kuznets, S. 1955. Economic growth and income inequality. The American Economic Review 45: 1–28.

• Piketty, T. 2014. Capital in the Twenty-First Century. Cambridge, MA: Harvard University Press.

• Possehl, G.L. 2002. The Indus Civilization: A Contemporary Perspective. Walnut Creek, CA: AltaMira.

What they found was dense. Bones packed without obvious vertical structure, skulls clustered toward the jar’s edges, long bones bundled together. When the osteologists worked through the assemblage, counting from the most commonly repeated skeletal elements, they arrived at a minimum of 37 individuals. The highest count came from dentition: 469 observed teeth, producing an MNI of 37. The youngest was approximately 18 months old. Adults of varying ages were present throughout.

“The big jar we’ve found is unique,” Skopal told reporters, “and I’ve seen a lot of jars.”

That uniqueness matters. Hundreds of large stone vessels are scattered across more than 120 known sites on the Xieng Khouang Plateau. The most studied concentration is near Phonsavan, the so-called Plain of Jars, where some jars reach three meters in height and weigh several tons. The first systematic survey was done by French archaeologist Madeleine Colani in the 1930s. She rejected the idea that the vessels were storage containers and proposed a funerary role. A local legend holds that giants used them to brew rice wine. Scholarship since Colani has leaned toward mortuary use, but the physical evidence has been thin. A few jars contained ash and burned bone fragments. Some had burial pits dug nearby. Most were empty, which made the funerary hypothesis difficult to confirm with any confidence. If each jar was a tomb, why did so few of them hold human remains?

Jar 1 makes that question answerable, and the answer involves movement.

A Multistep Death

The critical detail about the bones inside Jar 1 is their disarticulation. Most joints had already separated before the remains were placed inside. The bones had not come from bodies laid directly into the jar. They came from bodies that had decomposed somewhere else first. This is secondary burial: the deliberate reinterment of already-processed skeletal material after an initial phase of decomposition away from the final deposit.

The paper’s most consequential observation follows from this. About 500 meters west of Jar 1, the team found a second cluster of seven smaller stone jars. None of them contained human remains. The ceramics found near these smaller vessels are consistent with the assemblage from Jar 1’s surrounding context, suggesting the two locations were used at the same time by the same community. The working hypothesis is that the smaller jars served as the initial stage: a newly dead person would have been placed in one, the flesh allowed to separate from the bone, and then the cleaned skeletal remains transferred to the large jar.

Skopal described the logic to Science News:

“Maybe they used those [smaller] stone jars to ‘distill’ the bodies, so when someone died, they might have put the body in there so all the flesh came off. Then they took the bones and they put them in this big jar, so it’s almost like a crypt.”

But the process may not have ended there. The team noted a discrepancy between the minimum individual count from dental remains and the lower count derived from cranial and postcranial elements alone. Bones appear to be missing. Some may have been selectively removed from Jar 1 after deposition, possibly for reinterment at a habitation site or place of worship. Stone slabs found inside the jar are comparable to those used as markers for secondary bundle burials at other Plain of Jars sites. Comparable practices of secondary processing and relocation of skeletal remains are documented across northeast India, Myanmar, Thailand, and Vietnam during the Iron Age and Historical periods. The large jar, on this reading, was not the end of the journey. It was a stage in an ongoing process.

The radiocarbon dates establish how long that process ran. Eight bone and tooth samples returned results spanning approximately cal AD 890 to 1160, a range of roughly 270 years. Charcoal from pottery deposited in an adjacent trench dated to cal AD 890 to 1020, consistent with the early end of the skeletal dates. The jar was not filled in a single event. Generations of a family, or an extended kin group, returned to it repeatedly.

“The number of individuals also suggests the jars were owned by family or extended family groups,” Skopal noted in the press release accompanying publication. “They likely served as places where ancestral rites were performed over generations.”

Four skulls were found with their mandibles still in close association, a small departure from the general pattern of disarticulation. The jaw is typically among the first elements to separate during decomposition. Their preservation together raises the possibility that in some cases, a recently dead or only partially processed individual was placed inside the jar directly, perhaps with greater ceremony. The osteological analysis is ongoing.

One other detail is worth holding onto. At least two individuals show deliberate tooth ablation: the intentional removal of specific teeth during life. One person lost both upper lateral incisors; another lost both lower central incisors. Tooth ablation is documented widely across prehistoric and historical mainland Southeast Asia and southern China, where it appears associated with social identity and life transitions. Its presence at Site 75 places this community within a broad pattern of shared bodily practice across the region.

