prehistoric questions

The pebbles of contention

Archeologists can’t reach consensus on peopling of the Americas

Bernardo Esteves

 versão em português (para assinantes)

The French archeologist Antoine Lourdeau was excavating a site in the Serra da Capivara mountain range, in the Brazilian state of Piauí, when he encountered a large pebble. He was only able to see one side but the stone’s size and shape suggested it would fit ideally in an adult hand. He suspected it had been fashioned as a tool. Using a brush and a trowel he delicately unearthed the object. Little by little, he revealed the other side, where in fact, chips had been removed. Lourdeau had no doubt. “This was a marvelous tool, a very fine piece of work,” he said. “I won the day.”

Last winter, Lourdeau unearthed hundreds of stone fragments which he believed to be tools made by prehistoric peoples to cut, scrap, dig, and work leather or wood. Many of the pebbles had chips removed on only one face, and since the artifacts took advantage of the natural shape of the stone, their appearance is not always what one might expect of a tool. For Lourdeau however, their identification is elementary. “When you see an object like one of these, there is no doubt,” he says.

At a restaurant near the beach at Boa Viagem, in Recife, the archeologist explained that fracturing of rock to produce stone tools always follows the same principles. “Nature breaks rocks, but man has an objective in mind when he chips a rock. If you follow the sequence of removals, it is possible to know whether they reflect a predefined objective. If there is a logic to the removal of material, you can disregard natural forces.

Lourdeau is thin and has long hair; and he appears younger than his 31 years. At the University Paris X, he specialized in the study of stone tools in the pre-history of Brazil, training under Eric Boëda, an expert in the field. Ten days after defending his doctoral thesis, Lourdeau solicited a position at Federal University of Pernambuco in the Brazilian city of Recife. He was hired, and in 2011 he moved to the Recife neighborhood of Boa Viagem. He has been participating in the excavations organized by a French-Brazilian team under the leadership of Boëda. The group has continued the work begun by the Brazilian scientist Niède Guidon in the Serra da Capivara, an ongoing research project since the 1970’s.

Lourdeau lead the dig at the site Toca da Tira Peia, a 25 square meter area excavated to a depth of 2.5 meters. The density of material uncovered was not particularly significant: between 2008 and 2011, the site produced only 113 pieces interpreted as undisputable artifacts. The results began to be published last year in the Journal of Archaeological Science, the highest impact factor journal in the field. The paper concluded that a number of the artifacts are at least 22,000 years old. If this interpretation is accepted by the archeological community, it would make these the oldest evidence of human presence on the American continent.


We are all foreigners in America. With the exception of Antarctica, which remains largely uninhabited, the American continent was the latest to be occupied by Homo sapiens since the species’ first appearance in Africa, between 150,000 and 200,000 years ago. Remains left by prehistoric humans allow us to trace their movements as they spread throughout the Middle East, and from there to Europe and South East Asia, Oceania and Siberia and ultimately, the Americas.

At the time when the first humans conquered the world, the planet was experiencing a prolonged period of cooling and glacial expansion. Sea levels were more than 100 meters below those of the present day, islands existed which are today submerged, and the contours of the continents looked quite differently. The Bering Strait, the 90 kilometer ribbon of sea separating Alaska from Siberia, did not exist. Eurasia and the Americas formed a continuous land mass. At the end of the last ice age, a particularly long-lived and cheerful errant hunter-gatherer might well have wandered from the Iberian Peninsula all the way to Tierra del Fuego entirely on foot, had it suited him.

Scientists estimate that Beringia, the land bridge between Asia and America, remained emerged from 27,000 to 10,000 years ago. The bridge is believed to be the point of arrival for the first Americans. Recently, this hypothesis was reinforced by DNA testing of human fossils, which revealed genetic similarities between Native American and Asian populations. Researchers proposing other models of population – via the Atlantic or Pacific – have not achieved general acceptance. And until recently, the leading hypothesis proposed that the first peoples arrived in the Americas around 15,000 years ago – a number which is incompatible with the suggestion that humans were knapping flints in Piauí 7,000 years prior.

The first author of the article published in the Journal of Archaeological Science is Christelle Lahaye, a thirty five year-old French physicist, and a specialist in dating archeological sites. The most common dating technology used by researchers is radiocarbon dating, which measures the presence and quantity of the isotope carbon-14. Radiocarbon dating however, can only be used for organic matter, such as the remains of fire wood, and fossilized skin or bone with preserved collagen. No organic material of this kind was found at the Toca da Tira Peia site.

The team relied on photoluminescence dating, Lahaye’s area of concentration at the University of Bordeaux 3. The technique allows scientists to measure a material’s last exposure to sunlight – in other words, the approximate date it was buried underground. Lahaye explained that radiocarbon dating has less margin of error, but cannot always be used, and is only accurate up to 50,000 years. “Photoluminescence allows us to date events which could not be dated by other methods,” she explained; “however, in case of controversy, it is important to combine the two methods.”

