Multiple archaeological sites across Europe preserve Neanderthal bones bearing unmistakable signs of butchery by other Neanderthals, including stone-tool cut marks, deliberate bone-breaking for marrow extraction, and processing patterns that mirror the way these same groups handled animal carcasses. The evidence is not universal, though. At some famous sites, what was once called cannibalism turned out to be something else entirely, and the picture that emerges is less one of routine human-eating and more one of varied, site-specific behavior shaped by circumstances researchers are still working to untangle.
The Sites Where the Evidence Is Strongest
The clearest case comes from Moula-Guercy, a cave site in the Ardèche region of southeastern France. In 1999, a team led by Alban Defleur published findings in Science showing that Neanderthal bones recovered from the cave had been processed in ways virtually identical to the deer bones found alongside them. The inference of cannibalism rested on three converging lines of evidence: the spatial distribution of the human and animal bones overlapped, both bore modifications from stone tools, and the representation of skeletal parts matched patterns seen when carcasses are being systematically dismembered for consumption rather than simply discarded.1PubMed. Neanderthal cannibalism at Moula-Guercy, Ardèche, France When researchers later examined the environmental context, they found that the cannibalism episodes at Moula-Guercy coincided with significant ecological upheaval, including a depletion of prey animals at the start of the Upper Pleistocene.2Journal of Archaeological Science. Impact of the last interglacial climate change on ecosystems and Neanderthals behavior at Baume Moula-Guercy, Ardèche, France
The second landmark site is Goyet, a cave system in Belgium that provides what has been called the first unambiguous evidence of Neanderthal cannibalism in Northern Europe. The bones there show clear evidence of butchery, but Goyet adds a twist: at least four Neanderthal bones had been repurposed as retouchers, tools used to resharpen the edges of stone implements. That makes Goyet the first site where multiple Neanderthal bones were used this way.3PubMed Central. Neandertal cannibalism and Neandertal bones used as tools in Northern Europe The researchers who described the Goyet material emphasized that this find also highlighted something important beyond the cannibalism itself: considerable diversity in how late Neanderthal groups in the region treated their dead, all during the period just before Neanderthals disappeared.
A third site, Les Pradelles in southwestern France, rounds out the picture with a more subtle pattern. Neanderthal remains there show evidence of processing, but the scale was small compared to how these same groups handled reindeer. Only five Neanderthal bone fragments were identified using protein fingerprinting (a technique called ZooMS), versus abundant reindeer remains that bore tool marks and were heavily fragmented. While the human bodies were processed on multiple occasions across the site’s occupation layers, the reindeer bones carried many more surface modifications and were far more thoroughly broken apart.4Journal of Quaternary Science. Integrating ZooMS and taphonomy further highlights differential processing of animal and Neandertal carcasses in the Quina Mousterian at Les Pradelles (Marillac‐le‐Franc, southwest France) That asymmetry matters: it suggests the human processing at Les Pradelles was marginal, something that happened occasionally rather than as a core subsistence activity.
How Researchers Tell Cannibalism Apart from Burial Rites
The single hardest problem in studying prehistoric cannibalism is that cutting flesh off a bone with a stone tool leaves marks whether the purpose is eating the person or preparing them for burial. Many human societies, past and present, practice secondary burial, where a body is left to partially decompose, then the remaining tissue is scraped away and the cleaned bones are reburied or kept. The surface damage from this kind of processing can look, at a glance, very similar to butchery.