What the Beads Know

Twenty glass beads came out of the jar during wet-sieving of the sediments. Orange, turquoise-blue, dark blue, black, red-over-orange. Twelve were selected for chemical analysis by laser ablation mass spectrometry at the Field Museum in Chicago, and the results upend any assumption that the Xieng Khouang Plateau was a remote backwater.

Nine of the beads matched a mineral soda-high alumina composition produced in eastern South India, a glass type that circulated across South and Southeast Asia from roughly the end of the first millennium BC through the eleventh century AD. Two beads had a soda plant-ash composition whose chemistry aligns most closely with Mesopotamian glass, the type that appears at Southeast Asian sites dating from the eighth or ninth century through the eleventh century AD, before later sites begin yielding Egyptian glass instead. The team notes that the temporal overlap between the Mesopotamian glass signature and the radiocarbon dates from the jar is consistent. One dark-blue bead had a potash-rich composition more commonly dated to 300 BC through AD 300, most likely from southeastern China or northern Vietnam. The team treats it as a curated antique, a much older object that had been kept, traded, or passed down.

No glassmaking tradition is known from the Lao highlands. These beads arrived from elsewhere, through exchange networks that extended from the Laotian interior to the Indian subcontinent and the Near East. They are not rare luxury goods in the way that term sometimes gets misused in archaeology; they are trade goods, the kind of thing that moved in bulk along commercial routes. Their presence in a burial jar in upland Laos means the people using that jar were connected to those routes.

The timing sharpens the point. The period cal AD 890 to 1160 sits within an era of expanding commercial connectivity across Asia. The Song Dynasty in China was actively intensifying long-distance trade. The Khmer Empire was at its height. The Dali Kingdom in Yunnan extended into parts of northern Laos. The Đại Việt and Champa kingdoms controlled Vietnam. The Pagan Kingdom held Myanmar. All of these states were embedded in overland and maritime exchange networks that moved goods, technologies, and cultural practices across vast distances. The team proposes that the jar sites may have sat at nodal points along routes connecting East and Southeast Asia, the plateau elevated enough to serve as a waypoint, the communities there integrated into the wider system. The Mesopotamian beads in Jar 1 are physical evidence for the highland end of that connectivity.

Miriam Stark, an anthropologist at the University of Hawaii not involved in the study, told Science News that the collective mortuary assemblage was exactly what she had hoped to see discovered. But she pressed on the gap that the data still cannot close:

“I do wonder, where did these people live?”

No settlements associated with the Plain of Jars tradition have ever been found. No houses, hearths, or refuse deposits. The entire archaeological record of these communities consists, almost entirely, of what they did with their dead. The jar sites are the only material evidence of who they were, and most of those sites are empty.

Researchers already knew this. What they didn’t know, until now, was when.

A new study published in Proceedings of the National Academy of Sciences1 addresses that question directly, using oxygen isotope analysis of the shells themselves. The chemistry is elegant in its logic: as a mollusk grows, it incorporates oxygen from surrounding seawater into its shell carbonate, and the ratio of heavier to lighter isotopes shifts with water temperature. Reconstruct that variation across the shell’s growth rings, and you get a record of seasonal temperature change. You get, in other words, a calendar.

Lead author Asier García-Escárzaga and colleagues from the Universitat Autònoma de Barcelona, the University of Burgos, and the University of Cantabria applied this method to the Los Aviones assemblage at unusually fine resolution. What came back was not random. The shells told a story of deliberate, repeating behavior: collection concentrated in late autumn, winter, and early spring, with a falloff in the warmer months.

Read more

They weren’t gone. Not for a very long time.

A new study led by geochemist Kendra Chritz at the University of British Columbia, published in the Proceedings of the National Academy of Sciences,1 draws on isotopic data from 111 individuals spanning roughly 9,500 to 230 years before the present across Kenya and Tanzania. The researchers analyzed stable isotopes preserved in tooth enamel and bone collagen — chemical signatures that accumulate as teeth form and that record, with surprising specificity, what a person was actually eating. They supplemented this skeletal evidence with leaf wax isotopes and lipid residues extracted from ancient ceramic cooking pots.

The picture that emerged doesn’t fit the transition model neatly. Eastern Africa’s earliest pastoralists weren’t eating like pastoralists.

Read more

Then someone buried him.