Lahaye went to Piauí to collect samples and survey the context in which they were removed. Material is captured using a PVC tube inserted into the wall of an excavation; only the material inside the tube, protected from exposure to light, can be dated. The analysis does not determine the age of the artifacts themselves but of the layer of sediment in which they were buried.

It was not the use of photoluminescence dating however, which generated the criticism the paper received upon its publication. When the article appeared, Science News magazine gave voice to two archeologists unconvinced by the results. One said the fragments could have been the result of erosion – the natural fracturing of the rocks. American archeologist Stuart Fiedel proposed another possibility: capuchin monkeys or other primates might have produced the tools.


At the start of the last century, a violent summer storm caused flooding in the city of Folsom, New Mexico, in the U.S. Southwest. Upon overturning the earth, the rain brought to the surface the bones of a giant bison, an extinct species of giant mammal which had been abundant in North and South America at the end of the last ice age. In 1927, during an excavation of the site, two ribs of the animal were found with a stone flint wedged between them. The finely made point had grooves or flutes indicating it may have been attached to the shaft of a spear.

The digging at the site was interrupted immediately to telegraph archeologists at various national research centers. The specialists arrived within days and gave their verdict: the find was an unequivocal sign that the bison had been killed by hunters. This was the first verifiable indicator of humans on the American continent in the midst of the last ice age.

A few years later, other spear points even larger than those found at Folsom, although less sophisticated, appeared in the city of Clovis, New Mexico, this time alongside the bones of giant mammoths. Soon, similar projectile points began to be found at other sites in the United States. Apparently, the central plains of North America had been inhabited by populations hunting large animals. They became known as the Clovis people.

In instances where artifacts appear associated unequivocally with fossils, as with the points found at Folsom and Clovis, age can be determined by radiocarbon dating of the adjacent organic material. Only in 1964, with the advent of radiocarbon technology, was it possible to determine with any precision the provenance of the two points: 13,000 and 13,500 years ago.* Scientists suspect that around that time, a great corridor had been opened between the immense glaciers covering the territory that is today Canada. This had been something like an avenue for the migration of hunter gatherer peoples coming from Asia, perhaps in pursuit of large game. It appeared that these hunters may have been the first Americans, and this model of population became known as “Clovis First”.

Discovered at hundreds of locations, the Clovis points became objects of devotion in the U.S., inspiring publications, excursions, exhibitions and workshops related to stone tools. The rarest examples can be traded for tens of thousands of dollars. The flints became something of a founding myth of the culture of innovation: in his book First Peoples in a New World, the archeologist David Meltzer suggests that the technology of the spear point would have been the first American invention.

There was however, a problem to be faced by the proponents of this model. The Clovis points all date to a rather narrow window of time and occur in a similarly restricted geographical region – Panama is the southernmost point at which they appear. Nevertheless, throughout the rest of the continent, south all the way to Tierra del Fuego, there are other signs of the presence of humans which are nearly contemporary with the Clovis culture. If the Clovis people arrived first, they were very quick to cover so much territory. Additionally, the stone tools attributed to southern peoples bear little resemblance to the Clovis and Folsom points. The Clovis First model does not account for this technological diversity, nor the similar age of sites south of the Equator, and yet, these discrepancies have not impeded the theory’s general acceptance.


The incredulity with which some archeologists received the news of the French-Brazilian finds at Piauí echo a similar resistance towards the results of earlier excavations in the region. The supposed tools of 22,000 years ago from Toca da Tira Peia are not the first artifacts older than the Clovis points to appear in the Serra da Capivara – nor the oldest.

At the center of the controversy is the archeologist Niède Guidon, a 1,50 meter tall woman born in Jaú, in the Brazilian state of São Paulo, to a French father and Brazilian mother. She studied natural history at the University of São Paulo and pre-history at the Sorbonne, with André Leroi-Gourhan and Annette Laming-Emperaire, seminal figures of French archeology. Niède will be 81 in March. Since the 1990’s she has lived in Piauí, in the city of São Raimundo Nonato, where she directs the Fundação Museu do Homem Americano, the foundation responsible for managing the Serra da Capivara National Park. She works at her house, next to the foundation, in a glass-walled, climate controlled veranda.

During our interview at her home, Niède explained how she had come to find out about the archeological treasure in Piauí, when in 1963, she organized an exhibition of prehistoric paintings at the Museu Paulista. She was approached by a visitor looking to speak with the curator of the exhibition. “He said that near his land there were indigenous paintings and he showed me some photographs”, she said. “I could see that this was something completely different.”

A few years passed before Niède was finally able to visit the paintings at Piauí. In the interval Brazil experienced a military coup and she decided to move to France. She became a researcher at the CNRS, the French national research center, and later, at the School of Advanced Studies for the Social Sciences in Paris. On a research trip to Brazil, she found an opportunity to visit Piauí and photograph a few sites with paintings. The photographs helped her to obtain financing for further study of the region, in 1973.

