Human Differences

Following the Evidence: Human Genetic Diversity, Intelligence, and the Need for Honest Science

A simplified version of this article can be found here.

---

Human biological variation and behavior have long been contentious topics. In academia, a prevailing ideological resistance has often stymied objective scientific discussion about human taxonomy (the classification of human populations) and behavioral genetics (the genetic basis of traits like intelligence). This report conducts a rigorous analysis of two interrelated issues: (1) the reluctance to apply standard taxonomic classifications (such as subspecies or even phylogenetic “species”) to human populations despite genetic evidence, and (2) the suppression or misrepresentation of behavioral genetic research (especially on cognitive abilities) due to political or social sensitivities. We examine how measures of genetic differentiation like Jost’s D provide insight into human population differences, and compare human genetic distances to those separating recognized species/subspecies in other mammals. We then review key findings from twin studies, adoption studies, polygenic score research, and other approaches in psychometrics – showing the strong genetic contribution to intelligence and group differences – and how these findings have repeatedly confirmed predictions made by hereditarians (as in The Bell Curve), even as critics (like Stephen Jay Gould in The Mismeasure of Man) were later found flawed. Next, we document case studies of academic suppression: how influential figures such as Richard Lewontin, Stephen J. Gould, and others (e.g. James Flynn) allowed ideological beliefs to drive opposition to genetic research, leading to biases like the near-taboo on describing human populations in zoological terms. We draw historical parallels to other instances where ideology impeded science – from the refusal to see birds as reptiles due to emotional biases, to religious opposition against human evolution, to political doctrines (e.g. Lysenkoism) that rejected genetic explanations. In each case, we refute the standard counterarguments (such as “Lewontin’s fallacy” about within-group variation) that are used to dismiss human genetic differentiation, explaining why these rebuttals are misleading. Finally, we argue for the paramount importance of intellectual integrity in science: the evidence must lead us, wherever it goes, and only by fearlessly confronting facts can science and society progress. All claims are backed by peer-reviewed research and primary sources, with an unapologetic commitment to scientific objectivity over comfort.

1. Genetic Differentiation and Human Taxonomy Under the Phylogenetic Species Concept

Scientific taxonomy classifies living organisms into species and subspecies based on genetic and phenotypic differentiation. Humans, however, have been conventionally treated as a monotypic species (no subspecies) – largely for social reasons – even though our global populations exhibit measurable genetic structure. The Phylogenetic Species Concept (PSC) defines species as the smallest diagnosable monophyletic units (i.e. distinct evolutionary lineages). Under this concept, one could argue that certain human populations (e.g. continental groups) constitute distinct lineages. Indeed, an analysis by Woodley (2010) made “a case for the hypothesis that H. sapiens is polytypic” – meaning it contains multiple distinct population lineages – “and in this way is no different from other species exhibiting similar levels of genetic and morphological diversity”​ ^[1]. Woodley demonstrated that humans actually show high levels of morphological diversity, genetic heterozygosity, and intergroup differentiation (measured by \(F_{ST}\), a standard population genetic metric) compared to many animal species that are acknowledged to be polytypic (having subspecies) ^[1]. In other words, the degree of variation within Homo sapiens is not exceptionally low; on the contrary, it is on par with or greater than variation in many species that taxonomists routinely split into subspecies.

1.1 Measuring Genetic Differentiation: Jost’s D vs. \(F_{ST}\)

A key issue in evaluating human population differences is how genetic differentiation is measured. For decades, scientists cited Lewontin’s (1972) finding that about 85% of human genetic variation occurs within local populations, and only ~15% between major populations (“races”). This low between-group fraction (15%) led many to conclude that human groups are too genetically similar to merit taxonomic distinctions​ ^[2]. The metric underlying that inference is \(F_{ST}\) (and related statistics like \(G_{ST}\)), which measures the proportion of total genetic variance attributable to differences between populations. However, it is now recognized that \(F_{ST}\) can severely underestimate differentiation when overall genetic diversity is high. Jost (2008) introduced a statistic D (now called Jost’s D) specifically to address \(F_{ST}\)’s limitations​ ^[3]. The crux is that \(F_{ST}\) is constrained by within-population heterozygosity: if a species is very polymorphic (as humans are), even populations fixed for completely different alleles may show a moderate \(F_{ST}\) value well below 1.0​ ^[4]. For example, with highly variable markers (like microsatellites), the maximum possible \(F_{ST}\) might only be 0.2 even if two groups share no alleles​ ^{[5] [6]}. This is not because the groups are “not very different” genetically, but because the abundant internal variation depresses \(F_{ST}\). Jost’s D, by contrast, directly measures allelic differentiation: it calculates the fraction of genetic diversity that is unique to different populations. Unlike \(F_{ST}\), Jost’s D can reach 1.0 (100% differentiation) when populations share no alleles, regardless of within-group diversity​ ^[7]. In essence, \(F_{ST}\) reflects a relative measure (accounting for within-group variance), whereas D reflects an absolute measure of population differences​ ^[7]. As one population genetic guide puts it: “D is best suited for describing the allelic differentiation among populations, while \(F_{ST}\)…is better suited for demographic processes”​ ^[8]. Thus, Jost’s D is considered a more accurate gauge of genetic differentiation for our purposes, because it is not artifically deflated by high within-group variation. Using metrics like D can reveal that human populations are more genetically differentiated than the raw \(F_{ST}=0.15\) suggests. Indeed, even Lewontin’s original data, when interpreted properly, showed that one can almost perfectly classify individuals into their population of origin if one considers enough genetic loci simultaneously – a point we revisit in Section 5.1.

1.2 Human Genetic Distances vs. Other Mammals (Are Humans “More of the Same”?)

To objectively judge whether human populations qualify as subspecies (or even “phylogenetic species” in the PSC sense), we can compare genetic distances among human groups to those between recognized subspecies of other animals. The surprising truth is that humans are not unusually homogeneous; many animal species that are split into multiple taxa have equal or less genetic divergence than that found among major human lineages. For instance, the average \(F_{ST}\) among continental human populations is on the order of 0.15 (15%) for autosomal DNA​ ^[1]. This value lies in the range that zoologists consider “moderate” differentiation. In fact, there are animals with far lower between-group \(F_{ST}\) that nevertheless are divided into subspecies. A striking example: the Canadian lynx shows an \(F_{ST}\approx0.033\) between populations – only about 3% genetic differentiation – yet is classified into three subspecies ^[1]. The African buffalo has \(F_{ST}\approx0.059\) (~6% differentiation) and is split into five subspecies ^[1]. By comparison, humans’ between-group differentiation (≈15%) is substantially higher, but humans are officially recognized as having zero subspecies in current taxonomy. Table 2 of Woodley (2010) lists many such comparisons: e.g. chimpanzee populations have \(F_{ST}\) around 0.32 with 4 subspecies recognized; gorillas \(F_{ST}\approx0.38\) with 2 subspecies; African elephants ~\(0.17\) (\(\geq4\) subspecies)​ ^{[1] [1]}. Humans, at ~\(0.156\) on comparable analyses, fall in the upper-middle of the animal spectrum – yet we alone are often declared “monotypic”​^{[9] [1]}. In terms of heterozygosity (overall genetic diversity), humans actually have relatively low diversity compared to many animals (e.g. rhesus macaques have ~2.5× more genetic diversity than humans)​ ^[10]. Low overall diversity could indicate less structure, yet despite that, clear structure exists in our species. The human species consists of multiple genetically distinguishable clusters corresponding largely to geographic ancestry (Africa, Europe/Middle East, East Asia, Oceania, and the Americas being the broadest clusters identified by tools like STRUCTURE analysis​ ^[11]). These clusters are statistically robust: Rosenberg et al. (2002) famously showed that when analyzing multilocus genotypes, nearly all individuals could be assigned to their continent-of-ancestry cluster with high accuracy despite the modest \(F_{ST}\)​ ^[11]. Under the phylogenetic species concept, each of these major lineages can be seen as a distinct evolutionary unit – in fact, Woodley (2010) argues that “the least inclusive monophyletic units exist below the level of species within H. sapiens, indicating the existence of a number of potential human phylogenetic species”​ ^[1] (though under the biological species concept humans are defined as one species since all populations interbreed freely). The refusal to recognize any human subdivisions is thus inconsistent with standard zoological practice. As one zoologist bluntly observed, in non-human animals even “a few distinctive features” correlated with geography are “enough to call them different races (or even subspecies)… The same logic should be used to humans, right?”​ ^{[12] [13]}. Yet a “systematic bias” exists whereby scientists apply one standard to animal taxonomy and a far stricter standard to humans. This bias appears to be driven not by science, but by political and ethical concerns – a theme explored further in Section 3. For now, the key point is: by objective genetic criteria, human populations are sufficiently differentiated to be considered taxonomic subspecies or phylogenetic species, if we were to follow the precedent set in other mammalian species ^{[1] [14]}.

