On the biology of race

the-different-human-races-from-natural-history-of-the-animal-kingdom2by Massimo Pigliucci

The biology of human races is back in the news, big time. This is because of a new book by former New York Times journalist Nicholas Wade, A Troublesome Inheritance: Genes, Race and Human History [1].

The basic thesis of the book is that human races are real, and that their genetic differences — which according to Wade evolved rapidly after the invention of agriculture — account for much of the behavioral differences among human groups, as well as for the success of some and the failure of others.

Here is a sample of quotes from Wade himself, to give you an idea of what he is up to [2]:

[Trying to explain why Western societies have been capable of developing advanced democracies while others haven’t] “Conventionally, these social differences are attributed solely to culture. But if that’s so, why is it apparently so hard for tribal societies like Iraq or Afghanistan to change their culture and operate like modern states? The explanation could be that tribal behavior has a genetic basis.” (Indeed, but what explains the difference between the cultures of North and South Korea, which — as Allen Orr notes in his review of the book (see below) — are certainly genetically very close to each other?)

[Attempting to show that the Industrial Revolution was the result of a sort of trickle down genetics from the rich to the lower classes] “The essence of the Industrial Revolution was a quantum leap in society’s productivity. … Given the correctness of Darwin’s theory, there is no reason to doubt that natural selection was working on the very English population that provided the evidence for it. … The burden of proof is surely shifted to those who might wish to assert that the English population was miraculously exempt from the very forces of natural selection whose existence it had suggested to Darwin.” (Except that the very source that Wade cites in support of his hypothesis goes on to show that there was very little differential production between the higher and lower classes, if one expands the time frame from one generation to two or more.)

[Suggesting that Jews are genetically superior in terms of intelligence] “A second instance of very recent human evolution may well be in evidence in European Jews, particularly the Ashkenazim of northern and central Europe. In proportion to their population, Jews have made outsize contributions to Western civilization.” (Too bad that intelligence is both vaguely defined and known to result from complex gene-environment interactions, and that “Jews” is much more of a cultural than a genetic group anyway.)

[While trying to explain the West’s superiority to China and Islam] “One candidate is the rise of the West, which was prompted by a remarkable expansion of European societies, both in knowledge and geographical sway, while the two other major powers of the medieval world, China and the house of Islam, ascendant until around 1500 AD, were rapidly overtaken. … Do Chinese carry genes for conformism and authoritarian rule? May Europeans have alleles that favor open societies and the rule of law? Obviously this is unlikely to be the case. But there is almost certainly a genetic component to the propensity for following society’s rules and punishing those who violate them.” (This despite a glaring lack of actual data. Also, if Wade had written his book 500 years ago, he would have been forced to conclude that Westerners were the inferior race, which would have explained why they were at the mercy of the Ottoman Empire. Then again, had he written it 2000 years ago, he would have concluded that the Romans had the best genetic complement of them all. I quite like that one, myself.)

No need to go further, you get the gist. There are several excellent reviews of the book [3], and I will briefly quote from one of the best, by my colleague Allen Orr, an evolutionary biologist at the University of Rochester [4]. I’m afraid I really couldn’t do any better myself.

“Wade says that biologists realized only in the last few years that natural selection might change a trait by causing slight changes in the frequencies of variants at many genes instead of a large change in frequency at one gene. In fact, the former hypothesis is the traditional view of evolutionary change and is nearly a century old. It would be unfair to suggest that these sorts of mistakes undermine Wade’s main claims in the first part of A Troublesome Inheritance. But they do suggest that he is not the surest guide to a technical literature. … At one moment, he will concede that he writes in a “speculative arena” and, at the next, he will issue pseudofactual pronouncements (“social behavior, of Chinese and others, is genetically shaped”). This strategy lets Wade move in a kind of intellectual no-man’s-land where he gets to look like he’s doing science (so many facts about genomes!) while covering himself with caveats that, well, it’s all speculative. … it’s also true that some behaviors or institutions persist for purely cultural reasons. The English have used a currency called the “pound” since Anglo-Saxon times. And Western music has been built on a diatonic scale since the Renaissance and probably much earlier. So why doesn’t Wade conclude that differences in currency and musical scale reflect differences in genes? … there is a species of bravado here, as though demonstrating that he, unlike others, is tough-minded enough to face unpleasant facts. But surely there is a difference between facing facts that are unpleasant and spinning tales that are improbable.”

