From an early age, Anna Curtenius Roosevelt knew she wanted to be an archaeologist. Her mother, one of her biggest influences, was painter Frances Blanche Webb Roosevelt (1917–1995), who spent part of her childhood living among archaeologists in Arizona and New Mexico, USA, and later took her daughters—Alexandra, Susan, and Anna herself—to visit the region’s archaeological sites. “That’s how I became interested in the field,” says Anna Roosevelt. Now a professor of anthropology at the University of Illinois, Chicago, she is recognized as one of the most prominent experts on Amazonian archaeology. Her work in Brazil, however, is viewed with reserve by some archaeologists. One criticism is that she never used the opportunity to train Brazilians in the field. Another is that she has taken excavated material to the United States.
Anna Roosevelt initially thought about dedicating herself to classical archaeology and studying ancient cultures in the Middle East. While studying her bachelor’s degree at Stanford University, however, an internship at the Museum of Natural History in New York opened her eyes to South America. During her PhD at Columbia University, she studied the communities that lived on the floodplains of the Orinoco River before European colonization, in present-day Venezuela. Based on her work there and at archaeological sites in Santarém and on the island of Marajó, both in the state of Pará, she reinterpreted how the Amazon and the American continent were historically occupied.
In collaboration with researchers from the Emílio Goeldi Museum of Pará, Roosevelt identified rock paintings almost 13,000 years old at archaeological sites in the municipality of Monte Alegre, near Santarém, one of which is considered the oldest solar observatory in the world.
The great-granddaughter of US President Theodore Roosevelt Junior (1858–1919), who went on two expeditions to the Amazon at the beginning of the twentieth century, Anna Roosevelt still regularly returns to Brazil, where she plans to carry out further excavations in the coming years. She spoke to Pesquisa FAPESP by telephone on October 11.
Field of expertise
University of Illinois, Chicago
Bachelor of Arts in History, Classics, and Anthropology from Stanford University (1968) and PhD in Anthropology from Columbia University (1977)
Roughly 100 scientific articles, as well as numerous books and chapters
Your work helped change our understanding of prehistoric human settlements in South America. How was it that you were able to reach such different conclusions to the archaeologists who preceded you?
The researchers in the 1950s who hypothesized on Amazonian archaeology actually changed what was the prevailing theory before that time. At the end of the nineteenth century, naturalists such as Domingos Soares Pereira Penna [1818–1888] and Emílio Goeldi [1859–1917] understood the complexity of Amazonian societies. Archaeologists who later worked in the region, such as Betty Meggers [1921–2012] and her husband, Clifford Evans [1920–1981], both Americans from the Smithsonian Institution, believed that only desert regions with big rivers like the Nile or the Valley of Mexico could support civilizations or complex cultures. But here’s the interesting thing: we know that human beings arose in tropical forests in Africa, not savannas. Even though we were tropical forest animals in our early history, Meggers and Clifford concluded that humans could not survive in tropical forests because of the many diseases and the difficulties of farming there. For them, the only civilizations that existed in the Amazon were those that originated in other regions, which would have quickly disappeared.
What made you think differently from them?
Early in my career, I was not dealing with the issue of tropical forests. But I was interested in agriculture, and I knew that the floodplains of the Orinoco and Amazon rivers were even larger than the Nile’s, and were full of good soil. My reasoning at the time was that this would have been sufficient to support complex cultures. During my PhD, I showed that in the Orinoco region, people practiced intensive agriculture during population growth before European colonization. My focus was on floodplain societies. As I learned more about tropical forests and the geology of the Amazon, I realized that there were many different types of forest, even in areas of dry, solid ground [known as terra firme], where the soil, generated by erosion of the Andes, was good enough to support agriculture. This kind of soil could be intensively cultivated, provided that certain measures were taken, such as building terraces to retain the soil or using organic waste as a fertilizer, as was done with Amazonian dark earth. In Southeast Asia, in many areas where complex cultures like the Khmer emerged, intensive agriculture took place in tropical rainforest areas. So I realized that in the Amazon, complex societies could have existed in terra firme regions, depending on the nature of the soil and the agricultural strategies used. But native peoples did not clear the forest or burn it. They developed a form of agroforestry. On the floodplains of large rivers, such as the Llanos de Moxos region in Bolivia, the Ucayali river in Peru, and the Orinoco in the Guianas, people raised and drained the land, they did a lot of earthworks, and they used the land intensively for open cultivation. Inside the rainforest, they cultivated orchards.
