Composite image of earliest humans and wooly mammoths
Composite image of earliest humans and wooly mammoths

New Evidence Complicates the Story of the Peopling of the Americas

New techniques have shown that people reached the New World far earlier than the long-standing estimate of 13,000 years ago, but scientists still debate exactly when humans arrived on the continent—and how.

emma yasinski
Emma Yasinski

Emma is a Florida-based freelance journalist and regular contributor for The Scientist.

View full profile.

Learn about our editorial policies.

May 2, 2022

When the National Park Service’s Daniel Odess was invited in 2017 to work with an archaeological team at White Sands National Park in New Mexico, it was to help preserve fossilized mammoth trackways before they eroded. But upon his first visit to the site in person, he saw that along with the large mammoth impressions, there were human footprints. 

They were “like a film clip,” he recalls. “You get to see what happened during that walk. And sometimes you get to see individual humans interacting. A woman carrying a child, setting the child down, picking the child up. And sometimes you get to see how animals react to people and people react to animals. So, you know, it was kind of a light bulb moment where I went, ‘this is a chance to see things that we never get to see.’”

It was also “pretty intriguing” to see the...

It was also “pretty intriguing” to see the human footprints side by side with the tracks of now-extinct animals such as mammoths, Odess notes. Humans are often blamed for hastening mammoths’ extinction shortly after arriving on the North American landmass, which, according to what was once the prevailing hypothesis, occurred some 13,000 years ago via the Bering Land Bridge from modern-day Siberia to Alaska. But the footprints suggest the two species may have lived side by side for thousands of years, Odess explains. Last year he and colleagues published radiocarbon dating of the layers of sediment where the human footprints were found that determined their age to be between 21,000 and 23,000 years old.

The findings weren’t the first to challenge the so-called Clovis-first model, named after the Clovis people, who were thought to have been the first to pass through Beringia. The White Sands findings were, however, the most conclusive, agree researchers who spoke with The Scientist. “For a long time, there’s been a consensus view that [human-occupied] sites that were 13,000 years old or so were legitimate,” but that archaeological finds dating back further in time were potentially erroneous, says Odess. The idea that humans had arrived in the Americas before that time “was controversial,” he explains, “because it was putting people here before the glaciers opened up the path”—an event estimated to have occurred around 13,000 years ago, several thousand years after the peak of the most recent ice age, the so-called Last Glacial Maximum. “But White Sands changes everything.”

When you have data that doesn’t fit a model, you don’t change the data. You have to adjust the model.

—Daniel Odess, National Park Service

The footprints lent credence to various archaeological digs in both North and South America that had unearthed stone tools and butchered animal bones predating Clovis tools. They also supported the claims of Indigenous scholars who have long attested that their ancestors predated the Clovis, statements bolstered by recent findings from geneticists who had started using a combination of ancient and modern DNA from humans and commensal microbes to trace multiple human migrations that date back as far as 24,000 years ago. Today, researchers agree that this growing trove of data has poked large holes in the Clovis-first model, while simultaneously opening the field to many more mysteries about how and when the earliest groups of people arrived in North America and spread across the landmass and beyond. 

A series of contradictions

From 1980 to the early 2000s, archaeologists uncovered habitation sites in both North and South America that predated Clovis times, but many of the sites were discounted by other scientists. Bones that appeared to have been butchered by humans might not have been, some argued; tools that looked deliberately carved could have been broken naturally.

The first convincing chink in the Clovis-first model’s armor came back in 1988, when archaeologist Tom Dillehay, then at Austral University of Chile, discovered artifacts at a site called Monte Verde in southern Chile that he and his team estimated to be as old as 33,000 years. In addition to their potential age, the artifacts were distinct from the well-known Clovis spear points found elsewhere in the Americas, suggesting that at least one, if not several, separate human groups infiltrated the continents before the Clovis people. The Clovis-first model “was refuted effectively in the ’90s with this archaeological site of Monte Verde in Chile that was accepted as a ‘true’ pre-Clovis site,” says Lorena Becerra-Valdivia, a radiocarbon dating scientist at the University of Oxford. More-recent excavations of tools, remnants of campfires, a possible shelter, and food scraps preserved in peat at the site support Dillehay’s initial finding that people lived there more than 12,500 years ago, but narrowed the window to a maximum of about 19,500 years ago (the 33,000-year dating from the deepest levels of the site could neither be verified nor falsified).