2. Behavioral Genetics, Intelligence, and Psychometrics: Nature’s Consistent Verdict

If human populations do show taxonomically significant genetic differentiation, an even more charged question is whether there are behavioral or cognitive differences influenced by genetics. Over the past century, a vast body of evidence from behavioral genetics – including twin studies, family studies, adoption studies (especially transracial adoptions), and most recently genome-wide polygenic scores – has converged on a consistent conclusion: individual differences in intelligence (cognitive ability) are substantially influenced by genes, and population-group differences in cognitive outcomes are partially genetic in origin. These empirical findings have withstood intense scrutiny and numerous attempts at purely environmental explanations. In fact, many of the core claims of Herrnstein and Murray’s The Bell Curve (1994) have been vindicated by later research, despite that book’s controversial reception. We summarize the major lines of evidence here, showing how they reinforce each other. Notably, these findings emerged despite a climate of resistance – scientists who produced unwelcome results were often attacked or marginalized. Yet the data, replicated across decades, countries, and methodologies, speak clearly.

2.1 Twin and Adoption Studies: High Heritability of IQ and Persistence of Group Differences

Classic twin studies compare identical (monozygotic) twins to fraternal (dizygotic) twins to estimate heritability – the proportion of variance in a trait due to genetic differences. Decades of twin research have consistently found intelligence to be one of the most heritable behavioral traits. In childhood, heritability of IQ might be moderate (~40-50%), but by adulthood it rises to ~75% or higher in many studies (as genetic potential fully manifests)​ ^[15]​ ^[15]. A review of “dozens of twin, adoption, and family studies” confirms the “high heritability of intellectual and behavioral traits” within populations​ ^[16]. This means that in a given population, the majority of variation in IQ is due to genetic differences among individuals. Moreover, the influence of the shared family environment on IQ is strong in early childhood but fades by late adolescence, often becoming minimal by adulthood​ ^[16].

It is crucial to understand what high heritability implies – and what it doesn’t. A high heritability of IQ within a group tells us that genetic differences substantially influence individual differences in that group’s IQ. It does not by itself prove genetic causes of average differences between groups. However, it sets an important context: if genes strongly affect IQ, and if groups differ in relevant gene frequencies, then genetic influences could contribute to group differences too. Adoption studies provide a powerful test of nature vs. nurture by examining children of one ancestry raised in families of a different ancestry. The most informative are transracial adoption studies. These have consistently shown that while enriched environments can boost children’s IQ somewhat, they do not erase inborn group differences. As noted by behavioral geneticists, transracial adoptions represent a “massive intervention” that should equalize outcomes if environment were solely decisive​ ^{[16] [16]}. Yet results show this is not the case.

One famous example, the Minnesota Transracial Adoption Study, followed black, white, and mixed-race children adopted into white upper-middle-class families. At age 7, the adopted black children had IQs somewhat below the adopted white children; by age 17, a clear gradient emerged aligned with genetic ancestry. In the follow-up, the children with two white biological parents scored about 106 IQ on average; those with one black and one white biological parent around 99; and those with two black biological parents around 89 ^{[16] [16]}. All groups were raised in similar elite environments from infancy. The fact that the black adoptees still, by late adolescence, scored lower on average (despite early advantages and even some initial gains at age 7) strongly suggests a genetic component to the difference. Indeed, the black adoptees’ mean of 89 at age 17 was only slightly above the average for black teenagers in the general U.S. population (≈85-87)​ ^[16], illustrating how little lasting impact a privileged upbringing had on closing the gap. Meanwhile, studies of East Asian children adopted into Western families show the opposite: despite often poor starts (e.g. orphanage malnutrition), they tend to score above the host population mean in IQ as they grow up ^[16]. For example, Korean adoptees in the U.S. and Europe reach IQs noticeably higher than the white norm of 100, even if as infants they were undernourished ^[16]. This again indicates that genetic lineage, not just upbringing, shapes cognitive outcomes.

Another revealing phenomenon is regression to the mean in mixed ancestry individuals. In the Minnesota study, children with mixed black-white ancestry scored intermediate to the two parental groups, as expected under genetic inheritance​ ^[16]. Such patterns are hard to explain except via heredity. Overall, the evidence from adoption and twin studies converges: by late adolescence, shared environmental advantages (like being raised in a wealthier, educated family) have only a limited effect on IQ, whereas genetic influences are potent and lasting​ ^{[16] [16]}. As researchers summarized, by adulthood the between-family variance in IQ is tiny compared to within-family (genetic) variance, meaning that an adoptee tends to resemble their biological parents/siblings in IQ more than their adoptive ones​ ^[16]. No serious scientist today doubts that IQ is substantially heritable; the debate (to the extent it exists) is about the causes of group differences, not the existence of genetic influence itself.

2.2 Polygenic Scores and Genomics: DNA Confirms the Genetic Basis of Cognitive Differences

The genomic era has provided direct confirmation of the earlier behavioral genetic findings. Genome-wide association studies (GWAS) of educational attainment and intelligence, using hundreds of thousands of individuals, have identified many specific genetic variants associated with cognitive ability. While each variant has tiny effects, their effects can be summed into a polygenic score that predicts a person’s propensity for higher or lower intelligence. As of 2018, with large sample sizes, scientists could account for about 20% of the variance in intelligence using DNA alone ^[17]. In other words, purely from analyzing a person’s genome (and without knowing anything else), researchers could predict ~20% of the differences in IQ – a remarkable achievement that directly validates the earlier heritability estimates (since ~50% of IQ variance is genetic in total, and 20% has already been identified from common DNA markers)​ ^[17]. The predictive power of polygenic scores continues to improve as sample sizes grow; recent scores for educational achievement, for example, rival traditional socioeconomic factors in predicting life outcomes. The take-home message is that the molecular genetic evidence now unequivocally shows a biological basis for cognitive ability differences. As Plomin (2018) notes, intelligence is “highly heritable” and thousands of DNA variants of small effect contribute to it, such that aggregated genome-wide scores can significantly predict intelligence and academic success​ ^{[17] [18]}. This directly refutes past claims that IQ differences were somehow artifacts of test bias or could not possibly be “in the genes.”

Furthermore, genomic studies have started to examine whether allele frequency differences across human populations affect traits like IQ. While this is a nascent and controversial area, initial findings indicate that the average polygenic scores for educational attainment/IQ do differ by continental ancestry (with East Asian and European populations tending to have higher average scores than African populations, in line with group IQ patterns) – though research is ongoing and must control for assortative mating and other factors. What’s critical is that if such differences exist, they provide a biological mechanism that could contribute to the well-documented group differences in cognitive performance. This is precisely what hereditarians hypothesized decades ago, and now the genomic data are starting to support it.

