The first unequivocal evidence that American Indians practiced cannibalism has been discovered by researchers studying a small Anasazi settlement in what is now southwestern Colorado that was mysteriously abandoned about 1150 A.D.
As many as 40 sites scattered around the Southwest contain human bones that show distinctive evidence of having been butchered and cooked - signs consistent with cannibalism. Until now, however, most archaeologists have shied away from conceding the evidence proves cannibalism - favoring explanations such as ritual burial or the execution of people believed to be witches.
The new, conclusive evidence comes from preserved pieces of human excrement that were found at the site. The pieces contain human proteins that could be there only if the subjects had eaten human flesh. Researchers believe that if cannibalism has been definitively proven at this one Southwestern site, it is overwhelmingly likely that cannibalism was common enough to have taken place at the other sites where butchered bones have been found.
The report in Nature is certain to add fuel to a bitter argument among scholars and Native Americans over cannibalism among the Anasazi, regarded as the ancestors of the modern Hopi, Zuni and other Puebloan peoples in the North American Southwest.
"Fur is going to fly over this," said anthropologist Tim White of the University of California, Berkeley.
Anthropologist William Lipe of Washington State University said many archaeologists have been reluctant to accept the existence of cannibalism "because it carries a lot of negative implications."
But, "There are certainly plenty of other cases from history where humans have engaged in cannibalism for whatever reason. The next challenge is to understand what those reasons were," he said.
Archaeologists have long suspected the existence of cannibalism in the Southwest, but the modern descendants of the Anasazi, the Hopi and Zuni, have a firmly entrenched moral code that forbids the practice. The tribes have insisted that their ancestors were not cannibals, and archaeologists have largely bowed to their beliefs.
Controversy erupted last year when physical anthropologist Christy Turner of Arizona State University published a book called "Man Corn: Cannibalism and Violence in the Prehistoric American Southwest."
Turner argued that a "band of thugs" - Toltecs from Central Mexico - used both cannibalism and violence to terrorize the Anasazi over a period of perhaps 200 years. He viewed the practice as a political tool that was used for expansion of an empire. Turner could not be reached.
But the new study, Lipe said, suggests there was a breakdown of social order, quite likely induced by drought and famine.
Why did they turn to cannibalism? "They were hungry," White said.
The new evidence was found at a site in Cowboy Wash in southwestern Colorado.
The findings paint a detailed and grisly picture.
Nine centuries ago, the little village was home to three families with as many as 15 people. They had lived there about 25 years, eking out a living growing corn, squash, beans and other crops.
But times were tough. The region was in the midst of a long-term drought. Evidence found at the site shows it was early spring, and the three families' own food supplies were running low when a small group of raiders killed them.
The bodies of four adults, one adolescent about age 11 and two younger children were killed and eaten. Afterward, one of the marauders went into the third pit house and defecated on its cold hearth. The fecal remains, called a coprolite by archaeologists, clinched the case. After hearing about the discovery of the coprolite at a scientific meeting in 1997, Dr. Richard Marlar of the Denver Veterans Affairs Medical Center and the University of Colorado School of Medicine offered his assistance. An amateur archaeologist, he told the researchers that he could devise a test to determine if human tissue was present in the coprolite.
The test focused on myoglobin, a protein that is used by the heart and other muscles for transporting oxygen. Marlar was able to show that the coprolite contained human myoglobin not normally present in feces.