Not everyone was as convinced as Becerra-Valdivia, however, in large part because artifacts can be moved by wind and water. It was possible, critics said, that the artifacts that Dillehay’s team had identified were actually younger than 13,000 years, but that they’d been moved into older sediment layers by weather events or other physical factors. But footprints can’t be moved, Odess explains, and they can’t be made by other natural phenomena, meaning that the findings at White Sands were nearly impossible to refute. “A four-year-old can look at that and say, ‘Yep, that’s a person. Yep, I make tracks like that when I play in the mud.’ Those are human—unquestionably human. And there are lots of them.” He adds that the team hasn’t yet excavated and tested the oldest layers of footprints, which might push the date back even further than 21,000 to 23,000 years ago. 

Still, Ben Potter, an archaeologist at the University of Alaska Fairbanks, isn’t completely sold. When Odess’s study was published in 2021, Potter told The New York Times that, because the carbon dating could still be influenced by water, he thought that more evidence was required to build a conclusive case for such an early occupation of the Americas. In a recent email to The Scientist, he suggests that the White Sands discovery was compelling, but not definitive. “The recent White Sands paper provides some evidence for a possible earlier expansion of humans.” 

Other findings have emerged that continue to support a pre-Clovis habitation of the Americas. One study published in 2017 by a mostly US-based group of archaeologists even suggests that date may be as old as 130,000 years ago, as evidenced by mastodon bones found in San Diego that appeared to have been processed by humans. However, the bones were the only evidence found, and some scientists wrote a rebuttal saying that they may have been damaged in recent years by heavy construction equipment, rather than by the stone tools of ancient humans. Other recent studies have gotten similar pushback. But perhaps even more convincing than the accumulation of archaeological evidence are recent DNA analyses of ancient and modern genomes that further confirm an early occupation of the Americas.


For decades, scientists subscribed to the Clovis-first model of the peopling of the Americas, the idea that the earliest humans on the landmass had crossed the Bering Land Bridge after the Last Glacial Maximum when glaciers began to recede, about 13,000 years ago. These people spread widely throughout North and South America, as evidenced by the leaf-shaped spearheads they left behind. Some discoveries have begun to poke holes in the model, however. The Clovis people were in the Americas 13,000 years ago and did spread widely, but it’s becoming more and more clear that they weren’t the first humans to inhabit the New World.

Illustrated map showing where evidence was found of the earliest humans
Illustration of bacteria

18,000–20,000 years ago

Genetic sequencing of H. pylori bacteria from the stomachs of modern humans suggested that an ancient population of people survived in the area during the Last Glacial Maximum, when the land was covered in ice. 

Illustration of DN

11,500 years ago

Comparing DNA from the bones of a girl with sequences of DNA from other ancient populations and with genomes of modern Native Americans showed that she was part of a previously unknown group that separated from modern Native Americans about 20,000 years ago, suggesting that multiple groups migrated into and through North America. 

Illustration of DNA

8,700–21,000 years ago

Fifteen ancient genomes spread across the Americas revealed two distinct migrations from North to South America, highlighting the complexity of early human migrations and suggesting that early people had wide access to the continents that allowed them to rapidly spread throughout, despite obstacles such as glaciers.

Illustration of a foot

21,000–23,000 years ago

Footprints found in White Sands National Park revealed human activity in the area for thousands of years during the Last Glacial Maximum.

Illustration of animal fossils

14,500–19,000 years ago

Stone artifacts, animal remains, and burn sites uncovered in 1988, with new samples revealed and analyzed in 2015, were the first conclusive evidence that humans were in the Americas before 13,000 years ago. 