2.3 Long-Term Outcome of Environmental Interventions: The “Fadeout” Effect

Perhaps the most sobering evidence for the primacy of genetics comes from the failure of intensive environmental interventions to produce lasting IQ gains. If intellectual disparities were easily malleable by improving environments, early childhood education programs and other interventions should permanently boost IQ and eliminate gaps. Numerous large-scale programs (Head Start, the Perry Preschool Project, Abecedarian, etc.) have been attempted. Often, initial IQ gains are seen in children while the program is ongoing. But these gains almost always fade out in the years after the intervention ends. A comprehensive meta-analysis by Protzko (2015) examined 44 randomized controlled trials of early childhood interventions aimed at raising IQ. The conclusion: any boosts to IQ were temporary. Once the special intervention ceased, children “lost their IQ gains” – essentially converging back to where they would have been without treatment​ ^[19]. The experimental group’s advantage decayed rather than the control group catching up, demonstrating that the initial improvements were not sustainable​ ^[19]. By elementary school age, the IQ scores of the treated children were indistinguishable from controls in most studies. This well-replicated “fadeout effect” shows that early environment can speed up cognitive development, but it cannot permanently alter the trajectory set by genes ^[20]. As one researcher put it, interventions like Head Start give a short-term boost, “but when you take that away and put [children] back in with everybody else, they’re going to adapt to that new system” and lose the edge​ ^[20]. These findings are important because they mirror what behavioral genetics predicts: moderate improvements are possible, but the strong pull of genetics reasserts itself over time, re-establishing prior differences. This also aligns with the observation that the heritability of IQ increases with age – genetic effects become more expressed as a child grows up, eventually overshadowing early environmental boosts​ ^{[15] [15]}. The inability of intensive, costly interventions to eliminate cognitive gaps (despite often substantially improving school readiness and other non-IQ outcomes) reinforces the conclusion that cognitive inequalities have deep genetic roots that are not easily mitigated by environment.

2.4 The Bell Curve Thesis Revisited: Enduring Validity and Misconceptions

When The Bell Curve was published in 1994, its authors were excoriated for arguing that (a) IQ is a real, important predictor of life outcomes; (b) IQ is highly heritable; (c) social stratification by cognitive ability was increasing; and (d) group differences in IQ likely have a genetic component. A quarter century later, every one of those propositions is supported by mainstream research. IQ’s predictive power for education, income, job performance, and even health is now beyond question (it is one of the single best predictors of many life outcomes)​ ^{[18] [21]}. The paradigm of “meritocratic cognitive stratification” that Herrnstein and Murray described is plainly visible in modern societies. As we saw, the genetic basis of IQ is firmly established – the debate has shifted to how much genes contribute to gaps, not whether they do. Notably, large-scale studies have found that when you control for IQ, many disparities in social outcomes between racial groups shrink (though not disappear), consistent with IQ being a mediating factor​ ^{[22] [23]}.

One often-misrepresented aspect of The Bell Curve is that it did not claim certainty about the genetic portion of the Black-White IQ gap; it said genes were a plausible hypothesis. Since then, evidence (from transracial adoption, genomic data, etc.) has if anything made the genetic hypothesis more plausible, not less. Even James Flynn (for whom the Flynn effect is named), originally a strong proponent of purely environmental explanations, conceded later in his career that he could not rule out some genetic contribution to group IQ differences. The “Flynn effect” – rising raw IQ scores over the 20th century – is often cited to argue environment can change IQ. But Flynn himself noted that these gains likely reflect improved test-taking and abstract problem-solving, not increases in real cognitive ability relevant to group differences (which show a different pattern). In fact, the Flynn effect has decelerated or reversed in some developed countries recently, suggesting it was tied to modernization factors that eventually plateau. What remains are persistent differences that are much more resistant to change.

In short, the core message of The Bell Curve – that cognitive ability is a meaningful, genetically influenced trait that profoundly shapes individual and societal outcomes – has been powerfully validated. Critics often mischaracterized the book as claiming genetic determinism or as advocating social Darwinism. In reality, the authors called for compassionate policy responses given the biological reality of differences, much like how we accommodate any inherent differences among people. Unfortunately, much of the critique (e.g., Stephen Jay Gould’s) was driven by political aversion rather than fair appraisal (more on Gould in the next section). Modern psychometric consensus affirms that general intelligence (“g”) is a real, measurable construct and that tests are not biased against any group in predicting outcomes – they measure genuine differences​ ^[24]. Massive empirical efforts, such as an APA task force report in 1996, corroborated that IQ tests predict academic and work performance across racial groups equally well and that no one has found an environmental intervention that can eliminate the racial IQ gap in a lasting way​ ^{[25] [19]}. In sum, the scientific evidence consistently supports a strong genetic basis for intelligence and its variation – both within and between groups – despite longstanding ideological resistance to these findings. The next sections will detail how and why that resistance arose, and how it has distorted scientific discourse.

3. Academic Suppression and Ideological Resistance in Science

The study of human differences has often been politicized, with some academics actively working to discredit or suppress research that conflicts with egalitarian or anti-racist ideals. In the 20th century, a group of influential scientists – notably geneticist Richard Lewontin, paleontologist Stephen J. Gould, psychologist Leon Kamin, anthropologist Ashley Montagu, and others – took strong public stances against the hereditarian view. Their motivations were explicitly ideological: many were Marxists or left-wing activists who believed that emphasizing genetic differences would fuel racism or social inequality. As a result, they sometimes misrepresented data, attacked the integrity of researchers, and imposed double standards to shut down lines of inquiry they found objectionable. This section examines case studies of such suppression and resistance, along with the personal ideological motivations behind key figures. We also discuss the systematic bias against acknowledging human subspecies or races in taxonomy – a bias that stems more from political correctness than from scientific evidence.

3.1 Lewontin, Gould, and the Politicization of Science

Richard C. Lewontin was not only a population geneticist but also an avowed Marxist who saw his scientific work as intertwined with political struggle. In 1972, Lewontin published the famous paper claiming the insignificance of racial differences (the 85% within vs 15% between variation statistic). It’s now known that his interpretation was flawed (see Section 5.1), but it became a rallying cry that “race is just a social construct” with “virtually no genetic or taxonomic significance”​ ^{[2] [2]}. Why was Lewontin so adamant? Historical analyses indicate that he viewed hereditarian research as a dangerous attack on equality. In the late 1960s, as the civil rights movement made gains, Lewontin felt these gains were fragile and under threat from scientists like Arthur Jensen (who in 1969 suggested genetic involvement in IQ differences)​ ^[26]. Partly out of a sense of duty “because of the social and political stakes,” Lewontin launched “withering attacks” on Jensen and Herrnstein – not just disputing their data but impugning their motives and attempting to undermine their authority ^{[26] [27]}. Historian of science Segerstrale documents how Lewontin, Kamin, and Gould formed an “anti-hereditarian” coalition that used both academic critiques and popular writing to discredit any research that hinted at biological differences in behavior.

This politicization wasn’t subtle. E.O. Wilson, the Harvard biologist who founded sociobiology, was himself a liberal but became a target of Lewontin and Gould because his work implied genetic bases for social behaviors. When Wilson published Sociobiology (1975), Gould and Lewontin famously wrote that it was paving the way for racist and sexist ideologies. Wilson later pointed out that Lewontin’s fervent opposition was rooted in Lewontin’s “hardline Marxist” beliefs – i.e. Lewontin’s ideology was guiding his claims, rather than empirical science ^[28]. Indeed, Wilson was taken aback that Lewontin would publicly vilify a colleague’s work on essentially political grounds, accusing Wilson of promoting genetic determinism as a form of right-wing propaganda. Lewontin and colleagues went so far as to organize the infamous letter in the New York Review of Books (1975) denouncing Wilson, and during a scientific conference, protesters (including fellow scientists) poured a pitcher of water over Wilson’s head while chanting that he was promoting Nazi-like science. This is an extreme example of academic thuggery, where intellectual debate was supplanted by ad hominem attacks and intimidation – all because Wilson’s data on animals (and tentative extension to humans) conflicted with Marxist egalitarian doctrine. Lewontin’s behavior was driven by his political commitment; as Wilson and others noted, Lewontin admitted his Marxist framework influenced his science​ ^[28].

Stephen Jay Gould likewise mixed science with ideology. Gould was a brilliant writer and popularizer of science, but when it came to human abilities, he was a staunch egalitarian who refused to accept that any group could be biologically “better” at anything. His 1981 book The Mismeasure of Man is a scathing critique of intelligence testing and hereditarian theory. However, many of Gould’s arguments in that book have since been debunked. A telling example is Gould’s attack on 19th-century scientist Samuel Morton, who had measured skull sizes of different races. Gould accused Morton of systematically biasing measurements to fit racist preconceptions – essentially scientific fraud. This accusation stood for decades as a prime example of “unconscious bias” in research. But modern re-examination proved Gould wrong. A 2011 peer-reviewed study re-measured Morton’s skull collection and carefully reviewed Morton’s data and Gould’s claims. The results “resolve[d] this historical controversy, demonstrating that Morton did not manipulate data to support his preconceptions, contra Gould”​ ^[29]. In fact, Morton’s data were basically sound; it was Gould who, perhaps driven by his assumption that all such research must be biased, made errors and cooked his analysis to accuse Morton unjustly ^{[30] [31]}. The PLOS Biology paper concluded that the Morton case actually shows science’s objectivity prevailing over bias, and that Gould’s harsh allegation was unwarranted​ ^[29]. This revelation was quite shocking: a revered critic of scientific racism had himself fallen victim to confirmation bias, seeing what he wanted to see.

