The attackers probably struck the sleeping pueblo at dawn. Dozens of warriors, moving as silently as the rising sun in the cold desert air, climbed to the flat roofs of the tightly clustered multistory dwellings. The pueblo's 150-odd rooms had neither windows nor ground-level doors; a ladder propped against a small hole in the roof was the only way in or out of each dwelling. The invading warriors quickly pulled out the ladders, trapping the occupants inside. They lit shredded juniper bark and highly flammable greasewood and threw the flaming bundles into the rooms. The warriors fed the growing fire with more and more wood; some snatched bunches of chilies that were drying on the pueblo walls and tossed these into the inferno, creating smoke that would have stung like tear gas. Soon the flames reached the roof beams, and the attackers had to jump from the roofs before they collapsed. All around the pueblo, roofs were caving in, silencing the screams within.
    On a late summer afternoon seven centuries after that massacre, Salmon Ruin, as the ancient, long-abandoned pueblo is now known, is serene, its stillness broken only by the cackle of scrub jays. Named after an early homesteader, Salmon Ruin lies a few miles east of Farmington in northwestern New Mexico. Exactly how the conflagration there unfolded can never be known, but Harvard archaeologist Steven LeBlanc believes the scenario he pieced together based on oral legends of similar battles of the Hopi, who live about 150 miles to the west, is plausible. He points to some blackened stone on a ruined wall. "You can see where it burned. And pueblo rooms don't burn easily," he says.
    Here, on the quiet outskirts of a small town, it's difficult to imagine the carnage. In 1973 archaeologists digging at the site uncovered the bones of 33 children who were apparently burned alive inside a kiva, a large circular structure used for religious ceremonies. The children must have been hiding there during the attack, which occurred sometime between 1263 and 1300, according to tree-ring dating of wood at the scene. Archaeologists studying the children's skeletal remains noted signs of what they call green-bone burning—bone that has burned with flesh still attached.
    Casual visitors to the site will learn none of this. Inside the small museum at Salmon Ruin, no mention is made of the slaughter. The omission does not surprise LeBlanc, the director of collections at Harvard's Peabody Museum of Archaeology and Ethnology. After three decades spent studying similar sites in New Mexico, Colorado, Utah, Arizona, and northern Mexico, he is convinced that most archaeologists have completely misread the region's archaeological record. Many of his colleagues even deny that a bloody battle was fought here. Instead, they attribute the deaths to accidental burning or ritual sacrifice.
    Archaeologists have long believed that the pueblo-dwelling corn farmers who populated the southwest before Europeans arrived in the 16th century were a peaceful people. On the contrary, LeBlanc argues, prehistoric warfare in the Southwest was common, prolonged, and at least as deadly on a per capita basis as the most violent conflicts of the 20th century. Much of what we take to be characteristic of pueblos today—the magnificent, haunting cliff dwellings, the sophisticated architecture—is testimony to intense warfare and the deaths of entire tribes.
    Clear evidence of prehistoric conflict can be difficult to discern in other parts of the world, but the arid climate and sparse settlement in the Southwest have left pueblo ruins, cultural artifacts, and even human remains remarkably intact. "My real interest is, How did people live in the past on a worldwide basis?" says LeBlanc. "What the Southwest provides us is probably the best-worked-out archaeological database in the world. We use it as an example, a place to test ideas. But ultimately it's not unique; it's not even particularly bloody. It's no different from anyplace else in the world.
    "When I first began studying this, I asked, Was there any warfare among prehistoric cultures? Now, of course, my conclusion is that that's a completely stupid question. The real question that any archaeologist has to face is, Was there any peace?"

In 1972, when LeBlanc was a 29-year-old assistant professor at Wichita State University, he received a grant from the National Science Foundation to study a series of ruins in El Morro Valley in western central New Mexico. The valley is beautiful, an expanse of sweeping plains and mesas clothed in ponderosa pine. But at 7,200 feet above sea level, the cold nights and scant rainfall make it unsuitable for agriculture. Yet during the late 13th century, the valley had one of the largest pueblo populations in New Mexico. Some pueblos had more than 1,000 rooms; many had hundreds.
    At a small hotel in Dolores, Colorado, where LeBlanc is preparing for a lecture at a nearby museum, he talks about that early work at El Morro: "What were these people doing at 7,200 feet in such a crummy place, trying to be corn farmers? That was our question. We wrote an application to the National Science Foundation and never mentioned the word warfare once."
    What they found at the site raised other interesting questions.
    "About the second day we were at El Morro, we walked over the site and could see evidence of burning—everywhere. So we began to excavate. We found rooms like this," he says, pointing to a slide showing the freshly excavated interior of a room at the pueblo. Broken pottery litters the floor, probably smashed when roof beams collapsed in a fire.
    "The people obviously ran out the door in a hurry," says LeBlanc. "Everything was sitting on the floor just like they left it. Virtually every room in this site, which consisted of 20 distinct little pueblos, ranging from 5 to 10 rooms to maybe 30 rooms, spread out over a long ridge, had been catastrophically burned."
