“Don’t touch that tree,” Balée said.

I froze. I was climbing a low, crumbly hill and had been about to support myself by grasping a scrawny, almost vine-like tree with splayed leaves. “Triplaris americana,” said Balée, an expert in forest botany. “You have to watch out for it.” In an unusual arrangement, he said, T. americanaplays host to colonies of tiny red ants—indeed, it has trouble surviving without them. The ants occupy minute tunnels just beneath the bark. In return for shelter, the ants attack anything that touches the tree—insect, bird, unwary writer. The venom-squirting ferocity of their attack gives rise to T. americana’s local nickname: devil tree.

At the base of the devil tree, exposing its roots, was a deserted animal burrow. Balée scraped out some dirt with a knife, then waved me over, along with Erickson and my son Newell, who were accompanying us. The depression was thick with busted pottery. We could see the rims of plates and what looked like the foot of a teakettle—it was shaped like a human foot, complete with painted toenails. Balée plucked out half a dozen pieces of ceramic: shards of pots and plates, a chipped length of cylindrical bar that may have been part of a pot’s support leg. As much as an eighth of the hill, by volume, was composed of such fragments, he said. You could dig almost anywhere on it and see the like. We were clambering up an immense pile of broken crockery.

The pile is known as Ibibate, at fifty-nine feet one of the tallest known forested mounds in the Beni. Erickson explained to me that the pieces of ceramic were probably intended to help build up and aerate the muddy soil for settlement and agriculture. But though this explanation makes sense on engineering grounds, he said, it doesn’t make the long-ago actions of the moundbuilders any less mysterious. The mounds cover such an enormous area that they seem unlikely to be the byproduct of waste. Monte Testaccio, the hill of broken pots southeast of Rome, was a garbage dump for the entire imperial city. Ibibate is larger than Monte Testaccio and but one of hundreds of similar mounds. Surely the Beni did not generate more waste than Rome—the ceramics in Ibibate, Erickson argues, indicate that large numbers of people, many of them skilled laborers, lived for a long time on these mounds, feasting and drinking exuberantly all the while. The number of potters necessary to make the heaps of crockery, the time required for labor, the number of people needed to provide food and shelter for the potters, the organization of large-scale destruction and burial—all of it is evidence, to Erickson’s way of thinking, that a thousand years ago the Beni was the site of a highly structured society, one that through archaeological investigation was just beginning to come into view.

Accompanying us that day were two Sirionó Indians, Chiro Cuéllar and his son-in-law Rafael. The two men were wiry, dark, and nearly beardless; walking beside them on the trail, I had noticed small nicks in their earlobes. Rafael, cheerful almost to bumptiousness, peppered the afternoon with comments; Chiro, a local figure of authority, smoked locally made “Marlboro” cigarettes and observed our progress with an expression of amused tolerance. They lived about a mile away, in a little village at the end of a long, rutted dirt road. We had driven there earlier in the day, parking in the shade of a tumbledown school and some old missionary buildings. The structures were clustered near the top of a small hill—another ancient mound. While Newell and I waited by the truck, Erickson and Balée went inside the school to obtain permission from Chiro and the other members of the village council to tramp around. Noticing that we were idle, a couple of Sirionó kids tried to persuade Newell and me to look at a young jaguar in a pen, and to give them money for this thrill. After a few minutes, Erickson and Balée emerged with the requisite permission—and two chaperones, Chiro and Rafael. Now, climbing up Ibibate, Chiro observed that I was standing by the devil tree. Keeping his expression deadpan, he suggested that I climb it. Up top, he said, I would find some delicious jungle fruit. “It will be like nothing you have experienced before,” he promised.

From the top of Ibibate we were able to see the surrounding savanna. Perhaps a quarter mile away, across a stretch of yellow, waist-high grass, was a straight line of trees—an ancient raised causeway, Erickson said. Otherwise the countryside was so flat that we could see for miles in every direction—or, rather, we could have seen for miles, if the air in some directions had not been filled with smoke.

Afterward I wondered about the relationship of our escorts to this place. Were the Sirionó like contemporary Italians living among the monuments of the Roman Empire? I asked Erickson and Balée that question during the drive back.

Their answer continued sporadically through the rest of the evening, as we rode to our lodgings in an unseasonable cold rain and then had dinner. In the 1970s, they said, most authorities would have answered my question about the Sirionó in one way. Today most would answer it in another, different way. The difference involves what I came to think of, rather unfairly, as Holmberg’s Mistake.

