This is the first of three guest posts by Professor Jonathan Kaplan of Oregon State University.  Professor Kaplan is a noted philosopher of biology who has published extensively on biological race and IQ among other topics.

Part One: Introduction and Misleading Comparisons

In this series of blogposts, I discuss three relatively recent papers that, in one way or another, defend the so-called hereditarian hypothesis – the claim that genetic differences between human populations identified as ‘races’ are causally responsible for realized differences in cognitive abilities as measured by I.Q. test-taking ability, in a straightforward way via developmental effects on e.g. neurobiology (rather than through e.g. racism and the legacies of racism associated with racial ascriptions). But my main point in these posts is not to engage directly with the arguments put forward in defense of that hypothesis; these have been confronted many times before, and found seriously wanting. Rather, my goal is to think more about the structure and stated aims of those papers, and the ways in which the papers fail to take seriously their responsibility to honestly present their own arguments, and to honestly engage with the relevant literature.

The first paper that I’ll address (Winegard, Winegard and Anomaly’s “Dodging Darwin”) is, unlike the other two, at least honest about its goal – providing a defense of hereditarianism. So that’s something! Its main problem, in my view, is that it does this through the strained lens of evoking Darwin (as opposed to, for example, evoking modern evolutionary biology in all its richness). The paper appeals to some vague “Darwinian Research Program” and pretends that the kind of work in which the proponents of the hereditarian hypothesis are engaged is relevantly similar to work done by e.g. evolutionary biologists and ecologists, when it simply is not. To make their own work seem more like the work done by serious biologists, the paper fails to adequately represent the latter, pretending that it shares the weaknesses of their own work, when it does not. Along with this general complaint – the paper misrepresents the work of both hereditarians and serious biologists, pretending that these are more alike than they really are – I argue that the paper’s treatment of some of the literature that it cites to be at best sloppy.

Part of my complaint here is, I must admit, personal. When I first saw this paper, and saw that I was cited in it, I assumed that I would be cited in my rather long-standing role as a critic of hereditarian hypotheses. Instead, I found that one of my co-authored papers had been cited as defending a claim that is at best quite distinct from, and at worst the opposite of, the claim we in fact defended. A quick perusal finds other places where what the authors claim other research says, and what it actually says, differ in important ways. Given the author’s cavalier treatment of evolutionary biology and ecology, it should perhaps not be surprising that they treat the literature they cite with a similar lack of seriousness.

In later posts, I’ll address a newer paper by Cofnas,¹ and a somewhat older paper by Anomaly. In both cases, I want to suggest that what the papers claim to be doing, and what they are actually doing, diverge, sometimes quite radically. This divergence, I want to suggest, hints at an ugly intellectual dishonesty at the heart of this work.

Part of the reason for this divergence is perhaps obvious, on reflection. What all of these authors are really doing is arguing for hereditarian claims, on the basis of the same old widely discredited evidence (discredited at least in the sense that it cannot support the claims made). But there isn’t really a point to making those arguments now – certainly, there is nothing much in these pieces that wasn’t in Rushton and Jensen’s 2005 “review” of the literature. So they need to find an angle – a way to get a paper published that was at least pretending to not just rehash the same old arguments. Their goal is not necessarily to convince anyone, but rather to make these claims seem more plausible and therefore more acceptable to endorse than a reasonable assessment would make them.

As noted above, the first paper I will address in these posts is a bit different from the other two, in that “Dodging Darwin: Race, evolution, and the hereditarian hypothesis” is at least upfront in its claim that it will be defending some version of the hereditarian hypothesis. In the abstract, the authors acknowledge that there is not “decisive evidence about the causes of differences in cognitive ability,” however they claim that “a partial genetic hypothesis is most consistent with the Darwinian research tradition” (p.1). Here, again, the actual evidence that the paper cites in support of its thesis is fundamentally misguided, and the argument given is, as usual for hereditarian pieces, pretty shoddy. But while the paper is at least honest about its fundamental goals, it fails to honestly pursue those goals, opting instead for a variety of shortcuts that end up simply misrepresenting other people’s research and research projects.

