Saturday, August 1, 2015

Guess who first came to America?

Semang from the Malayan Peninsula, Wikicommons


Before the Europeans came, the Americas were settled by three waves of people from northeast Asia: the oldest wave beginning some 12,000 to 15,000 years ago, which gave rise to most Amerindians, and two later waves, which gave rise respectively to the Athapaskan and Inuit peoples of northern Canada and Alaska. That's the conventional view.

Kennewick Man. An earlier form of Northeast Asian?

There is growing evidence, however, for earlier waves of settlement. There's Kennewick Man, who lived nine thousand years ago in the American northwest and who looked more European than Amerindian, the closest match being the Ainu of northern Japan. He also looked a lot like Patrick Stewart.

Nonetheless, a DNA study has found him to be closer to Amerindians than to any other existing population in the world (Rasmussen et al., 2015). He was apparently descended from the same Northeast Asians who would later become today's Native Americans. Those earlier Northeast Asians looked more European because they lived closer to the time when these two groups were one and the same people. It may be that the Ainu best preserve the appearance of this ancestral population that would later develop into present-day Europeans, East Asians, and Amerindians.

But why would Kennewick Man be closer anatomically to an Ainu while being closer genetically to an Amerindian? The answer is that the genes that shape our anatomy are a tiny subset of the entire genome. Most genes are of low selective value, often being junk DNA, so they change at a steady rate through random processes. Taken as a whole, the genome thus provides a "clock" that can measure how long two populations have been moving apart since their common ancestors. Genealogically speaking, Kennewick Man is closer to present-day Native Americans than he is to the Ainu. Anatomically speaking, the reverse is true ... probably because his ancestors had escaped the extreme Ice Age conditions that affected northeast Asia 20,000 - 15,000 years ago by retreating to an ice age refugium on the Northwest Pacific Coast. The Ainu may have similarly sat out the Ice Age in another refugium on the other side of the Pacific.

[...]  ancient plant and animal remains found on several offshore islands provide evidence that some areas of land on the outer coast remained unglaciated and habitable during the Ice Age. These ice-free areas are called refugia, and evidence for their existence has been found off the Pacific coast from Alaska to southern British Columbia.

Although there is no direct evidence for human occupation of these refugia during the mid-glacial period, it is clear that a chain of habitable environments existed along the Pacific Northwest Coast, and that these environments could have supported people as they made their way down the coast.

If people moved down the West Coast, and then into the interior from there, where and when did this inward movement occur? Is there any archaeology suggesting that populations on the coast began moving inland? 

A few sites from the interior areas of Washington State, Oregon and Idaho may demonstrate this. Stemmed projectile points are found in a site along the Snake River in Washington State, with dates ranging from 8,800 to 10,800 years ago. Another site in south-central Oregon, Fort Rock Cave, contained a layer of gravel that had two obsidian points within it. Dates from this layer are as old as 13,000 years BP. Wilson Butte Cave from Idaho also contains human made artifacts dating to between 14,500 and 13,000 years ago. Perhaps these sites are examples of early people moving in-land; however the small number of sites uncovered so far makes it hard to determine definitively whether the early settlers came from the coast, or from the east. (VMC, 2005)

Kennewick Man may thus have been part of an earlier wave of people moving into the Americas, which was long confined to the coastal Northwest. With the end of the ice age, circa 12,000 years ago, another wave of settlement opened up via an ice-free corridor running from Alaska to Montana along the eastern side of the Rockies. This second wave, associated with the Clovis culture, brought more people than the first one and ultimately contributed the most to present-day Amerindians.

There were humans even earlier

But the story doesn't end here. There seems to have been another people before the Amerindians and even before the older and more European-like Kennewick humans. A recent study has looked at the gene pool of Native Americans from the Amazon. Not surprisingly, most of it closely matches that of Northeast Asians. But a tiny portion is like what we see in the natives of Australia, Papua New Guinea, and Melanesia (Skoglund et al., 2015)

It would be easy to dismiss this finding as a fluke, were it not for other evidence of a very different people who once lived in the Amazon basin 16,000 to 9,000 years ago (Roosevelt et al., 1996). While overlapping in time with the Clovis culture, they show none of its emphasis on big game hunting, as seen in the well-known Clovis projectile point and other hunting tools. In fact, they were much like tropical foragers of central Africa or Papua-New Guinea. And their earliest remains precede the Clovis culture by at least three thousand years, even though the Amazon rain forest should have been one of the last areas to be penetrated by former denizens of the Arctic.