Researchers have developed several criteria to distinguish the two. A 2016 study systematically compared bone modifications at four prehistoric sites, including ones with known cannibalism and ones with funerary processing. The findings showed that cannibalized remains present a uniform cut-mark distribution associated with disarticulation of all joint types, along with signs of scalping and filleting of muscles. In secondary burials involving defleshing of partially decomposed bodies, disarticulation marks are less common, and marks at joints that naturally fall apart early in decomposition are rare. Crucially, microscopic analysis of the cut marks themselves revealed measurable differences between marks made on fresh tissue and marks made on decaying tissue.5PubMed. Cannibalism versus funerary defleshing and disarticulation after a period of decay: comparisons of bone modifications from four prehistoric sites
In practical terms, this means researchers look for a constellation of features rather than any single smoking gun. Cannibalism tends to produce bones that were cracked open for marrow, that show percussion damage on the shafts, that bear tooth marks from humans chewing on them, and that were cut in locations consistent with removing meat for consumption. A skeleton cleaned for reburial shows a different pattern: cut marks concentrated where soft tissue would cling stubbornly to bone, fewer breaks, and no tooth marks.
The Krapina Lesson
For much of the twentieth century, the Croatian cave site of Krapina was held up as a premier example of Neanderthal cannibalism. The bones there are highly fragmented, many bear surface marks, and the overall pattern seemed, to early researchers, consistent with violent consumption. But a closer look changed the story.
A reassessment of the Krapina material found that none of the damage patterns present in the collection could be explained solely as the products of cannibalism. The supposed evidence, including cranial fragmentation, splitting of long-bone shafts, marks on the surface, patterned breakage, and burnt bone, all had alternative explanations. Importantly, the frequencies of skeletal part preservation suggested the Krapina Neanderthals were buried, either by natural processes or by other Neanderthals, soon after death.6Journal of Human Evolution. Cannibalism and burial at Krapina
A separate study focused specifically on the striations, the fine surface marks that had been interpreted as butchery evidence. When compared to known Mousterian-era butchery marks on reindeer, the Krapina marks did not match. They did, however, closely match authenticated cut marks on modern human skeletons that had been defleshed with stone tools after partial decomposition, as preparation for secondary burial.7PubMed. Mortuary practices at the Krapina Neandertal site Krapina is now widely considered an example of funerary processing rather than cannibalism, and it serves as a cautionary tale about jumping to dramatic conclusions when the bone-surface evidence is ambiguous.
Was It Hunger, Ritual, or Something Else?
The most natural question people ask about Neanderthal cannibalism is why. The honest answer is that the evidence, being entirely skeletal, cannot directly reveal motivation. But a few lines of research offer clues.
The environmental context at Moula-Guercy points toward ecological stress as at least a contributing factor. The cannibalism episodes there coincided with the depletion of prey biomass at the transition to the Upper Pleistocene, suggesting that Neanderthals may have turned to consuming members of their own species during times when their usual food sources collapsed.2Journal of Archaeological Science. Impact of the last interglacial climate change on ecosystems and Neanderthals behavior at Baume Moula-Guercy, Ardèche, France Isotopic analysis of Neanderthal bones from multiple European sites consistently shows they were top-level carnivores who derived almost all of their dietary protein from large herbivores.8PubMed Central. Out of Africa: modern human origins special feature: isotopic evidence for the diets of European Neanderthals and early modern humans9PubMed. Neanderthal diet at Vindija and Neanderthal predation: the evidence from stable isotopes A group this dependent on animal protein would have been particularly vulnerable when large herbivore populations crashed.
But starvation cannibalism does not explain all the evidence. At Goyet, the reuse of Neanderthal bones as tools suggests the bodies were not simply consumed and discarded. Some researchers point to the diversity of mortuary behaviors across contemporary Neanderthal populations as evidence that cannibalism, where it occurred, might have been one practice within a broader repertoire of ways these groups dealt with their dead, some of which we might call ritualistic and others pragmatic.
