Secrets in the Soil
There are, unsurprisingly, mysteries buried in the dark earth of the Amazon.
The soil doesn’t simply hide fragments of the region’s past. According to Eduardo Neves, it’s a narrative in its own right. The terra preta (literally, “black soil”) of the Amazonian basin—which derives its distinctive color from the charcoal, bone, and manure worked into it by indigenous peoples over thousands of years—can, when properly studied, serve as a catalog of agricultural history.
Neves, a professor of archaeology at the University of São Paolo, Brazil, recently delivered a public lecture at Dumbarton Oaks that outlined the history of Amazonian archaeology and the suppositions that have driven it up until now. At the same time, his talk proposed new theoretical perspectives from which to approach the field. Seeking to “interrogate archaeology,” Neves fought back against the “paradigm of marginality” he believes has wrongly cast the region as an infertile zone unable to support large populations.
Neves began the lecture by describing the incredible diversity of the Amazon basin. Occupying roughly the same amount of land as the continental United States, the basin plays host to a variety of biomes and seven distinct language families that comprise among themselves hundreds of native languages. This parallel between the environment and its human inhabitants was, in a way, the crux of Neves’s larger argument; as he would go on to assert, the lush biodiversity of the region is partly a result of the diverse human activities undertaken there in the Pre-Columbian past.
Neves then took a brief detour to outline previous scholarship, focusing on Betty Meggers’s 1971 text Amazonia: Man and Culture in a Counterfeit Paradise, a pioneering work in the field of cultural ecology. Meggers had argued that because the tropical soil in the basin was so acidic, the most effective approach to cultivation was slash-and-burn agriculture. This, combined with the main crop of manioc, or cassava, which grows quickly but is low in protein, forced early populations to move about frequently, preventing the establishment of large settlements.
The new consensus, one that Neves supports, contends that modern biomes in the Amazon basin are formed by ancient populations, and that the landscape itself, not merely the soil, was shaped by indigenous peoples. Much of the evidence for these claims, according to Neves, starts to appear in the stratigraphic record roughly 2,500 years ago, as the result of population growth and a settlement boom. Singling out the occupations at Pocó-Açutuba, Neves emphasized the stability and fertility of the terra preta, which contains ceramic sherds. According to Neves, Pre-Columbian indigenous peoples didn’t necessarily have to bow down before environmental limitations—they were very much capable of overcoming them.
But what, exactly, is the evidence for these claims? When excavating, Neves searches for both organic macroremains—chunks of preserved plant matter, like a corncob, that are visible to the naked eye—and microremains, miniscule fragments of wild rice or squash that require the aid of a microscope to discern. An even more telling trace comes in the form of phytoliths, small mineral bodies (most often of silica) that form inside a plant and are later fossilized, allowing them to survive when other organic evidence has decayed.
By searching for evidence like this, Neves has been able to discover signs of plant cultivation stretching back to the mid-Holocene period (6000–2000 BC). At Teotônio, a site located in the Upper Rio Madeira region of Brazil, Neves and his colleagues found evidence of the non-domesticated management of palms from approximately 6,500 years ago—findings that push back the oldest proven occupation date at the site by some three thousand years.
Neves spent perhaps the most time discussing another mid-Holocene site, Monte Castelo, located on the Guaporé River. Still occupied by the Tupari people, the remote site’s extensive shell midden was first excavated in 1983, though it wasn’t until thirty years later, with the aid of grant money, that Neves was able to visit the site.
In the wet season, as the high grasses flood, the midden is turned into an island; Neves and his team were forced to paddle to the large mound, but the effort was worth it. Over time, the large amount of shells buried in the midden have created a relatively neutral pH level in the surrounding soil, Neves explained, lending it remarkable preservative properties. Organic remains abound, and ceramic discoveries that appear to date from roughly 5,200 years ago would be among some of the earliest in the Americas.
After an intense discussion of the evidence, Neves offered a simple segue: “So what?” Monte Castello, as Neves explained, is not unique; sites like it are to be found throughout the tropical lowlands. The consequence of these findings, Neves believes, is that the old unified narrative of the Neolithic period is falling apart. “Ceramics, we are beginning to see, are not necessarily tied to farming,” Neves explained. They often predate the development of agriculture, and evidence of their production can be found far from traditional agricultural cradles.
This argument flowed naturally into a larger distinction Neves evinced, that between agriculture and domestication. “Domestication and cultivation may not be processes that, necessarily, lead to the development of agriculture,” Neves contended. Rather than way stations on a clearly defined road of cultural development, they might be ends in themselves. To encapsulate this state, Neves coined the phrase “the permanent incipient.”
It’s a conceptual turn that, Neves is convinced, would go a long way toward overturning “the notions of absence, uncompletedness, and emptiness” that seem to undergird the study of Pre-Columbian societies. When the nineteenth-century Brazilian historian Francisco Adolfo de Varnhagen declared that “for such people, who still live in childhood, there is no History, only Ethnography,” he was speaking within a developmental framework that Neves considers obsolete.
Neves ended his talk by letting loose, so to speak, and examining other sites with a more casual, broadly interrogative tone. He dwelled on the magnificent goldwork discovered in Tolima, Colombia, and displayed LiDAR images (a method of surveying that uses laser light to create highly detailed maps) of a site in northern Colombia where clearly designed manmade shapes are visible in the earth. The images, projected onto a screen, gradually zoomed out, and the individual geoglyphs gave way to a sprawl of overlapping shapes like a jumbled cipher.
As Neves evinced, there are still mysteries in the soil.