And so on. Check it out, it’s worth reading in full. But this essay isn’t about Wade’s book per se, it is about the broader concept of human races as biological (as opposed to cultural) entities. I feel somewhat qualified to write about it for two reasons: as a biologist, I worked on gene-environment interactions (albeit on plants!), i.e. precisely on the question of nature vs nurture [5]. As a philosopher, I have published three peer reviewed papers on human races [6], though my principal interest lies in the philosophy of science (and pseudoscience).

Before we proceed, let me state very clearly that Wade is absolutely right on one thing: the science of human races is entirely orthogonal to the ethics of the issue. Precisely because we cannot derive ought from is, it simply doesn’t matter — from a moral perspective — what science finds out about races, since nothing of moral import will ever follow from that. If you are not convinced, just ponder for a minute that we accord full person status to severely mentally deficient human beings, and that we are beginning to consider the idea of expanding that status to include at the least some other species of primates, which are most definitely and unequivocally biologically inferior to us in terms of intelligence.

The question, then, is what is the best interpretation of the available evidence concerning the existence and alleged biological foundations of human races. That genes are capable of influencing behavior, including in humans, is a fact. That relatively small genetic differences can cause detectable phenotypic differences (i.e., differences in appearance and/or behavior) is also a fact. And that geographically separated human populations differ (slightly) in their genetic makeup is a fact too. But there isn’t much else to go on as far as empirical evidence is concerned, unfortunately. Wade-style talk of genes “for” democratic tendencies, financial smarts or capitalist innovation is just that, talk [7]. Not only do we know of no such genes, but we also have excellent reasons to believe that if they exist their effects are small and completely intertwined with cultural (and other environmental) effects. To pretend otherwise is intellectual travesty or naiveté, not bravery in the face of political correctness.

The major reason for this rather unsatisfactory state of affairs is that it is inordinately difficult to study gene-environment interactions, especially in long-lived species like humans, quite outside of the obvious ethical problems raised by the necessity to carry out controlled genetic crosses between “races” and to raise the resulting offspring in controlled separate environments. It’s not that we don’t know how to do these things, indeed I carried out plenty of “GxE” (as they are called) experiments in my lab for several years, and so do plenty of other people. But it is next to impossible to do it with humans.

Here is how it ought to be done, however. There is a classic paper by R.M. Cooper and J.P. Zubek [8] where the authors were able to quickly (seven generations) select for “dull” and “smart” rats, with the speed of running a maze being used as a proxy for intelligence. So far, nothing extraordinary, as Cooper and Zubek simply demonstrated a genetic basis to intelligence, and therefore the predictable ability of the trait to respond to selection. Things became more interesting when they reconstructed parts of what is known as the “norm of reaction” for intelligence in their populations [9]. They exposed both dull and smart rats to either an enriched environment (full of visual and tactile stimuli) and a poor environment (deprived of stimuli), as opposed to the “standard” cage environment in which the selection process had been carried out. The results were as clear as stunning: the genetic difference between dull and smart rats was essentially erased in both the enriched and the dull environment, thereby showing a high degree of “phenotypic plasticity” for intelligence in the genetically selected lines of rats.

Since we can’t do Cooper-Zubek type experiments with humans, what we are left with is that we just don’t have any good scientific evidence that there are systematic differences in cognitive traits among human races that are genetically based, the result of natural selection, and that cannot be substantially overridden by changes in the cultural environment. That, however, doesn’t mean that there is no sensible way in which we can talk about human races. In fact, there is at the least one such way, I think, but let me first debunk one common type of biological race talk.

“Race” is a rather fuzzy concept in biology to begin with, even outside of the human case. More often than not, however, in biology a race is a subspecies, or incipient species, i.e. a geographically isolated population that is on its way — unless the process reverses itself because of new gene flow to/from nearby populations — to becoming a new species. Here is how the great evolutionary biologist Ernst Mayr famously put it: “a subspecies is a geographic race that is sufficiently different taxonomically to be worthy of a separate name” [10]. In that sense, then, there most definitely are no human races, for the simple reason that there is no deep phylogenetic separation (no evolutionary branching) among human populations. This, in turn, is the result of the fact that modern humans evolved from a genetic bottleneck (i.e., are the descendants of a very small population) that occurred in Africa about 50,000 years ago.

If human races are not subspecies, is there any scientifically meaningful way to recover the concept of “race” as it applies to our species, and how does this square with the “folk” concept of species (i.e., with talk of “blacks, Caucasians, Asians,” and so forth)?