The site is in Kinchega National Park in western New South Wales, about 100 kilometers southeast of Broken Hill, along a meander bend of the Baaka where a palaeochannel marks the river’s older course. The dingo — known as garliin Barkindji language — came to light in the early 2000s when roadworks cut into the edge of a riverside midden and erosion began exposing his skeleton. Barkindji Elder Uncle Badger Bates and National Parks and Wildlife Service archaeologist Dan Witter identified the remains. Subsequent flooding in 2021 dislodged the skull. In September 2023, a salvage excavation was carried out by researchers from the University of Western Australia, the Australian National University, the Australian Museum, and the University of Sydney, in close partnership with the Menindee Aboriginal Elders Council. Before the excavation began, Barkindji Elders conducted a smoking ceremony. Earlier this year, Garli was returned to Country.

The results, published in Australian Archaeology1 in May 2026, are striking in several respects. Garli represents the first directly dated dingo burial from the Baaka system, pushing the known range of this practice further north and west than previously documented. Radiocarbon dating of a caudal vertebra placed his burial between 963 and 916 years ago. But what makes this case genuinely unusual is not the burial itself. It is what apparently happened afterward.

The midden in which Garli was interred continued to receive additions of freshwater mussel shells (Velesunio sp.) for roughly five centuries after his death. Radiocarbon dates on shells recovered from different parts of the midden are consistently and substantially younger than the dingo’s bones, some by several hundred years. The youngest shell dates cluster between roughly 542 and 495 years ago. Barkindji Elders interpret this sequence as a feeding ritual — an ongoing practice of honoring the buried garli as an ancestor, maintained across multiple generations. The research team believes this is the first time such a post-burial feeding practice has been identified archaeologically anywhere in the world.

Read more

The inexplicably part has long been the problem. There is no shortage of proposed explanations. Language lateralization, tool use, social organization, diet, brain size, habitat. Pick your hypothesis; there are papers defending each of them. What has been missing is a framework rigorous enough to test them all at once, across enough species, while controlling for the fact that closely related species share traits simply because of shared ancestry. A study published in PLOS Biology1 by Thomas Püschel, Rachel Hurwitz, and Chris Venditti does exactly that.

Their dataset spans 2,025 individuals across 41 anthropoid species. Their method is a Bayesian phylogenetic comparative meta-analysis, which is the kind of approach that lets you ask, genuinely, whether a trait is unusual given where a species sits on the tree of life, and which variables actually explain it. When they applied this framework, H. sapiens stood out immediately as a statistical outlier. Not slightly unusual. Dramatically, conspicuously outside the range the model would predict. Then, when two variables were added to the model, brain size and the ratio of arm length to leg length, humans stopped looking anomalous. Their handedness became statistically predictable.

That arm-to-leg ratio, called the intermembral index, is a standard anatomical proxy for locomotion. Humans have an unusually low value because our legs are so much longer than our arms. That is a signature of bipedalism. Among other primates, a low ratio correlates with a leftward hand bias; among hominins, the freeing of the upper limbs from locomotor function appears to have done something very different. The hands, no longer needed for weight-bearing and quadrupedal movement, became available for an entirely new range of fine, asymmetric behaviors. The selective pressure to favor one hand, and favor it consistently, intensified.

Read more

It is, according to a new database assembled by Jennifer French, Somaye Khaksar, Agustín Fuentes, and Marc Kissel, the oldest known mobile container in the archaeological record.

That fact alone is worth sitting with. The bark tray at Kalambo Falls is not a pot. It predates pottery by several hundred thousand years. It is not a bowl from a settled farming village. It is a portable object, made to be carried, by a hominin whose name we may never know, at a time so deep in our evolutionary past that most of what those individuals made, thought, and carried has entirely ceased to exist.

The new database, published in the Journal of Anthropological Archaeology,1 documents 739 Pleistocene mobile containers drawn from 210 sites across Africa, Asia, Europe, and the Americas. The previous published estimate of Pleistocene containers stood at 22. That gap is not primarily a reflection of new discoveries. It is a reflection of what happens when researchers go looking systematically for something that has mostly been ignored.

Read more

This is the kind of detail that sits awkwardly inside a prevailing narrative, and a new review paper published today in Science Advances1 by David Anthony, Martin Trautmann, and Volker Heyd intends to press on the awkwardness. Their argument, built from archaeozoology, ancient DNA, osteology, and a careful reading of what the genetics literature actually says versus what it has sometimes claimed, is that horse domestication did not happen at 2200 BCE. It was well underway centuries, possibly millennia, earlier.