More broadly, Gould’s Mismeasure of Man took aim at IQ tests and the concept of general intelligence “g”, arguing they were nothing but reifications used to justify social hierarchies. Yet almost every major claim Gould made has been refuted by experts. For instance, Gould argued that factor analysis (the statistical method yielding “g”) was a subjective sham; in reality, decades of research have shown g is an extremely robust phenomenon predicting many real-world criteria​ ^[24]. Gould cherry-picked historical cases of scientists with racist ideas to insinuate that the entire field of mental testing was corrupt. It was a classic example of ideological motive driving scientific critique – Gould felt that if intelligence was largely genetic and differed between people, it would undermine his leftist humanist ideals, so he fought that conclusion by attacking the science, often unfairly. As one review summarized, Gould’s critique “contained errors of its own” and “his criticisms were largely without merit” on the technical points​ ^[32]. He was a passionate advocate, but not an objective assessor, and his work misled many readers into dismissing solid scientific findings about IQ.

Both Lewontin and Gould enjoyed considerable media influence and used it to sway public opinion. They framed themselves as defenders of truth and justice against racist “pseudoscience,” which made it difficult for nuanced scientific discussion to take place. Scholars who found evidence of genetic differences were pigeonholed as closet racists or reactionaries. This had a chilling effect: self-censorship became common. Scientists knew that certain results (e.g. on racial IQ differences or hereditary personality traits) would invite career-damaging controversy, so many avoided those topics. Research funding in these areas dried up; for example, after Jensen’s 1969 article on IQ, there was effectively a ban (formal or informal) by funding agencies on studies of race and IQ. Journals became extremely reluctant to publish anything that could be construed (often willfully) as supporting “genetic determinism.” The result was a de facto suppression – not through government censorship, but through academic taboos and fear of reputational damage.

3.2 Ideological Motives and Bias – The Cases of Flynn and Others

Another influential figure was James R. Flynn, who discovered the Flynn effect (rising IQ scores over time). Flynn was deeply concerned about racial injustice and initially hoped that environmental improvements could explain away the Black-White IQ gap. He corresponded with Arthur Jensen and set out to prove Jensen wrong. While Flynn’s work on secular IQ gains was valuable, he too approached the race-IQ debate with a strong prior belief that genes must not be involved. For many years, Flynn argued the gap was entirely environmental, citing phenomena like his namesake effect. However, to Flynn’s credit, he was more open-minded than Gould or Lewontin; in his later years, after reviewing all the evidence, Flynn admitted that a genetic component to group differences could no longer be excluded. In a 2012 interview, Flynn even said if pressed he would “have to say ‘more likely than not’ some portion of the racial IQ gap is genetic” (paraphrasing). This was a startling turnaround, and it highlights how compelling the data had become – even a longtime environmentalist was changing his mind. Nonetheless, during the decades when Flynn’s environmental arguments were celebrated, there was a tendency in academia to seize on any possible environmental explanation (no matter how speculative) and treat it as proven, simply because it was ideologically palatable. On the other hand, any genetic explanation (no matter how well supported) was treated with extreme skepticism or outright hostility. This asymmetric scrutiny is itself a bias. The motivations of figures like Flynn were often ostensibly “well intentioned” (reducing racism), but it led to confirmation bias: they accepted weak environmental arguments uncritically while demanding impossible proof for any genetic argument.

In the social sciences more broadly, numerous scholars promulgated the idea that “race is a social construct” and that human biological variation is trivial. The American Anthropological Association (AAA) issued statements essentially denying the existence of biological races in humans. These were ideologically driven pronouncements, not reflective of consensus in human biology. In fact, surveys show a split between biologists and social scientists. As Woodley (2010) notes, in a survey of 1,200 academics, only 16% of biologists disagreed with the statement “There are biological races in H. sapiens,” whereas 53% of sociocultural anthropologists disagreed​ ^[33]. This stark difference points to the role of ideology: anthropologists (especially in the late 20th century) were heavily influenced by Franz Boas and others who made it almost a dogma that human races are fictitious. Many anthropologists viewed any concession to biological differences as lending support to racism, so they took an absolutist “no biological races” stance. This is a textbook case of Normative Ideology intruding into a scientific question. As Woodley observes, the “problem with social constructivism [in this context] is that it attempts to engage racial classification on a normative rather than a scientific level,” essentially arguing that because the concept of race has been misused or is associated with racist ideology, it must be scientifically meaningless​ ^[34]. That is an appeal to consequences or appeal to motive fallacy: it doesn’t actually refute the biological reality, it just says “we don’t like where this might lead, so we’ll declare it invalid”​ ^[34]. Such reasoning held sway in academia for many years, heavily coloring how textbooks and courses presented human variation (usually downplaying or denying genetic structuring).

To summarize, the latter half of the 20th century saw a concerted effort by politically motivated scholars to suppress or distort research on human differences. Sometimes this was done by rigorous critique (which is fair game in science), but in several famous instances it crossed into outright obstruction: data were ignored or misrepresented, researchers were slandered, and a climate of fear prevailed. The “standard social science model,” as psychologist Steven Pinker calls it, reigned – attributing human behavior almost entirely to environment and culture – and anyone challenging that was in for a rough ride. This context is essential for understanding why certain fallacies (like Lewontin’s) persisted uncorrected for so long, and why, even today, discussing human biological differences is often met with reflexive denial.

3.3 Bias Against Taxonomic Classification of Human Populations

One manifestation of ideological resistance is the refusal in scientific circles to apply the same taxonomic terms to human populations that are routinely applied to animal populations. As discussed in Section 1, by biological criteria humans are a polytypic species (multiple regional variants with partial reproductive isolation in our evolutionary past). Yet calling these variants “races” or “subspecies” is taboo. The rationale given is usually that human variation is clinal (gradual across geography) and individuals can’t be cleanly separated into discrete races. However, this rationale rings hollow because many species with clinal variation are still taxonomically subdivided. Real-world taxonomy often acknowledges fuzzy boundaries – subspecies in nature intergrade, yet taxonomists still find it useful to name subspecies for conservation and study. The near-universal rejection of any human subspecies classification is driven by political correctness, not empirical necessity. As one commentator dryly noted, “there is NO SCIENTIFIC REASON to say there are no human races, if we’re using zoological reasoning. (Unless we are not animals anymore!)”​ ^[13].

Case in point: ornithology recognizes about 2 subspecies per bird species on average​ ^[13]. Birds often have slight gradations in plumage or song across regions – yet those regional variants are named and studied. In humans, there are clear continent-level clusters (as genetic studies show), along with differences in frequencies of traits (physical and even some behavioral traits). Early physical anthropologists did label human subspecies (e.g. “Caucasoid,” “Mongoloid,” “Congoid” for Africans, etc.), but this fell out of favor entirely in the late 20th century under pressure from social views. Today, suggesting that “races” are equivalent to subspecies invites accusations of racism. The irony is that denying the existence of human races does not erase racism – it only muddles scientific clarity. A 1985 survey found most biologists did not buy the “race does not exist” claim, but sociologists and anthropologists (those most influenced by ideology) largely did​ ^[1]. The systematic bias in academia thus leans towards rejecting human taxonomic classification to signal opposition to racism, regardless of the biological evidence. In practice, this has led to awkward circumlocutions. Researchers speak of “population structure” or “continental groups” or use ad-hoc terms like “Eurasian ancestral components” – anything to avoid the taboo words “race” or “subspecies.” This linguistic evasion is a symptom of the larger issue: when ideological fears dictate how scientists are even allowed to describe reality, science loses its objectivity. As Woodley (2010) argued, the notion that human races are “not real” because they are socially sensitive is not a scientific argument but a moral one​ ^[1]. We can and should separate misuse of a concept from the empirical truth of it. Admitting the truth of a matter is not the same as committing harm. Unfortunately, the climate of suppression has often conflated the two.