The Nutritional Puzzle
One way researchers have tried to distinguish survival cannibalism from other motivations is by asking whether eating another human would actually have been worth it from a caloric standpoint. A 2017 study systematically estimated the nutritional value of a human body and compared it to the prey species typically found at Paleolithic sites. Human skeletal muscle has a caloric value broadly in line with other animals of similar body weight, but it is significantly lower than many of the large fauna frequently found alongside processed human remains at archaeological sites.10PubMed Central. Assessing the calorific significance of episodes of human cannibalism in the Palaeolithic
In other words, if a group of Neanderthals already had access to bison, horses, or reindeer, consuming another Neanderthal would not have offered a superior caloric return. That finding has been interpreted two ways. Some see it as evidence that Paleolithic cannibalism was more often social or ritual in nature than driven by raw hunger, since the nutritional payoff was not especially high relative to available alternatives. Others argue the comparison only holds when game is available; during a famine, another human body becomes a meaningful food source regardless of its relative ranking.
A study of Homo antecessor, an earlier hominin species whose members cannibalized each other at Gran Dolina in Spain roughly 800,000 years ago, took a different analytical approach. Using a prey choice model from optimal foraging theory, the researchers found that hominins were actually a high-ranked prey type when all acquisition costs were considered. The argument is that while a human body contains fewer calories per kilogram than, say, a horse, another human is easier to catch and kill, making the overall cost-benefit calculation more favorable than the raw calorie numbers suggest.11PubMed. Does optimal foraging theory explain the behavior of the oldest human cannibals?
When the Marks Came from Carnivores, Not Neanderthals
Not every damaged Neanderthal bone is evidence of cannibalism. A complicating factor that has received increasing attention is carnivore damage. A taphonomic study of hominin fossils bearing animal-tooth marks found that carnivore damage on Neanderthal bones is much more common than previously thought. The damage could result from either predatory attacks on living Neanderthals or scavenging of their bodies after death, and distinguishing between these scenarios has important implications for understanding Neanderthal behavior and vulnerability.12International Journal of Osteoarchaeology. Hunted or Scavenged Neanderthals? Taphonomic Approach to Hominin Fossils with Carnivore Damage
The presence of carnivore tooth marks can muddy the interpretation of a site. If hyenas or wolves gnawed on Neanderthal bones after they were discarded, the resulting damage might be confused with human-caused modifications, or it might overwrite evidence of earlier human processing. Researchers increasingly apply detailed criteria to separate carnivore marks from stone-tool marks, but at older or poorly preserved sites, the picture can remain genuinely ambiguous. The finding at Les Pradelles is instructive here: the newly identified Neanderthal remains at that site lacked the anthropogenic surface modifications seen on the reindeer bones, which means some Neanderthal bone fragments at the site may have ended up in the deposit through processes other than butchery.4Journal of Quaternary Science. Integrating ZooMS and taphonomy further highlights differential processing of animal and Neandertal carcasses in the Quina Mousterian at Les Pradelles (Marillac‐le‐Franc, southwest France)
Cannibalism Across the Human Family Tree
Neanderthals were not unique in this behavior. Evidence for cannibalism appears at multiple points in the hominin lineage, suggesting it is a recurring, if uncommon, feature of human evolution rather than something peculiar to one species.
The earliest well-documented case comes from Gran Dolina in Spain, where Homo antecessor, a species living roughly 800,000 years ago, left behind bones of other H. antecessor individuals showing butchery marks identical to those on deer remains from the same deposits. The cannibalism at Gran Dolina mainly involved infants and immature individuals, and the researchers inferred that one group of hominins periodically hunted and consumed members of another group.13PubMed. Intergroup cannibalism in the European Early Pleistocene: the range expansion and imbalance of power hypotheses That detail, the targeting of vulnerable individuals from a rival group, echoes patterns of lethal intergroup violence documented in chimpanzees.
