What were the ancient societies of Amazonia like before the first European sailors clambered ashore in the mid-16th century carrying strange new diseases and the seeds of war and conquest? For years, archaeologists have duelled over different views. Initially, most saw the vast Amazon Basin—which sprawls across Brazil, Peru, and five other countries in South America—as home to sparse bands of pre-Columbian hunters and gatherers who preserved the region's forests as wilderness. Then, archaeological discoveries in the 1990s revealed large villages and complex societies in eastern Amazonia, giving rise to the theory that prehistoric agriculturists had cleared and intensively managed forests across the Amazon Basin for thousands of years.
Much rides on this assessment. If the Amazon, with its great diversity of plant and animal life today, was once extensively cleared and supported a large population of prehistoric people, that bodes well for the forest's powers of recovery. But if the region harbored only small, scattered populations, then today's ecosystem is truly virgin forest and perhaps very vulnerable to human activities.
Now a new study, published online today in Science, suggests that large ancient populations never cleared and tamed the western Amazon. By analyzing soil samples from 55 locations in central and western Amazonia, a team of American and Brazilian researchers led by paleoecologist Crystal McMichael, who is now at the of the University of New Hampshire in Durham,have found that pre-Columbian bands ranged in small numbers over the region and gathered food without slashing, burning, or cultivating large tracts of the Amazon. "Most of the forest we sampled has never been intensively disturbed," McMichael says.
McMichael and her colleagues journeyed by riverboat, small aircraft, and four-wheel-drive trucks across an area measuring 3 million square kilometers, taking soil cores from river bluffs close to known archaeological sites and from inland sites such as randomized localities along a transect from Porto Velho, Brazil, to Manaus, Brazil. Back in the lab, McMichael analyzed each of the 247 cores to determine the abundance of charcoal, an indicator that humans set the forest ablaze to clear trees and enrich soils for crops, since natural fires are rare in Amazonia. In addition, archaeobotanist Dolores Piperno of the Smithsonian Institution National Museum of Natural History in Washington, D.C., extracted and identified phytoliths -- characteristic microscopic silica fossils that are produced by many plant species and that trap minute amounts of carbon. Then the investigators sent samples of the charcoal and phytoliths out for radiocarbon dating.
The results took the team by surprise. In the river-bluff cores, they found bits of charcoal, phytoliths of burned grasses, and phytoliths of herbaceous plants often found in disturbed areas—but no phytoliths of common Amazon crops. This suggested that humans had cleared and disturbed some parts of the forest there, but there was little evidence of farming. And in the inland cores, the team detected relatively few traces of charcoal and little to no sign of burned phytoliths, disturbance-related phytoliths, or crop phytoliths. "If humans were in those areas, they didn't stay very long, and they didn't farm," says Piperno.
Moreover, this picture of small groups of hunters and gatherers in the western Amazon seems to fit with some data concerning ancient agriculture in the region. Before the arrival of Europeans, Amazonia's inhabitants had only stone axes at their disposal, and clearing large rainforest tracts with such tools would have been enormously time-consuming. A study conducted in the 1970s, for example, showed that a worker equipped with a stone ax needed 115 hours to cut down just one tree. So it is possible that slash-and-burn agriculture arose in the western Amazon only after Europeans arrived with metal tools.
The team's findings are bound to stir controversy and debate, however. Anna Roosevelt, an archaeologist at the University of Illinois, Chicago, thinks the team failed to gather sufficient data to reach any conclusions. She says that the 120-centimeter-long soil cores that team members took with a handheld auger may not have been sufficient to find deeply buried evidence of ancient human agriculture. "You really need to study the geomorphology of each sampling site," says Roosevelt, "because in Amazonia there's a lot of erosion and redeposition [of sediments by water], and you need to find out exactly where the human occupation surfaces are."
But Deborah Pearsall, an archaeologist at the University of Missouri in Columbia, sees little to criticize in the study, calling it "really solid." The findings, she notes, make a lot of sense: the study areas in the western Amazon receive as much as 3500 mm of rain a year, far more than eastern regions where ancient farmers clearly tended a variety of crops. "So maybe early agriculturalists are not favoring this [wetter western region] because there are higher pest loads there and more highly leached soils."
Michael Heckenberger, an archaeologist at the University of Florida in Gainesville, adds that the study serves as a badly needed reality check to much of the blue-skying of the past, when researchers developed sweeping models for the entire Amazon region, an area roughly the size of the continental United States. "We've really just scratched the surface in this region [archaeologically]", Heckenberger concludes, "and I think we need to be very cautious in creating highly generalized models of what such a vast areas would have been like."