Sometimes, science tells us things we would rather not know. Most of us can turn away. The scientist cannot. Brian Billman would prefer to be talking about his work in Peru, where he studies the origins of political systems among pre-Columbian people. But as soon as the September 7 issue of Nature appeared, he and his colleagues were news. National news. They didn’t want it that way. They hadn’t set out to alarm anyone, or offend anyone. They had only been doing their jobs.

For Billman, assistant professor of anthropology, the story began in 1997. Soil Systems, Inc., the archaeological-consulting company he worked for at the time, was under contract to excavate an ancient settlement at Cowboy Wash, which is on a floodplain in the Mesa Verde area of southwestern Colorado. Having resolved a water-rights dispute with the federal government, the Ute Mountain Ute Tribe was planning a large-scale irrigation project on tribal land. But first, the tribe would have to unearth and conserve what remained of the past.

“Our first objective,” Billman says, “was to get human remains out of the way of destruction so that they could be treated with respect and reburied.”

Preliminary work had found traces of a twelfth century community of Anasazi, ancestors of modern Puebloan people. The community had included as many as 130 people, farming families who lived in small clusters of pithouses—keyhole-shaped dwellings dug eight or 10 feet deep into the ground, roofed with timbers and sod. Life in the pithouse revolved around an open hearth, and smoke from the cooking fires left the structure just as the people did—through a hole in the roof. Hundreds of pithouses had been excavated in the Southwest, so Billman knew—or thought he knew—what to expect.

But as the team reached the remains of a pithouse roof, Billman suspected that something was amiss. There was too much good wood in the roof. Anasazi people typically did not leave perfectly good building materials lying around. As the team dug deeper, through an infill of mud, they uncovered something else they weren’t expecting: human bones. The bones lay scattered on the pithouse floor. Broken, tossed aside, and left.

This was a crime scene, and Brian Billman knew it.

When we found broken bone, I said, ‘This isn’t going to be the usual pithouse site,’” Billman remembers. “You just don’t normally find broken human bone on a pit-structure floor. We told the tribal leaders this immediately.”

And so, at that moment back in 1997, the tribe could have said “Stop,” could have turned away and left well enough alone. After all, this was their project, their land. But instead, they said, “Go ahead.” Terry Knight, a tribal leader who administered the contract and oversaw work at the site for the Ute Mountain Utes, has been quoted as saying that he hoped the research would help anthropologists revise an overly simplistic view of complex Native American cultures. He wanted the truth.

It was not a pleasant truth. In a cluster of three neighboring pithouse sites lay more than 1,000 broken bits of human bone. The fragments had belonged to at least seven human beings—men, women, and children. Several of the bones bore the mark of stone tools, the grooves where someone had cut through the tendon and peeled back the flesh. Joints of bone had been charred at both ends from the coals. Skulls had been broken and roasted. Pieces of bone showed “pot polish,” smoothed from having been stirred in a pot.

The evidence of this crime scene, sealed tight by flash floods that had plugged the pithouse with sediment, was remarkably complete. For Billman, it was a rare chance to reopen a moment in time.

“Most archaeological sites we study are built up over generations of occupation, sometimes hundreds or thousands of years,” he says. “So you have to consider averages. We don’t see the behavior of any specific individual or event. We see the result of the behavior of many people over long periods of time. In this case, we have this scene, this moment, with very little subsequent disturbance at the site.”

As the evidence emerged from the soil, Billman could see that it would not be telling a story about last-resort cannibalism—the necessary sacrifice of a few to feed a starving many. In starvation cases, mature men and young children tend to die first, over a prolonged period. At this site, there was no such pattern. The victims had died suddenly, probably within hours or days of one another, and they seemed to have been part of an extended family, including children, adolescents, and mature adults. Patricia Lambert, a bioarchaeologist from Utah State University, helped the team reconstruct, in minute detail, the victims’ fate. Yes, the killers processed their victims for food, just as they would have game animals. But deer and elk did not endure the kind of brutality Lambert was finding in her studies of the human victims. Here were people who had been maimed and battered, their faces crushed, their teeth broken out. For the research team, this scene wasn’t about witchcraft or ancestor worship, both of which had been proposed to explain other episodes of cannibalism in the Southwest. And it wasn’t about the desperate need for food. It was about violence, sudden and extreme.