©, Daria_Andrianova, SurfUpVector; the scientist staff
See full infographic: WEB | PDF

DNA in America 

DNA sequences acquired from a handful of ancient human remains found at archaeological sites in the Americas suggest that the peopling of the Americas is far more complex than a single population of Clovis people crossing Beringia some 13,000 years ago. In 2018, one group of researchers found evidence of four separate migrations from North America to South America, with more-recent groups displacing older ones. That same year, another study pointed to a split of Native American ancestors between 18,000 and 14,000 years ago into two separate branches, which then split and mixed again and again as they radiated into the Americas. A couple of years ago, Becerra-Valdivia and colleagues developed a model that integrated data from 42 different archeological sites in North America and Beringia with emerging evidence from ancient DNA, and found support for the idea that people were indeed in the Americas 26,500–19,000 years ago. “I would say the biggest development is in nuclear genomic analyses, which add an independent line of inquiry into movements of people, migrations, and admixture among populations,” says Potter. 

But the genetic data raise more questions than answers. One notable quandary comes from the genomes of a small modern population in Brazil, which contain a signal that José Víctor Moreno-Mayar, a geogeneticist at the University of Copenhagen, says he cannot explain. About 2 percent of the DNA in some South American natives appears to be closely related to that of ancient Australasian remains from Southeast Asia, and a similar sequence hasn’t been found in any other living Indigenous American population. In the 2018 study that showed two waves of migration to South America, Moreno-Mayar and his team sequenced 10,000-year-old genomes from Brazil and were shocked to find this same genetic signature.9 “That’s really, really puzzling,” he says; it’s as if there was a population that magically appeared in Brazil sometime before 10,000 years ago, without having traveled through North America to get there. At that time, “there was absolutely no way in which it was possible to make it through the Pacific,” he says. And even if ancient people did cross the ocean, they’d have had to then traverse the Andes and the Amazon, a similarly challenging journey, without leaving any trace behind. “Every new explanation we come up with is worse than the previous one.” 

Doing this work in the right way means developing long-term relationships with tribes and developing trust.

—Jennifer Raff, University of Kansas

The story gets even more complex as scientists look further back in time. Another 2018 study in which Moreno-Mayar and his colleagues sequenced the DNA of a child who lived in Alaska 11,500 years ago revealed that she’d been part of an isolated group of ancient Beringians that separated from modern Native Americans about 20,000 years ago. “DNA shows isolation, but where were they isolated?” asks Jennifer Raff, an anthropological geneticist at the University of Kansas who was not involved in that study. Southern Siberia and East Asia were home to populations other than those ancient Beringians 20,000 years ago, she notes. “South Central Beringia would have been a really good place for that isolation event to have taken place, but we don’t have any archaeological evidence that’s really secure putting people in that spot at that time. It’s all underwater [now],” as melting glaciers raised water levels and sank the land bridge. 

Because sequenceable ancient human DNA is hard to isolate and extract, some researchers have turned to the genetic sequences of gut bacteria. About 20 years ago, Mark Achtman, a now-retired microbiologist then at the Max Planck Institute for Infection Biology in Berlin, realized that the Helicobacter pylori (the bacterial species that can cause ulcers) in a person’s gut differed based on where they were from. This gave Achtman a “crazy idea,” recalls Yoshan Moodley, a zoologist at the University of Venda in South Africa and a former postdoc in Achtman’s lab. “It could be that this bacteria has been in our stomach since before we even left Africa,” and could thus provide insights into early human migrations. Gut bacteria aren’t only transferred from parent to child but are also shared among people who interact, and bacterial genomes evolve more quickly than do human ones. These factors mean that bacterial sequences could provide hints about how different groups may have traded or otherwise associated with one another, Achtman and his colleagues reasoned.  