    Natural fires in a stone pueblo quickly burn out or the community extinguishes them, usually before more than one room is damaged. The torching of an entire pueblo, LeBlanc's team concluded, must have been an act of war.
    LeBlanc and his crew also noticed that many walls they excavated were missing large numbers of stones. This puzzled them until they uncovered a much bigger pueblo, consisting of 500 rooms and enclosing a central plaza, just a few hundred feet away. The second site was completed no later than 1279. According to tree-ring dates, the neighboring smaller pueblos were burned in 1276.
    "What I believe happened was that the survivors dismantled those small scattered pueblos, which were not defensible, and they basically built a fortress," LeBlanc says. "These people were very afraid of something." At first, he recalls, "I explained it in terms of a very local, accidental kind of thing, like the Hatfields and the McCoys, that didn't have any broad explanation."
    Only in the last several years has LeBlanc come to believe that El Morro represents just one episode from an era of violence that swept the Southwest, and probably much of North America, at the close of the 13th century. The evidence for widespread catastrophe is overwhelming, he says, from palisaded Iroquois villages in New York to fortified sites in the Pacific Northwest.
    "It took me 25 years to come to grips with what was really going on," he says.

Graphic by Matt Zang

LeBlanc is 5 feet 4 inches and built close to the ground.
    He easily spots pottery shards where the hapless reporter at his side sees only sticks and pebbles. He'll pause to gaze down at a spot and then move quickly on, like a shorebird dodging waves. Brown hair flecked with gray and a trim beard fringe a face that seems primed for debate. Behind his glasses, the eyes look at once challenging and puzzled, as if wondering why the rest of the world can't see the evident truth of his arguments. He sometimes shouts words or whole sentences for emphasis. He does that now, looking out over Hovenweep, the spectacular ruins straddling the border between Utah and Colorado that were abandoned more than seven centuries ago.
    "WHY DID YOU DO THAT?" LeBlanc yells, haranguing not an obstinate colleague but the long-vanished people who built a series of massive stone towers on the edges of opposing sheer cliffs and on top of high, narrow pinnacles. Why would the people of Hovenweep, like their contemporaries at El Morro and elsewhere in the Southwest, choose to live in the most inaccessible places they could find? Why would anyone have settled here, where water is so scarce, where the sunlight almost impales you?
    "People have tried to explain these towers in a variety of ways—as storehouses or observatories. These were built at exactly the same time the fortress was being built at El Morro. The towers defended the canyon right where the water supply comes out of the rocks. They're not a random series of towers. They are clustered in patterns to provide mutual defense, one to the other."
    Hovenweep was one of the sites that prompted LeBlanc to reconsider the fate of the burned pueblo at El Morro. For more than 20 years after his work there, he busied himself with a wide variety of archaeological pursuits. He was a curator at the Maxwell Museum of Anthropology in Albuquerque and became involved in efforts to preserve archaeological treasures across the Southwest. "I was traveling around the Southwest about five or six years ago, and I began to see basically the same pattern: sites being abandoned, fortresses being built, lots of burning, and I said, whoops, maybe I got the scale of events wrong at El Morro. Maybe this was a pan-Southwestern phenomenon. I can't explain what happened at El Morro by a simple Hatfields and McCoys kind of thing. This is a bigger issue." The scope of the conflict proved larger than LeBlanc had imagined.
    "By the mid 1200s, you begin to find, all over the Southwest, fortification after fortification. And by 1275 or 1300, depending on where you're looking, everybody was living in a defensive site."
    A survey by LeBlanc showed that nearly every pueblo or cluster of pueblos in the late 13th century was surrounded by a no-man's-land, like the lethal terrain that separated opposing armies in the trenches of the First World War. Small isolated pueblos were abandoned in favor of large, fortified settlements, leaving wide swaths of territory unoccupied. The deserted tracts, LeBlanc says, must have become buffer zones between enemy pueblos, turf too dangerous to tread.
    The trigger for all this bloodshed, LeBlanc believes, was the onset of a mini ice age, a period of global cooling known to have inflicted famines in Europe. The little ice age came on the heels of 300 years of generally benign weather known as the medieval warm period. In the American Southwest, says LeBlanc, beginning in about 900, a warming trend moderated the usual harsh dryness of the region and brought more rain. Crop yields increased, and so did the human population. A mysterious and sophisticated pueblo culture known as the Anasazi flowered around Chaco Canyon in northwestern New Mexico. The Anasazi built multistory dwellings and a network of roads that crisscrossed parts of New Mexico, Arizona, Colorado, and Utah. It was a relatively peaceful age, says LeBlanc, with little evidence of warfare between 900 and the late 1200s. Most people lived in undefended sites on valley floors near their crops and water sources.