In summary, political/ideological resistance in academia has: (a) cast undue doubt on the reality of human genetic differentiation (to the point of denying human races as a biological category), and (b) impeded research into genetic bases of behavioral differences. However, truth tends to resurface. As we saw, even some former skeptics (Flynn) have come around, and studies continue to appear (albeit sometimes outside the most mainstream journals) supporting the once-taboo ideas. The next section draws parallels between this episode in science history and other instances where prevailing ideologies fought scientific findings – ultimately unsuccessfully. Science, when done properly, is self-correcting and transcends ideological wishful thinking.

4. Parallels with Other Ideologically Suppressed Scientific Topics

The saga of human taxonomy and behavioral genetics is not unique. Throughout history, there have been many cases of scientific truths encountering fierce resistance because they clashed with entrenched beliefs or emotional comfort. Here we discuss a few instructive parallels: the reluctance to accept that birds are reptiles (a fact of evolutionary history) due to emotional bias in classification; the resistance to human evolution from religious quarters; and the extreme case of Lysenkoism in the Soviet Union, where genetic science was quashed for ideological reasons. These examples illustrate a common theme: reality remains true regardless of whether people want to accept it, and suppressing science for ideology’s sake only delays the inevitable (while often causing harm in the interim). They also reinforce why it’s vital to “follow the evidence wherever it leads,” as the ethos of science demands.

4.1 Birds as Reptiles: Emotional Resistance to Taxonomic Truth

Modern biology has revealed that birds evolved from theropod dinosaurs, and under cladistic classification (grouping by common ancestry), birds are a subset of reptiles. This is a straightforward implication of evolution: the traditional class “Reptilia” gave rise to birds, so unless one includes birds in Reptilia, the latter group is not a complete lineage. Today, many textbooks and museums reflect this, referring to the clade of birds + traditional reptiles (excluding mammals) as “Sauropsida” or simply stating that birds are reptiles in the cladistic sense​ ^[35]. However, initially there was considerable pushback against this idea – not from religious folks, but from some scientists and educators who found it counter-intuitive or upsetting to call birds “reptiles.” After all, birds are warm-blooded, feathered, and generally seen as beautiful and musical, whereas “reptile” conjures images of cold, scaly, sluggish creatures. This is an emotional bias rooted in human categorization: people didn’t like lumping together things that seemed so different and valued so differently. For a time, one could hear arguments like “birds are so unique they should remain their own class, we shouldn’t call them reptiles even if technically true.” Yet, as scientists adopted a more rigorous phylogenetic framework, this resistance gave way. The denial of birds’ reptilian nature was never based on evidence – everyone knew the fossil record and comparative anatomy linked birds with dinosaurs – but on a reluctance to change mental models and perhaps a bit of sentimentality. It’s a small parallel to human taxonomy: just as some recoil at labeling human groups as subspecies (because of negative connotations), some recoiled at labeling birds as reptiles (because it felt disrespectful to our beloved birds). In both cases, science had to cut through psychological bias. The emotional discomfort was not a valid scientific argument. Over time, the truth about birds became widely accepted and the terminology adjusted. One can draw hope from this: that similarly, the truth about human biological variation can eventually be discussed soberly once emotional reactions are set aside.

4.2 Resistance to Human Evolution: Religious and Ideological Opposition

When Charles Darwin proposed the theory of evolution by natural selection (1859) – and especially when he later suggested that humans shared common ancestry with apes (The Descent of Man, 1871) – there was an uproar from religious and ideological quarters. The idea that humans were not separately created but rather emerged from the same natural processes as animals was deeply offensive to many Victorian-era people. This resistance was fundamentally ideological: it stemmed from the prevailing religious belief in special creation and the immaterial soul. Accepting human evolution seemed to undermine the biblical creation story and humanity’s privileged status​ ^{[36] [37]}. For decades, anti-evolutionists attacked the theory on moral and theological grounds (often cloaking their critiques as scientific). In the famous 1925 Scopes “Monkey” Trial in Tennessee, a teacher was prosecuted for teaching human evolution, reflecting laws passed under religious pressure to ban such instruction. The opposition to evolution was explicitly about preserving an ideological worldview – in this case, a literal interpretation of Genesis and a belief in human exceptionalism. Scientists like Thomas Huxley (“Darwin’s bulldog”) and others had to fight not just scientific debates, but public opinion and religious influence, to get evolution accepted. Over time, the evidence became overwhelming and most educated people (and even many religious denominations) made peace with evolution. But even into the 21st century, surveys (especially in the U.S.) show a significant fraction of the population rejects human evolution due to religious beliefs​ ^{[38] [39]}. This ongoing conflict underscores how ideology can create a filter that makes people impervious to evidence. The evolution case parallels the race/IQ debate: in both, certain facts are resisted because they supposedly have “broader social implications” that some find dangerous​ ^[40]. In evolution’s case, opponents feared it would lead to atheism or nihilism (and some even argued it would erode morality). In the human biodiversity case, opponents fear it will lead to racism or eugenics. The common pattern is the belief that accepting a scientific fact will unleash evil social consequences – a consequentialist objection. Yet, as history shows, denying reality does not truly protect us from bad outcomes; it usually just leaves us less informed. It’s better to accept what is true and then shape ethical policies with that truth in mind.

4.3 Lysenkoism: A Cautionary Tale of Political Suppression of Genetics

Perhaps the most extreme example of ideology suppressing science is the case of Trofim Lysenko in the Soviet Union. In the 1930s-50s, Lysenko rose to power in Soviet biology by promising an agricultural revolution in line with communist ideology. Lysenko rejected Mendelian genetics – which was at that time a flourishing field – because it didn’t fit with Marxist dialectical philosophy. Mendelian genetics emphasized genes, inheritance, and limits to how quickly one could change organisms; Lysenko instead favored the idea (a form of Lamarckism) that environment could directly and heritably change organisms in unlimited ways, which resonated with the Marxist vision of malleability and utopian transformation. With Stalin’s backing, Lysenko effectively banned genetic research in the USSR. In 1948, classical genetics was officially proscribed; scientists who dissented were silenced, fired, even imprisoned or executed. This led to a 17-year annihilation of Soviet genetics research ^{[41] [42]}. Lysenko’s unscientific methods (such as planting seeds very close, “educating” plants through environmental conditioning, etc.) caused crop failures and set Soviet biology back decades​ ^{[43] [42]}. By the time Lysenko was discredited (mid-1960s), enormous damage had been done. “Lysenkoism” has since become a byword for the politicization of science and the dangers of state-enforced dogma. It was a case where the reigning ideology (communism) demanded a certain scientific outcome (unlimited biological egalitarianism and environmental primacy), and any evidence to the contrary (like genes mattering) was literally outlawed. The parallel to the Western experience with race and IQ is less direct (no one was jailed in the West for saying genes influence IQ), but the chilling effect and academic ostracism were very real in the West even if less extreme. In both scenarios, ideologically inconvenient research was stigmatized and halted. The Lysenko affair stands as a stark reminder that suppressing science for political ends can lead to disastrous consequences. The USSR not only failed to improve crops by denying genetics, but also fell behind scientifically. In the West, by suppressing open inquiry into human differences for so long, we similarly risked falling behind in understanding how nature and nurture truly work in shaping societal outcomes.

Lysenkoism also illustrates that reality is not optional. Despite propagandists proclaiming genetics a “bourgeois pseudoscience,” peasants still starved when crops failed. Similarly, denying genetic differences won’t make everyone equal in ability; it will just leave us puzzled why, despite equalizing environment ever more (as has been done in many countries), performance gaps persist or re-emerge.

4.4 Other Examples and the Lesson

There are other instances too: the initial rejection of plate tectonics in geology (some old-guard geologists ridiculed Wegener’s continental drift hypothesis for decades due to inertia and perhaps nationalistic bias), the opposition to the “germ theory” of disease by proponents of miasma theory (partly ideological in clinging to established wisdom), and so on. In each case, evidence eventually triumphed, but only after often bitter disputes. The common thread is ideological or psychological resistance – people have commitments (religious, political, emotional) that make certain ideas unthinkable, and so they fight those ideas, sometimes by disingenuous means. Science, ideally, should be an apolitical search for truth, but scientists are human and can fall prey to biases and external pressures. The best antidote is a culture of openness and debate, where ideas can be tested without fear or favor.