Jonathan Kaplan and I (first reference mentioned in [6]) think there is. We suggested that human races are really what biologists call “ecotypes” [11], geographically distinct populations that have evolved some special way of adapting to their local environment. Classical examples of ecotypes are Alpine plants, which differ genetically very little from their lowland conspecifics, but have developed a more compact appearance to better withstand the harsh conditions under which they live (e.g., by limiting exposure to damaging winds, as well as reducing water loss). The concept applies to animals as well: there are two ecotypes of Tundra reindeer, for example, a migrant and a sedentary one.

The idea, then, is that certain superficial physical characteristics in humans — chiefly, but not exclusively, skin and eye color — did evolve genetically, likely in response to natural selection for adaptation to certain types of environments, most obviously dark skin to protect people from high levels of UV radiation, and light skin to minimize inability to synthesize vitamin D in places where high incidence of sunlight is not the most pressing problem.

If Jonathan and I are correct (and we do have our critics in the professional philosophical literature), there are a few consequences of our suggestion that are worth considering. To begin with, there is no reason whatsoever to think that selection on, say, skin color, was accompanied by selection on any cognitive trait. It is possible, of course, but the burden of proof is on those making the claim. Second, “race,” then, would literally be skin deep: human races, as far as we can tell, are characterized only by rather straightforward differences in skin color, eye color and a few related traits. Lastly, and most importantly, thinking of human races as ecotypes means that the scientific image does not match with the folk image of race. For instance, light skin evolved following different genetic pathways in Western and Eastern Eurasian populations. Moreover, although South Asian populations retained the ancestral dark pigmentation, its genetic modulation appears to be very different from that of, say, West African populations [12]. The upshot is that when people think of “blacks” as a race, they are not actually picking a scientifically coherent unit, only one that is defined by a mishmash of small and superficial set of biological traits (skin color etc.) and a convoluted cultural history.

To be as clear and explicit as possible, then, here is the full last section of my paper with Jonathan, which summarizes our views and how they differ both from the “races are fundamental biological entities” and the “races are entirely a social construct” views so common in this debate:

“As we have seen, insofar as biologically meaningful races are conceptualized as populations more like ecotypes than like incipient species, many of the arguments purporting to show that there are no human races miss their mark. While in nonhuman biology the term ‘race’ has been and is being used in a variety of ways, the best way of making sense of systematic variation within the human species is likely to rely on the ecotypic conception of biological races. In this sense, there are likely human races (ecotypes) of biological interest. But again, biology provides no support for the very strong, essentialist-style conception of ‘race’ that has, both historically and at present, underwritten racism (of both the individual and institutional varieties), and indeed, biology reveals that the assumptions underlying such a conception of race are false.

This does not, of course, imply that our folk conception of race is not significant — while it does not pick out populations of biological interest, it does pick out populations of deep social and political interest. These populations do not, in fact, have many of the features they were historically supposed to have, but that does not prevent the application of the folk concept of race. Nor, we believe, should it. As long as the folk racial category to which one happens to belong is systematically related to other important aspects of one’s life, there is obviously still a need to pay attention to race in formulating, for example, social policy. And, it need hardly be said, it is. In the U.S., and in at the very least many other contemporary societies, one’s (folk) race is systematically related to one’s chances of acquiring most (if not all) important goods — everything from education to money to self-respect.

While it is valuable for biologists to note that the essentialist conception of human races has no support in biology whenever particular claims are made that seem predicated on such a conception (e.g., Herrnstein and Murray’s 1994 work on race and intelligence [13]), they should not fall into the trap of claiming that there is no systematic variation within human populations of interest to biology. Studying human ecotypes could yield insights into our recent evolution, and perhaps shed increased light onto the history of migrations and gene flow. To some extent, this is already happening (see Cavalli-Sforza et al. 1994, etc [14]). However, the ambiguity surrounding definitions of ‘‘race’’ and the politically charged atmosphere surrounding race in humans has hampered research into these areas, a situation from which neither biology nor social policy surely benefit.”


Massimo Pigliucci is a biologist and philosopher at the City University of New York. His main interests are in the philosophy of science and pseudoscience. He is the editor-in-chief of Scientia Salon, and his latest book (co-edited with Maarten Boudry) is Philosophy of Pseudoscience: Reconsidering the Demarcation Problem (Chicago Press).

[1] A Troublesome Inheritance: Genes, Race and Human History, by N. Wade, Penguin Press.

[2] An excerpt of the book, if you don’t want to support the author by purchasing the whole thing, appeared in Time magazine.

[3] E.g., The Paradox of Racism, by A. Gelman, Slate; On the Origin of White Power, by E.M. Johnson, Scientific American blogs.

