Origin of Native America

The origin and early history of Native American people has always been an issue of debate and contention. There has never been a moment when all, or even most, interested parties agreed on anything close to a single story. New research published in the Open-Access journal PLoS Genetics tends to support a very traditional (among archaeologists) view of a single relatively simple migration from Siberia across the New World, more or less from north to south.

Studies of genetic variation have the potential to provide information about the initial peopling of the Americas and the more recent history of Native American populations. To investigate genetic diversity and population relationships in the Americas, we analyzed genetic variation at 678 genome-wide markers genotyped in 29 Native American populations. Comparing Native Americans to Siberian populations, both genetic diversity and similarity to Siberians decrease with geographic distance from the Bering Strait. The widespread distribution of a particular allele private to the Americas supports a view that much of Native American genetic ancestry may derive from a single wave of migration. The pattern of genetic diversity across populations suggests that coastal routes might have been important during ancient migrations of Native American populations. These and other observations from our study will be useful alongside archaeological, geological, and linguistic data for piecing together a more detailed description of the settlement history of the Americas.
[From the paper's Author's Summary]

Now, as you read about this new research, you will see it placed in the context of the Bering Land Bridge model of the peopling of the New World. You will also see reference to the question of how many different sources there may have been for the Native American population (single vs. multiple migrations), as well as the question of "pre-Clovis" occupation. Within the context of the "pre-Clovis" discussion you will see two kinds of discussion, what one might call "pre-Clovis" and "Older-than-Clovis" ... meaning, people who were not yet but would be bearers of the typical Clovis stone tool culture, vs. people who lived before, and possibly unconnected to, the Clovis culture.

All of these links are pathways to misunderstanding, or in some cases, over interpreting and almost always over simplifying Native American history.

What I want to do is to very briefly note what are clearly the three most important things about this new paper, then address each of these other issues in turn, mainly by way of expressing my opinion. If you want a more detailed discussion with references and bits of evidence, that's easy to obtain: Go read a few thousand pages of pertinent literature and synthesize it. Let me know how that goes, I'm sure you will find it interesting!

The PLoS paper employs a new analysis of a bigger than ever and better-analyzed than ever genetic data set and from this makes these three critically important points:

1) By and large, Native Americans come from Asia, most likely Northeastern Asia (Siberia).

2) The spread across the New World was what would be expected with a more or less unified migration from the Northwestern regions of North America across North America, through Central America, and into and across South America. Like in this figure from the paper:

You can visit the original paper to read the details, but you can get the point easily from this graphic, which is color coded to show some measure of genetic distance from a Northwestern origin.

3) The movement of people was probably often along coastal routes, so that in a given region, you will see people living in coastal region moving into the interior, more or less. This does not mean that people did not move in all sorts of ways, across all sorts of landscapes over this many thousands of year time period, but there is a reasonably strong signal of coastal to inland movement. Relative to the first two main findings (above) this is somewhat new and a tiny bit of a surprise (depending on what you were previously thinking).

This graphic demonstrates the thinking on this:

This analysis is a bit conjectural in a sense. What this is showing is that there are stronger correlations between coastal and inland areas than between other randomly paired points in the data set.

So, a big rolling wave of Siberians working their way across the New World, with something of a bias towards moving along coastal regions and from there, inland, probably over thousands of years.

You can stop reading now and you'll have everything I'm reporting from the article. Following are my additional comments:

This inland movement signal could be the result of movement near the beginning of the migration across a given region or from later movements. I would assert that the latter is probably supported in the archaeological record of the Laurentian Shield and the Late Archaic and early Woodland of the East Coast, for instance.

What does the Bering Land Bridge have to do with this? Probably nothing at all. The Bering Land Bridge is one possible way in which people could have gotten from Siberia to Alaska/Canada without using boats most of the way. Or, they could have used boats. The simple fact is that the Pacific Ocean is a big, huge, wide piece of water that no one between 10 and 20 thousand years ago was going to cross in a boat, except way up north where the ocean is so narrow that you can see across it on a clear day. We don't need no stinking land bridges. If there was a land bridge at the right place and the right time, fine. If not, I don't see that as an issue.