In a 2021 study, Moodley and an international team of scientists took samples of H. pylori from the stomachs of more than 500 humans currently living in different regions of Siberia and Mongolia.  The researchers then compared the genetic sequences of these bacteria to those of representative strains of the bacteria sampled throughout the world, and found that the genetic lineages of the bacteria split during the Last Glacial Maximum: one group seemed to have remained isolated in northeast Asia, while another showed new mutations that researchers believe were introduced when people left Siberia during the Last Glacial Maximum, then returned thousands of years later. 

The results suggest that some people survived in Siberia, despite the extremely harsh environment during the Last Glacial Maximum, with subsequent generations harboring their ancestral strain of H. pylori, a finding supported by fossils in the region. “The bacteria that these people would have had in their stomachs 24,000 years ago, it’s still alive and well [in people] in Siberia,” says Moodley. While the ancient people may no longer exist as a distinct population, he adds, “the bacteria that originally populated the stomachs of these ancestral northern Eurasians [is still there].”


For much of the last ice age, massive glaciers covered the northern part of the Americas. After the last glacial maximum, that ice started to melt, and by 13,000 years ago, an ice-free pathway opened through modern-day Canada that scientists know the Clovis people to have traveled. However, recent archeological and genetic evidence suggests that some groups of people migrated to the Americas before the last glacial maximum, when there was no ice-free corridor, suggesting that they may have traveled along the western coastline. If so, many of the artifacts and remains they left behind may be under water.

Illustration showing migration of the earliest humans
the scientist staff
See full infographic: WEB | PDF

Challenges ahead

These current genetic analyses offer many clues to how people moved through the continents, but considered in isolation they can only tell us so much. Unfortunately, melting ice has left many ancient sites underwater, drowning the artifacts and other evidence that might have shed light on ancient peoples’ paths through Beringia and possibly down the West Coast of North America, where they would have had to travel if they entered the continent while glaciers still covered the more easterly routes thought to have been followed by the Clovis people.

Aside from the limited amount of physical evidence, one of the biggest challenges facing researchers seeking to revise the story of the peopling of the Americas, they say, is intellectual inertia. Eric Boëda, an archaeologist at University Paris Nanterre, told The Scientist in 2020 that there is “denialism” in the field, and that the long-standing belief that humans didn’t live in the Americas until 13,000 years ago causes people to discount older artifacts found there. Odess says that when the White Sands paper was published, some scientists rejected it because it didn’t fit their models. While it’s important to confirm the validity of new data, he says, “when you have data that doesn’t fit a model, you don’t change the data. You have to adjust the model.” Since the findings first came out, Odess adds, most experts he knows of have come to accept them as valid.

Another aspect of the challenge is that many different types of scientists approach the question of migration into the Americas differently, and they don’t always work together. Scientists “from archaeology, from chemistry, and from genetics are kind of all coming together and approaching this really interesting piece of history from these multiple angles. And naturally, you get different data sets from these different fields,” says Raff. “I would almost critique us geneticists for not steeping our work in the archaeological data as well as we should. I think that that’s been a real problem.”

Another obstacle to an accurate telling of the peopling of the Americas is a tenuous relationship between the scientific establishment and the very populations central to the story. Researchers have historically failed to collaborate with Indigenous populations that could be affected by their work, breeding deep mistrust, experts agree. “Scientists have tended to kind of disregard their perspective on these things because they don’t see the ‘scientific proof,’” says Odess.

Many investigators who spoke with The Scientist agreed that conducting genetic analyses in an ethical way is both crucial and one of the biggest challenges they face. When human remains are found, tribes often want to leave them undisturbed or give them a traditional burial rather than allow DNA to be extracted for research. “Doing this work in the right way means developing long-term relationships with tribes and developing trust,” says Raff. “And that is difficult when so many times, again and again, their trust has been violated.” 

Interested in reading more?

May 2022, issue 1 cover image


Become a Member of

Receive full access to TS Digest, as well as digital editions of The Scientist, feature stories, more than 35 years of archives, and much more!
Already a member?