Prehistoric paintings are abundant on the sandstone walls of the Serra da Capivara. There are more than a thousand identified sites in the region. Unlike the cave paintings of Europe, like those found at Altamira in Spain or Lascaux, in France, the paintings at Piauí are found on exterior rock walls, exposed to the open air. They depict armadillos, emus, monkeys, and lizards, as well as an abundance of human figures, something rare in the prehistoric paintings of Europe. There are figures hunting, playing, dancing, fighting, making love, and giving birth.

Since the pigments used were not made from organic material, it is not possible to obtain a direct measure of their age. Other means of dating chips of paint found at archeological sites indicate that the paintings became abundant around 10,000 years ago, although they may have begun long before that – some of them have been dated closer to 20,000 years. The cave paintings at Chauvet, the oldest of which we have evidence, are close to 30,000 years old.


In 1978, Niède Guidon convinced the French government to establish an archeological mission for the study of prehistory at Piauí. Since the Napoleonic area, France has supported archeological excavations in various corners of the globe. Expeditions in dozens of countries are financed by the French Ministry of Foreign Relations, which sees in archeology an instrument of its “scientific diplomacy”.

It was on her first expeditionary foray that Niède discovered the largest – and most controversial – archeological site in the Serra da Capivara: the Toca do Boqueirão da Pedra Furada, at the foot of a giant sandstone wall, some seventy meters wide and nearly seventy meters tall, inclined so as to form a kind of shelter at the base. The wall is covered with prehistoric paintings – more than a thousand individual figures have been catalogued. The site is a good indicator that the area was frequented by peoples in the past. When excavations began, Niède encountered stone fragments which appeared to have been shaped by humans and charcoals that indicated intentional fires. “These were fires that had been built; they arranged a ring of stones and built the fire inside”, she explained.

At the end of a season in the field, Niède sent samples of charcoals to a French lab for dating. When she received the results by mail she was in disbelief: the samples were 26,000 years old. She immediately called the laboratory to register a complaint with the researcher in charge of the analysis: “You’ve confused my samples. In America there’s nothing that old!” The response on the other side of the line was: “Go back and continue your excavation, because these coals are yours.” Niède returned to the field and confirmed that her colleague was right. The world would hear about the Boqueirão da Pedra Furada in 1986; an article in the journal Nature sustained that humans were already living in Piauí 32,000 years ago.

From the wooden guard rail installed for the tourists who visit the paintings, the immense volume excavated by the investigation becomes evident. By Niède’s estimate, the dig “covers an area of about 60 by 15 meters, and reaches a depth of 8 meters.” Excavation lasted ten years, from 1978 to 1987. At the end, work at the site was led by the Italian archeologist Fabio Parenti, Niède’s doctoral student at the Sorbonne.

Parenti’s aim was to prove that the artifacts discovered could not have been produced by a natural accident. Fragments of quartz – the primary material for the tools – occur naturally in the soil at the top of the bluff overlooking the site. For critics, the edges which give the fragments the appearance of tools are no more than an accident, the result of erosion by wind or water, removing quartz from the bluff and causing it to fall to the canyon below and fracture upon impact. In order to test this hypothesis, the Italian made an analysis of 2,000 quartz fragments collected at the base of the elevation and found that they had a pattern that was quite different from the material excavated from the site. For Parenti, these remained indisputable artifacts.

By the end of the excavation, Parenti’s team had catalogued nearly 600 objects they believed to be created by peoples predating the Clovis culture. The dating of samples, using remains of charcoal, goes up to the limit which can be measured by radiocarbon technology, close to 50,000 years. Other tests, capable of measuring beyond the limits of carbon-14, have led Niède to defend an even older population at the site.

If the scientists working at Piauí are correct, then humans were living in Brazil’s North East tens of thousands of years before hunters following migratory herds spread out across North America. In that case, the time of arrival for the first Americans defended by archeological consensus would suffer a profound setback, and a radically different model for the peopling of the Americas would need to be formulated. Niède has suggested an arrival directly from Africa, by way of the South Atlantic, taking advantage of lower sea levels and a favorable current: “They would have arrived at the level of the Parnaíba river delta”, she explained. Few of her colleagues however have given much credit to her hypothesis.


Even if an arrival 50,000 years ago is contested by nearly the entire archeological community, the same thing cannot be said for other theories defending a human presence in the Americas prior to the Clovis. Evidence of earlier peoples has existed since the nineteenth century, and began appearing with increasing frequency around 1970.

At the end of that decade, Tom Dillehay, then a professor at the Universidade Austral de Chile, in Valdivia, was presented with samples collected by timbermen near a creek in Chile’s Monte Verde region, 800 kilometers south of Santiago. The specimens included a mastodon bone and stone artifacts along with other objects. During excavation at the site, Dillehay uncovered an impressive range of particularly well-preserved remains. There were artifacts in wood, stone, bone and ivory, preserved fruits, roots, seeds, and sea weed, locks of hair, fossilized feces and three footprints approximately 13 centimeters in length. A layer of peat had isolated the remains from oxygen and impeded the growth of bacteria, preventing their decomposition.