Although the Sirionó are but one of a score of Native American groups in the Beni, they are the best known. Between 1940 and 1942 a young doctoral student named Allan R. Holmberg lived among them. He published his account of their lives, Nomads of the Longbow, in 1950. (The title refers to the six-foot bows the Sirionó use for hunting.) Quickly recognized as a classic, Nomads remains an iconic and influential text; as filtered through countless other scholarly articles and the popular press, it became one of the main sources for the outside world’s image of South American Indians.

The Sirionó, Holmberg reported, were “among the most culturally backward peoples of the world.” Living in constant want and hunger, he said, they had no clothes, no domestic animals, no musical instruments (not even rattles and drums), no art or design (except necklaces of animal teeth), and almost no religion (the Sirionó “conception of the universe” was “almost completely uncrystallized”). Incredibly, they could not count beyond three or make fire (they carried it, he wrote, “from camp to camp in a [burning] brand”). Their poor lean-tos, made of haphazardly heaped palm fronds, were so ineffective against rain and insects that the typical band member “undergoes many a sleepless night during the year.” Crouched over meager campfires during the wet, buggy nights, the Sirionó were living exemplars of primitive humankind—the “quintessence” of “man in the raw state of nature,” as Holmberg put it. For millennia, he thought, they had existed almost without change in a landscape unmarked by their presence. Then they encountered European society and for the first time their history acquired a narrative flow.

Holmberg was a careful and compassionate researcher whose detailed observations of Sirionó life remain valuable today. And he bravely surmounted trials in Bolivia that would have caused many others to give up. During his months in the field he was always uncomfortable, usually hungry, and often sick. Blinded by an infection in both eyes, he walked for days through the forest to a clinic, holding the hand of a Sirionó guide. He never fully recovered his health. After his return, he became head of the anthropology department at Cornell University, from which position he led its celebrated efforts to alleviate poverty in the Andes.

Nonetheless, he was wrong about the Sirionó. And he was wrong about the Beni, the place they inhabited—wrong in a way that is instructive, even exemplary.

Before Columbus, Holmberg believed, both the people and the land had no real history. Stated so baldly, this notion—that the indigenous peoples of the Americas floated changelessly through the millennia until 1492—may seem ludicrous. But flaws in perspective often appear obvious only after they are pointed out. In this case they took decades to rectify.

The Bolivian government’s instability and fits of anti-American and anti-European rhetoric ensured that few foreign anthropologists and archaeologists followed Holmberg into the Beni. Not only was the government hostile, the region, a center of the cocaine trade in the 1970s and 1980s, was dangerous. Today there is less drug trafficking, but smugglers’ runways can still be seen, cut into remote patches of forest. The wreck of a crashed drug plane sits not far from the airport in Trinidad, the biggest town in the province. During the drug wars “the Beni was neglected, even by Bolivian standards,” according to Robert Langstroth, a geographer and range ecologist in Wisconsin who did his dissertation fieldwork there. “It was a backwater of a backwater.” Gradually a small number of scientists ventured into the region. What they learned transformed their understanding of the place and its people.

Just as Holmberg believed, the Sirionó were among the most culturally impoverished people on earth. But this was not because they were unchanged holdovers from humankind’s ancient past but because smallpox and influenza laid waste to their villages in the 1920s. Before the epidemics at least three thousand Sirionó, and probably many more, lived in eastern Bolivia. By Holmberg’s time fewer than 150 remained—a loss of more than 95 percent in less than a generation. So catastrophic was the decline that the Sirionó passed through a genetic bottleneck. (A genetic bottleneck occurs when a population becomes so small that individuals are forced to mate with relatives, which can produce deleterious hereditary effects.) The effects of the bottleneck were described in 1982, when Allyn Stearman of the University of Central Florida became the first anthropologist to visit the Sirionó since Holmberg. Stearman discovered that the Sirionó were thirty times more likely to be born with clubfeet than typical human populations. And almost all the Sirionó had unusual nicks in their earlobes, the traits I had noticed on the two men accompanying us.

Even as the epidemics hit, Stearman learned, the group was fighting the white cattle ranchers who were taking over the region. The Bolivian military aided the incursion by hunting down the Sirionó and throwing them into what were, in effect, prison camps. Those released from confinement were forced into servitude on the ranches. The wandering people Holmberg traveled with in the forest had been hiding from their abusers. At some risk to himself, Holmberg tried to help them, but he never fully grasped that the people he saw as remnants from the Paleolithic Age were actually the persecuted survivors of a recently shattered culture. It was as if he had come across refugees from a Nazi concentration camp, and concluded that they belonged to a culture that had always been barefoot and starving.