At the other end of the timeline, modern humans at the Upper Paleolithic site of Gough’s Cave in England, dating to roughly 14,700 years ago, practiced what appears to have been ritualistic cannibalism on a striking scale. The human remains there were extensively processed: defleshed, disarticulated, chewed, and broken open for marrow. Human tooth marks on many postcranial bones provide what the researchers described as incontrovertible evidence for cannibalism.14PubMed. Upper Palaeolithic ritualistic cannibalism at Gough’s Cave (Somerset, UK): The human remains from head to toe Perhaps more remarkable, three cranial vaults had been meticulously shaped into skull-cups, with careful removal of facial bones and smoothing of the edges.15PubMed Central. Earliest directly-dated human skull-cups
A human radius from Gough’s Cave added yet another layer to the picture. Alongside the expected cut marks, percussion damage, and tooth marks indicating cannibalism, the bone bore a set of unusual zig-zagging incisions that could not be explained by filleting. Detailed analysis showed these marks were intentional engravings consistent with Magdalenian design motifs, meaning someone deliberately decorated the bone as part of the cannibalistic ritual. The sequence of modifications suggested the engraving was a purposeful component of the practice rather than an unrelated act, implying a level of ritualistic complexity never previously recognized for the Paleolithic period.16PLoS ONE. An Upper Palaeolithic engraved human bone associated with ritualistic cannibalism The treatment of the dead at Gough’s Cave has parallels at other Magdalenian sites in central and western Europe, suggesting that cannibalism among these Upper Paleolithic modern humans was not isolated but part of a broader cultural practice in the region.
The Prion Disease Hypothesis
One of the more speculative ideas connected to Neanderthal cannibalism involves prion diseases, the class of fatal brain conditions that includes Creutzfeldt-Jakob disease and kuru. Among the Fore people of Papua New Guinea, the practice of consuming the brains of deceased relatives during funerary rituals led to an epidemic of kuru, a transmissible spongiform encephalopathy, in the twentieth century. A 2008 paper in Medical Hypotheses proposed that something similar could have happened among Neanderthals. The argument is that if Neanderthal groups engaged in cannibalism that included brain tissue consumption, prion diseases could have spread through the population, with further transmission possible through continued cannibalistic activity or even through the shared use of contaminated stone tools.17PubMed. A potential role for transmissible spongiform encephalopathies in Neanderthal extinction
This idea remains firmly in the realm of hypothesis rather than established fact. There is no direct evidence of prion disease in any Neanderthal fossil, and prion proteins do not survive in the archaeological record in a way that would allow detection. The hypothesis also depends on brain consumption being widespread enough to sustain a chain of transmission, something the archaeological evidence does not confirm one way or another. Still, the hypothesis illustrates how cannibalism connects to questions about Neanderthal health and population decline in ways that go beyond the simple question of whether it happened.
Neanderthal Bone Processing in a Broader Context
The debate over Neanderthal cannibalism fits into a larger picture of how these groups interacted with bone as a material. Neanderthals were skilled at extracting every bit of nutrition from animal carcasses. At Neumark-Nord in Germany, dating to roughly 125,000 years ago, Neanderthals processed at least 172 large mammals at a lakeside site, transporting partial carcasses there specifically for the extraction of within-bone nutrients, particularly bone grease. Researchers described the site as a kind of specialized processing location devoted to rendering fat.18Science Advances. Large-scale processing of within-bone nutrients by Neanderthals, 125,000 years ago This level of organized, large-scale bone processing tells us that when Neanderthals did apply similar techniques to human bones at sites like Goyet and Moula-Guercy, they were drawing on well-established skills they used every day for animal carcasses. The processing was not clumsy or haphazard; it reflected deep familiarity with how to break bone and extract what was inside.
That context makes the contrast at Les Pradelles all the more interesting. At that site, Neanderthal bodies were processed on multiple occasions but far less intensively than reindeer. The pattern hints at something more nuanced than either pure subsistence cannibalism or pure ritual: a practice where human bodies were occasionally processed using the same toolkit and techniques applied to animals, but without the systematic thoroughness that marked everyday hunting and butchery. Whether that reflects taboo, limited opportunity, or simply the fact that human bodies were rare compared to reindeer is anyone’s guess.