NOT SO long ago there was a simple and seemingly incontrovertible answer to the question of how and when the first settlers made it to the Americas. Some 13,000 years ago, a group of people from Asia walked across a land bridge that connected Siberia to Alaska and headed south.
These people, known to us as the Clovis, were accomplished toolmakers and hunters. Subsisting largely on big game killed with their trademark flint spears, they prospered and spread out across the continent.
Advertisement
For decades this was the received wisdom. So compelling was the Clovis First model that few archaeologists even contemplated an alternative. Some with the temerity to do so complained of a “Clovis police”, intent on suppressing dissent.
No longer. Thanks to recent discoveries, the identity of the first Americans is an open question again. Clovis First is not quite dead, but most researchers now accept it is no longer a good fit for the evidence. And so the question must be asked again: when were the Americas first settled, and by whom?
The colonisation of the Americas has long fascinated and frustrated archaeologists. It was the last great human migration, the final leg of our journey out of Africa to lay claim to Earth’s habitable continents. Big-game hunters from Asia were always considered likely candidates, but it wasn’t until the mid-1960s that this idea was formulated into the Clovis First model, primarily by archaeologist C. Vance Haynes of the University of Arizona in Tucson.
According to Clovis First, around 13,500 years ago, near the end of the last ice age, a brief window of opportunity opened up for humans to finally enter North America. With vast amounts of water locked up in ice caps, sea level was lower than today and Siberia and Alaska were connected by a now-submerged land bridge called Beringia. As the world began to warm, the huge ice sheets that blocked entry into North America began to retreat, parting like the Red Sea to create an ice-free corridor to the east of the Rockies (see map). The Clovis walked right in.
The presence of throughout the US and northern Mexico supports the theory, as does the timing of an extinction that wiped out more than 30 groups of large mammals including mammoths, camels and sabre-toothed cats. This coincides neatly with the arrival of Clovis hunters, and could have been their handiwork.
But over the years inconvenient bits of evidence have piled up. In 1997 a delegation of 12 eminent archaeologists visited Monte Verde, a site of human habitation in southern Chile that was first excavated in the 1970s and was claimed to be 14,800 years old. That, of course, contradicted Clovis First. The trip was a pivotal moment: most of the visiting archaeologists changed their minds, and prehistory started to be rewritten.
Many other pre-Clovis sites have also been found, some producing more credible evidence than others. Monte Verde is the most widely accepted; a recent of 132 archaeologists found that around two-thirds believe it is pre-Clovis.
Compare the DNA
Recent DNA studies also contradict the old orthodoxy. By comparing the genomes of modern Asian and Native American people and estimating the amount of time it would take for the genetic differences to accumulate, geneticists estimate that people entered the Americas at least 15,000 years ago – 1500 years earlier than in the Clovis model.
There are archaeologists who still embrace Clovis First. One of them is Stuart Fiedel of the consulting firm Louis Berger Group. He points out that if people really had arrived on the scene earlier we would see evidence of human-induced stresses or extinctions, but we don’t. He also describes the alternative models and evidence as “incoherent and contradictory”.
But a majority of scholars now accept that people occupied the Americas before Clovis. “We’re past the tipping point,” says Michael Waters, director of Texas A&M University’s . “The evidence right now is just too robust.”
So if Clovis First is wrong, how and when was America colonised? Despite the shift, some things stay the same. Most researchers still think that the first Americans migrated from Asia, a conclusion based largely on DNA research. A recent study led by David Reich of Harvard Medical School, for example, compared DNA from Native American populations scattered from the Bering Strait to Tierra del Fuego with DNA from native Siberians. The conclusion? The Americans are descended from Siberians who arrived in at least three waves ().
That might not sound too different from Clovis First. But other research tells another story. The DNA of Native Americans is sufficiently different from the Siberians’ DNA to suggest that the populations went their separate ways around 30,000 years ago.
That could indicate a much earlier entry. Some archaeologists claim that they have found evidence of human habitation as far back as 50,000 years ago, but they are not widely believed.
A more likely scenario is that the settlers did not come directly from Asia, but from a population that settled in Beringia 30,000 years ago and stayed put for 15,000 years before pressing on to Alaska. This group would have become isolated from the ancestral population in north-east Asia, perhaps by ice, and built up 15,000 years’ worth of genetic differences. This scenario is dubbed the “Beringia standstill”.
“Perhaps the settlers did not come directly from Asia but from a population that settled in Beringia 30,000 years ago”
Not everybody, however, clings to the idea that the first Americans arrived from Siberia. One of the problems thrown up by Monte Verde is its remoteness from Beringia. It is about 12,000 kilometres from the supposed entry point; even if people colonised North America 15,000 years ago, it is a stretch to imagine them reaching southern South America just 200 years later.
Enter the second alternative to Clovis First – coastal migration. Some researchers have suggested that instead of walking across Beringia, the first Americans hopped on boats and sailed along the Pacific coast.
It is mightily difficult to test this scenario. The melting of the glaciers approximately 10,000 years ago submerged the ancient coast along with any archaeological evidence it holds. “The coastal migration theory has been marginal until relatively recently,” says University of Oregon archaeologist Jon Erlandson, a leading proponent of the idea.
Nonetheless, some evidence exists. Reich’s DNA study suggests that the first wave of colonists moved south along the Pacific coast. And we know that ancient East Asians were accomplished seafarers, reaching the isolated Ryukyu Islands between Japan and Taiwan roughly 35,000 years ago.
“We know that east Asians were accomplished seafarers, reaching isolated islands between Taiwan and Japan roughly 35,000 years ago”
There are also archaeological finds that lend credence to the idea. Erlandson has worked on the Channel Islands of California for decades and has uncovered evidence of an advanced 12,000-year-old culture there. It includes numerous barbed stone points and crescents that display remarkable workmanship. The points were conceivably used to spear fish while the crescents were likely mounted on a shaft and thrown at birds.
Though these weapons are younger than any Clovis artefact, they are also so different that Erlandson suggests there is no connection between the two types. He and his colleagues propose that they were made by East Asian seafarers who travelled on a coastal “kelp highway” of seaweed stretching from Japan to South America.
More tentative evidence in favour of a coastal route comes from an unlikely source: human excrement. A number of human coprolites – the most ancient of which is 14,300 years old – have turned up in the Paisley Caves in Oregon, one of the claimed pre-Clovis sites. Dennis Jenkins of the University of Oregon spent six seasons excavating the caves. In 2008 he announced that the coprolites contained human DNA.
Several experts expressed doubts when Jenkins published his results in Science (), but further DNA analysis supports the conclusion. The DNA also suggests that the people originated in Siberia or East Asia.
That in itself says nothing about coastal migration. But last year Jenkins announced another important discovery: a type of spearhead known as , found in the same geological layer as a 13,200-year-old coprolite. This style of weapon was known already; the spear points are markedly different from Clovis ones and were assumed to be the handiwork of a later culture. Now it seems the technology was around at the same time as Clovis ().
Who, then, made the spear points? One plausible answer is coastal migrants. The caves are quite close to rivers which feed into the Pacific. “Mariners could have easily followed these rivers inland,” says Jenkins.
There is also a resemblance between Western Stemmed and a style of point known as Tanged, made in Japan about 15,000 years ago, according to Erlandson. He adds that the crescent points from the Channel Islands also resemble Western Stemmed.
Bladelets and scrapers
Support for coastal migration also comes from Waters. He is leading the excavation of Friedkin, a pre-Clovis site in central Texas that he describes as a “game changer”. Since digging began in 2006, over 15,000 artefacts have been uncovered – more than found at all other pre-Clovis sites combined – dating from 15,500 to 13,200 years ago. The great majority are offcuts from toolmaking, but there are also choppers, scrapers, hand axes, blades and bladelets.
In Waters’s opinion, these could be the precursors of Clovis technology. The blades, bladelets and scrapers are the types of tools they used. “You have the technology that could have become Clovis,” he says.
If the dating is right, it adds further credence to a coastal migration. Waters says it is uncertain whether the ice-free corridor was open 15,000 years ago, suggesting the people who made the tools may not have entered via Beringia. “There’s any number of ways people could have come,” he says – but he favours the coastal route.