    "During the medieval warm period it was a great time to be a corn farmer—and a great time to be a wheat farmer in Europe too," says LeBlanc. "That's when the Europeans started building Gothic cathedrals. Chaco Canyon was built at the same time they were building Gothic cathedrals. This was a good time for everybody, and lo and behold, there's plenty to eat, and warfare declines. Then the warm period ends, and you get the beginning of the little ice age. The peak of the little ice age wasn't until the 1500s—that's when the glaciers came out of the Alps and demolished villages—but people were dying of famine because the crops didn't mature in 1317 in England, pretty darn close to the period of warfare in the Southwest. What happened? The population had grown immensely. In bad times, there just weren't enough resources to go around. People competed and the best organized survived. It's a classic pattern that occurs everywhere on Earth."
    LeBlanc estimates that before 1300, at least 20 distinct alliances of pueblos occupied western New Mexico, eastern Arizona, and the southern parts of Colorado and Utah. Today, excluding a few pueblos along the Rio Grande, only four groups survive—the Hopi in Arizona and the Zuni, Laguna, and Acoma in northern New Mexico. The others were wiped out in war, the survivors absorbed by other tribes.
    "The fundamental, underlying reason for warfare is competition over scarce resources," says LeBlanc. "It's not to prove your manhood, it is not to gain prestige, it is not because of inherent genetic blood lust, or anything like that. You're fighting for survival because there are always more people than there are resources. And the only times you have peace are when there are a lot of resources."

Finding someone to criticize LeBlanc's ideas is easy—just call any university's archaeology department. Jeff Dean, an archaeologist at the University of Arizona at Tucson who studies Anasazi settlement patterns, was happy to oblige. Dean believes that warfare played no significant role in the Southwest before Europeans arrived. "Armies marching up and down fighting each other? There never was any of that," he says. "There was raiding for women, food, and loot, but is that warfare?"
    Charles Adams, an archaeologist at the University of Arizona who studies a cluster of prehistoric pueblos in northeastern Arizona, is also skeptical. He argues that people originally gathered in pueblos for efficiency: "You can't simply say that the reason people moved into cliff dwellings is because of warfare. These very large villages are a very successful adaptation for pueblos. That's why they went to them in the 1200s, and why they're still living in them."
    With large numbers concentrated in pueblos, Adams says, people could pool their resources and farm much larger plots of land than individual families could. "Instead of having five acres to farm," says Adams, "suddenly the village has a couple of thousand acres, over a wide area."
    As for the burned pueblos and bodies found at some sites, Adams and others attribute them to purification rituals, where buildings contaminated by disease, death, or sorcery were cleansed by fire or abandoned. "Some things LeBlanc considers to be due to conflict are probably more accurately a result of witchcraft persecution," Adams says.
    LeBlanc, of course, scoffs. Early pueblo dwellers were certainly capable of fielding large numbers of warriors, he argues, and cites late-16th-century reports from Spanish conquistadors to support his case. "According to Coronado's account, when he attacked the Zuni, the Zuni lined up by division, blew trumpets, and charged."
    Moreover, he says, even intermittent raids can be an efficient way to kill, with long-term death rates far higher than conventional warfare. Where good ethnographic data exist, as they do for some tribal groups in New Guinea and South America, the mortality rate among males from raiding approaches 25 percent.
    "I once computed the deaths at Waterloo—they came out to a fraction of a percent of the adult male population of Europe at the time," says LeBlanc. Yet when archaeologists find evidence for similar numbers of violent deaths at a site, they dismiss the toll as insignificant.
    And if pueblos were not built as defensive structures, asks LeBlanc, why were they built on cliff walls or on the tops of sheer mesas? It would have been much easier to live on valley floors near the cornfields that sustained the pueblos. Besides, he says, "you can't use the Zuni or Hopi today, who are being protected by the federal government, as an analogy. Even in the 19th century, the Zuni repeatedly tried to set up farming communities away from the main pueblo because they needed more good farmable land. And they couldn't do it, because the Navajo kept getting them. It was just too risky. It wasn't until peace came that they were able to establish those outlying communities."
    Archaeologists ignore the evidence of warfare because of empathy for their subjects, says LeBlanc. "You cannot help being impressed with the ingenuity, skills, and determination of the ancient people you are studying. My people could not have slaughtered an entire village or burned people in their kivas."
    In addition, emphasizing the region's violent past might offend tribes that control access to ruins. And many anthropologists and archaeologists still see prehistoric peoples as peaceful groups living in harmony with nature.
    David Wilcox, an archaeologist at the Museum of Northern Arizona in Flagstaff, has been exploring the role of warfare in the collapse of Anasazi culture. "These ancient puebloans were people like us. They were not inherently warlike. But if you say they never had any conflict in their lives, then you're not talking about people, you're talking about a myth that you've created to abet your own sentiments. And you're not any longer treating the Indian people as real people. So there's a larger philosophical dimension to this," he says.
    LeBlanc doesn't particularly like the portrait of the past he has reconstructed, any more than he likes to read about genocide in Serbia or Rwanda, or for that matter, the genocide of American Indians in 19th-century America. But he feels it's important to face up to the truth about our past.
    "That's my story, that this is something all humans share, and that we have shared for a long time, and we are fools to think that this isn't part of our human past. To pretend that it isn't is rather dangerous and leads to a lot of false assumptions. It's better to come to grips with reality."