[4] Stretch Genes, by H.A. Orr, The New York Review of Books.

[5] See: Phenotypic Plasticity: Beyond Nature and Nurture, by M. Pigliucci, Johns Hopkins University Press.

[6] On the concept of biological race and its applicability to humans, by J. Kaplan and M. Pigliucci, Philosophy of Science; Human races, by G. Barbujani and M. Pigliucci, Current Biology; and What are we to make of the concept of race? Thoughts of a philosopher–scientist, by M. Pigliucci, Studies in History and Philosophy of Biological and Biomedical Sciences.

[7] Jonathan Kaplan and I have also written about the very concept of “genes for”: Genes ‘for’ phenotypes: A modern history view, Biology and Philosophy.

[8] Effects of enriched and restricted early environments on the learning ability of bright and dull rats, by R.M. Cooper and J.P. Zubek, Canadian Journal of Psychology. Unfortunately, it is behind a paywall. If you email me I can lend you my pdf version.

[9] A norm of reaction, or reaction norm, is a graphic or mathematical function illustrating how a given genotype produces a range of phenotypes in response to a range of environmental conditions. If the norm of reaction is flat, i.e. parallel to the environmental axis, the genotype is said to be non-plastic; otherwise, it shows phenotypic plasticity. Here is a graphic representation of what norms of reaction look like.

[10] Populations, Species, and Evolution: An Abridgment of Animal Species and Evolution, by E. Mayr, Belknap Press.

[11] Here is the Wiki entry on ecotypes.

[12] Here is an in-depth Wiki article on human skin color variation, its genetics and its evolution.

[13] Herrnstein, Richard J., and Charles Murray (1994), The Bell Curve: Intelligence and Class

Structure in American Life. New York: The Free Press.

[14] Cavalli-Sforza, Luigi Luca, Paolo Menozzi, and Alberto Piazza (1994), The History and Geography of Human Genes. Princeton: Princeton University Press.


101 thoughts on “On the biology of race

  1. I don’t want us to get too side-tracked into arguments about Fst values. There is no firm cut-off re: Fst values in the literature for identifying subspecies that is widely accepted, and about this, Templeton and Hoffman for example are, indeed, just wrong. Part of this is because Fst is a bad measure of the kind of differentiation associated with subspecies status, and part of this is because “subspecies,” as a category, is simply a conceptual mess. Nevertheless, in at least some nonhuman cases, where there are discussions of subspecies, Fst values of under .3 or so are taken to be surprisingly low (see e.g. “Gorilla Biology: A Multidisciplinary Perspective,” page 117, on the “Morphological Differentiation of Subspecies” where Fst values of .24-.3 are said to be “relatively low” and therefore “unexpected” given the other evidence for there being well-defined subspecies). But yes, there are other cases were much lower Fst numbers are said to be compatible with subspecies status.

    Part of the problem is that Fst values (as a measurement, roughly, of excess homozygosity) are sensitive to total variation. But even so-called true diversity measures, when they partition diversity into within-population and between-population components, hide the fact that equally high between- as opposed to within- population variation can be associated with very different amounts of total variation…

    So. Given the very low levels of genetic diversity in the human species, the very low levels of morphological diversity, and the well-established fact that most of the variation that exists, both genetic and morphological, occurs within rather than between populations, I think it would be odd to try to pick out subspecies of the human species. Not impossible, given, again, the different ways that the term gets used, but certainly odd. But here is what I think is the critical point: IF one were to do so, *absent* current notions of folk-racial categories, I can see no good reason that one would settle on the populations we currently identify as races in our ordinary discourse.

    Why, I ask, would someone, not already in the thrall of current ordinary social racial thinking, pick out those particular populations as “subspecies”? There was always extensive gene flow between the so-called “big three”; while there were also partial discontinuities (steeper gradients, etc.) due to geography, that doesn’t change the fact that what we had was still, in essence, isolation by distance. And note that focusing on the discontinuities associated with the so-called “big three” of course ignores the other places where there were barriers that reduced gene flow within the particular populations usually identified as one of the ‘major’ races in this context.

    A biologist obsessed with finding subpopulations — a classic splitter — could no doubt split up the human population into different many different “subspecies,” but why think they would resemble the populations we now pick out as “races” in social discourse? There was never a time when the so-called “major races” as usually identified were separate from each other (where there wasn’t gene-flow) for an extended period of time. To the best of my knowledge, every recent study that has looked for it has found evidence for extensive gene flow between the so-called “major races,” and, conversely, has shown that population structure — differential gene-flow — goes “all the way down” to tiny geographic regions. There just isn’t anything *biologically* special about the populations usually identified as “races” in ordinary discourse.

