How the human races got their stripes

By Razib Khan | September 27, 2007 5:53 pm

In my post The new races of man I tried to offer a verbal exposition of my current thinking as to how and why human physical variation shows the patterns we see around us. In short, I believe that powerful selective forces have reshaped a subset of the human genome in similar and different ways across a range of populations over the past 10,000 years. Empirically, I would predict that the physical appearance we denote as stereotypically “Chinese” or “Swedish” or “West African” might be rather recent ecotypes, adapted to new circumstances, both environmental and cultural. The types we see are simply a collection of correlations of salient phenotypic featuers (e.g., skin color, nasal form) which are the outcomes of selection driving evolutionary change on a small set of loci in the recent human past. But a great proportion of the genome might be relatively unaffected. And I believe this proportion is what we generally look toward to ascertain the phylogenetic characteristics of a population (generally you examine selectively neutral genes assuming that their evolution has proceeded via a constant molecular clock). Therefore, their may be a discordance between what a small number of recently selected alleles and their salient phenotypic outcomes may imply, and what the total genome content suggests.
Below the fold is my attempt to illustrate all that with a simple schematic representation.

So what’s going on here? The eyes are pretty cheesy, but I’m trying to get across that selection may “see” only a small proportion of the genome at any given time. Like light filtering down into the ocean most of the genome is dark and perturbed by constant but slow moving forces (neutral forces). In contrast the surface waters are bathed in sunlight and relatively dynamic in terms of their ecological output (selective forces). Additionally, winds and all sorts of extra-oceanic inputs can roil the surface. Though the oceanic depths are not trivial, as human beings we tend to see the surface and make our judgments based upon them. We classify waters as tropical based on their surface temperature, though at the greatest depths the variation across latitudes is sharply diminished. Similarly, we judge humans based on gross outward phenotypes, even if they don’t necessarily match the phylogenetic relationships which trace ancestry. Old style anthropologists classified the Ainu as “Caucasoid,” while many peoples of Oceania were classified as “Melanesians” or “Oceanic Negroes” because of their superficial physical resemblances with the peoples of Africa.
With the rise of cheap sequencing technology and more sophisticated analytic tools the broad outlines of the human tree of life are coming into focus. The Ainu were in fact an ancient East Asian people, and in terms of their ancestry shared more with the other peoples of Northeast Eurasia than they did with the distant Europeans. The peoples of Melanesia are not related to Africans, they are a branch of the large cluster of peoples of the western fringe of the Pacific. And so on.
What is occurring here? 10,000 years ago I believe there was relatively little difference between the Ainu and the peoples of mainland East Asia. But around the region which has now become China selective forces led to the emergence of a new phenotype, what old style anthropologists would term “classical Mongoloids.” Nevertheless, the neutral markers drifted at a slower pace, so in terms of total genome content the Ainu and the other peoples of East Asia are not particularly distant. Instead, on a subset of loci subject to selection powerful divergent evolutionary forces have taken the Ainu and the Han on radically different tracks.
So what is more important? The surface, or the total sum of the depths? That depends on your perspective. For example, in Brazil there is some evidence that the non-European ancestry among white Brazilians is far higher than they suspect. Their physical appearance doesn’t necessarily reflect their total genome content. How could this be? It seems plausible that assortative mating might have worked to separate out the traits associated with “whites” and “blacks” from an admixed population which exhibited the full phenotypic range. But the non-white ancestry of white Brazilians might mask the presence of alleles such as sickle cell.
In the end, it’s all a matter of emphasis. What is your interest? What is your passion? Nature is wide and deep, we can never truly comprehend it. We shouldn’t straight-jacket it into our narrow categories but attempt to capture its fullness and subtlety.
Note: Obviously phenotypes “visible” to selection don’t have to be literally visible. For example, the human immune system is always roiling with evolutionary activity.

  • agnostic

    Their physical appearance doesn’t necessarily reflect their total genome content.
    Only if you look at their skin color — other places would suggest to anyone with eyes to see that they have non-trivial African ancestry. I’d post a YouTube clip to substantiate this, but that would be prurient.