The diversity of materials seemed to suggest a collection assembled by humans. Among the plant species encountered were varieties occurring only many kilometers away, including alimentary and medicinal plants used by local indigenous peoples. In Dillehay’s interpretation, the site was likely a prehistoric camp, suddenly abandoned due to a flood or some other natural disaster. Carbon-14 dating of the samples placed them at 14,600 years old.

Since they were not compatible with the Clovis First model, the results of the study, published in 1989, were initially received with skepticism. It took almost a decade before wider acceptance arrived. In 1997 the site was inspected in loco – as had been the practice at the time of Folsom and Clovis – by a panel of twelve archeologists, including long standing critics of Dillehay’s work. All agreed that the site and the materials recovered by Dillehay’s team attested to a definite human presence at Monte Verde over 14,000 years ago. The rare consensus made headlines; The New York Times affirmed that archeology had crossed a threshold, similar to the breaking of the sound barrier for aviation.


For an archeological find to be accepted without controversy, a number of pre-requisites must be fulfilled. It is essential that the site be found in good condition, without signs of disturbance. It also needs to be reliably dated, demonstrating a clear succession of layers, from the most recent to the oldest. None of this was a problem at the Boqueirão da Pedra Furada. But the site did not pass a third test, also essential for its acceptance: it must demonstrate an unequivocal human occupation. The fragments discovered at Piauí by Niède Guidon and Fabio Parenti were hardly considered tools by their peers.

Extraordinary claims require extraordinary evidence, so goes the saying popular among scientists. Archeologists who have claimed to have proof of very ancient cultures in the Americas have always been systematically discredited by skeptics among their colleagues. In a 1990 article, the journal Science called the archeologists defending the status quo on the peopling of the American continent “the Clovis Police”.

In the case of laboratory science, a scientist who harbors doubts regarding assertions by a colleague can always attempt to replicate the results of an experiment in the lab. In archeology, of course, the replication of a discovery is impossible. The tables, photographs, drawings and descriptions found in an article are not always sufficient to convince the most rigorous, and in many cases acceptance of a theory comes only after peers can visit a contested site. If eminent archeologists had not seen, with their own eyes, the flint lodged between the ribs of the prehistoric bison at Folsom, perhaps the find would have been received with the same skepticism which has greeted new models explaining the peopling of the Americas.

It was with this in mind that Niède Guidon organized a 1993 symposium on the peopling of the Americas in São Raimundo Nonato. She invited North American colleagues to visit the Boqueirão da Pedra Furada. Among those whom accepted the invitation were Tom Dillehay, David Meltzer and James Adovasio, an archeologist who had explored a pre-Clovis site in the North East of the United States – Meadowcroft – with human remains up to 20,000 years old, a site also contested by some of his peers. The visit however, was insufficient to convince them. The inspection was reported in an article for the journal Antiquity, in which the authors questioned the human origin of the charcoals tested by Niède and Parenti and doubted whether the fragments of quartz were true artifacts.

Fabio Parenti assumed a regretful tone when discussing the visit in a recent conversation. He was clearly disillusioned by the passing judgment his colleagues made regarding the work of many years. “They were superficial”, he said, speaking from Rome over Skype. “It is not possible to judge such a complicated situation in four or five hours of observation.”

Niède Guidon, for her part, published a reply to the article by Meltzer, Adovasio and Dillehay in which she dismissed their criticism as either false or superficial. She has never lost sleep over the rebuke from her North American colleagues. She has often said that it was never her priority to weigh in on when the continent was populated. “I came to Piauí on behalf of the cave paintings. I was never in search of what was oldest.”

Whoever asks her about the peopling of the Americas will believe there is no controversy at all; only some researchers who persistently refuse to accept new evidence. In her view, more than a question of interpretation, what is at play is a collision between the French school of archeology, open to new ideas, and the American, incapable of letting go of Clovis – a theoretical model which, for her, should hardly have been taken seriously. More than convince her colleagues to the north, the Brazilian archeologist is concerned with consolidating the legacy she has constructed in the backlands of Piauí over decades.


Since her retirement, Niède no longer coordinates the French-Brazilian archeological mission. She invited Eric Boëda, researcher at the CNRS and professor at the Université Paris X, to assume leadership. Boëda has edited and written numerous books precisely on the subject of prehistoric flints. The most recent, which came out last September, is a treatise on ancient lithic cutting tools.

Boëda has lead excavations in Europe, Africa and Asia. He worked in Syria for more than twenty years. At the site Um el Tlel, on the steppes bordering the Euphrates River, his team excavated to a depth of 22 meters, uncovering more than a million years of pre-historical evidence left by Homo sapiens and their ancestors in the region. Since 1997 he has participated in excavations in China to investigate the tools used by the early peoples of East Asia. He found that both primary materials and technologies were very different from those used in Africa at the same time.

Presented with samples of the quartz fragments found at Piauí, Boëda noticed immediately that they had little resemblance to stone tools found elsewhere in the Americas. At the same time, they were similar to lithic technologies he had encountered in East Asia. In his appreciation, only someone unfamiliar with the prehistoric tools of other peoples would fail to recognize the anthropic character of the fragments from the Serra da Capivara. “If the stones I encountered in China were man made, I could not say that the ones found at Piauí were not.”