    In short: Biology simply does not force the “races” identified in our ordinary racial discourse on us as obvious divisions of our species. Biology cannot explain why we identify *these* populations as races rather than others; cannot explain why we use this set of categories rather than a different set. Other ways of sorting people into subpopulations are equally well, and in some cases perhaps better, justified by our best biological practices.

    (Again, many of the biologists cited as suggesting that there are biological human “races” are using that term to mean something very weak. The populations so-identified would not necessarily resemble those identified in social discourse, and the biological facts used to support racial realism would not be sufficient to explain why we pick out these populations in social discourse rather than others. Again, *that* there is population structure, no one denies. Given population structure, the ability to identify subpopulations will always be possible, and where there are even partial barriers to gene-flow, or local adaptive regimes, there will be ‘natural’ breaks that make the identities of those subpopulations not-completely arbitrary. Granted. But none of that forces the races usually identified in ordinary discourse upon us, or even strongly suggests that those are the natural divisions…)

    As far as the term ‘race’ is concerned, I remind you that MOST of the peer-reviewed, serious biological literature on nonhuman organisms that addresses population structure, local adaptations, speciation, etc., gets by just fine without using the term ‘race.’ It is too ambiguous, and better, more careful, terminology exists, and is regularly deployed. Not always, but the terminology of “races” in the non-human animals literature is clearly on the way out, and has been for some time. Talking about “race” in non-human organisms is perhaps still optional, but since it is more likely to confuse than actually using more careful and specific terminology, increasingly avoided. In those areas where the term is still sometimes used, it is used differently in different places, and there would be nothing less “natural” about referring to a “high-altitude-adapted race” than about a “geographic race.”

    On the other hand, talking about — using the term — “race” in humans remains critical, because of the history and existence of racism, and the vital importance of social ascription of “race” to individuals. And if one wants to carefully discuss what particular biological research projects have shown about e.g. the history of particular human populations, to use the term “race” seems misguided, unless it is referring *explicitly* to the social ascriptions rather than to the biological facts. People are discriminated against today (and still suffer because of past discrimination) based on socially ascribed racial designations, based on “race,” where this designation has little to do with the kinds of biological facts (understood narrowly, at least) that are used to understand population structure and adaptation.

    The above is important, because it gets at, in part, why so many researchers are so frustrated with Wade’s book (when both Orr and Marks are against one’s interpretation of evolutionary biology and recent work on human population genetics, one has got a problem!). The hypotheses he spins seem to presuppose certain kinds of assumptions about e.g. gene flow and selection, and he helps himself to these assumptions by misrepresenting the state of knowledge about the history of human migrations, and about human population structure more generally. That’s a problem.

    So when I hear someone say “race, in humans, is not biologically real,” I interpret this, based on the history of the arguments, to mean that biology cannot explain why we pick out these populations as “races” worthy of attention rather than others, nor, especially, can it explain the important social facts about these populations. Have some people said things like that and meant much stronger (and hence false) things by it, such as that there are no biological differences between the populations identified in social discourse as “races” at all? Probably, but if so, they were wrong. Some people that say “evolution is true” think this means a variety of false things; this doesn’t mean we shouldn’t say “evolution is true”!

    (On another note, if you are really not outraged and disgusted by people advocating horrible positions designed to harm others, especially where those people in harm’s way are already disadvantaged, I strongly urge you to reconsider your moral compass. I note that below you consider the wisdom of maintaining a professional relationship with someone with morally appalling views in a way that makes it seem like the only issues are prudential — will people’s views of your own work be influenced by your known association with someone horrible? But the issues must surely go beyond that — to not condemn (let alone not even be outraged by!) morally abhorrent views, is I think, a moral failing, and an important one. To fail to find arguments premised on there being a nefarious “homosexual agenda” morally appalling, and to fail to strongly condemn the people who put forward such arguments, strikes me as at least similar to failing to condemn someone who regularly cites the “Protocols of the Elders of Zion.” That holds, even if one were never to cite that person’s particular views on “the Jews” but only on international banking policy. While I know that we are all very tired of the “fake outrage” that is so popular these days, to fail to be outraged by outrageous claims that support real harm in the world — there are still many countries where homosexuals are regularly beaten and killed simply for being homosexual, and ignorant assholes who spread vicious lies about “homosexual agendas” have some of that blood on their hands, after all! — can’t be the right answer, either.)


Comments are closed.