As the team carefully analyzed the contents of each pithouse, evidence mounted in extraordinary detail.

“In addition to finding the body parts of these seven individuals, we also found a full range of domestic artifacts that had been left in place—things like bowls, a cooking pot, jars, stone tools, and grinding tools,” Billman says. “We found two stone tools that had blood residue on them—human blood residue—tools that were probably used in the processing of the human remains. The hearths were overflowing with ash from the last cooking. There were ash dumps adjacent to hearths where so much cooking had been done that the hearths were cleaned out and then refired and then burned until they overflowed again with ash, all in the same short time. There were ornaments that were probably on the bodies of the people who had been consumed.”

While all of this added up to a strong circumstantial case for cannibalism, Billman and his colleagues knew that skeptics would not be convinced unless they built an air-tight case. Traces of human blood on the shards of a cooking pot were compelling, but not in themselves conclusive.

Billman had heard for years the contentious debate over cannibalism in the Southwest. Beginning in the mid 1960s, Christy Turner of Arizona State University and his late wife, Jacqueline Turner, had been assembling evidence about the practice, but their work had met resistance and disbelief. More than once, academics had declared that the evidence for cannibalism was shaky and unconvincing. Even in 1992, when Tim White of Berkeley published an exhaustively detailed book about cannibalism at Anasazi sites from Mancos, a small pueblo on the Colorado Plateau, a few critics continued to doubt. And so while Billman knew that many archaeologists working in the Southwest were ready to accept the evidence he and his colleagues were about to reveal, he also knew that others would not.

“There are a few archaeologists who have staked a significant part of their careers on saying that this doesn’t exist,” Billman says.

The debate was distracting and divisive, and he knew it would go on and on unless there was direct, incontestable evidence of the consumption of human flesh. But what sort of direct evidence could they find?

Sometimes, the treasures of archaeology are solid gold. Sometimes they are dung.

The coprolite—prehistoric excrement—Brian Billman and his colleagues found preserved in the cold ash of a pithouse hearth was, they think, the first ever discovered on a hearth in the American Southwest—with good reason. A hearth was literally and figuratively the center of Anasazi life, and it was almost inconceivable that anyone who had lived in that house would defile it so. For Billman, this coprolite seemed to represent the ultimate insult, the final gesture of contempt.

But he and his colleagues also knew that the coprolite could be a kind of biological time capsule, containing the possibility of direct evidence of consumption of human flesh. So when Richard Marlar, a pathologist from the University of Colorado, took the coprolite into his laboratory, he knew what was at stake. There was no room for ambiguity. As Billman moved on to UNC-Chapel Hill, continuing his analysis of the site, Marlar prepared for his biochemical studies of the coprolite. He needed to find a protein that existed in human muscle and nowhere else, and then an antibody that would identify that protein and nothing else. Simply detecting human DNA in the coprolite would not be sufficient, because the human digestive system sheds and excretes its own cells. So the human DNA would have to come from muscle tissue.

After three years, Marlar found his antibody and tried it with the samples. Sure enough, the coprolite contained human myoglobin, a protein that transports oxygen in muscle cells and is found only in skeletal muscle and the heart. The only way human myoglobin could have found its way into the coprolite is if someone had eaten human flesh. And in this case, flesh seems to have been all that was eaten for at least 12 hours before the final insult dropped onto the hearth. While other human coprolites from the period usually contain starch from meals of maize, along with the fibrous remains of fruits and vegetables, this coprolite showed only meat.

So here at last was the long-awaited evidence. And as work continued in the community, each newly excavated site yielded more broken human bones, more artifacts, more butchering tools with human blood on them.