Where did you start your fieldwork before coming to Brazil?
My dissertation was based on work I did in the Orinoco River region in Venezuela. Then I applied for a grant from the US government to analyze museum collections containing material from the Amazon in North America, South America, and Europe. It was a great opportunity to travel through Peru and Colombia. On my first trip to Brazil, in 1980 or 1981, I got to see Amazonian dark earth in Santarém and I realized that it was formed of garbage from large settlements. At the National Museum of the American Indian, in the USA, and at the Emílio Goeldi Museum of Pará and the National Museum in Brazil, I saw projectile points from Paleoindian sites [cultures that existed between 13,000 and 10,000 years ago], some of which were collected by archaeologist Mário Simões (1914–1985). That’s when I realized that the sequence of these cultures in the Amazon was much longer than the textbooks said.
How did you manage to get a permit for your first excavations?
It was through geologist José Seixas Lourenço, director of the Goeldi Museum at the time, who was working in Marajó. I gave him a copy of my PhD thesis, and he must have thought that my work would validate his, which it certainly did. Although Simões and I were friends, he may not have given me a permit to dig at the site. Simões did not recognize the projectile points as Paleoindian: he believed they were from an archaic culture, which began around 10,000 years ago, following the Paleoindian era. This is because at the time, everyone was very influenced by American archaeologists who claimed that the only Paleoindian culture that could have existed was the Clovis culture, in New Mexico, USA. The points found in the Amazon, however, were not like Clovis points. They were triangular and had a peduncle [a small stem]. Other researchers studying Paleoindian cultures in the Amazon, California, and Peru came to the same conclusions as I did. The transition that had been imagined from the Clovis culture, when people hunted large animals, to the archaic period, when people fed on plants and fish, never happened. We found that in coastal areas and along rivers, people were never big game hunters—they were broad spectrum foragers. Of course, many researchers, such as Brazilian archaeologist Oldemar Blasi [1920–2013] and American Wesley Hurt [1917–1997], who had worked in Lagoa Santa, Minas Gerais, already knew this. What we discovered was not particularly different to what others were observing. Looking at the complete sequence of cultures in the Amazon allowed us to improve our understanding of how the region was historically occupied and how Paleoindians lived in other parts of the continent.
Were there any differences between the way of life of Paleoindians in the Amazon and those elsewhere in what is now Brazil?
No. They fed on small fish and palm oils and fruits, such as the murici, and legume pods, such as the jatobá [Hymenaea courbaril]. They also ate very small animals. They loved turtles.
Was it important to start your studies with museum collections?
I learned a lot from these collections. The collection at Harvard includes shellfish and pottery collected in Brazil in the 1870s by Canadian geologist Charles Hartt [1840–1878]. Hartt and Pereira Penna both described Amazonian middens from the archaic period. I used the shells to date the midden in Taperinha, near Santarém, before I even went there. It was almost 6,000 years old. When we excavated the deepest part of the Taperinha midden, which is more than 6 meters deep, we found even earlier pottery, about 7,000 years old. The pottery from Pedra Pintada, in Monte Alegre, was older still, almost 500 years older than the earliest pottery found in Taperinha. My initial plan was to start my fieldwork in Santarém, which I felt had been home to more cultures. It has Paleoindian material as well as archaic pottery. This pottery, in fact, represents a completely new form of adaptation and contributed to our understanding of the peoples of the Americas. It indicates that very early, humans in Amazonia assumed sedentary occupations, with diets based on fish and shellfish.
Was it the most important place where you worked in the Amazon?
It was where we established the presence of Paleoindians in the Amazon. The archaeological material also reveals a complete sequence of cultures that existed in the region.
The complex societies of Marajó were comparable to those of early Egyptian cultures
What does this sequence of cultures tell us?