For the topic at hand – human taxonomy and behavioral genetics – the historical parallels give us confidence that truth will prevail in the long run. Already we see the edifice of denial cracking: new genetic data are harder to ignore, and younger scholars are more willing to discuss human variation without the 20th-century ideological baggage. The lesson from history is clear: suppressing the discussion or study of factual realities (whether evolution, genetics, or human differences) due to ideological anxiety is both intellectually dishonest and ultimately futile. The truth does not disappear because one refuses to see it. Birds were always reptiles by descent, even when textbooks said otherwise; humans have always been an evolved species with biological variation, even when anthropologists pretended otherwise. Facing reality squarely is not only more honest – it also allows us to make informed decisions. In the final section, we reinforce why intellectual integrity is paramount and address directly some of the remaining counterarguments that the opposition often raises (like “Lewontin’s fallacy” and misinterpretations of within-group variation).

5. Refuting the Counterarguments: Clarity on Lewontin’s Fallacy and Other Misconceptions

The ideological opponents of human genetic research have frequently used a set of recurrent counterarguments to dismiss findings. Many of these arguments sound plausible at first glance, but on closer analysis they are flawed or irrelevant. Here we systematically refute the key points often raised: Lewontin’s fallacy (the idea that high within-group variation makes between-group differences meaningless), the claim that “within-group variation swamps any between-group variation” (misunderstanding how classification works in biology), the notion that human populations are not discrete enough to classify (true in a trivial sense, but not a barrier to useful taxonomy), and various other red herrings (“race is socially constructed,” “intelligence can’t be measured,” etc.). The goal is to clear away these misunderstandings so that the conversation can proceed based on evidence and logic, not slogans or statistical fallacies.

5.1 Lewontin’s Fallacy: Within-Group vs. Between-Group Variation

Richard Lewontin’s famous 1972 result – that ~85% of human genetic variation is within populations, ~7% between populations within a race, and only ~8% between races – has been endlessly cited as “proof” that human races have no genetic basis. This argument is often phrased as: “Any two random individuals from different races can be nearly as genetically different as two individuals from the same race.” The fallacy here is a misunderstanding of what those percentages mean for classification. Lewontin (and those who parrot his argument) assumed that because the majority of variation is within groups, the variation between groups is negligible. However, this logic is fundamentally flawed, as it confuses the proportion of variation with its taxonomic significance.

One of the simplest ways to refute this argument is to recognize that the same pattern—most genetic variation occurring within groups rather than between them—is observed in nearly all animals. This includes species where subspecies, breeds, or even distinct species are widely recognized. If the existence of high within-group variation meant that races or subspecies were meaningless, then by the same reasoning, no subspecies or breeds could exist in any animal species. But we know this is not true. Taxonomists routinely classify animals into distinct subspecies and breeds even when most genetic variation is found within populations. The same principle applies to humans; the presence of high within-group variation does not erase the meaningful and systematic genetic differences between groups.

Beyond that prima facie invalidation, geneticist A.W.F. Edwards demonstrated in a seminal 2003 paper that Lewontin’s argument was based on a misunderstanding of how genetic differences accumulate across multiple loci ^{[2] [1]}. Even if groups differ only slightly in allele frequencies at individual loci, the combination of small differences across many loci allows for near-perfect classification of individuals into their ancestral populations ^[44]. Lewontin’s analysis examined variation one gene at a time, which ignored the way multiple loci together form a genetic signature. Organisms are characterized by multivariate genomes, meaning that it is the pattern of variation across many genes that distinguishes populations, not any single locus on its own. Edwards showed that when considering multiple loci simultaneously, the “signal” of population structure emerges clearly, while the “noise” of within-group variation is greatly reduced ^{[1] [1]}.

Edwards provided a simplified example demonstrating that even if within-group variation is 85% at each individual locus (Lewontin’s figure), one could still classify individuals into the correct group with extremely high accuracy by examining enough loci​ ^{[2] [1]}. This directly undercuts the notion that high within-group variation makes races or subspecies meaningless.

To illustrate: imagine two populations that differ only slightly in allele frequencies at 100 loci. At any single locus, individuals from different populations might often share the same allele, making differentiation difficult. However, when examining the full profile of all 100 loci, the probability that two individuals from the same population share a more similar genetic profile than those from different populations becomes overwhelming. The “correlation structure” of the data – the pattern of how allele frequencies co-vary across loci – contains the information that distinguishes populations ^[2]. Lewontin ignored this and treated each locus independently, thus missing the forest for the trees. Edwards called this oversight “Lewontin’s fallacy” because it led to a completely incorrect conclusion about taxonomic significance​ ^[2].

In technical terms, within-group variation being large does not preclude clear between-group differences. A useful analogy: Consider two very broad and overlapping bell curves – say the height distribution of men and women. Perhaps 90% of height variation is within each sex and only 10% between sexes. Yet, the average man is taller than 95% of women, and if you know someone’s height plus a few other traits, you can predict their sex with high accuracy. Similarly, human groups overlap in most traits but still have different means and allele frequency distributions that allow classification. Indeed, researchers have quantified that using ~20 genetic loci, one can classify humans by continent with ~99% accuracy​ ^{[2] [2]}. The probability of misclassifying someone’s continental origin becomes “infinitesimal” as more loci are used ^[2]. This directly rebuts Lewontin’s implication that because misclassification is 30% with one locus (a point he noted himself), it means races aren’t real ^{[1] [1]}. With multiple loci, misclassification drops essentially to 0%​ ^[1]. Lewontin’s error was focusing on a single-locus perspective to draw conclusions about a multilocus reality – a classic statistical fallacy.

To reinforce this point, Edwards (2003) quoted an earlier scholar, who presciently wrote: “local races may be very different as populations, although individuals may overlap in all characters” ^[2]. In other words, even if no single trait perfectly separates groups (overlap in each trait), the overall combination of traits can distinguish the groups. This is exactly the case in human races: you can find some Europeans taller than some Africans, some Chinese darker-skinned than some Europeans, etc. – overlap in individual traits. But if you consider enough traits together (skin color, facial structure, genetic markers, etc.), you can tell a person’s ancestry with extremely high reliability. That’s why forensic anthropologists can determine race from a skeleton or why genetic tests can report ancestry percentages. The science works, which is proof that Lewontin’s hand-waving was misleading.

In summary, Lewontin’s argument is fallacious because it confuses the proportion of variation with the usefulness of variation for classification. An analogy given by others: imagine 95% of the variance in a book’s text is within chapters and only 5% between chapters; that doesn’t mean the chapter divisions are arbitrary – by looking at patterns of words you can still easily tell which text belongs to which chapter​ ^[44]. The remaining 15% between-population genetic variation in humans, far from being trivial, is sufficient to allow robust classification into populations (and has significant phenotypic effects too). For instance, that 15% includes things like skin color genes, lactose tolerance genes, disease resistance alleles, etc., which certainly differ between populations. So Lewontin’s 15% isn’t “nothing” – it’s enough to tell populations apart and is likely meaningful for many traits. Moreover, many other species have a similar or even lower between-group variation fraction and yet have well-defined subspecies (as noted earlier with lynx, buffalo, etc.), which shows that taxonomists do not consider a <20% \(F_{ST}\) to negate the existence of subspecies​ ^[1]. The insistence that it should in humans is inconsistent and special pleading.

