In 1973, a French-Brazilian team pushed their excavations into a rock shelterRock shelterA shallow cavity at the foot of a cliff or under a rocky overhang, offering natural shelter; a favoured site of prehistoric habitation and rock art.→ in the Serra da Capivara National Park, in Brazil's northeastern Piaui state. What they found would ignite one of the fiercest debates in world prehistoryPrehistoryThe span of human history before the invention of writing, from the Palaeolithic to the Metal Ages, known mainly through material remains.→: charcoal dated between 32,000 and 48,000 years before present, and quartzite fragments that looked disturbingly like knapped tools. If these dates are correct, humans were living in South America long before the ClovisClovisA Palaeoindian culture of North America (c. 13,000 years ago), recognizable by its fluted stone points; long believed the oldest on the continent, no longer so.→ people, long considered the "first Americans" who crossed the Bering land bridge some 13,000 years ago.
\n\nSince then, Pedra Furada has divided archaeologists worldwide. On one side, Niede Guidon and her international team defending dates going back as far as 60,000 years. On the other, North American researchers such as David Meltzer, James Adovasio and Tom Dillehay, who dismiss the evidence as the product of natural processes. The debate, far from settled, touches on one of prehistory's most fundamental questions: when and how did humanity populate the last great habitable continent on Earth?
\n\nA rock shelter in a landscape of stone
\n\n
Pedra Furada means "pierced rock" in Portuguese, referring to a natural quartzite arch that towers over the Brazilian caatinga scrubland. The word caatinga is a Tupi term designating a biome unique to northeastern Brazil, characterized by thorny dry vegetation, intense droughts and outcrops of sandstone and quartzite in shades of ochreOchreA red or yellow mineral pigment (iron oxides), used from prehistory for adornment, funerary rites and art.→ and red. This arid but visually spectacular landscape forms the natural setting for one of the largest rock art complexes in the Americas.
\n\nSerra da Capivara National Park, a UNESCO World Heritage Site since 1991, contains more than 800 archaeological sites spread across 100,000 hectares. Engraved walls, shelters adorned with vivid paintings depicting humans, animals and hunting scenes: Serra da Capivara is one of the largest prehistoric rock art complexes in the Western Hemisphere. The sheer density and quality of preservation of the paintings immediately attracted scientists' attention from the first expeditions into the region in the 1960s and 1970s.
\n\nBrazilian archaeologist Niede Guidon, trained in France and affiliated with the School for Advanced Studies in the Social Sciences (EHESS), has been the driving force behind the excavations since the 1970s. Born in Sao Paulo in 1933, she spent most of her career in Paris before settling permanently in Piaui to direct work on site. She assembled a multidisciplinary team uniting researchers of every nationality: palaeoanthropologists, geologists, radiocarbon specialists, zoologists, botanists. Under her leadership, the Fundacao Museu do Homem Americano (FUMDHAM) was founded to protect, study and promote the site.
\n\n
The main Pedra Furada shelter offers exceptional preservation conditions. The overhanging rock face shields deposits from weathering, allowing charcoal, bones and lithic tools to survive across dozens of successive stratigraphic layers. Guidon identified 15 distinct levels corresponding to three major cultural phases: the Pedra Furada phase, the oldest and most contested, with dates reaching as far back as 60,000 years; the Serra Talhada phase (12,000 to 7,000 BP), with universally accepted quartzite tools; and the Agreste phase, contemporary with rock paintings dated 11,000 to 5,000 years old.
\n\nThe site's stratigraphyStratigraphyThe study of the superimposed layers (strata) of an archaeological site; each layer corresponds to a phase of occupation and yields a relative chronology.→ is remarkably continuous. Sediments accumulated gradually, layer by layer, at the pace of successive occupations and natural inputs. This continuity is one of Guidon's key arguments for the site's archaeological integrity: if the charcoal in the deepest layers came from natural fires transported by wind, one would expect random, discontinuous distributions rather than regular concentrations associated with circular combustion structures bounded by carefully arranged stones.
\n\nThe rock paintings of the Nordeste Tradition
\n\n
Even if Guidon's excavations attracted worldwide attention for their controversial dates, Pedra Furada is first and foremost an exceptional rock art site. The paintings adorning the shelter walls and hundreds of other sites in the park represent one of the richest and best-preserved cultural heritages in the Western Hemisphere.