What about the number of migrations? One or many? This is a question that starts to look a little dumb once you realize what we are asking. Try this on for size: We divide the New World Native sample into, say, five genetically separable parts and determine that each came from a different place, A, B, C, D, and E. So, is that five migrations into the new world? No, of course not, because in my particular thought experiment, I'm imagining all five populations to have differentiated in Canada. Or, maybe I'm imagining that A, B and E arose in Siberia, and C and D arose in Canada. Indeed, maybe there is some population in Siberia that descends from ancient native Canadians. Why not? Is there some rule that people can only go in one direction? If you believe that, why do you believe that? Are you nuts? Well, maybe you are nuts, but if you read the literature, this one-wayness is an essential assumption made by almost everybody.

Regarding pre-Clovis. Here's the thing. The Clovis Culture is identified mainly by the presence of a certain kind of projectile point (probably a spear point) known as "The Clovis Point." It is one of the coolest looking stone points you'll see in the North American toolkit. Archaeologists knock each other over the head over the question of pre-Clovis sites ... are there pre-Clovis sites or not? Well, the evidence has been mounting that a) yes there are pre-Clovis sites in the New World, but b) the really really oldest of the pre-Clovis sites may be questionable.

But what does this have to do with the genetic and migration paper at hand? Very little. The timing on this is very hard to pinpoint at the level necessary to test the pre-Clovis vs. Clovis-first hypotheses. First, the coalescence of all of the Native American data sets on a particular point in time may be fairly inaccurate, and second, that coalescence point is irrelevant to the pre-Clovis question . Refer to the A-E model above. If you could accurately date, with mutation rates, etc., the divergence of all of the populations sampled in this paper and found that this divergence dated to, say, 12,500 years ago plus or minus 200 years, what would that tell you about Clovis? Nothing, really. That coalescence point may or may not be related to the "crossing" into the New World. It may predate the crossing. It may post date the crossing. And for the most part, you can't tell.

Finally, there is no rule in archaeology or genetics that says that the genetic signature of a region is related to every archaeological manifestation. If there is a pre-Clovis archaeological manifestation, it could be unrepresented in the genetic signal. They all died off. Or they moved back to Siberia. Seriously, they could be anywhere. Indeed, the Clovis itself may be unrelated to the present genetic study. Why would you assume it was? Wishful thinking? Forgetting about the fact that almost all groups of people die out without issue?

As we look at both genetic and archaeological signals, it is important to remember that just because we have two signals means that they are correlated in a particular way, or correlated at all (especially for the earlier periods). When we look at an archaeological sequence, it is usually a matter of politics and not science when it is interpreted as the result of a migration or an indigenous development. When we identify genetic coalescence points, it is too easy to assume that these genetic "events" are actual events, and correlate with archaeological or biogeographical events. In truth, there is almost never a way, or even a need, to connect the two.

One final comment on this complex issue: It is believed by many that an important rule in science is the Law of Parsimony. The simplest solution is the most correct. Well, not only is that not the law of Parsimony, but I guarantee you that no form of the Law of Parsimony was drafted by historians.

WANG, S., LEWIS, C. M., JAKOBSSON, M., RAMACHANDRAN, S., RAY, N., BEDOYA, G., ROJAS, W., PARRA, M. V., MOLINA, J. A., GALLO, C., MAZZOTTI, G., POLETTI, G., HILL, K., HURTADO, A. M., LABUDA, D., KLITZ, W., BARRANTES, R., BORTOLINI, M. C., TIRA, SALZANO, F. M., PETZL-ERLER, M. L., TSUNETO, L. T., LLOP, E., ROTHHAMMER, F., EXCOFFIER, L., FELDMAN, M. W., ROSENBERG, N. A., RUIZ-LINARES, A. & S (2007): Genetic Variation and Population Structure in Native Americans. PLoS Genetics, 3(11), e185 doi:10.1371/journal.pgen.0030185.