5.2 “No Clear Boundaries” and “Clinal Variation” – Not an Obstacle to Classification

Another common argument: human variation is continuous, without sharp boundaries, so any division into discrete groups is arbitrary. It is true that human genetic differences tend to vary gradually over geography – a concept called clinal variation. However, clines are fully compatible with the existence of clusters. If you sample people from around the world, you’ll find that those from nearby locales are more similar (genetically and phenotypically) than those from far apart locales – a gradual change. But because human migration was not uniform, there were long periods of relative isolation between continents, leading to clustered clines. Think of it like colors in a rainbow spectrum: it’s a continuous gradient, yet we can still identify blue vs. green vs. red regions of the spectrum. Similarly, human variation is not random or uniform; it’s geographically structured. Cluster analysis of genetic data (as by Rosenberg et al. 2002) identifies clusters that correspond to traditional notions of races​ ^[11]. The boundaries are fuzzy (exactly where is the line between Europe and Asia? perhaps somewhere in Central Asia, but somewhat arbitrary). Yet fuzzy boundaries do not invalidate the reality of the clusters themselves. In biology, many species consist of populations that intergrade – for example, the so-called ring species concept, where neighboring populations can interbreed in a chain but the ends of the chain are distinct species. Even there, taxonomists do draw lines; nature is messy, but we impose categories for understanding. The question is whether those categories capture real structure. In humans, they do: the clusters are statistically real, not an artifact. Whether one calls them “races” or something else is semantics; the underlying structure exists.

It’s often said that “any two individuals from different populations can be genetically more similar than two from the same population.” This is only true if you are looking at just one or two genetic loci, due to the large amount of genetic variation within populations. However, this fact is misleading when taken out of context. On average, individuals from the same population are far more genetically similar to each other than to individuals from different populations, which is why ancestry classification works so well. The oft-cited claim that “there are more differences within groups than between them” does not mean an African and a European are no more different than two Europeans—it simply means that at any single genetic locus, the difference might not be apparent, but across the whole genome, the distinction is clear. A 2007 study by Witherspoon et al. confirmed that when comparing a small number of loci, occasional misclassification can occur, but once 10 or more loci are used, the odds of two individuals from different continents appearing more similar than two from the same continent become almost zero ^{[45] [46]}. This is why genetic clustering is so accurate. The “more similar across race than within” argument is misleading because it ignores the full genomic picture. The same reasoning dismantles the claim that racial differences are meaningless because they are based on allele frequency differences rather than the presence or absence of unique genes. Yes, human groups share almost all the same genes, but at different frequencies—which is exactly how subspecies work in nature. Subspecies of mammals are rarely defined by completely unique genes; rather, they are distinguished by consistent and non-trivial differences in allele frequencies. For example, many mammal subspecies share all the same alleles but differ in their frequencies, yet taxonomists still classify them separately. Humans are no different. Claiming that “we all share 99.9% of our DNA” is a trivial statement that does not negate the significant functional differences found in the remaining 0.1%, which still consists of millions of base-pair differences. Even small genetic differences can have major effects—for example, humans and chimpanzees differ by only ~1-2% of DNA, yet that small percentage results in enormous biological and cognitive differences. Likewise, the 0.1% genetic difference between human populations includes genes affecting disease susceptibility, metabolism, physical traits, and potentially cognitive abilities, making it far from insignificant.

5.3 “Race is Only a Social Construct” – Clarifying Terminology

It is true that the concept of race as used in society is partly socially constructed – societies choose certain visible traits (skin color, etc.) to define races, and different societies have classified people differently. But to leap from that to “race has no biological basis” is a fallacy of equivocation. The social labels may be imprecise or arbitrary in border definition, but they correlate with real biological ancestry. For example, “Black” in the US corresponds to largely West African ancestry; “East Asian” corresponds to ancestry from China/Korea/Japan; etc. These correspondences are why medical outcomes can differ (e.g. different rates of certain genetic diseases by race) and why genetic ancestry tests work. Yes, the folk taxonomy doesn’t capture all complexity (it ignores some populations, mixes others), but it’s not pure fiction either. Saying “race is a social construct” is often a way to dismiss the idea of genetic differences, but one must distinguish between the folk concept of race and the scientific concept of population structure. The latter is absolutely real and measurable; the former is an imperfect mapping onto the latter. It’s similar to how the concept of “species” is sometimes blurry (ring species etc.), yet nobody would claim species aren’t biologically real. In short, the phrase “social construct” doesn’t mean an entity is imaginary – it means a human-defined concept applied to a real pattern. Money is a social construct, yet money has real effects. Race as a social construct is based on real lineage differences, it’s just interpreted through a cultural lens. So this slogan should not be used to reject genetic findings. Instead, we can refine it: yes, the definitions of races are culturally chosen, but the underlying genetic clusters are real.

5.4 On Intelligence: “IQ is Just a Number” or “IQ Doesn’t Capture Real Abilities”

Critics like Gould often argued that intelligence is too multifaceted or culturally loaded to be summed up in an IQ score, implying that group IQ differences are meaningless or due to bias. Decades of psychometric research have answered this. IQ tests, when properly constructed, are not culturally biased in predicting academic or job performance; they have similar validity for all groups​ ^[24]. The “g factor” that IQ tests measure is a real biological phenomenon, correlated with brain information processing efficiency, brain volume, and other neural parameters. It’s not an artifact of test design – it emerges from factor analyses of hundreds of cognitive tasks. While humans have multiple cognitive abilities, they tend to intercorrelate positively, and g is the best predictor of broad cognitive performance. Thus, differences in g (and hence IQ) are consequential. The failure of massive educational efforts to equalize outcomes (discussed earlier) further shows IQ differences reflect something robust, not a testing artifact. Gould’s assertion that IQ only measures how well one takes IQ tests has been thoroughly discredited by the predictive validity of IQ: it predicts things far outside the testing room (years of education, income, occupational status, risk of incarceration, health literacy, etc.)​ ^{[18] [21]}. If group A scores higher on IQ than group B, that difference tends to manifest in real-world achievement gaps, which is exactly what we see in societies. So one cannot wave away IQ gaps as meaningless. Moreover, neuroscience and genetics confirm that IQ is tapping a biological reality – twin studies, polygenic scores, neuroimaging all converge on the same construct. Therefore, critiques like Gould’s are considered deeply flawed by experts (indeed, in a survey of experts in intelligence, Gould’s book The Mismeasure of Man was ranked as the worst book on intelligence, full of errors and misrepresentations ^[32]).

5.5 Other Counterarguments and Rebuttals

Sometimes it’s claimed that environmental factors not yet considered could explain all differences (for instance, stereotype threat, bias in testing, etc.). While it’s important to investigate all possibilities, the weight of evidence has increasingly ruled these out as major factors for longstanding gaps. Stereotype threat, for example, has turned out to have very fragile effects and cannot account for large, consistent differences observed from early childhood (before kids even are aware of stereotypes). Another argument: “even if differences are genetic, we can’t define races clearly – so it’s useless.” As discussed, fuzzy boundaries do not mean the absence of differences. We could use another approach: treat ancestry as continuous (percent ancestry from various continental groups) rather than discrete races. Even on that continuum, the ends of the continuum differ genetically and phenotypically. One doesn’t need a sharp boundary to discuss differences (e.g., people of 100% African ancestry vs. 0% African ancestry clearly differ on average in certain traits; those intermediate will, unsurprisingly, be intermediate). So the existence of admixture and gradients doesn’t invalidate analysis, it just requires statistical methods to handle (which we have).

Critics often invoke past racist abuses, such as 19th-century scientific racism and eugenics, to argue that we should not explore these topics today. While history should be acknowledged, this reasoning commits a moralistic fallacy—the idea that if a fact has been misused for harmful purposes, it must either be false or too dangerous to discuss. But the truth of a scientific claim is independent of how people might misuse it. Gravity is real, even though it enables bombs to fall. Likewise, if human genetic differences are real, suppressing knowledge of them does not prevent racists from making their own arguments; it simply leaves the conversation open to distortion and misinformation.

In fact, many argue convincingly that open and honest discussion is the best defense against false, exaggerated, or malicious interpretations. When mainstream scientists refuse to address genetic differences candidly, they create a vacuum that can be filled by misleading or ideologically driven narratives. It is far better for responsible scientists to lead the discussion, ensuring accuracy and nuance rather than allowing half-truths or outright falsehoods to dominate the conversation.

Moreover, science and morality serve different purposes: science describes reality, while morality determines how society should respond to that reality. Acknowledging that genes influence IQ, and that average IQ differs between populations, does not dictate what policies should be adopted—it simply establishes the factual foundation upon which policies can be debated. The real danger lies in crafting policies based on false premises, such as the assumption that all disparities must be due to discrimination. This mistaken belief has led to failed interventions, misplaced blame, and harmful policies, such as accusing White people of causing Black crime rates or academic underperformance. In extreme cases, this flawed logic even justifies overt discrimination against Whites under the guise of “equity.”