\n\nThese paintings, grouped under the term "Nordeste Tradition", date from 11,000 to 5,000 years before present. They stand apart from other rock art traditions in the Americas through their narrative character: rather than the schematic or symbolic representations found elsewhere, the Serra da Capivara paintings describe scenes of daily life with remarkable precision and vitality. We see hunters setting traps, groups engaged in collective activities, dancing and ceremonial scenes. The animals depicted -- deer, armadillos, capuchin monkeys, snakes, birds -- testify to an abundant fauna that populated this now-arid region.
\n\nThe pigments used are primarily red and yellow ochres based on iron oxide, plus black (burned bones) and white (clay). Technique varies by period: finger traces appear in the simplest figurations, while more elaborate compositions used brushes made from animal hair or plant fibres. Chemical analysis of the pigments found that artists prepared their colours on concave sandstone palettes recovered in the excavations. Guidon estimates the Nordeste Tradition developed over some 6,000 years, during which the style evolved continuously from schematic to increasingly refined representations.
\n\nDates that push every limit
\n\nIt is the Pedra Furada phase that concentrates the bulk of scientific debate. The first radiocarbon analyses, conducted in 1985 at the CNRS Centre des faibles radioactivités in Gif-sur-Yvette, France, produced spectacular results: charcoal from the deepest layers yielded ages of 32,000 to 48,000 BP. Publication in 1986 in Nature[1] by Guidon and Delibrias sent shockwaves through the international scientific community.
\n\nTo overcome the limitations of standard radiocarbon on such ancient samples, researchers applied a newer method: the ABOx-SC protocol (acid-base-wet oxidation followed by stepped combustion), developed by Andrew Bird in 1999. Used with an accelerator mass spectrometer (AMS) at the Australian National University, this technique allows more rigorous purification of charcoal samples and avoids some contamination. Applied to seven distinct hearth samples, the protocol confirmed and extended the dates: 55,000 to 60,000 BP for the oldest levels, near the upper limit of radiocarbon dating.
\n\nA third contribution comes from the nearby site of Toca da Tira Peia, 17 kilometres from Pedra Furada. In 2013, a team led by Christelle Lahaye of the University of Bordeaux applied optically stimulated luminescence (OSL)Luminescence (OSL)Optically stimulated luminescence dating: measures the last exposure of sediment grains to light.→, a dating method independent of radiocarbon[2]. OSL measures how long quartz grains have been shielded from light, allowing direct dating of the sediments enclosing artefacts rather than the artefacts themselves. Results indicated human occupation from 22,000 BP, during the Last Glacial MaximumLast Glacial MaximumThe peak of the last glaciation (c. 26,000 to 19,000 years ago), with ice sheets at their greatest extent; it pushed populations towards southern refuges.→, a period when the "Clovis First" model cannot explain human presence this far south.
\n\nIf the oldest dates (55,000 to 60,000 BP) are accepted, the implications are staggering. This period corresponds to Europe's Middle PalaeolithicMiddle PalaeolithicA Palaeolithic period (c. 300,000 to 40,000 years ago) associated mainly with Neanderthals and early Homo sapiens, marked by Levallois tools.→, when NeanderthalsNeanderthalsA fossil humanity of Eurasia, robust and cold-adapted, extinct around 40,000 years before present.→ still inhabited the continent. Anatomically modern humans would have reached South America before their African cousins colonised Europe under the standard chronology. This requires a radically different migration route, perhaps maritime along Pacific shores or even transatlantic from AfricaAfricaThe cradle of humankind: the continent where the first hominins appeared, then Homo sapiens around 300,000 years ago, before the expansion to the rest of the world.→ -- a hypothesis Guidon herself raised, drawing fierce criticism from colleagues.
\n\nThe case for the sceptics: natural fires and monkey tools
\n\nThe most classic objection, raised in 1994 by Dillehay, Meltzer and Adovasio in Antiquity[4], concerns the origin of the charcoal. They argue that natural vegetation fires could have deposited charcoal in the shelter without any human involvement. The caatinga is prone to intense droughts and spontaneous fires; wind-borne carbonised particles could easily have been carried into rock crevices over tens of millennia. In this scenario, the "hearths" of Pedra Furada would be nothing more than natural accumulations of charcoal integrated into sediments by ordinary geological processes.