The papers they are responding to are significant. Between 2021 and 2025, a series of high-profile studies from equine genetics teams led by Ludovic Orlando and William Taylor argued that domestication, in any meaningful sense, should be equated with the emergence of favorable mutations in a horse lineage called DOM2, concentrated between roughly 2200 and 2100 BCE. One mutation, at the GSDMC locus, is associated with back endurance and front leg strength during riding. Another, at ZFPM1, relates to fear and anxiety responses, potentially making horses calmer around humans. When these mutations spread and DOM2 horses rapidly replaced other horse populations across Eurasia after 2200 BCE, the geneticists argued, that was domestication. What came before was taming, at best.

Anthony and colleagues do not dispute that this genetic transformation happened, or that it mattered. What they dispute is the interpretive move of treating a specific genetic signature as the definition of domestication, and then reasoning backward from its absence in earlier populations to conclude that horses were not yet domesticated, not yet ridden, and not yet integrated into human economies in ways that would qualify as domestic.

Read more

We know this because the tooth survived. It came out of Chagyrskaya Cave in southwestern Siberia as specimen Chagyrskaya 64, a lower second molar from an adult individual, and when paleoanthropologist Alisa Zubova of the Russian Academy of Sciences and her colleagues examined it, something was immediately wrong. There was a hole in it. Not a small one — a large, irregular concavity measuring 4.2 mm long, 2.8 mm wide, and 2.6 mm deep, positioned at the center of the occlusal surface and reaching all the way down to the floor of the pulp chamber. Around the edges, visible at magnifications up to 500x, were fine linear striations. The kind left by a rotating tool.

“We were intrigued by the unusual shape of the concavity on the tooth’s chewing surface,” said Zubova in a statement accompanying the study. “It differed from the normal morphology of the pulp chamber and did not match the typical pattern of carious lesions seen in Homo sapiens. Moreover, distinctly visible scratches suggested that the concavity was not the result of natural damage but of intentional actions.”

Their analysis, published May 13, 2026 in PLOS One,1 presents a compelling case that Chagyrskaya 64 carries the oldest known evidence of deliberate dental treatment in human evolutionary history — predating the previous record holder by more than 40,000 years.

What the Tooth Actually Shows

Before accepting the drilling interpretation, the team had to work through the alternatives. A deep concavity in an old molar could plausibly result from severe wear, a fracture, or natural carious decay. None of these fit.

Dental trauma leaves sharp, fractured edges. The concavity walls in Chagyrskaya 64 are smooth and rounded. Extreme attrition can expose the pulp chamber, but it cannot widen it — and the upper portion of the concavity is broader than the pulp chamber below, which rules out wear as the cause. The comparison sample from the same stratigraphic layer, another Chagyrskaya molar recovered from identical sediment, shows intense occlusal wear but a completely flat chewing surface, no concavity, and uniformly mineralized dentin.

Micro-CT imaging told a different story for Chagyrskaya 64. The tooth showed pervasive demineralization of primary dentin, corresponding to Grade 4 or 5 on the Downer diagnostic scale — severe, deep caries extending into both inner and outer dentin. The absence of secondary dentin in a heavily worn tooth is itself anomalous; normally, the pulp responds to attrition by laying down reparative tissue. Here it hadn’t, because the infection had already destroyed it. There was no secondary dentin to form.

The morphology of the concavity also doesn’t match anything caries produce naturally. Bacterial decay works by chemical demineralization, creating soft tissue breakdown and irregular cavities. It does not excavate clean, sub-rounded depressions with parallel microstriations on the walls. Those striations, analyzed under scanning electron microscopy, are oriented parallel to the concavity walls and exhibit a corrugated, ridged structure consistent with a rotating tool edge. They appear in at least two discrete locations, at slightly different depths, suggesting the tool was worked from multiple angles.

And critically: over the striations, there is polish. Ante-mortem wear, from years of chewing after the procedure was done.

Drilling in the Dark

To test whether a jasper perforator could actually produce this kind of concavity, Zubova and colleagues ran a series of experiments on three modern Homo sapiens molars. Two came from undocumented Holocene archaeological collections. The third — an upper left third molar with an untreated cavity and no previous dental work — was contributed by one of the authors. The tools were fabricated from local jasperoid raw material, matching what was available at Chagyrskaya Cave.

The experiments were conducted by a single researcher experienced in Paleolithic stone knapping, with a small amount of water added to simulate the oral environment. The technique that worked was manual rotation — holding the pointed perforator between two fingers and drilling with a twisting motion, not scraping. Scraping spread force too broadly and produced only surface striations. The rotational method disrupted the tooth surface within seconds.