At 59, Boëda is short, with long, curly black hair and a full, white beard. Speaking from Paris over Skype, he explained that, like the pebbles at Piauí, the tools found in China were made from only partially knapped stones. The production of the implement began with the selection of a suitable specimen. “You choose a stone which has the ideal shape, volume and surface for the tool, and then shape only that part which will be the active zone” he said. “It’s like sculpting a bust from a block where there is already a head and ears: it would only be necessary to add the missing features.” In Chinese prehistory, as at Piauí, stone tools of this type are the norm. In this respect, they are different from the artifacts made by the Clovis people and other cultures: when selecting a stone to fashion a flint, they made use of the by-products of fracturing – these fragments were then worked to produce an implement.

Boëda believes that many of his colleagues have refused to accept the human origin of these artifacts because they cannot recognize them. For the critics, the fragments are what the French archeologist calls “objects without memory”, in other words, they do not pertain to the conceptual repertoire of the beholder and do not evoke any association. “If you operate only by analogy, you would say that they are not man made, because they are unlike other tools.” Boëda, who is also trained as a physician, compared the situation to that of a doctor who is unable to make a diagnosis. “If you are unfamiliar with the disease, what do you say? That the patient is not sick? And this is exactly what is happening.”


Toca da Tira Peia, the site analyzed in the latest article by the French-Brazilian team, was baptized in honor of a snake found during the excavations – tirapeia is the local name for a native species of pit-viper. To arrive at the site, a ten kilometer dirt road with a number of precarious stretches leaves from the municipality of Coronel José Dias. Unlike other excavations in the region, Tira Peia is outside the perimeter of the Serra da Capivara National Park.

On the way to the site, the geographer Gisele Daltrini Felice described how the Serra da Capivara park lies on the border of two distinct geological formations. The Toca da Tira Peia is part of a limestone massif, a geological formation that is completely different from the sandstone formations inside the park. Like the Boqueirão da Pedra Furada, the site is also located at the base of a high rock wall, but here the escarpment is not covered with the quartz fragments so abundant at the other site. “We are 15 kilometers from where these rocks occur naturally”, Gisele said. “And yet the critics continue to say that the fragments simply fell from the cliff above.”

Gisele was born in Bragança Paulista and studied geography in the Brazilian state of Rio Grande do Sul. She fell in love with the geographical diversity of the Serra da Capivara when she visited the region in the 1980’s. She moved to Piauí in 1995 and lives in São Raimundo Nonato since ’99, where she is a professor at the Federal University of Vale do São Francisco. Her master’s research, under the direction of Niède Guidon, investigated the hypothesis that the coals found at the Boqueirão da Pedra Furada could have been produced by wild fire. If this were the case, she explained, the coals should have been found over a wider area. To test this hypothesis, Gisele excavated other sites in the region. She did not find any evidence of an extensive fire.

She did find, on the other hand, localized charcoals she believes to be man-made, at a site just dozens of meters in front of the rock wall at the Pedra Furada – the Vale da Pedra Furada. The excavations made at this new site rendered hundreds of artifacts in layers of rock almost 25,000 years-old. The results will be made public this year in an article for the journal Antiquity. Like the site at Toca da Tira Peia, 22,000 years old, the Vale da Pedra Furada is yet another sign of human occupation in Piauí predating the Clovis culture. Neither site however, has confirmed the 50,000 year-old dates at the Boqueirão da Pedra Furada.

Standing in front of the dig at Toca da Tira Peia, Gisele is wearing a blue jersey that says ARCHEOLOGY. She explains that members of the team working at the excavations in Piauí have no doubt about the age of the sites. “Human presence here around 20,000 years ago is undisputable”, she says. “Like it or not, Clovis is neither the oldest culture nor the essential reference for the peopling of the Americas.”

The discovery of human remains with datable organic material could put the controversy over the age of these sites to rest. However, the acidity of the sandstone soils where the majority of archeological sites in the Serra da Capivara are found, is not suitable for preservation of fossils. Ancient human skeletons are a rare find in the region. One, baptized as Zuzu, served to reconcile Niède Guidon with a former critic, the biological anthropologist Walter Neves, at the University of São Paulo.

Neves is best known for his description of Luzia, an 11,000 year-old human skeleton unearthed in the 70’s at Lagoa Santa, Minas Gerais, by Annette Laming-Emperaire, Niède’s academic godmother. Neves showed how aspects of the well preserved skeleton more closely resembled features of skeletons from Africa and Oceania than those from Eastern Asia, which in turn demonstrate more similarities to the skeletal structure of modern indigenous peoples. Neves proposed a model of two separate occupations of the continent by genetically distinct populations, both arriving by way of Beringia. The first would have been the culture of Luzia, which left no descendants. The ancestors of the Clovis and of modern indigenous peoples would have arrived in a second migration.