There's more. A site in central Brazil has yielded several skulls dated to between 8,200 and 9,500 years ago. They don't look at all Amerindian: 

[...] they exhibit strong morphological affinities with present day Australians and Africans, showing no resemblance to recent Northern Asians and Native Americans. These findings confirm our long held opinion that the settlement of the Americas was more complicated in terms of biological input than has been widely assumed. The working hypothesis is that two very distinct populations entered the New World by the end of the Pleistocene, and that the transition between the cranial morphology of the Paleoindians and the morphology of later Native Americans, which occurred around 8-9 ka, was abrupt. [...] The similarities of the first South Americans with sub-Saharan Africans may result from the fact that the non-Mongoloid Southeast Asian ancestral population came, ultimately, from Africa, with no major modification in the original cranial bau plan of the first modern humans. (Neves et al., 2003).

Similar findings have emerged from analysis of skulls from Mexico dated to between 9,000 and 11,000 years ago and skulls from Colombia dated to between 7,500 and 8,300:

[...] only 6 out of 25 comparisons displayed in Table 3 tend to tie an early Mexican specimen to an Amerindian sample. Conversely, 19 of the 25 comparisons reflect the greatest similarity to Africans (6/25), Paleoindians (5/25), Australians (3/25), Polynesians (3/25), South Asians (1/25), or the Ainu (1/25). When first-place positions are explored, all five are circum-Pacific, either recent or early. Among second-place positions, 4 out of 5 are circum-Pacific, and the remaining one is African.

[...] To summarize, analyses of individual skulls against reference samples suggest that the early Mexican fossils studied do not share a common craniofacial morphology with Amerindians or East Asians, as reported elsewhere for South Paleoindians, some North Paleoindian specimens […] and some modern groups like Fuegian-Patagonians and the Pericúes from Baja California.

[...] This study does not support continuity between Early and Late Holocene groups in the Americas: Archaic remains from Colombia are not an intermediate point between Paleoamericans and modern groups. Moreover, the data presented here support the idea that the first settlers of the New World preceded the origin of the more specialized morphology observed in modern populations from Northeast Asia. (Gonzalez-Jose et al., 2005)

This shouldn't be too surprising. Here and there in Southeast Asia we find relic groups of small, dark-skinned, and woolly-haired hunter-gatherers: the Andamanese of India, the Semang of Malaysia, and the Aeta of the Philippines. They used to predominate throughout that region as late as four thousand years ago. Farther back in time, in prehistory, they may have also lived farther north, perhaps at one point the entire East Asian littoral ... and from there into the Americas. This would be before the last ice age, and probably before another wave of modern humans moved into northern Eurasia.

The past is another country, just as the future is another country. We unthinkingly assume that a place has always been home to a people who look a certain way, behave a certain way, and organize their lives a certain way. This is as untrue for the Americas as it is for anywhere else. Going back in time, we see people who look more and more ancestral not only to Amerindians but also to Europeans and East Asians. Eventually, those ancestral Eurasians disappear and we meet a very different sort of human.

What happened to those first inhabitants of the Americas? Did they go peacefully into the night when the newcomers arrived, retreating farther and farther into more remote areas? Or did the two groups fight it out? There was probably a range of scenarios—perhaps small numbers of newcomers initially worked out a modus vivendi with the natives, which later broke down as they became more and more numerous. In any case, the process matters less than the result. Those first Americans went into the night, peacefully or not.


Gonzalez-Jose, R., W. Neves, M. Mirazon Lahr, S. Gonzalez, H. Pucciarelli, M. Hernandez Martinez, and G. Correal. (2005). Late Pleistocene/Holocene Craniofacial Morphology in Mesoamerican Paleoindians: Implications for the Peopling of the New World, American Journal of Physical Anthropology, 128, 772-780

Neves, W.A., A. Prous, R. Gonzalez-Jose, R. Kipnis, and J. Powell. (2003). Early Holocene human skeletal remains from Santana do Riacho, Brazil: implications for the settlement of the New World, Journal of Human Evolution, 45, 19-42. 