It shows tremendous dynamism and creativity. There are examples of monumental art from the very first moments of occupation. Most of the paintings are over a meter long and many panels are several meters long. In Serra da Lua, there are 500 meters of paintings. In the Pedra Pintada cave and at Painel do Pilão, in Monte Alegre, both I and my former student Chris Davis found paint pigment in the oldest layers. So the first thing they did when they arrived in the region was to make these paintings. As Chris described in his PhD thesis, one society set up a solar observatory, which let them know what time of the year it was. This is the oldest known solar observatory on the planet, dating from the beginning of the Paleoindian period 13,000 years ago—the Paleoindians stayed in the region for one or two thousand years. Today the tradition of painting pottery continues among the Shipibo, in Peru, but it really began in Marajó at the beginning of the Christian era. People were very curious and intelligent. From the first Paleoindians to current cultures, small fish have always been an important staple food, as well as palm fruits and other tree fruits. They used and still use the forest intelligently. In some places, they cultivated the floodplains intensively, producing corn and other crops. At the same time, they planted trees that generate more nutritious fruits, such as palm fruits, pequi, brazil nuts—they produced a lot of food. You can produce a lot more food and for a longer time that way than growing soybeans or raising cattle.
What did your work in Marajó involve?
We analyzed several mounds, and found that several had been domestic occupations, not just ceremonial sites. They had garbage and hearths. By the end of the 19th century, Hartt’s students already understood this. Archaeologists of the early 20th century, however, did not use this information. Our excavations revealed that a large population lived in Marajó for almost a millennium.
In your book, Moundbuilders of the Amazon, you describe these people as being part of a civilization somewhat similar to other ancient civilizations, such as the Egyptians. What are the similarities between them?
The complex societies of Marajó are comparable to those from the beginnings of Egyptian culture or to the first complex societies in the Near East, such as the Çatalhöyük, in Turkey. What we call Egyptian culture is a combination of millennia of development. The early complex societies existed long before the first pharaonic dynasties, which built the pyramids. That’s what Marajoara culture looked like. It would also have been similar to the formative cultures of Mexico, Peru, or Asia. They go through similar stages of adaptation and development. They created ceremonial monuments, although the way their settlements were organized varied widely. Marajó has many more sites than the Çatalhöyük culture, for example, and they are much bigger too.
What took you to the south of Pará after having worked in Marajó and Santarém?
I went to investigate what happened to the Paleoindian culture in interfluvial areas [regions of terra firme between rivers]. An airplane pilot who was interested in archaeology brought a photograph of a projectile point to the Goeldi Museum and one of my associates there, Mauro Vianna Barreto, now a researcher at the Federal University of Pará [UFPA], sent it to me. I noticed it was the same type of projectile point that I had found in Pedra Pintada. So I made some calls to find out where the point actually was, and it was in the hands of a hotel owner in Castelo dos Sonhos, in the Xingu. He told me that an elderly gold miner called Waldemar had found it.
Did the miner help you?
I went to speak to him with Maura Imazio da Silveira, an archaeologist and longtime friend of mine from the Goeldi Museum. We asked him if he had found any wooden objects or basketry, because they are often preserved in water, and he said: “Wait a minute. I want to show you something.” He came back with a piece of wood that appeared to be a harpoon foreshaft. It was one of the most wonderful moments of my career. I had already hypothesized that the projectile point was part of a harpoon. It dates from a time when the archaeological site Curupité was a rapids and it was probably used to catch fish during breeding seasons. The region is submerged under about 12 meters of water now. Waldemar got on a small plane with me and Maura and we flew over the region where he found it in the Curuá River. We marked the coordinates of the location on the GPS so that we could later return by boat. The second time we were there, Waldemar’s son-in-law Ney dived with us and showed us where the point was found.
Did you return to Curupité? Have you found anything else there?
We have been several times, and we need to go back. We made a good map of the underwater archaeological site, but we need to go back and excavate to get a better understanding of the culture that lived there. Curupité is important because it shows that there are submerged sites in the Amazon, which are very important because they preserve the materials so well. There may also be Paleoindian sites along the Rio Negro region of Amazonas. Geologist Elena Franzinelli, from the Federal University of Amazonas, was given some projectile points from the region by local indigenous people, and they are very similar to the one from Curupité. I need to work with an oceanographer to study the region. One of the interesting things about Amazonian cultural history is that in certain places, women seemed to be more important than men, as reported by Spanish missionary Gaspar de Carvajal (1500–1584). For example, I have found only one image of a man represented in the art of Marajó. Most of the human figures are female.
Is there any evidence that women played a more important role in these cultures?