“We think,” Billman says, “that the entire community was extinguished in this event, that everyone was killed, captured, or driven off. There is no evidence that people came back. At other massacre sites, there’s evidence that people came back to the site and cleaned it up and buried people who had been killed and mutilated. In this case, everything was left on the pit-structure floors.” The victims seem to have been recent immigrants, and patterns of pottery exchange suggest that the newcomers had no allies in their new home. They may have been isolated and vulnerable.

“The ceramics don’t resemble the ceramics of the Mesa Verde area,” Billman says. “A significant percentage look like ceramics from about 60 miles to the south. They’re Anasazi people but they’re from another region. So we think they moved into the Mesa Verde area around 1125 or 1130, established a community, and then were wiped out.”

But who wiped them out, and why? Billman and his collaborators considered various explanations. The site was a few miles south of the border between farmers and hunter-gatherers, so it was conceivable that hunters from the north had raided the community for food. But the M.O., in this case, didn’t seem to match hunter-gatherer behavior. So the team also considered the possibility that the attackers were neighbors. A day’s walk from Billman’s research site, researchers had uncovered what Billman calls “a potential perpetrators’ site.” Here, tool-marked human bones appeared in trash deposits, not on the floors of houses—as if the occupants had brought their victims in to be killed and eaten, then tossed their remains away with the trash.

“So there is one line of evidence—not real solid—pointing in the direction that it was other groups, other Anasazi in the area,” Billman says.

The raid had the earmarks of a territorial conflict, a purge. At the time, having extra territory may have meant survival. Tree-ring records indicate that the period from about 1150 to about 1200 suffered the second-worst drought in the last 2,000 years. The drier the climate, the more land required to grow food.

And the Mesa Verde area wasn’t the only region under pressure. Globally, times were tough. The onset of the “little Ice Age” brought climatic changes that threatened human survival in many parts of the world. While the Plague decimated Europe, major population centers in the Americas fell apart. In Chaco Canyon, about 150 miles to the south of the Mesa Verde area, a stronghold of Anasazi culture collapsed, scattering its population across the Southwest.

“So you have thousands of people moving around, looking for places to settle and raise crops,” Billman says. “The people of the Mesa Verde region, in some places, may have turned to a strategy of raiding and cannibalism as a means of intimidating people, of keeping people out.”

While climate stress may help explain why cannibalism flared up in the last half of the twelfth century, it doesn’t explain why it stopped. When a second severe drought struck during the late 1200s, raiding resumed but cannibalism apparently did not. This is the mystery that interests Billman most.

“After 1200, there may be one or two cases of it, over the next 500 years,” Billman says, “but there’s basically very little evidence of cannibalism among historic Puebloan people, or any other people in the Southwest area. Somehow, this group of people figured out a way to end this particular form of terrorism and violence. Today, Cannibalism is something Puebloan people regard with contempt. It is one of their strongest taboos.”

Billman regards this as a significant accomplishment, one that not every culture has managed on its own. “In the South Pacific—in highland New Guinea or Fiji, for example—cannibalism continued until Western pacification, when outsiders began to police it,” he says.

Billman understands why today’s Puebloan people become upset when they hear about the findings from Cowboy Wash, but he is baffled by some of the outrage he’s heard from other quarters. A response to his paper in American Antiquity went so far as to suggest that the research would aid and abet racism and genocide.

He shakes his head. “What do they want us to do? Just flat out ignore these things? I don’t have a particularly pessimistic view of human nature, but I think I’ve got my eyes open. I’ve lived in the twentieth century. Wherever archaeologists work, if they work there long enough they’re going to find evidence of human violence. Recently, there have been finds of Neanderthal cannibalism in Europe. Cannibalism has been practiced around the world for a variety of reasons. In some cultures, it’s a form of ancestor worship. Our medieval ancestors practiced cannibalism for medical purposes.

“So, as archaeologists, we face a question: Are we going to present a sanitized view of the human past, or are we going to directly confront what people have done?”