This idea is popularized by Charles Mann's book, 1491: New Revelations of the Americas Before Columbus; you can read a shortened, pie-in-the-sky version here.
Obviously, it's a touchy subject among ecologists and conservation biologists who have traditionally held the area as the quintessential wilderness, among the last on the planet. Researchers from the Florida Institute of Technology recently published evidence that supports the traditional view, adding new facets to the debate.
"We don't contradict that there were major settlements in key areas flanking the Amazon Channel -- there could have been millions of people living there," says Mark Bush, a British-born paleo-ecologist who travels to extremely remote rain forest locations to collect core samples from ancient lakes. He then analyzes those samples for pollen and charcoal and thus is able to conclude with a high degree of accuracy the extent of human settlement in that region.
"What we do say is that when you start to look away from known settlements, you may see very long-term local use," he says. "These people didn't stray very far from home, or from local bodies of water for several thousands of years. We looked at clusters of lakes and landscapes where people lived, and asked, did they leave their homesite to farm around other nearby lakes? No they didn't. These findings argue for a very localized use of Amazonian forest resources outside the main, known, archaeological areas."Bush says the evidence comes from a geographically diverse area: three districts, each with 3 (in two cases) or four lakes.
"In each we have one lake occupied and used, and the others little used or not used at all," he says. "So this is a total of 10 lakes that provide three separate instances -- one in Brazil, one in Ecuador and one in Peru, where there is evidence of long, continuous occupation of more than 5,000 years that did not spread to the adjacent, 8 to 10 kilometer distant lakes."
"These data are directly relevant to the resilience of Amazonian conservation, as they do not support the contention that all of Amazonia is a 'built landscape' and therefore a product of past human land use," Bush says. "Most archaeologists are buying into the argument that you had big populations that transformed the landscape en masse. Another group of archaeologists say that transformation was very much limited to river corridors, and if you went away from the river corridors there wasn't that much impact. That's what our findings tend to support."
Bush doesn't expect that his new findings will settle the debate, however.
"There's just too much passion on this issue. People who are inclined to believe what we're talking about will say this is very strong evidence, and say 'let's have more.' The archaeologists will say this study only examines two districts."
Those damn archaeologists. Look what else they say:
"While the majority of archaeologists argue the rivers were the major conduit for populations," he adds, "there is an increasing vocalization that there was much more widespread habitat transformation; that you still had a bulk of people along the river but their influence extended deep into the forest. It's still nebulous, and difficult to get people to map stuff, or put hard numbers on it, but there is a sentiment that the Amazonia has been disturbed and that the view of the Amazonian rainforest as a built landscape is gaining momentum. There are extremes at either ends, and the majority of people are in middle but there's a tendency of drifting toward the high end."
Bush is implying a split between scientific disciplines on the issue. I don't want to generalize, but it's probably safe to say that proponents at either end are wrong and clinging fast to dynamic ideas that they like.
The idea that native peoples were responsible for creating Amazonia in its entirety is just as ridiculous as rejecting the notion that those same peoples had a lasting localized effect on the ecosystem, and perhaps did stimulate growth and development on some level.