Indeed, it was a gap between what Winegard et. al. claim a paper said, and what it actually said, that first drew my attention to this paper. As I note above, they cite my co-authored article “Prisoners of Abstraction? The Theory and Measure of Genetic Variation, and the Very Concept of ‘Race’” (with Rasmus Grønfeldt Winther) in support of the claim that “Since the 1950s, many intellectuals have assailed the concept of race, arguing that it is a dangerous fiction and a deeply deceptive construct that creates an illusion of patterned variation that just does not match reality.” Now, while I suppose that I happen to believe that something like this claim is true, I can see no way of getting that claim out of the paper of ours that they cite. Indeed, the explicit thesis that our paper defends is almost the opposite of this. Our abstract starts with our primary claim: “It is illegitimate to read any ontology about ‘race’ off of biological theory or data” (emphasis in original). Our point in this paper was not in fact to defend any particular take on the existence of (biologically real) human races, but rather to explore why using biological data to try to settle questions about the existence of race was so fiendishly difficult, and why different people were able to make use of biological facts (often the same biological facts!) in such radically different ways (some to argue in favor of biological racial realism, and some to argue against it). In our paper, we note that there are many different ways of thinking about within and between population genetic variation, and that it was important to distinguish carefully between measures of genetic diversity, measures of genetic differentiation, and measures of heterozygosity, as these in fact measured different things, and were best deployed for different purposes (none of which, in any obvious sense, is to determine whether ‘race is really biologically real’). How this got translated into the claim that “race… is a dangerous fiction and a deeply deceptive construct” is frankly beyond me.

Now, one might claim that my work with Rasmus is just difficult to interpret (or easy to misinterpret); after all, in a recent paper Holly Dunsworth and her co-authors (“Human races are not like dog breeds: refuting a racist analogy,” 2019) claim that another paper in that series (see here and here) suggests that “clustered human variation demonstrates the reality of a biological concept of race,” (14), which is, again, pretty much the exact opposite of the point that Rasmus and I were trying to get across. So maybe this is in part my own fault, and not the fault of Winegard et. al. But there are other papers where it is harder to see the gap between what they claim the paper says and what it says as a simple misunderstanding. For example, it is true that of the several papers the authors cite are supporting the “harsh conditions select for intelligence as an adaptation” in fact support a thesis at least like that. One, however, instead suggests that more intelligent species are better able to colonize harsh environments, not that harsh environments select for more intelligence within a population that has already colonized it. These are not entirely unrelated claims, of course, and the former might provide some evidence for the latter, but they are not the same claim, either, and to cite a paper defending the former as if it were straightforwardly defending the latter is at least sloppy.

More serious are the claims made that are more or less entirely unsupported by contemporary research, and seem wildly out of touch with reality. For example, one passage from the paper which has been shared extensively on social media reads:

Nevertheless, even if discrimination against Blacks in some countries is widespread, this would not explain IQ and achievement gaps between ethnically homogenous countries, such as Iceland and Haiti, or between different demographic groups within multi-ethnic countries in which racial discrimination is not widespread, such as the UK, France, or Sweden.

This passage is remarkable both for its complete lack of citations or references, and for its quite astonishing ignorance about the actual state of the world. So for example, the literature surrounding anti-Black racism in France, and its relationship to French colonial policy, is extensive (see here, here, or here). And there is a significant literature on the ways in which French racism play out in via discrimination in e.g. the labor and housing markets. Sweden, along the great things it is justifiably famous for, is also quite famous for the amount of racism it harbors, which includes serious and wide-spread anti-Black racism (see here, here, here, here, or here) It is remarkable that not one of the three authors knew or suspected that Sweden had problems with racism! There seems to be some debate about whether racism against Blacks has reduced, increased, or stayed about the same in recent years in the U.K., but there is no serious researcher who would suggest that it isn’t still both very serious and more or less ubiquitous (here, here, here,or here) .

The claims made by Winegard, Winegard, and Anomaly are exactly the sort about which I complained in my “Ignorance, Lies, and Ways of Being a Racist” article – claims so out of touch with reality that they are either meant to deliberately mislead readers (lies), or that reveal a level of willful ignorance that is best explained by viewing the world through a lens deeply colored by racism. (Their assertion that discrimination against blacks in the U.S. is not “as pervasive” as usually claimed also demands ignoring the vast body of published research, on discrimination in hiring, housing, policing, etc.)