It was quite common for women to be more prominent in earlier complex societies. In Egypt, some of the early dynasties were matrilineal, and they were often ruled by two people: a man and his mother, the queen mother. I call this dual rule, which also existed in many African kingdoms in tropical regions during the colonial period. The Shipibo, a current Amazonian society in Peru, represents women in pottery and paintings of shamans. It is a matrilineal culture where women are the heads of the family home, according to Ronald Weber. On the Rio Negro, however, most societies are patrilineal. But in interviews with anthropologists, many current peoples of the Rio Negro say that they are descendants of a society run by women on a large island at the mouth of the Amazon. For me, that’s Marajó.
When you look at their art, they use the anaconda, which is the spirit animal of the “woman-shaman,” in their pottery and paintings. There is a hypothesis that in areas of the Amazon with poor soil, such as around the Rio Negro, people formed patrilineal societies, although it is not clear why. But those on the Amazon floodplains were matrilineal, like the ancient peoples of Marajó and the Shipibo, who live to this day on a floodplain of the Ucayali River in Peru, a tributary of the Amazon. Women had a greater role in the floodplain regions than in other areas. In the 17th century, Jesuit missionary Samuel Fritz (1654–1728) reported on a polychrome culture [those who paint their pottery with red, white, and black ink] in Amazonas, where it was the women who made the pottery. When he asked them about the meaning of the patterns they painted on the pottery, which are the same as observed in Marajó, they told him it was the anaconda. Shipibo women also say the patterns on their huge pots represent the woman shaman. They call it the great anaconda. In the Marajoara culture, many images show women with the clothing and tools of shamanism. In addition, the houses are centered around hearths, which is common to matrilineal societies. In these communities, the families were led by the mother, whose daughters continued living in the house after they got married, with the husband moving in. The daughters grew up cooking with their mother, so their hearths are together. In patrilineal cultures, such as those on the Rio Negro, the cooking hearths are separate, because the women in each house are not related—it is the men who are related.
Were women influential in other ancient communities around the world?
Yes. In early Greek societies, women’s graves were richer than men’s. The women were wealthier. It’s very common in certain periods of development for women to be more prominent. When empires develop and militarism becomes more important, men start to assume a more predominant role.
Your education was influenced by the women in your family. What did you learn from them?
I learned from the women in my family and from the societies I studied that were ruled by women or dual ruled by a man and a woman. I was raised in a matriarchy. My paternal grandmother, Eleanor Butler Roosevelt [1888–1960], was the head of the family. She and my mother were widowed young and moved in together. Then there were me and my two sisters. My grandmother was very intelligent and moderate, she commanded quietly. All the men involved—my uncle and the man who took care of the property on which we lived—were very receptive to female leadership. Later, when I read about the Assante people of Ghana, I learned that the men were very respectful of the Queen Mothers. But matriarchies can be harsh sometimes too. I did a literature review about the Warao people from Venezuela, Guyana, and Suriname, who are also matrilineal and live in stilt houses. In this society, men come from other groups to marry the daughters, and they can have a difficult life. Some young men suffer abuse at the hands of their wives. It is not uncommon for them to be taken to clinics having been beaten. So not everything women do is wonderful. But on the whole, women are less likely to abuse substances, which allows them to make better decisions. And according to studies, they are usually less violent, although they can be combative.
Do you have any plans to return to Brazil?
I intend to excavate at the underwater Paleoindian site in Curupité, and at the Guajará archaeological site in Marajó. There are some very well-preserved cemeteries there. But I would like to go back to Curupité first, because the mining activities there may be changing the landscape and putting the site at risk. I would also like to work on the Rio Negro, as soon as I can find an oceanographer with a boat or a submarine to collaborate with. The river is so deep that it’s not possible to scuba dive. I also have a colleague, archaeologist Alexandre Guida Navarro of the Federal University of Maranhão, who has invited me to explore the old stilt-house villages in Maranhão. First, however, I need to do interviews and excavate in Africa—either in Angola or the Congo. There is a culture there called Sangoan, which in my opinion, was the first early human culture of Homo sapiens sapiens [this culture produced stone tools and existed from about 130,000 to 50,000 years ago, spreading across most of tropical Africa at the time]. It is associated with the tropical forest. The theory that the tropical forests in the region became savannas is not correct. There is no evidence that savannas existed in the Congo River basin during the Pleistocene, a geological period that extended from 2.58 million to 11.7 thousand years ago. Stone tools produced by this culture have been found well preserved under water. I hope to excavate a site near the border between Angola and the Congo. If you find well-preserved wooden objects and plant remains, you can get an idea of what the environment was like and what tools they used.