In a telephone interview from his office, in São Paulo, Neves related how he had been initially skeptical about the age of the Boqueirão da Pedra Furada. He changed his mind when he got to know the site and the materials from other excavations in the Serra da Capivara, at the invitation of Niède. “I was surprised by what I saw”, he admitted. “I left 99% convinced that those sites are very old.” Neves returned to Piauí to study the skeleton of Zuzu with the bio-anthropologist Mark Hubbe, formerly his doctoral student. They demonstrated that the skeleton is the oldest yet encountered in the region, more than eleven thousand years old. The analysis reinforced Neves’ hypothesis. “The few skeletons which have been found there have the same morphology as the Luzia people”, he explained.


The number of sites allegedly older than Clovis is in the hundreds. In South America the list includes Taima Taima in Venezuela, the Cueva Pikimachay, in Peru, Cerro Tres Tetas, in Argentina, and Santa Elina in the Brazilian state of Mato Grosso, where artifacts have been dated at 25,000 years old. The majority of these sites remain contested by archeologists. Monte Verde in Chile is still the only site older than Clovis to be widely recognized. A 2012 survey of scholars studying the peopling of the Americas found that only 10% continue to contest the age of the Monte Verde site. Such consensus however applies only to the 14,600 years dates. Stone artifacts and charcoals found in another part of the site have been dated at almost 40,000 years old. Wary of the difficulties in convincing colleagues of far less, Tom Dillehay doesn’t even bother trying to achieve a consensus around the almost embarrassingly early date.

If “Clovis First” seems to have been surpassed, another model capable of explaining the peopling of the continent, one taking account of new evidence, has yet to be found. The site now recognized as the oldest in the Americas is located in the Southern Cone, 13,000 kilometers from the supposed point of entry for the first peoples. To have been in Monte Verde 14,000 years ago, the ancestors of prehistoric Americans must have crossed Beringia millennia before – perhaps 16 or 20 thousand years ago. If the evidence of 20,000 year old anthropic artifacts in Piauí were to become widely accepted, it would be necessary to further revise the date of an arrival by way of the Bering Strait.

Antoine Lourdeau, the young French professor at Federal University of Pernambuco, hopes that the question will be resolved sooner rather than later. For Lourdeau, the fixation on establishing a date of arrival distracts from the questions which are truly important. “The real goal is to try to understand who was here, how they lived, and how they traveled and what their relation may have been to the societies which came afterwards”, he explained.


Last October, researchers in the peopling of the Americas met in Santa Fe, New Mexico, only a few hundred kilometers from the historic sites of Clovis and Folsom. The conference, entitled Paleoamerican Odyssey, was held at the municipal convention center, a low adobe building, like the majority of the architecture in the city’s historic center. Held over three days, the event drew hundreds of students and researchers, as well as amateur archeologists and prehistory buffs (wearing wide brimmed hats and jerseys, many of the attendees appeared as if they were on their way to an excavation). Besides the presentations, delivered in a high-ceilinged auditorium, the event included also an exhibition of spear points, flints and other archeological artifacts.

In 1999, another large conference on the peopling of the Americas was held at the same location. The Monte Verde site had already been confirmed two years earlier, but the event – “Clovis and Beyond” – was still held under the aegis of the then decades-old theoretical model. In the years separating the two conferences, important contributions were made towards the understanding of human prehistory on the American Continent.

Today, archeologists know more than they did fourteen years ago. On the other hand, there is less certainty regarding how the peopling of the Americas really happened. In a hallway conversation during the recent conference, the Brazilian archeologist Mark Hubbe, professor at Ohio State University, explained that presently there are four central questions to be answered regarding the occupation of the continent: who were the first peoples to live here, where did they come from, when did they arrive, and how did they live. “Already we have spent millions in research and we still can’t answer a single one of them”, he said.

One of the organizers of Paleoamerican Odyssey was the geoscientist Michael Waters, professor at Texas A&M University and the director of the Center for the Study of the First Americans. Recently, Waters lead a team that uncovered a projectile point lodged in the ribs of a 14,000 year old mastodon skeleton in the state of Washington, yet another site dated before the supposed spread of the Clovis people across North America. Waters speaks with a distinct Texan accent, and in an interview at the end of the first day of the conference he said that now is a formidable time to be studying the origin of the first peoples of the Americas. “Today it is opened up again. We are free of the Clovis First model and we can explore new ideas. We are seeing the end of a paradigm.”

Waters was using the terms consecrated by Thomas Kuhn in his 1962 book The Structure of Scientific Revolutions, an enduring influence for the philosophy of science. In the terms proposed by Kuhn, we are living a moment in which the complex of presuppositions shared by the scientists of a given discipline – a paradigm – is no longer able to explain an ever increasing number of discrepant data. “A new model must emerge”, continued Waters.