This brings us to the paper’s evocation of Darwin and the “Darwinian Research Tradition;” there is a profound dishonesty lurking in the way that this term is used, and what its use hides. It is worth noting right at the start that the idea of there being a “Darwinian Research Tradition” in the sense of a set of contemporary research practices that actual evolutionary biologists deploy is odd at best. The phrase is sometimes used, but usually in historical / philosophical contexts; no one working on e.g. the kinds of local physiological adaptations to which Winegard, Winegard, and Anomaly compare their hypothesized psychological adaptions explicitly refers to their work as being within a “Darwinian Research Tradition,” probably because that’s not the way that practicing biologists think about their work. Rather, most practicing biologists think of their work as situated within biology, and those who make use of evolutionary theory explicitly think of their work as situated with evolutionary biology (and related fields like ecology, ethology, etc.). Darwin, for all the brilliance of his insights and the evidence he gathered to support his views, published Origin over 160 years ago. Contemporary biology is, of course, much richer and more complex than anything Darwin envisioned.

While the authors evoke Darwin’s name repeatedly, and refer to the “Darwinian Research Tradition” in numerous places, no where do they actually say what they mean. They have a longish quote from Darwin which they claim supports the view that Darwin at least considered the possibility that “recent evolution might have produced cognitive differences that help explain the relative prosperity of different societies,” (5). The section in question is read more naturally as being focused on what Darwin supposed were the “dysgenic” tendencies of some successful societies, but Darwin wrote enough in The Descent of Man about the ways in which selection could modify the average intelligence of populations that despite the poor choice of supporting text, there is no real doubt that Darwin in fact argued things like this.

Speculative claims made by Darwin, with all the usual attendant historically situated racism, aside, the main use to which the authors put the “Darwinian” label is to criticize the study of humans as incompletely Darwinian, because while a number of physical traits that vary in frequency between human populations are given adaptive explanations, psychological traits that they suppose also so-vary are not. But there is a key element here that the authors either deliberately ignore or actively try to hide. Critically, the examples of relatively local adaptations (human ecotypes) that most biologists find compelling are not “Darwinian” in the sense of marshaling (only) the kinds of evidence that would have made sense to Darwin. The examples that biologists in fact find compelling are those that are engaged fully with modern evolutionary (and molecular) biology and ecology. In none of the examples of purported physical adaptations that the authors bring up is the only evidence a difference in realized populations averages and an adaptive story; indeed, the less evidence of other sorts that there is, the more those cases are thought of as “speculative.” There is, in other words, a rich research tradition the goal of which is to understand the adaptive significance of traits in local populations; where the tools of this research project have been successfully brought to bear on human adaptations, researchers can be and often are convinced that a particular adaptive story is at the very least on the right track. But the fact that putative psychological adaptations in local human populations are not accepted has little to do with resistance to hypothesizing about psychological differences, and everything to do with the lack of relevant evidence.

At one extreme, take the authors’ example of skin color. Skin color might seem a pretty straightforwardly physical trait, but the genetics turn out to be rather complicated (see here or here). Adaptive stories involving a trade-off between vitamin-D production and sunburn are common, but are likely only a part of a more complicated picture. Skin color variation within the same latitudinal region is significant, and cannot be easily explained under the simple adaptive account. The authors implicitly accept this uncertainty, but do not seem to realize the significance of it. If we are unable to fully account for variation in skin color in local populations, where the developmental biology and genetics are at least relatively straightforward, what hope have we for accounting for e.g. differences in how “collectivistic,” on average, different human populations are? The latter trait cannot even be straightforwardly measured, the genetics are almost entirely unknown (though I suppose that there is probably some relationship between whatever the authors mean by having “collectivistic” tendencies and “agreeableness,” which shows the typical massively polygenic pattern with very weak associations that is pretty standard for personality traits, but even that relationship is hardly straightforward), and development is entirely obscure. And yet the authors feel comfortable spinning tales about the ways in which local adaptive forces might have resulted in a situation in which “Northeast Asians are more collectivistic than other populations” (4).