\n\nThe second objection concerns the tools themselves. The lithic pieces of the Pedra Furada phase are sharp-edged quartzite flakes but show little retouch. Critics see them as "geofacts" -- fragments produced naturally by rocks falling from the cliff above the site, broken by gravity, frost or rainwater. The site's quartzite walls fracture naturally along cleavage planes that can produce sharp-edged flakes morphologically similar to knapped tools.
\n\nJames Adovasio summarised the sceptical position bluntly: "The Pedra Furada material is not even up to capuchin standards." This quip references a discovery published in Nature in 2016[3]: black-striped capuchin monkeys (Sapajus libidinosus) in Serra da Capivara were observed smashing pebbles against rocks to extract quartz dust that they lick for its minerals. The flakes they produce are morphologically similar to the "tools" attributed to humans in the oldest Pedra Furada layers. If modern monkeys can produce flakes indistinguishable from prehistoric tools, how can we tellTellAn artificial mound formed by the accumulation of successive layers of settlement remains at the same spot, typical of the Near East. Each destruction-rebuilding event adds a stratum.→ a human lithic industry from an ancient simian "workshop"?
\n\nSceptics also raise a geographical argument: if humans were present in northeastern Brazil 48,000 years ago, they should have left traces elsewhere on the continent. Yet no archaeological site firmly dated to this period has been found anywhere else in the Americas. The absence of contemporary sites is difficult to explain if one accepts such an ancient human presence.
\n\nGuidon's counter-arguments
\n\nGuidon has never yielded under pressure. Her main counter-argument against the natural fire thesis is structural: the charcoal is concentrated not randomly through the sediments but in circular structures bounded by carefully arranged stones. "The charcoal is not from a natural fire. It is only found inside the sites. You don't get natural fires inside the shelters," she told The Guardian. She also notes that the charcoal shows evidence of adjacent wall scorching, a sign of intense, focused heat impossible to achieve through aeolian accumulation of ash.
\n\nLithic technology specialist Eric Boeda, professor at the University of Paris Nanterre, analysed the Pedra Furada phase pieces in detail. His conclusions are unambiguous: some pieces display intentional retouch, that is, successive, oriented removals that cannot result from simple mechanical fracture by rockfall. He identifies notched pieces (coches) and backed tools, two tool types recognised in all Lower and Middle Palaeolithic industries in the Old World. According to Boeda, the typological variety of the pieces exceeds what any natural or simian process could produce.
\n\nJacques Pelegrin, research director at the CNRS and one of the world's foremost specialists in prehistoric lithic technology, takes a nuanced position. He grants that natural processes can produce flakes resembling Pedra Furada specimens, but considers the natural explanation "very unlikely" given the stratigraphic consistency of the finds and the continuous human occupation proven in the upper levels of the site. He notes that the capuchin argument, relevant for the oldest levels, does not apply to the Serra Talhada and Agreste phases, which yielded unambiguously anthropic tools associated with the same types of combustion structures.
\n\nPedra Furada and the peopling of the AmericasPeopling of the AmericasThe migrations that led Homo sapiens to occupy the American continent, from Beringia; its date and routes remain debated.→
\n\nThe Pedra Furada controversy cannot be separated from the grand debate on how humans populated the Americas. For decades, the "Clovis First" theory dominated without challenge. Formulated in the 1930s following the discovery of bifacially flaked lance points associated with PleistocenePleistoceneThe geological epoch of the great ice ages (c. 2.6 Ma–11,700 BP), spanning most of human prehistory.→ animal remains near Clovis, New Mexico, the theory posited that the first Americans arrived from Asia across BeringiaBeringiaA vast land bridge emergent between Siberia and Alaska during the last glaciation, at the site of today's Bering Strait; a cold steppe through which the first Americans passed.→ and swept through both continents in a few millennia through an ice-free corridorIce-free corridorAn inland passage between the Laurentide and Cordilleran ice sheets, open and viable only around 14,000-13,000 years ago, too late for the earliest arrivals.→ that opened around 13,500 years ago.