Penetrating through to the pulp chamber took between 35 and 50 minutes of continuous drilling. The research team notes this is probably a conservative estimate for the original procedure. Working inside a living person’s mouth, with limited visibility, constrained access, and an uncooperative patient, could easily double the time.

Neanderthals have thinner enamel than Homo sapiens, distributed over a larger volume of coronal dentin. The infection in Chagyrskaya 64 had already softened the dentin considerably. Both factors likely made the drilling faster. Still: somewhere between one and two hours of someone boring into your tooth with a sharp rock.

Raman spectroscopy of the tooth surface detected no residue of plant material, resins, or analgesics. The team notes this doesn’t rule out their use — if the individual lived and chewed on the tooth for years afterward, any herbal packing would long since have worn away. But it leaves open the possibility that no painkiller was used at all.

The experimental results matched the archaeological tooth closely. The groove profiles on both share a similar V-shaped morphology, with gently sloping outer walls. The width of the experimentally produced grooves (0.249 mm) closely approximates what’s preserved on Chagyrskaya 64 (0.294 mm). Macrotraces are more pronounced in the experimental specimens — expected, given that they hadn’t then spent 59,000 years in sediment with subsequent wear on top. Microtraces on the original molar are accordingly faint, but present.

The concavity itself is not a single depression. Micro-CT shows three partially overlapping depressions that together occupy the entire pulp chamber volume. Whether they were drilled in separate episodes or represent the practitioner widening the cavity as they worked isn’t clear from the preservation. The team’s experimental replication of a three-depression configuration suggests the smallest of the three was probably an early stage, with the two deeper ones expanded after initial penetration. One possibility is that the third depression targeted a separate area of carious tissue. Another is that it was an error — stone tools aren’t precision instruments.

The Broader Picture

Caries are rare among Neanderthals. Fewer than ten cases have been identified across the entire fossil record, from sites including Palomas, Bau de l’Aubesier, De Nadale, and Kebara. All the previously known cases involve minor lesions confined to enamel or upper dentin, none reaching the pulp chamber. Chagyrskaya 64 is in a class by itself.

The Chagyrskaya population appears to have been particularly susceptible. Another tooth from the same cave, Chagyrskaya 18, is a naturally shed deciduous molar from a Neanderthal child of about 9 to 11 years. It carries two carious lesions. The smaller of the two has already breached through the enamel into the upper dentin. The larger is still confined to enamel but actively demineralizing. Two caries cases in a small population is unusual for Neanderthals; the team suggests it may indicate the presence of specific cariogenic bacteria in the local oral microbiome rather than a dietary shift, since isotopic evidence from Chagyrskaya places these individuals squarely within the dietary range of other European Neanderthals.

Chagyrskaya 64 also carries a second ante-mortem modification: a pronounced interproximal groove on its distal surface, with microscopic parallel striations characteristic of Stage 4 toothpick use-wear. A nascent groove is also forming on the mesial surface. The same tooth was toothpicked on both sides and then drilled. The toothpick groove sits precisely where demineralization is concentrated in the cervical area — over a second, smaller carious lesion. The drilling concavity sits over the larger lesion on the occlusal surface. Two interventions, two lesions, two tools.

Toothpicking behavior predates H. neanderthalensis considerably. Interproximal wear grooves appear on teeth attributed to Homo habilis from Olduvai Gorge, and the same wear pattern has been documented in Japanese macaques (Macaca fuscata). The behavior itself isn’t cognitively remarkable. But drilling is something else. The rotational technique requires coordinated fine motor control, sustained application of force in a confined space, and — given the documented grain of the striations — deliberate adjustment of the tool’s angle during the procedure. The team notes that analysis of bone retouchers from Chagyrskaya Cave shows Neanderthals there typically held tools between two fingers during knapping, a technique already requiring significant digital precision.

The previous oldest evidence of dental intervention belongs to an individual from Ripari Villabruna in northeastern Italy, dated to around 14,000 years ago. That tooth, a third molar from a late Upper Paleolithic H. sapiens, shows traces of enamel scraping around a carious lesion. Micro-CT of the Villabruna tooth shows substantial retention of demineralized tissue — the scraping was superficial, limited largely to the enamel surface. It would not have relieved the pain of deep infection. Chagyrskaya 64 represents not just an earlier intervention but a more effective one, by a population whose tool-use repertoire apparently included the insight that drilling into the pulp chamber — deliberately and completely — would stop the pain by destroying the nerve.