I see no evidence of human occupation in South America prior to 13,000 years ago
How long have you been working there?
Since 1997. What I learned from our excavations in Pedra Pintada led me to question the hypothesis of the emergence of primitive human beings in savannas in the Pleistocene. Because I am interested in environmental archaeology, I had realized that most of the conclusions about the environment during human emergence in Africa were wrong. So I wanted to work there to try to reveal the tropical forest aspect. In Africa, I applied a lesson I learned in Amazonia: it is very important to look at living societies to gain insight into the ancient ones, and vice versa. In the Amazon, people said that the indigenous peoples were primitive, that they lived on subsistence agriculture and had not developed a civilization. But look at the prehistory. It is very different. I learned that the conquest of Amazonia by Europeans and the establishment of governments by European nations and empires changed the way the natives lived. The people there not only adapted to the environment, they also adapted to historical and political events and processes.
In which museums or institutions are the artifacts that you excavated in the Amazon stored? Do other researchers have access to them?
Most of what we collect in excavations are small, broken, carbonized biological remains. Brazilian museums did not wish to acquire these. They are useful mainly for identification of species and for dating. Most of the cultural material from our Santarém excavations went to the local museum. Some went to archaeologist Denise Schaan (1962–2018), from UFPA. It consisted of small pottery sherds and lithic fragments produced when people create stone tools. We excavated mainly in ancient garbage. I heard that termites attacked the labels and boxes in the Santarém Museum. In Chicago, I have the charcoals, animal bone, and some flakes that were with the sherds that went to UFPA. The projectile points from Pedra Pintada are in the Goeldi Museum, as are the few decorated or rimmed sherds from Taperinha. At first, the Marajoara material was kept in a Goeldi room in Rocinha [the location of the museum’s administrative headquarters]. The whole pieces are still there. The director asked me to remove the small, broken materials because the museum was unable to store it in its facilities. At the museum, the material had eventually been placed in an outdoor enclosure, which visitors sometimes got into. So we moved it first to a storeroom I rented in Santarém, with the family of WIlton Hagman, the owners of the Taperinha site, and then eventually to the University of Illinois at Chicago. Scholars are free to come study these materials, but only Alexandre Guida Navarro and José Oliver have come to see the pieces so far. I know that some researchers have visited the collections in Belém.
There is evidence of human occupation 25,000 years ago in what is now Mato Grosso, and 15,000 years ago in Serra da Capivara, in Piauí. What do you think about these dates, and what might they tell us about the occupation of South America?
The very early dates are not based on secure cultural material [objects made or modified by humans] in either case. The purported lithics are not tools, to me. I see no evidence of human occupation in South America prior to 13,000 years ago.
Are new technologies, such as genomics, useful in Amazonian archaeology?
Yes. More than 10 years ago, DNA analysis was carried out on a thousand-year-old sample from the lower Amazon. It was one of the many results from the Americas that showed great genetic diversity in the prehistoric period. Previous data suggested that the genetics of living groups in the Amazon and elsewhere in the New World was uniform and that there was a lack diversity. They hypothesized at the time—wrongly—that there was a population bottleneck at the Bering Strait, that only a few people crossed it and their descendants spread widely across the continent, resulting in a low genetic diversity. When this hypothesis was put forward, genetic material had not yet been extracted from ancient skeletons. We now know that the idea is incorrect. Genetic material from Paleoindians who lived around 12,000 years ago and from the Windover archaeological site in Florida, dated around 6,000 years, show that the diversity is similar to that observed in Northeast Asia, from where the first human beings to arrive in America came from. But this diversity is no longer found in indigenous groups today. The real population bottleneck occurred in 1492, when the Europeans arrived. It is common for researchers to attribute the development of indigenous cultural behaviors and characteristics to nature, but it can result from human influence.
You are the great-granddaughter of President Theodore Roosevelt (1858–1919), who himself participated in two expeditions to the Amazon. Have you visited the places he went to?
No. His first expedition was more to the south and the second much further southwest of the places I was interested in. When I read about his second expedition, I told myself that my digs would never end with someone dead [on Theodore Roosevelt’s expedition to the River of Doubt in 1914, one person drowned and another was murdered; a third was left behind]. His expedition was not well organized. He became sick and very emotional. At one point, he told the group to go on without him. To me that seems narcissistic. You have to have a strong personality on a dig, but you cannot be narcissistic.