\n\nThis model has been progressively dismantled. Monte Verde in Chile, excavated by Dillehay himself (one of Pedra Furada's critics), produced solid evidence of human occupation at 14,800 BP -- roughly 1,300 years before the ice-free corridor was passable. That chronology implies the ancestors of Chile's first inhabitants followed a coastal Pacific route. Other sites pushed limits further: Meadowcroft Rockshelter in Pennsylvania (16,000 to 19,000 BP), the Buttermilk Creek site in Texas, and most dramatically the White Sands footprints in New Mexico, dated between 21,000 and 23,000 BP and published in Science in 2021, sent shockwaves through the field.
\n\nGenetic data provide complementary but inconclusive evidence. Ancient DNAAncient DNAGenetic material preserved in old remains, often degraded, sequenced with cutting-edge techniques.→ analyses trace the divergence of Amerindian populations from Asian populations to roughly 15,000 to 25,000 years ago. This range is compatible with White Sands (23,000 BP) and Monte Verde (14,800 BP). But it is difficult to reconcile with a presence of 48,000 years at Pedra Furada, unless one invokes extreme genetic drift or multiple migration waves whose last swept away genetic traces of the first. This is in fact the most serious objection today: the populations that would have inhabited Brazil 60,000 years ago do not appear to have left identifiable genetic descendants.
\n\nScientific legacy and heritage stakes
\n\nThe Pedra Furada controversy is not only scientific -- it is deeply political. The tensions that animate it reflect divides between North American and South American archaeologists that go beyond the question of dates. Guidon has repeatedly denounced what she sees as a cultural bias: "If I had found evidence of a 48,000-year occupation in a cave in France or Spain, no one would have contested it in the same way," she said in several interviews. She notes that the "Clovis First" theory was born within the North American academic tradition, which long dominated world palaeoanthropologyPalaeoanthropologyThe science that studies human evolution from the fossil remains of hominins (bones, teeth, footprints) and their context, to reconstruct our biological origins.→, and that any finding challenging this model from South America has systematically attracted more pronounced scepticism.
\n\nThe heritage and conservation dimension of her work is inseparable from its scientific dimension. FUMDHAM, founded by Guidon in 1986, created the Museum of the American Man in Sao Raimundo Nonato, which welcomes thousands of visitors a year and helps fund ongoing research. Guidon, who obtained Brazilian citizenship and settled permanently in Piaui, became a national figure honoured with numerous awards including the Ford Conservation and Environment Award and the Prince Claus Award from the Netherlands. In 2005, she was among the 1,000 women nominated for the Nobel Peace Prize.
\n\nA debate far from over
\n\nIn 2026, no consensus is in sight on Pedra Furada's oldest claims. Most prehistory textbooks list the site as controversial, whose most extreme dates are not accepted by the majority of specialists. But research continues: Brazilian and French teams are still excavating the deepest levels of the park. In 2020, a team from the University of Sao Paulo announced the discovery of new hearths at an adjacent site, Toca do Sitio do Meio, with preliminary dates of around 25,000 BP -- more conservative than Guidon's but still incompatible with "Clovis First".
\n\nThe methodological toolkit has improved enormously. New-generation mass spectrometers can analyse infinitesimally small charcoal samples, reducing contamination risks. Electron microscopy can analyse use-wearuse-wearMicro-traces left on a tool by its use, studied through use-wear analysis to reconstruct its function.→ traces on lithic pieces at unprecedented resolution. Ancient molecular biology can now extract and sequence DNADNAThe molecule carrying genetic information, used to reconstruct kinship between species.→ from remains in climatic conditions that would have seemed impossible a decade ago.
\n\nSerra da Capivara remains, regardless of the dating controversy, one of the most extraordinary prehistoric sites in the Americas. Its Nordeste Tradition rock paintings, 11,000 years old, are a fascinating testament to the lives of the first known inhabitants of the New World. They represent an irreplaceable cultural heritage protected by national park status and UNESCO inscription. Whether its deepest layers were frequented by humans 48,000 years ago or by improvising capuchins changes nothing about the importance of the question they pose: how long has humanity inhabited the Americas? The answer, whatever it may be, will rewrite a chapter of our species' history.
No comments yet. Be the first.