Whether this constitutes knowledge in any formal sense, or something closer to problem-solving arrived at under extreme duress, is a question the tooth cannot answer. What it does answer is whether the capacity existed. The individual who drilled Chagyrskaya 64 identified a source of pain, selected an appropriate tool, executed a controlled invasive procedure, and their patient then went on to chew with the treated tooth for long enough to wear it smooth.

“What struck me, and continues to strike me, is what an incredibly strong-willed person this Neanderthal must have been,” said Lydia Zotkina, co-author of the study. “Now, every time I go to the dentist, I think about that guy.”

Further Reading

• Oxilia G, Peresani M, Romandini M, Matteucci C, Spiteri CD, Henry AG, et al. Earliest evidence of dental caries manipulation in the Late Upper Palaeolithic. Scientific Reports. 2015;5:12150. https://doi.org/10.1038/srep12150

• Kolobova KA, Roberts RG, Chabai VP, Jacobs Z, Krajcarz MT, Shalagina AV, et al. Archaeological evidence for two separate dispersals of Neanderthals into southern Siberia. PNAS. 2020;117(6):2879–85. https://doi.org/10.1073/pnas.1918047117

• Mafessoni F, Grote S, de Filippo C, Slon V, Kolobova KA, Viola B, et al. A high-coverage Neandertal genome from Chagyrskaya Cave. PNAS. 2020;117(26):15132–6. https://doi.org/10.1073/pnas.2004944117

• Skov L, Peyrégne S, Popli D, Iasi LNM, Devièse T, Slon V, et al. Genetic insights into the social organization of Neanderthals. Nature. 2022;610(7932):519–25. https://doi.org/10.1038/s41586-022-05283-y

• Faux P, Ding L, Ramirez-Aristeguieta LM, et al. Neanderthal introgression in SCN9A impacts mechanical pain sensitivity. Communications Biology. 2023;6(1):958. https://doi.org/10.1038/s42003-023-05286-z

• Estalrrich A, Alarcón JA, Rosas A. Evidence of toothpick groove formation in Neandertal anterior and posterior teeth. American Journal of Physical Anthropology. 2017;162(4):747–56. https://doi.org/10.1002/ajpa.23166

That tooth, and five others like it from two additional sites across China, have just yielded the first informative molecular data ever recovered from Homo erectus. The results, published this month in Nature1 by a team led by paleogeneticist Qiaomei Fu of the Chinese Academy of Sciences’ Institute of Vertebrate Paleontology and Paleoanthropology, connect these 400,000-year-old individuals to the Denisovans -- and, through them, to people alive today.

The breakthrough didn’t come from DNA. It couldn’t. Ancient DNA degrades quickly; even under ideal preservation conditions, retrievable sequences rarely survive beyond a few hundred thousand years, and the specimens from Zhoukoudian, Hexian in Anhui Province, and Sunjiadong in Henan Province are far older than what DNA analysis can reach. Instead, Fu’s team turned to proteins.

Proteins survive longer than DNA because they are structurally sturdier, and because tooth enamel -- the hardest tissue in the vertebrate body -- protects them especially well. Enamel is a mineral matrix, and proteins trapped inside it can persist for millions of years in fragmentary form. They are not genetic material themselves, but they are produced by genes and carry the signature of the genetic variants that shaped them. As University of Copenhagen biochemist Enrico Cappellini, who was not involved in the study, put it in comments to Science: proteins are,

“a direct expression of genes, so in a sense they carry genetic information. It’s the best proxy for DNA.”

Retrieving those proteins from a prized and irreplaceable fossil is a different kind of problem. Most paleoproteomic studies drill or cut into bone or tooth to extract a powdered sample -- destructive work that curators of important specimens understandably resist. Fu’s team used a method called acid etching: the tooth is wrapped in a waterproof film, a patch of enamel just a few millimeters square is exposed, then briefly treated with acid. The technique removes just enough material for analysis while leaving the specimen physically intact except for slight surface discoloration. The teeth from Zhoukoudian, Hexian, and Sunjiadong emerged with their morphology preserved.

What came out of that process was a catalog of 11 ancient enamel proteins covering hundreds of amino acid positions. Across all six individuals, from all three sites, the team identified two amino acid variants in a single protein: ameloblastin, which plays a key structural role in enamel formation.

Two variants, two stories

The first variant, designated AMBN(A253G), has never been seen before in any other hominin or primate. It wasn’t in the 1.77-million-year-old H. erectus tooth from Dmanisi, Georgia -- the only previous specimen to yield H. erectusproteins, though those had been too degraded to reveal informative variants. It wasn’t in Homo antecessor from Atapuerca. Not in Neanderthals, not in Denisovans, not in any modern human population. The variant appears to be specific to this population of Middle Pleistocene H. erectus from East Asia.