Waters had been impressed by the presentation made just before by the evolutionary biologist Eske Willerslev, of the University of Copenhagen. The Danish scientist had shared the results of a study to be published the following month in the journal Nature: the analysis of genetic material from the 24,000 year old skeleton of a boy unearthed in Siberia – the oldest human DNA to have been sequenced. The child’s genetic material demonstrated similarities with indigenous peoples of the Americas, and also with peoples of Europe. This would suggest that the first inhabitants of the Americas were not only descended from peoples from East Asia, as prior genetic evidence had seemed to indicate, but also from peoples coming from Europe.

Waters has wagered that the paradigm to substitute Clovis will include important contributions from molecular biology. “At some point, the genetic evidence and the archeological evidence we take out of the ground will converge, and once we that happens we will have a novel story.” In his opinion, the script of that story does not appear to include a chapter in the Serra da Capivara. When I asked if he was familiar with the latest finds by the French-Brazilian team, he said he knew the material from the site at the Pedra Furada. The fragments “appear to be stones that were chipped and broken naturally and don’t have the appearance of human-made artifacts. It was a well excavated site, but I don’t think it will stand up to scientific scrutiny.”


At the exhibition of archeological objects organized in parallel with the conference, the Clovis points were the items most prominently exhibited. Some were displayed on cushions of leather or felt. Ensconced in a glass case was a dark piece of obsidian some 25 centimeters in length. It had been excavated at a site in Washington State and was presented as the largest fluted point yet found.

Also on display were fifteen quartz artifacts recently discovered at the Vale da Pedra Furada in layers of rock 23,000 years old. They had been brought by Eric Boëda, one of the event’s invited speakers. During the conference, the French archeologist circulated at the exhibition, on hand to attend questions from visitors. Yet he was ill at ease with the environment. He felt that the artifacts had been displayed on the basis of some aesthetic criteria. “This has all the appearance of a beauty pageant”, he gibed.

The pebbles from the site at Pedra Furada did in fact attract far less attention from visitors than the many projectile points, perhaps because their form does not immediately evoke any specific function. As legitimate objects without memory, a number of the artifacts brought by Boëda could hardly have been recognized as tools by an untrained eye. With very old artifacts, he observed, if we pay attention to the flakes, we can see knives, forks, spoons and other objects. I asked whether he could identify a tool simply by looking at it. “It’s like looking at a pen sitting in a plate of pasta”, he replied, looking at me over the rims of his oval-framed spectacles. “You recognize it immediately.”


Boëda’s presentation was included in a panel entitled “Archeology of Pre-Clovis”, which gathered a number of controversial finds. He presented the results of the continuation of the French-Brazilian mission in Piauí. Avoiding scientific geo-politics, he limited his talk to technical details, demonstrating the criteria used to classify the excavated objects and determine whether or not they were of anthropic origin. At the end of his presentation, he apologized for his heavily-accented English.

Language was not the only obstacle. Boëda faced also the stigma of the Boqueirão da Pedra Furada, the rejected site of 50,000 year-old artifacts which continued to cast a shadow of doubt over any further claims coming from Piauí. According to what I heard from one participant at the conference, whenever a researcher would present work from the Serra da Capivara, it was received with skepticism, independent of what was said. “The reaction was always the same: Here come those crazies from Piauí again.” The isolated character of the discoveries emboldened critics: if there had been such ancient human population in Brazil’s North East, why were the archeological finds all concentrated in a single area?

When he assumed leadership of the French-Brazilian investigation, Boëda made every effort to guard flanks where he felt their work might be vulnerable. He focused research on new sites, like the Toca da Tira Peia and the Vale da Pedra Furada – there are now seven sites with identified artifacts of ancient human occupation. The group also continued to publish in English, and in journals read by North American archeologists. And Boëda attempted to court strategic allies, capable of providing interdisciplinary support. From Barcelona, he recruited Ignacio Clemente Conte, a specialist in the microscopic analysis of stone fragments. Clemente’s studies confirmed that artifacts from the sites at Piauí had been used to cut meat and animal hides and work wood, bone and other materials.

As he finished a hamburger and a beer at the end of a day of the conference, Boëda explained how he had also sought out the Chilean geologist Mario Pino, who had been instrumental to the excavations at Monte Verde. Pino performed analysis on the sedimentation at the new sites in Piauí. “Pino convinced Tom Dillehay that the material was in fact of human origin” he said.

The next morning, Dillehay admitted, in a more thoughtful tone, that he had changed his mind about the material from Piauí. “Looking back at the sites explored by Niède Guidon and Eric, I am becoming more convinced that people were there 20 or 25 thousand years ago”, he said. “The accumulation of data through time is pretty consistent.” But he still expressed reticence regarding other objects of supposed anthropic origin found at the excavations. “More than the lithics, which I know the French-Brazilian team is focused on a lot, I am a little concerned about other aspects of human activity and signatures, like hearths and arrangements of rocks. And I said these things to Eric and his colleagues.” Since traces of the cave paintings have not been uncovered by the excavations, the paintings do not serve to corroborate the age of a human presence at those sites.