At the other extreme, take their example of lactase persistence. The contemporary understanding of the distribution and evolution of lactase persistence is a model of the kind of superb work that evolutionary biologists are capable of. Winegard Winegard and Anomaly write that:

For example, after humans began to domesticate animals, some dairying populations evolved the ability to digest lactose after childhood (lactase persistence). This is why the distribution of lactase persistence across the globe varies from population to population (Gerbault et al., 2011).

This is true as far as it goes, but notice the way in which this two sentence summary hides the wealth of evidence that went into forming these conclusions. The one article the authors cite here, “Evolution of lactase persistence: an example of human niche construction,” summarizes only a tiny fraction of that research, but even so it includes genetics, archaeological research, cultural anthropology, simulation studies, and more. To take one small example, that article confronts the question of whether populations with a higher percentage of lactase persistent individuals took up milk use because more people could digest it, or whether most populations were similar with respect to the frequency of lactase persistence before milk production, and only those that took it up selected for that trait; it argues that evidence from archaeology and population-genetics points towards the latter. But more generally, each of these areas (molecular genetics, population genetics, archaeology, cultural anthropology, developmental biology, chemistry, nutrition, but also research on gut biomes and epigenetic inheritance, and many more besides) represents a serious research project; since some of the earlier work on this topic in the late 1960s, literally thousands of articles have explored the geographic distribution of the trait, the archaeological evidence regarding when those populations developed a pastoral lifestyle and how their diets changed during those periods, the genetics underlying lactase persistence, the nutritional implications of being able to digest lactose including the caloric loss from converting lactose-rich milk into food-stuffs (cheese, yogurt) with reduced lactose, etc.

There is, quite obviously, nothing like this at all for any human psychological traits at all. In some ways, this isn’t the fault of the hereditarians; psychological traits are simply vastly more complex than traits like lactase persistence, high-attitude adaptations, malaria resistance, and the like. Both the genetics and development of psychological traits are obscenely complex, and the traits themselves are less easily individuated, operationalized, and measured. It is much easier with physical traits to know that a realized difference in population averages has a basis that can be traced clearly to differences in allele frequencies, and it is much easier to see how those differences are likely to manifest (or not) in different developmental environments.

So if we ask “why in humans do we know of a reasonable number of local adaptations in physical traits but not in psychological traits?” the answer is not “because researching into the latter is taboo” but rather “because the latter are more or less entirely inscrutable.” (On this topic, it is worth revisiting Lewontin’s 1998 chapter, “The Evolution of Cognition: Questions We Will never Answer.” Even if one believe that Lewontin is overly pessimistic in this piece, the challenges he puts forward ought to be taken very seriously indeed.)

This leads, I think, to one of the fundamental dishonesty at the heart of the hereditarian position. When proponents pretend that the evidence they have gathered, and the adaptive stories that they are telling, are similar to those in the case of lactase persistence, the HbS allele, etc., they are simply being dishonest. Hereditarians provide no evidence remotely similar in quality or quantity to that provided by biologists studying the bases of those local human adaptations that are physical, and the kinds of evidence that they do provide are utterly unlike the kinds of evidence that are required to support claims about ecotypic variations more generally.

When Jensen reinvigorated the hereditarian position in 1969², he had, to a first approximation, no evidence at all for his position; his argument, basically, was that he wasn’t convinced that the average developmental environment of Black Americans, in the 1940s, 1950s and 1960s, was so much worse than the average environment of White Americans, that the observed difference in average IQ scores between Black Americans and White Americans could be explained environmentally. Those that defend Jensen against the accusation that he naively assumed that within-population heritability implied between-population genetic differences by noting that he appealed to plausible differences in developmental environments are “right” – he did that. But once put this way – that in the 1940s, 1950s and 1960s, Black Americans’ developmental environment wasn’t that much worse, on average, than those of Whites’ – the pure obscenity of the argument becomes obvious.

¹ Egocentric editorial footnote by jpj: I’ve discussed Cofnas’s paper here.
² Another egocentric editorial footnote by jpj: I’ve discussed Jensen here.

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.