Although Waters published his work in Science (), a few archaeologists remain unconvinced the site is that old. Radiocarbon testing is generally the most accurate method for dating artefacts, but can only be used on organic material, which is absent at Friedkin. So Waters used a technology called optically stimulated luminescence.
One of the doubters is Dennis Stanford, a renowned First American expert at the National Museum of Natural History in Washington DC. He also thinks the site’s “pre-Clovis” artefacts could in fact be Clovis.
Stanford has another reason for doubting Waters: he believes that he has identified the predecessors of the Clovis. Stanford and Bruce Bradley of the University of Exeter, UK, are leading advocates of the most radical alternative of all: that the first Americans came not from Asia but from Europe.
Specifically they claim that the Clovis are descended from the Solutreans, a Palaeolithic people who flourished on the Iberian Peninsula approximately 24,000 to 17,500 years ago.
In their recent book , Stanford and Bradley present a scenario in which these people reached the Americas more than 20,000 years ago, travelling by boat along the edge of sea ice that stretched across the Atlantic Ocean from the north coast of Spain to the coast of the Americas.
This is a bold – some say preposterous – hypothesis, although not a brand new one. Over the past century, other archaeologists have proposed that Europeans could have preceded Columbus by millennia, citing a similarity between Clovis and Solutrean tools. Stanford and Bradley have embraced this hypothesis since the mid-1990s.
Stanford turned to Europe after spending about 30 years searching in vain for Clovis-like artefacts in Siberia. In 1996 he went to Solutré in France – the source of the term “Solutrean” – to attend an exhibition comparing Clovis and Solutrean artefacts. He eventually joined forces with Bradley, who was convinced of a Clovis-Solutrean connection.
The two argue that Solutreans were mariners and seal hunters who voyaged along the edge of the ice sheet in search of game. Some time between 23,000 and 19,000 years ago their wanderings led them to America.
Stanford also believes there were migrations from Asia, especially along the west coast of the Americas, but he maintains that the Solutreans were the precursors of Clovis.
Their hypothesis is buttressed by the recent discovery of two sites along Chesapeake Bay in Maryland, Miles Point and Oyster Cove, with Solutrean-style artifacts. Both sites have been dated at more than 20,000 years old. The pair also note that distinctively Solutrean-like artefacts have been recovered from two other claimed pre-Clovis sites in the eastern US, Meadowcroft Rockshelter in Pennsylvania and Cactus Hill in Virginia.
Stanford says Across Atlantic Ice has made some sceptics more amenable to his ideas. In an editorial in The Journal of Field Archaeology (), editors Curtis Runnels and Norman Hammond of Boston University write that Stanford and Bradley “make a plausible case”.
Open water
But others are dismissive. “It remains wild speculation,” says Lawrence Straus, a Solutrean specialist at the University of New Mexico, Albuquerque, who has argued against the hypothesis for years. “I can only say that there is no evidence of the Solutrean peoples being Atlantic navigators or seal hunters.”
What is more, a 2008 study by Kieran Westley, now at the University of Ulster in Coleraine, UK, and Justin Dix of the University of Southampton, UK, concluded that the ice sheet didn’t stretch across the ocean for much of the year. So, contrary to Stanford and Bradley’s scenario, the Solutreans would have had to navigate stretches of open water ().
There is also no DNA evidence, says geneticist Ripan Malhi of the University of Illinois, Urbana-Champaign. “The final nail in the coffin will likely come when we are able to sequence DNA from ancient individuals associated with Clovis points,” he says. As it happens, researchers at the Center for the Study of the First Americans are doing just that on the skeleton of an infant from Anzick, a 12,700-year-old Clovis site in Montana. These are the only confirmed remains of a Clovis human. The results will be announced later this year.
Whatever comes of that analysis, the old guard is unlikely to survive. “I think Clovis First is dead and we have to use our imaginations to come up with a new model,” says Erlandson. Exactly what that is remains to be seen, but he thinks it will consist of a mixture of coastal and land migrations. People may well have walked across Beringia into the Americas – but they were not the only ones, and probably not the first.