That exclusivity makes it useful. Fossils attributed to H. erectus across Asia are morphologically variable enough that researchers have sometimes disagreed about their relationships. The Hexian specimens, in particular, look morphologically closer to Indonesian H. erectus than to those from Zhoukoudian, and some researchers had proposed they might be more closely related to Denisovans. The AMBN(A253G) variant settles that debate. It appears in both Hexian teeth, just as it does in the Zhoukoudian and Sunjiadong teeth, clustering all six individuals together in a Bayesian phylogenetic tree with 100% posterior probability. Hexian belongs to H. erectus.

The second variant is where things get complicated. AMBN(M273V) had already been identified in Denisovans. The known Denisovan specimens include Denisova 3 from the cave in Siberia that gave the group its name, the “Dragon Man” skull from Harbin, the Penghu mandible from Taiwan’s seafloor, and a molar from Laos -- geographically scattered, morphologically distinct from Neanderthals and modern humans, linked almost entirely through ancient DNA and protein analysis rather than anatomy. When the Altai Neanderthal genome was analyzed, researchers found that the Denisovan lineage had received between 0.5% and 8% of its genome from a “super-archaic” hominin whose ancestors had diverged from the common lineage of Neanderthals, Denisovans, and modern humans more than a million years ago. They also found that roughly 15% of those super-archaic DNA regions later introgressed from Denisovans into populations in Asia and Oceania.

The AMBN(M273V) variant fits this pattern. It is absent from most modern human populations but present at 21% frequency in the Philippines, 1.17% in India, and 0.71% in Papua New Guinea -- exactly the populations expected to carry the highest proportions of Denisovan ancestry. It is homozygous in the more recent Denisovan specimens, Denisova 3 and Penghu 1, but heterozygous in two older Denisovans, the Harbin individual and Denisova 25 (dated to roughly 200,000 years ago). That pattern -- heterozygous earlier, homozygous later -- is consistent with an allele that entered the Denisovan lineage from outside and increased in frequency over time.

Fu’s team argues that the “outside” is H. erectus. The variant appears in all six H. erectus individuals across three sites spanning both northern and southern China. Its genomic region shows greater sequence divergence between modern Africans and Denisova 3 than between Africans and the Altai Neanderthal -- a signal that the region is older than the Neanderthal-Denisovan split, consistent with origin in a more diverged group. Denisovans diverged from Neanderthals approximately 380,000 to 470,000 years ago. The H. erectus specimens are dated to roughly 400,000 years ago. The timing, the geography, and the allele frequency pattern all point in the same direction: H. erectus populations in East Asia encountered Denisovans around or before this period, and something passed between them.

“Their shared habitats create opportunities for interactions,” Fu and colleagues write in the paper.

The scenario they reconstruct goes: H. erectus populations in East Asia, the same individuals whose teeth now sit in Beijing, Hefei, and Luoyang, passed the AMBN(M273V) variant to Denisovans through interbreeding, probably more than 400,000 years ago. Denisovans carried it forward, homozygosity increasing over time as it became more established in the population. Modern humans moving out of Africa and through Asia perhaps 50,000 years ago encountered Denisovans, interbred, and acquired it. A small proportion of people alive today -- primarily in island Southeast Asia and the Pacific -- carry a protein variant in their tooth enamel whose ancestry traces back to Homo erectus on the Chinese mainland almost half a million years ago.

What remains genuinely uncertain

Not everyone finds the causal chain fully convincing. Yale geneticist Diyendo Massilani, speaking to Science, noted that the study provides no direct evidence of admixture: each group could theoretically have evolved the M273V variant independently, through convergent mutation rather than shared ancestry. He characterized the admixture interpretation as hypothetical.

Kirsty Penkman at the University of York, a geochemist who specializes in ancient proteins and was not involved in the research, pushed back on that concern. Enamel proteins are structurally constrained -- they cannot mutate extensively without losing their function. A shared variant in an enamel protein is therefore less likely to represent independent evolution than a shared variant in a less constrained region of the genome might be,

“A biomineral protein can’t mutate too much,” she told Science, “because then it stops doing its job.”

The honest answer is that proteins cannot resolve this question the way genomic data can. Full H. erectus genome sequences would make the case definitively, but those sequences don’t exist and may never be recoverable. The DNA in these specimens is gone. What Fu’s team has done is work with what survives. John Hawks at the University of Wisconsin-Madison, also not involved in the study, made a point worth taking seriously,

“Scientists used to call this ‘the muddle in the Middle Pleistocene,’” he told LiveScience, “and now we know that muddling is just mixing.”