Far from being leftovers from the Stone Age, in fact, the Sirionó are probably relative newcomers to the Beni. They speak a language in the Tupí-Guaraní group, one of the most important Indian language families in South America but one not common in Bolivia. Linguistic evidence, first weighed by anthropologists in the 1970s, suggests that they arrived from the north as late as the seventeenth century, about the time of the first Spanish settlers and missionaries. Other evidence suggests they may have come a few centuries earlier; Tupí-Guaraní–speaking groups, possibly including the Sirionó, attacked the Inka empire in the early sixteenth century. No one knows why the Sirionó moved in, but one reason may be simply that the Beni then was little populated. Not long before, the previous inhabitants’ society had disintegrated.

To judge by Nomads of the Longbow, Holmberg did not know of this earlier culture—the culture that built the causeways and mounds and fish weirs. He didn’t see that the Sirionó were walking through a landscape that had been shaped by somebody else. A few European observers before Holmberg had remarked upon the earthworks’ existence, though some doubted that the causeways and forest islands were of human origin. But they did not draw systematic scholarly attention until 1961, when William Denevan came to Bolivia. Then a doctoral student, he had learned of the region’s peculiar landscape during an earlier stint as a cub reporter in Peru and thought it might make an interesting topic for his thesis. Upon arrival he discovered that oil-company geologists, the only scientists in the area, believed the Beni was thick with the remains of an unknown civilization.

Convincing a local pilot to push his usual route westward, Denevan examined the Beni from above. He observed exactly what I saw four decades later: isolated hillocks of forest; long raised berms; canals; raised agricultural fields; circular, moat-like ditches; and odd, zigzagging ridges. “I’m looking out of one of these DC-3 windows, and I’m going berserk in this little airplane,” Denevan said to me. “I knew these things were not natural. You just don’t have that kind of straight line in nature.” As Denevan learned more about the landscape, his amazement grew. “It’s a completely humanized landscape,” he said. “To me, it was clearly the most exciting thing going on in the Amazon and adjacent areas. It may be the most important thing in all of South America, I think. Yet it was practically untouched” by scientists. It is still almost untouched—there aren’t even any detailed maps of the earthworks and canals.

Beginning as much as three thousand years ago, this long-ago society—Erickson believes it was probably founded by the ancestors of an Arawak-speaking people now called the Mojo and the Bauré—created one of the largest, strangest, and most ecologically rich artificial environments on the planet. These people built up the mounds for homes and farms, constructed the causeways and canals for transportation and communication, created the fish weirs to feed themselves, and burned the savannas to keep them clear of invading trees. A thousand years ago their society was at its height. Their villages and towns were spacious, formal, and guarded by moats and palisades. In Erickson’s hypothetical reconstruction, as many as a million people may have walked the causeways of eastern Bolivia in their long cotton tunics, heavy ornaments dangling from their wrists and necks.


Flying over eastern Bolivia in the early 1960s, the young geographer William Denevan was amazed to see that the landscape (bottom)—home to nothing but cattle ranches for generations—still bore evidence that it had once been inhabited by a large, prosperous society, one whose very existence had been forgotten. Incredibly, such discoveries are still being made. In 2002 and 2003, Finnish and Brazilian researchers revealed the remains of dozens of geometrical earthworks (top) in the western Brazilian state of Acre where the forest had just been cleared for cattle ranches.

Today, hundreds of years after this Arawak culture passed from the scene, the forest on and around Ibibate mound looks like the classic Amazon of conservationists’ dreams: lianas thick as a human arm, dangling blade-like leaves more than six feet long, smooth-boled Brazil nut trees, thick-bodied flowers that smell like warm meat. In terms of species richness, Balée told me, the forest islands of Bolivia are comparable to any place in South America. The same is true of the Beni savanna, it seems, with its different complement of species. Ecologically, the region is a treasure, but one designed and executed by human beings. Erickson regards the landscape of the Beni as one of humankind’s greatest works of art, a masterpiece that until recently was almost completely unknown, a masterpiece in a place with a name that few people outside Bolivia would recognize.

If you find an error please notify us in the comments. Thank you!