Nataruk: How a Nature Paper on a Massacre Was Contested Within Months

In 2016 Nature reported a massacre of foragers at Nataruk, Kenya, c. 10,000 years ago. Months later specialists disputed it. Both readings.

12 November 2026 · Part of The Stone Age — a guide to 13 articles →

In January 2016 the journal Nature published an account of a killing ground on the western shore of Lake Turkana in Kenya, dated to roughly ten thousand years ago and presented as the earliest evidence of a massacre among hunter-gatherers. Ten months later, in the same journal, other specialists argued that the bones could not carry that weight. The dispute has never been settled, and it is a good case of how a prehistoric claim is made, tested and left open.

The site and the claim

Nataruk lies in West Turkana, in the arid northwest of Kenya. When the people there died, the area was not desert: it was the edge of a lagoon fed by a much larger Lake Turkana, in the wetter climate of the early Holocene. The remains were found in sediment that once lay at the water's margin.

The excavation team, led by Marta Mirazón Lahr of the University of Cambridge, reported the remains of at least twenty-seven individuals, of whom twelve were relatively complete, articulated skeletons. Ten of those twelve, in the team's reading, showed signs of a violent death. The team described blunt-force damage to skulls, wounds consistent with projectiles, and obsidian fragments lodged in bone. They also reported that the bodies had not been buried. In their account the dead were left where they fell, at the water's edge, and were later covered by sediment.

The team's conclusion was stated plainly in the paper's title and framing: this was inter-group violence, and in their view it was the earliest known case of a specific attack on a group of foragers. They also pointed to details they read as signs of a deliberate assault on a mixed group, including children and at least one woman they thought to be in the late stage of pregnancy.

Why the claim mattered

The significance did not lie in the number of dead. Nataruk was reported to be a group of nomadic foragers, not a settled community with stores, fields or a defended enclosure. Many arguments about the origins of war lean on the idea that organised killing needs property worth taking, so that violence follows farming. A massacre among mobile hunter-gatherers would count against that idea, or at least complicate it.

The paper was therefore read as a statement in an old debate. One side holds that deadly group conflict is ancient and widespread in human societies. The other holds that it is largely a product of settled life and rising social complexity. Nataruk was quickly cited by the first camp. For the wider record, see our look at the Linear Pottery mass graves, where the argument concerns farmers, not foragers.

The challenge

In November 2016, Christopher Stojanowski, Andrew Seidel, Laura Fulginiti, Kent Johnson and Jane Buikstra published "Contesting the massacre at Nataruk" in Nature. Their argument ran on several lines.

First, they said that much of the damage labelled as perimortem trauma could have formed after burial. Bone buried for ten thousand years is subject to pressure from the sediment around it, and they attributed lesions on three crania to soil compression rather than to blows. This is the same problem that sits underneath the whole field, which we discuss in why "perimortem" is the hardest word in osteology.

Second, they argued that the pattern of the bodies did not look like documented massacre sites. By their reading the skeletons were mostly articulated, spatially organised and not commingled, with little variation in body position. They contrasted this with the disordered deposits at places such as Talheim, where bodies were thrown together in a pit.

Third, they disputed details of body position, including how many individuals were prone, and suggested that positions might reflect a burial tradition tied to age and not the circumstances of death.

The reply

The excavators answered in the same issue. Their reply maintained that the critique was incorrect on both of its main counts and that inter-group conflict remained the best explanation for the assemblage. They did not withdraw the claim.

A reader should be careful here about what each side has actually shown. The critics did not demonstrate that no violence occurred; they argued that the published evidence was less secure than the headline and that some alternative explanations had not been excluded. The excavators did not claim certainty either; they argued that the whole pattern of injuries, taken together, was best explained by an attack. This is a disagreement over interpretation of the same skeletons, not a dispute over whether the skeletons exist.

Two readings side by side

Why the verdict stays open

Neither side can go back and examine a fresh deposit. Excavation destroys the context it records, and the lagoon sediments at Nataruk have already been dug. Further argument depends on re-examination of the bones, the photographs and the field records, and on how much confidence one places in the criteria used to separate perimortem breaks from later ones.

There is a useful contrast with an older site. At Jebel Sahaba in Sudan, a 2021 restudy found that many individuals carried healed wounds alongside fatal ones, which changed the reading from a single event to a long pattern of raiding; the story is told in what the restudy did to the oldest massacre. Healing is a clear marker of survival and therefore of an earlier injury. Nataruk offers no such marker, because all the injuries debated there are taken to be fatal or postmortem.

Other cases show the same split between clear and contested evidence. At Ofnet the question is whether severed heads reflect killing or ritual. At Herxheim members of the same excavation team disagree. At San Juan ante Portam Latinam the argument is about how many arrowheads in a rock shelter amount to a massacre.

How our record treats it

Our register of collective violence before 3000 BCE includes Nataruk, and it rates the certainty as low. That rating is not a statement that the critics were right. It records that the published reading is disputed, that the dispute turns on how to interpret the bone, and that a reader should see both references and not just the more famous one.

Conclusion

Nataruk may yet prove to be what its excavators said it was. It may equally prove to be a smaller or different event than the paper's headline suggested. What it already demonstrates is the speed and the manner with which a claim of this kind is tested: a major paper, a published challenge within the year, a published reply, and no resolution. That is peer review working in public, and it is a reason to write about such sites as contested, with both readings and their sources, and not as settled.

Frequently asked questions

Where is Nataruk and when did the events happen? Nataruk is in West Turkana, northwest Kenya, at what was once the edge of a lagoon connected to Lake Turkana. The excavators dated the remains to roughly ten thousand years ago, in the early Holocene.

What did the original Nature paper claim? It reported at least twenty-seven individuals, twelve of them in relatively complete skeletons, and argued that ten of the twelve showed signs of violent death. The authors concluded that this was inter-group violence among hunter-gatherers.

Why did other scholars dispute it? Stojanowski and colleagues argued that some skull damage could result from sediment pressure after burial, and that the arrangement of the bodies did not resemble known massacre sites. They did not claim that violence was impossible, only that the evidence as published was not conclusive.

Has the disagreement been resolved? No. The excavators replied in the same issue of Nature and kept their interpretation, and the two positions stand side by side. For that reason we record the case at low certainty and give both readings.

As an Amazon Associate I earn from qualifying purchases. Affiliate links in Recommended reading — and in Further reading on older articles — may earn Writesides of History a commission at no extra cost to you.