But he also raised a deeper question that the study implicitly surfaces: whether paleoanthropologists have been too loose with the label Homo erectus, grouping together fossils from the Middle Pleistocene in China that may actually represent Denisovan relatives or other populations not yet identified. The protein data confirms that the six specimens from Zhoukoudian, Hexian, and Sunjiadong belong together. It says less about where the boundaries of H. erectus as a category ultimately fall.

That is, for now, where the field sits. A small patch of acid-etched enamel, a handful of proteins, and two amino acid variants are the most informative molecular data ever extracted from a species that spent nearly two million years shaping the inhabited world. As Penkman put it: fewer than 20 hominins have ever yielded analyzable protein sequences. Each new specimen adds something. This study adds six at once -- and what they add is the outline of an encounter, written in the crystalline architecture of a tooth.

Further Reading

• Welker, F., et al. (2020). The dental proteome of Homo antecessor. Nature, 580, 235–238.

• Prüfer, K., et al. (2014). The complete genome sequence of a Neanderthal from the Altai Mountains. Nature, 505, 43–49.

• Hubisz, M.J., Williams, A.L., & Siepel, A. (2020). Mapping gene flow between ancient hominins through demography-aware inference of the ancestral recombination graph. PLoS Genetics, 16, e1008895.

• Tsutaya, T., et al. (2025). A male Denisovan mandible from Pleistocene Taiwan. Science, 388, 176–180.

• Fu, Q., et al. (2025). The proteome of the late Middle Pleistocene Harbin individual. Science, 389, eadu9677.

This is one of two double burials at the heart of a study published in 2026 in the International Journal of Osteoarchaeology,1 led by Alyson Caine and colleagues. The paper documents skeletal evidence of scurvy across multiple Late Holocene archaeological sites in California, and it is worth pausing on the specific strangeness of what it finds. California is not a place one associates with nutritional scarcity. The Bay Area and Central Valley supported extraordinarily diverse economies. Mussels and oysters line the margins of CA-ALA-11 itself. Berries, clover, tarweed, goosefoot, fresh fish, waterfowl, and deer were all available to the populations who lived there. And yet: scurvy. In infants who never drew an independent breath of air.

The puzzle is the point.

Read more

That context has now arrived, though the delivery mechanism is unusual: satellite imagery. A team led by Julien Cooper at Macquarie University, working under the Atbai Survey Project, has spent years scanning the desert between the Nubian Nile and the Red Sea from orbit. What they found, documented in a 2026 paper in African Archaeological Review,1 is a burial tradition of unexpected scale. Across the entire Atbai Desert, from the borderlands of Upper Egypt down to Eritrea, the survey identified 280 circular stone enclosure monuments. Twenty were already on record from earlier fieldwork. The remaining 260 had simply never been seen before.

The desert had been hiding them in plain sight.

Read more

At FwJj 80, a site within the Koobi Fora Formation in northern Kenya, researchers recovered and analyzed1 more than a thousand fossilized bone specimens dating to roughly 1.6 million years ago. The assemblage sits within what geologists call the KBS Member — a stratigraphic interval that has produced significant early Homo fossils over the decades but has, until now, yielded very little in the way of faunal material suitable for detailed zooarchaeological analysis. Most of what we knew about early Homo foraging behavior at Koobi Fora came from the younger Okote Member, some 40,000 to 200,000 years later. FwJj 80 changes that.

The bones are predominantly from antelopes and other grazers. Examined under high-powered magnification, many bear microscopic scratches and percussion marks consistent with stone tool use: cut marks from butchery, hammerstone damage from marrow extraction. The spatial patterning of those marks matters. Sharp cut marks positioned toward the midshaft of limb bones — where muscle attaches in greatest bulk — indicate that whoever was processing these carcasses got to them early, before carnivores had stripped the meat. The team’s analysis found minimal evidence of carnivore tooth marks in primary positions, which would suggest the reverse sequence: a predator kills, eats first, and hominins arrive at the leftovers.

That sequence, early hominin access followed by intensive processing, matches what has been documented in the Okote Member at Koobi Fora. It also matches the pattern at FLK Zinj in Tanzania, dated to around 1.84 million years ago, and at Kanjera South in Kenya, which pushes the record back to roughly 2.0 million years. Three sites, three different environments, spanning nearly half a million years. The consistency is what the researchers find significant.

Read more