  • razib

    other places would suggest to anyone with eyes to see that they have non-trivial African ancestry
    most ozzies can tell melanesians from africans…but i think many non-ozzies would get them confused.

  • HP

    Razib, I believe that Agnostic was referring to the incidence of steatopygia in fair-skinned Brazilian females.

  • hmm

    Interesting post, and now for a not-so-minute minutia:
    Didn’t we stop using ‘man’ to refer to humankind, like, decades ago?

  • razib

    Razib, I believe that Agnostic was referring to the incidence of steatopygia in fair-skinned Brazilian females.

    ah…i forgot the assman’s interests.

    Didn’t we stop using ‘man’ to refer to humankind, like, decades ago?

    you did, i didn’t.

  • hmm

    “you did, i didn’t.”
    So, an archaic expression that only refers to half the population is a good thing for you? It’s not even ‘political correctness’, it’s just accuracy.

  • Matt McIntosh
  • Ted Gemberling

    As a curious non-scientist, I wanted to raise a question about something I saw some months ago. I catalog old medical books. One subject that was quite popular in the early 20th century and has some relevance to this thread is eugenics. There’s a fascinating book I cataloged called Personal beauty and racial betterment (1920), by Knight Dunlap. I got into it expecting it to be ridiculous. The term “racial betterment” struck me as particularly ridiculous. But as I read quite a bit of it, I was struck by how generally lucid it is. It’s basically a book about what makes someone a good person to mate. Who will make the best contribution to the human gene pool, though I don’t think he used that terminology. When he uses the word “racial” in the title, it’s not referring to what we normally call “race,” but to the human race.
    His analysis is that people who are “well balanced” are the best candidates for mating. That’s what he calls “personal beauty.” One aspect of it is that a person with strengths in lots of areas, without perhaps being outstanding in any one, has greater fitness for mating.
    Now, this is where my question comes in. Nick convinced me some months ago that there is such a thing as “race” in the usual sense. I think he explained it as populations that have been isolated from each other and have developed separately, so they have genetic material they mostly don’t share with other populations. Forgive me if that misses the mark in some way. One thing I’ve noticed in old eugenic books is that they generally discourage racial mixing. Typically, they say it leads to “degradation.”
    However, in one collection of papers I looked at, there was one author that mentioned, without drawing any conclusions, that he’d seen cases where mixing different “breeds” of a species led to more vigorous offspring. It seems that eugenicists of the early 20th century generally did not distinguish between the social and biological causes of “degradation.” Apparently race mixing in itself is not a cause of degradation without social factors such as one ethnic group being stigmatized. While Dunlap seems to acquiesce in the idea of restricting race mixing, the popular idea of his time, I think he was too smart to miss that distinction.
    Are there any cases where you think race mixing is bad from a purely biological standpoint?

  • Ted Gemberling

    A comment to hmm on “man.” I think you can use intuition on that. If “man” is pluralized or preceded by an article (men or a or the man), then it is definitely masculine in reference. But if you speak of the “the good of man” or “the good of mankind,” without an article, it is not specifically male in reference. I’ll admit that “the good of woman” is of course strictly female in reference. But I think that’s just because we never use “woman” in the same generic way we use “man.”
    Of course you can always say it’s more “just” for “the good of man” to be male if “the good of woman” if female, but I just think intuition shows that’s not the case. That has nothing to do with male domination. It is just the way we use the words.


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Gene Expression

This blog is about evolution, genetics, genomics and their interstices. Please beware that comments are aggressively moderated. Uncivil or churlish comments will likely get you banned immediately, so make any contribution count!

About Razib Khan

I have degrees in biology and biochemistry, a passion for genetics, history, and philosophy, and shrimp is my favorite food. In relation to nationality I'm a American Northwesterner, in politics I'm a reactionary, and as for religion I have none (I'm an atheist). If you want to know more, see the links at


See More


RSS Razib’s Pinboard

Edifying books

Collapse bottom bar