Rasmussen, M., M. Sikora, A. Albrechtsen, T. Sand Korneliussen, J.Victor Moreno-Mayar, G. David Poznik, C.P.E. Zollikofer, M.S. Ponce de Leon, M.E. Allentoft, I. Moltke, H. Jonsson, C. Valdiosera, R.S. Malhi, L. Orlando, C.D. Bustamante, T.W. Stafford Jr. D.J. Meltzer, R. Nielsen, and E. Willerslev. (2015). The ancestry and affiliations of Kennewick Man. Nature, early view 

Roosevelt, A.C., M. Lima da Costa, C. Lopes Machado, M. Michab, N. Mercier, H. Valladas, J. Feathers, W. Barnett, M. Imazio da Silveira, A. Henderson, J. Silva, B. Chernoff, D.S. Reese, J.A. Holman, N. Toth, and K. Schick. (1996). Paleoindian cave dwellers in the Amazon: The peopling of the Americas, Science, 272, 373-384. 

Skoglund, P., S. Mallick, M.C. Bortolini, N. Chennagiri, T. Hunemeier, M.L. Petzl-Erler, F. Mauro Salzano, N. Patterson, and D. Reich. (2015). Genetic evidence for two founding populations of the Americas, Nature, early view. 

VMC (2005). A journey to a new land. Coastal Refugia


Anonymous said...

I was under the impression that the Kennewick DNA findings discredited the theory that he was part of a prior migration to the Americas. If indeed he his genome fits comfortably within the range of modern Amerindians, why posit population replacement at all?

There is a parallel study by Raghavan et. al. - titled "Genomic Evidence for the Pleistocene and Recent Population History of Native Americans" - in the latest issue of "Science," which calls into question the alleged Australasian affinities of groups like the Fuegians and the Pericues of Baja California, on both genetic and morphological grounds.

That said, the physical similarities between modern Northeast Asians and Native Americans are too close to be persuasively explained by parallel in-situ evolution. Since the classically "Mongoloid" facial structure does indeed appear to be heavily influenced by a small number of genes - EDAR370-A, I think, is one of them - it's entirely possible that one of the smaller late migrations to the New World - say, that of the Eskimo-Aleuts - enabled these traits to introgress into the older and more numerous Amerind groups without altering their overall genomic makeup.

This hypothesis would account for why Amerindians show considerable differences from, as well as similarities to, Northeast Asian peoples in their outward appearance. It might even explain some of the regional variation one still encounters in New World populations.

Malcolm Smith said...

I am reminded of the two racial types depicted by the Olmecs, the oldest civilisation in Mexico. The first is represented by the huge stone heads, which are distinctly Negroid in appearance ie flat noses, thick lips, and curly hair. Meanwhile, the reliefs on the stone walls depict "Uncle Sams" ie long faces, long noses, long hair, and goatee beards.

Anonymous said...

I still wonder who was camping out at Lake Lewisville 30,000 years ago...

Peter Frost said...


I don't think the Kennewick humans were replaced. It's just that they were submerged by a later and larger wave of migrants. I feel it's important to posit a separate and earlier wave because we are still faced with the difference in facial structure. Genetically, they were close to present-day Amerindians, but they seem to correspond to an earlier stage of evolution in northeast Asia.

I don't follow your suggestion that the Eskimo-Aleuts made the Amerindians to the south look more Mongoloid through gene diffusion. It sounds too much like the tail wagging the dog.


I had the same thought while writing this post. Were the Olmecs descended from these "First Americans"?


"During construction, members of the Corps of Engineers stumbled upon an archaeological site.[3] In 1956, Wilson W. Crook, Jr. and R.K. Harris announced that Carbon-14(14C) testing on artifacts from the site, including a Paleo-Indian Clovis projectile point, indicated that humans had lived there c. 36,000 BP.[4][5] This led to much controversy in the archaeological community.[6][7] It was not until 1978 that the water levels of the lake would go down far enough to access the site once again. Between 1978 and 1980, Dennis Stanford of the Smithsonian Institution performed a more thorough analysis of the site. He concluded that the original dating was probably due to a rare form of cross-contamination and that a date of c. 12,000 B.P. was probably more correct."

Derrick Bonsell said...

@Malcolm and Peter: A quick browse of le Wik says that they resemble people who live in and around Veracruz today.

XWindows said...

I see this has just come out: