Archaeology in Amazonia

An Anthropogenic Landscape

By Sadie L. Weber

In June of 2017, when I should have been writing my dissertation in Andean archaeology, I joined an archaeological project led by Helena Lima, researcher and curator at the Museu Paraense Emílio Goeldi in Belém, Pará. I am an archaeologist who, until recently, worked exclusively in the Peruvian Andes, which is to say, I wasn’t very familiar with archaeology in the lowlands. 

This Brazilian project in the Floresta Nacional de Caxiuanã and surrounding areas demonstrates that it was not only the fluvial regions of Amazonia that were densely populated, but rather settlements extended into the interfluvial spaces—in this case, the land between the settled areas of the Xingú and Tocantins Rivers. Traveling down the Amazon River itself and the smaller igarapés – tributary streams – it is impossible not to see the mark of humans, even if this mark is not apparent at first. Looking out the side of the boat, you see a kaleidoscope of green, sprinkled with açaí and burití palms. Eventually, palms dominate your view, and this is when you know that you're about to see a settlement.

Helena Lima places a scale bar for the photo of the profile in the archaeological excavation in Caxiuanã. The top, dark layer is terra preta, the white layer is crushed shell that was used as construction fill, and the rust-colored clay is the (more or less) original land surface.

Eventually our boat stopped at a cluster of houses on the banks of an igarapé. We unloaded our supplies and walked into the jungle on a dirt road, and the village of Gurupá-Mirim came into view. This would be home for the next two weeks while we carried out rescue archaeology so the village could install a more permanent electrical grid. To no one’s surprise, we found substantial evidence of ancient human occupation. This area had, and still has everything: animals for hunting, açaí, shrimp, manioc and fish. 

Amazonia holds a position as a hotspot of biodiversity in our modern world. This vast space, under threat of human overuse, fires, and illegal mining and deforestation, sits at the center of many battles—both political and physical—over how Amazonia should be used and by whom. However, this highly extractive, damaging human presence only came into play in the last 500 years. Before the arrival of Europeans to South America, Indigenous populations established settlements—some very large—across this landscape. This landscape merits the qualifier "anthropogenic," that is, created by humans. 

The debate over human use of Amazonia has vacillated between two concepts: the idea of Amazonia as a superficially rich environment that was so sparse that it could not sustain large populations and the cultural parkland that could have supported up to eight million people. On the two sides of this debate were Betty Meggers and Donald Lathrap, two archaeologists from the United States who were captivated early on by the past and contemporary cultures in Amazonia. But, these two disagreed profoundly on Amazonia's capacity to support complex, extensive settlement. For Meggers, the beautiful polychrome ceramics of Marajó Island at the mouth of the Amazon River looked similar to Japanese ceramics. They, therefore, came to South America with ancient Japanese settlers of the area, but for Lathrap, precocious settlement in Amazonia gave rise to the monumental societies of the Peruvian Andes. While both Meggers and Lathrap have since passed away, the legacy of their work lives on in contemporary Amazonian archeology. The question of the intensity of human use of Amazonia remains. However, if we examine the archaeological evidence that has emerged in the last 50 years, we can see that not only was Amazonia ideal for human settlement but that it is also not an untouched wilderness without a cultural history of its own. 

The ribeirinha community of Carrazedo, located on the banks of the Amazon River near the city of Gurupá.

Initial settlement of Amazonia 

While the questions of when and how humans arrived in the Americas remain contentious, we know that by at least 15,000 years ago, humans already lived in South America, in particular, on the coast of Chile at a site called Monte Verde. In the grand scheme of time, settlement of Amazonia occurred shortly after that, or at least archaeological evidence reflects as much. In what is today the state of Pará near the modern city of Santarém lies the site Caverna da Pedra Pintada, a rock shelter with paintings adorning its walls. First excavated by Anna Roosevelt and Edithe Pereira in the 1990s and early 2000s and more recently by Pereira and Cristiana Barreto, the site was inhabited more than 11,000 years ago. When these dates were initially published, Roosevelt received pushback from scholars working in the United States. At the time, it was generally accepted that the initial settlement of the Americas occurred only 10,000 years ago when nomadic hunter-gatherers (e.g., the Clovis culture) followed big game in the harsh post-glacial areas of western North America. Caverna da Pedra Pintada, and other sites like it, provided refuge for people during climatically unpredictable seasons. And, the people who occupied Caverna da Pedra Pintada did not fit the typical image of early settlers of the Americas chasing megafauna across the continent. These people had a close relationship with their landscape that included the use of aquatic animals and plants that remain culturally important even today, including Brazil nuts and Astrocaryum aculeatum (a palm species known as tucumã or cumare), among other tree and palm species.

While sites like Caverna da Pedra Pintada are relatively rare, early occupation marks a crucial establishment of human presence in Amazonia beyond general ideas of Paleoindians chasing big game. Settlements like this spurred the consistent, permanent occupation of Amazonia and a deeply rooted indigenous relationship with the landscape, which took the form of management of plants and animals as well as the physical modification of earth. From this initial settlement came separate innovations in plant cultivation, landscape modification, and technologies like ceramics. Indigenous peoples modified the landscape so much so that their mark is still left today in the form of terras pretas, earthworks, and the biological diversity of Amazonia itself. 

Terras Pretas

A bag of farofa – toasted manioc flour – surrounded by our excavation equipment and terra preta samples. Even today, manioc remains a staple in Brazil and worldwide.

The natural iron-rich clay sediments of Amazonia are nutritionally deficient, acidic and quickly depleted. However, the cultivation of plants was possible, even early on. While there is fierce competition for soil nutrients among plants in Amazonia, certain plant species are better suited to these acidic, nutrient-scarce environments than others; these include manioc, papaya, and açaí palm, among others. However, over time as humans increasingly cultivated plants, people began living in more permanent settlements; they formed terras pretas de índio or anthropogenic dark earths (ADEs). When these terras pretas were first recognized by the academic world, they were thought to be the result of wind-carried ash brought east from the Andes following volcanic eruptions. This purely environmental explanation for such an essential element of Amazonia highlights the ongoing expectation of Amazonian biological and cultural poverty as well as the fact that indigenous populations of Amazonia are continually discredited for their work in creating the landscape that we know today. Instead, it is now recognized that these anthropogenic soils were created by humans, both intentionally and unintentionally. 

The proof is contained in the soils themselves. They are often rich in stone tools, fragments of pottery, animal bone and burned plant remains. More bluntly stated, they were ancient trash or compost heaps. At the confluence of the Negro and Solimões Rivers near the modern city of Manaus, Brazil, Eduardo Neves, professor of archaeology at the Museu de Arqueologia e Etnologia at the University of São Paulo, identified deep deposits of terra preta, some of which date to nearly 2,000 years ago. While estimates vary, it is thought that terra preta accounts for 3-10% of the land across Amazonia. In some areas, terras pretas are the result of swidden, or slash-and-burn, agriculture. While the phrase "slash-and-burn" might set off an alarm for those accustomed to ecological conservation discourse, swidden agriculture is a traditional practice that can increase the biodiversity in an area over time by opening space for easily out-competed species of plants and animals. 

Today, terras pretas are sought-after resources for farmers and horticulturists, despite the fact that sites where ADEs are found are archaeological sites. People mine ADEs for use as fertilizer, or gardens are planted immediately on top of the archaeological sites. While this leads to the destruction of archaeological contexts and culturally significant places, not everyone who mines or uses these soils necessarily knows that they are sites. 

The tiny ceramic fragment in the center of the photo is about to be overrun by squash vines. The workers at the IBAMA station mined the terra preta for their gardens to supplement their food supply.

The workers at the IBAMA (Brazilian Institute of the Environment and Renewable Natural Resources) station near where we stayed in Caxiuanã had dug out terra preta to plant their garden. Next to their squash and onion plants lay pottery and stone tools. This small garden highlighted a more significant problem in conservation: the false dichotomy between biological and cultural preservation. 

Amazonian Monumentality

Often overlooked by the popular imagery and imaginaries of the Amazon, ancient Amazonian societies built monumental settlements extending from the Madre de Díos department of Peru into the Llanos de Mojos in Bolivia, and further east into Mato Grosso, Brazil. These earthworks, geoglyphs, and roads were only relatively recently noticed by farmers clearing their land in the 1960s and 1970s, who initially assumed that these were trenches left behind from the Acre War. It was thought that these trenches were too perfect and too massive to have been constructed by local indigenous peoples. However, later in 1977, when Alceu Ranzi, a geographer from Acre flew over, he observed massive rectangular and circular shapes on the landscape, some with roads connecting them, confirming that these structures were not natural geological formations. 

However, their discovery is bittersweet; they only become visible due to rampant deforestation aimed to clear land for cattle husbandry. In the state of Acre alone, there are over 500 known earthworks, with at least another 300 in the surrounding areas. However, many remain hidden due to different land-use policies in Peru and Bolivia. 

Square and circular geoglyphs are visible in Acre´s deforested landscape. Photo by Sanna Saunaluoma.

Pioneering work by the late Denise Schaan revealed the extent and complexity of these geoglyph networks. Recently, scholars have determined that the areas in which today earthworks are present, were not covered with the dense tropical forest that many people have in mind when they think of Amazonia. When many of the earliest geoglyphs were first constructed around 3,000 years ago, the area that today encompasses Acre and Rondônia in Brazil and northeastern Bolivia was a transitional zone between forest and savanna that was much drier than today.

In June and July of 2018, I worked with Harvard University Professor of Anthropology Gary Urton at Neves at Sol de Campinas, a geoglyph site just 60 kilometers east of the city of Rio Branco in the state of Acre in northern Brazil. Sol de Campinas is a ring of mounds surrounding a central plaza that roughly covers the 25,000 square meters, or about the size of a city block. This does not even include the three extant roads radiating from the ring. This type of settlement still exists today in indigenous reserves across the Brazil; it is still an important way to spatially organize a community.  

Our project aimed at identifying how the mounds were constructed as well as the settlement and land use patterns employed. While the students enjoyed their time, many felt disillusioned about the lack of forest around us. The site was in the middle of modern cattle pasture, and despite our warnings, the students expected that they would be digging in dense jungle. However, even when Sol de Campinas was built some 1,500 years ago, the landscape looked very different than even the popular imaginary of Amazonia. 

The Sol de Campinas not only incudes the ring of mounds but also the roads radiating out from the central plaza. Our excavation units (the white tents) are dwarfed by the sheer size of the site. Photo by Sanna Saunaluoma.

Work by Jennifer Watling from the Museu de Arqueologia e Etnologia in the Universidade de São Paulo has confirmed that the area was covered by bamboo and palm forests that were managed by the people who lived there. When an earthwork was constructed, a small area of the bamboo forest was temporarily cleared and then allowed to regrow. It was only around 1,700 years ago that the humid evergreen forest that we think of classically as classically Amazonian began to expand farther south. 

Further, the maximum extent of the Amazonian forest seen today is only due to rapid, dramatic post-1492 depopulation of the region. A cultural forest turned into a seemingly natural one. However, the effects of ancient human action can still be seen on the biodiversity of Amazonia today. Biological inventories of tree species across Amazonia demonstrated that of all the trees present in Amazonia, just 227 of the estimated 4,970 species are hyper-dominant; that is, they account for half of all trees in the area. Moreover, many of these hyper-dominant trees are useful to humans and are focused around archaeological sites, including açaí palms and cacao, to name some of the most recognizable. As such, it is thought that much of the Amazon forest that we know today was created by indigenous populations.

This apparent ephemerality of forested Amazonia is problematic. On the one hand, archaeological evidence proves that indigenous people have been integral to the formation of Amazonia as it is today; on the other, this research normalizes human use of the landscape. However, this human use that we saw in the past was nowhere near as destructive as modern soybean farming, cattle ranching, logging or mining. Instead, traditional uses of the land by indigenous and eventually quilombo populations increased the overall health and biodiversity of Amazonia. 

Cattle from ranch near Sol de Campinas visit the excavations; the trees dotting the horizon are almost exclusively Brazil nut and Astrocaryum palms.

The human effect 

In 1541, Francisco de Orellana traveled down the Amazon River, starting near what is today Quito, Ecuador, and likely ending near Marajó island. He reported seeing extensive, densely populated settlements on the banks of the river with roads connecting them. The people in the fortified cities produced art and ceramics that "rivaled those of Málaga."  These fortified settlements surprised Orellana, but years later, the cities had all but disappeared. Orellana's account was discredited not only for his fantastical accounts of Amazon warriors, yes, the warriors from Greek Mythology, but also because these populations had disappeared. However, archaeological evidence now confirms Orellana’s observations. Amazonia supported complex societies in the past.

This rapid depopulation of Amazonia, particularly around the major rivers, is what some scholars have argued that lead to the range at which the forested area of the Amazon Basin exists today—removing humans from the environmental equation allowed for the expansion of these forests. However, areas with particularly high biodiversity and "healthy" forests are those where humans remain practicing traditional lifeways. Rather than repeating depressing facts about how much forest is lost in the Brazilian Amazon, a forward-looking perspective is necessary. Indigenous management, and thus, the demarcation of indigenous and traditional community lands, are essential for the survival of this resource. The pattern of ecological conservation that excludes humans from the equation does not work for this part of the world. Here, cultural preservation is biological preservation.

A modern manioc and banana plot is planted over a terra preta archaeological site near Carrazedo.


Sadie L. Weber is a post-doctoral fellow in the Harvard Department of Anthropology and works in the Peru and Brazil. She is interested in traditional lifeways, food, and environmental archaeology. Sadie can be reached at