Dillehay is a professor at Vanderbilt University, in Tennessee. A tall man with a thin mustache, he looks like a maverick cowboy from an old western. In Santa Fe, he presented the results of an excavation he has been pursuing at a site in Huaca Prieta, on Peru’s northern coast. There, he encountered tools whose simplicity caught his attention, in terms of production techniques, these fragments had been only minimally worked – Dillehay noted the similarities with the artifacts found at Monte Verde, at Piauí and in Mato Grosso. The dating of the materials has suggested a human occupation at Huaca Prieta around 14,000 years ago.


The North American Stuart Fiedel, general of the Clovis brigade which had criticized the results of the French-Brazilian research in the Science News story, was the last speaker to take the stage on the second day of the conference. Fiedel has a grey beard and receding hair. He works as an archeologist for an environmental consulting company and is among the 10% who still contest the anthropic origin of the Monte Verde site. One colleague described him as “a pure conservative – maybe the last – certain to completely reject any possibility that Clovis might not represent the first American population.”

Fiedel criticized what he described as precarious evidence of human occupation prior to Clovis, like fossilized feces and foot prints of dubious origin. He mockingly noted that these were the same kinds of evidence presented to demonstrate the existence of the legendary Big Foot. Considering in detail the prints found by Dillehay at Monte Verde, he suggested that, by their size, the prints could only have been made by a one year old child, but do not demonstrate the characteristics of a human foot of that age. Fiedel concluded his talk with a vehement refutation of the evidence. Someone in the audience shouted “Yeah!”, lending to the reunion of scientists the air of a sports match.

At the end of the presentation, I asked Fiedel if the most recent evidence from new sites in Piauí seemed to him stronger than in the past. He returned to the argument he had used in the past: “The fragments could have been produced by capuchin monkeys, who make stone tools and live in exactly the same areas where those sites are located”, he said. When the interview was finished, a woman who had waited until the end handed Fiedel a pair of baby shoes and asked him to autograph the sole.

In the words of Thomas Kuhn, a new paradigm imposes itself gradually as it is accepted by an increasing number of scientists in a given field. But the old paradigm retains a residual minority and does not completely disappear. In the end, Kuhn suggests, a paradigm is only finally surpassed with the death of its last remaining adept.

Closing the conference at Santa Fe, Tom Dillehay, the leader of the excavations at Monte Verde which had marked the demise of Clovis First, asked for permission to make a digression. He defended the decolonization of scientific research and recommended that colleagues open their minds to new possibilities. He also addressed critics who had taken issue with his work and the research of others, making special reference to Stuart Fiedel’s remarks on the French-Brazilian excavations at Piauí. “What kind of monkey produces an archeological site?” he asked. “I hope that the next generation of scholars doesn’t have to go through the bullshit that some of us went through. I welcome the next generation of researchers.”

* – The results of radiocarbon dating need to be adjusted to arrive at the approximate age of the material tested. The dates provided in the text have all been accordingly processed.

Bernardo Esteves

Repórter da piauí, é autor do livro Domingo É Dia de Ciência (Azougue Editorial)

Leia também


Foro de Teresina #155: O apagão de Bolsonaro

O podcast de política da piauí discute os principais fatos da semana

Quinze anos de sigilo sobre “reunião da boiada” e cinco anos para relatórios do Ministério da Justiça sobre rolezinhos

Monitoramento de rolezinhos de jovens mereceu segredo por cinco anos; vídeo da reunião ministerial de abril de 2020, pelo triplo desse tempo

Candidato a animador

A não candidatura que promoveu Luciano Huck das tardes de sábado para as de domingo na Globo

Foro de Teresina ao vivo

Podcast de política da piauí terá edição especial em vídeo

Cinco anos de segredo para presos no Haiti e mortes no Jacarezinho

Sigilo de cinco anos protege documentos sem novidade, como sobre detentos haitianos, e outros sobre os quais pouco se sabe, como as mortes no Jacarezinho

“Sou um homem e amo outro homem”

No mês do orgulho LGBTQ+, o estilista Ronaldo Fraga conta como se apaixonou por um rapaz após dois casamentos com mulheres e critica retrocesso no governo Bolsonaro

Infelizes como nunca

Pesquisa mostra que o brasileiro jamais se sentiu tão infeliz como na pandemia – e que o país é desigual até na infelicidade

Cinco anos para proteger dados sobre manutenção de veículos da Marinha e pesquisas aeroespaciais

Marinha mantém em sigilo por cinco anos documentos sobre veículos – mesmo tempo pelo qual a Agência Espacial Brasileira (AEB) oculta pesquisas sobre foguetes

Mais médicos mal formados

Em duas décadas, Brasil abriu 247 cursos de medicina – uns de excelência e outros onde falta de tudo, de microscópio a paciente para o aluno aprender procedimentos complexos

Até cem anos de sigilo para vacina de Bolsonaro e 41 para documentos americanos sobre ditadura no Brasil

Carteira de vacinação do presidente ficará em sigilo por até cem anos; papéis americanos sobre ditadura brasileira foram mantidos ocultos por 41 anos e 11 meses

Mais textos