A more rational and evidence-based approach recognizes that both genetic and environmental factors contribute to group disparities. Accepting this reality allows for smarter, more effective policies that address root causes rather than wasting resources on misguided social engineering or fueling resentment through misplaced accusations and scapegoating.

Finally, the argument that discussing these things will harm society is countered by the idea that suppressing discussion breeds mistrust. When people eventually encounter genetic evidence (as will be inevitable with personal genomics on the rise), they may feel lied to by authorities who insisted “race is only a social construct” or “intelligence differences are just due to bias.” This can erode public trust in science. It’s better that scientists transparently communicate what is known and unknown, maintaining credibility. As the earlier parallels (Galileo’s heliocentrism vs. Church, etc.) teach us: truth will out, and institutions that fight it will lose esteem.

In conclusion, none of the standard counterarguments hold up under scrutiny. Within-group variation being high does not negate the reality of between-group differences or their taxonomic utility​ ^[1]. Lack of absolute boundaries doesn’t erase clustering. Social construction arguments mix up definition with existence. Fears of determinism or misuse are not refutations of facts. The misleading nature of Lewontin’s 85/15 statistic has been decisively exposed by Edwards and others ^{[2] [1]}, and we now know that human groups can be genetically distinguished with ease given sufficient data. Science has moved past these simplistic objections – it’s time for discourse to move on as well, freed from the shadow of Lewontin and Gould’s ideological science.

6. The Primacy of Scientific Integrity: Following Evidence Wherever It Leads

Throughout this report, one message has been clear: evidence and logic must guide our understanding of nature, not ideology or wishful thinking. Scientific inquiry is a noble endeavor precisely because it challenges our preconceived notions and forces us to confront reality, even if it is uncomfortable. The topics of human taxonomy and behavioral genetics have been uncomfortable for many, due to fears about how the knowledge might be used or what it might imply for society. But intellectual integrity demands that we do not distort or deny findings to suit political narratives. As the geneticist Theodosius Dobzhansky once said, “Facts are facts, no matter what our wishes.” Reality is under no obligation to cater to our moral preferences.

By rigorously examining data – from genetic distance measures to twin correlations – we followed the evidence. And the evidence led us to conclude that human populations are not genetically identical, that some meaningful structure exists (albeit with fuzzy edges), and that cognitive abilities are significantly influenced by genetic endowment. These conclusions are supported by decades of peer-reviewed research. To shy away from stating them plainly would be a disservice to scientific truth. It would also, ultimately, be a disservice to society: sound policy and social progress depend on accurate understanding of human nature. If we lie to ourselves about what causes differences in outcomes, we will prescribe the wrong remedies. For instance, if one insists that all group differences are caused solely by discrimination or environment, one might implement policies that fail to achieve equal outcomes and then erroneously blame people or systems, causing further social strife. On the other hand, recognizing that there are natural variations doesn’t mean accepting inequality as fated or justified – but it does mean our expectations and solutions can be more realistic.

Intellectual honesty is also essential for maintaining trust in science. When people see scientists suppressing findings for political reasons, it erodes public confidence in the entire institution. Conversely, when scientists are upfront—“This is what we found, it may be controversial, but here’s the evidence and here’s what it doesn’t mean”—it strengthens credibility. A historical example of intellectual courage was Francis Crick, co-discoverer of DNA, who in the 1970s openly speculated that science might eventually uncover genetic differences in mental capacities between human groups and warned that scientists should be prepared to face the truth, whatever it may be. He was maligned for his stance, but he was simply acknowledging where the evidence could eventually lead.

A more recent and disturbing case is that of James Watson, the other co-discoverer of DNA, who was publicly crucified and stripped of his titles, honors, and reputation simply for making statements about group differences in IQ that—while deemed offensive—are now largely supported by modern genetic research, polygenic studies, and psychometric data. Despite his towering contributions to biology, Watson was effectively erased from polite scientific society, demonstrating how even the most respected minds can be cast out for expressing truths that challenge prevailing ideological narratives.

Similarly, William Shockley, the inventor of the transistor, was vilified for his clumsy and politically naive advocacy of eugenics. However, his unfiltered willingness to engage controversial questions forced academia to respond with actual research rather than avoidance. Some scientists, rather than dismissing him outright, undertook more rigorous studies—such as Arthur Jensen’s later work—which ultimately advanced knowledge. Had academia simply stuck its head in the sand, we might still be debating these issues in a data vacuum, driven by ideology rather than evidence.

We should also acknowledge that following evidence includes exploring alternative hypotheses thoroughly. Scientific objectivity doesn’t mean we’ve decided all differences are genetic; it means we allocate causal weight based on evidence. In some cases (like height differences between populations) the genetic contribution is very clear. In others (like certain health disparities), environment might dominate. Each question needs open inquiry. The unfortunate climate of taboo around genes and race hindered even environmental research at times (because researchers were afraid even to stratify samples by race, etc.). Removing ideological blinders actually helps all research avenues, since nothing is off-limits and the best-supported explanations can emerge.

In championing scientific rigor, we can still maintain a commitment to human dignity and a respect for individual worth. Recognizing genetic differences does not require us to embrace a naive, blank-slate egalitarianism that ignores biological reality, nor does it mandate that every person or group must be treated as identical in all circumstances. While all humans share a common humanity, nations—like families—have the moral right to prioritize their own people’s well-being, just as a father naturally puts his own children first while still treating his neighbors with fairness. Equality under the law does not mean that every person on Earth must have equal standing in matters of citizenship, national policy, or cultural preservation. It simply means that within a given society, individuals should be treated fairly and justly according to that society’s own principles.

Recognizing human differences can actually help promote better policies that serve both national interests and social cohesion. Instead of assuming a one-size-fits-all approach, nations that acknowledge these realities can craft more effective policies in education, healthcare, and social organization—ones that respect the distinct needs and characteristics of their own people. For example, rather than blaming systemic discrimination for unequal outcomes, societies can develop targeted interventions that address real disparities in ability, temperament, and potential, ensuring that national resources are used efficiently and wisely for the benefit of their own people.

The pursuit of truth has historically led to great human advancement. When science overcame the ideological shackles of the medieval Church, we got the Enlightenment and leaps in knowledge. When biology overcame the dogmas of Lysenko, genetics and medicine flourished. In our current context, if we fully free the study of human diversity from stigma, we may unlock discoveries that benefit everyone – perhaps insights into cognitive development, educational techniques, or medical treatments tailored to genetic ancestry. Already, in biomedical research, acknowledging population differences is crucial (e.g., different populations respond differently to certain drugs due to genetic variants, so “race-based medicine,” while an imperfect concept, has some practical utility in personalized care). Intellectual honesty here literally can save lives – by, say, recognizing a gene variant more common in one group that predisposes to a disease and screening for it.

Ultimately, science progresses by confronting the unknown with an open mind. As the motto of the Royal Society says, “Nullius in verba” – take nobody’s word for it (i.e., rely on evidence, not authority or dogma). We have tried to do exactly that in this investigation. We questioned widely held claims, examined the literature, and cited primary sources to substantiate each point. The result is a picture that may not align with “politically correct” narratives, but it is what the science tells us. We see that human evolution has indeed left a signature on our genomes that differentiates populations; we see that genes play a large role in creating individual and group differences in traits like intelligence; and we see that denying these facts was a choice made out of ideological concern, not out of scientific necessity.

Going forward, it is incumbent on scholars and educators to communicate these truths responsibly. It’s a delicate balance, but not an impossible one. The key is sticking to data and clarifying what is known versus unknown.

In the end, the pursuit of knowledge should be “unapologetically committed to scientific objectivity,” as this report has aimed to be. That is not just a lofty ideal – it is a practical necessity for intellectual and societal progress. By following the evidence wherever it leads, we actually uphold the integrity of both science and society. Truthful understanding of human nature, coupled with humane values, arms us to create better policies and a more honest public discourse. Suppressing truth, conversely, leaves us blind and often exacerbates the very problems we wished to solve. As history’s suppressed truths eventually came to light (to the embarrassment of the suppressors), so too will the truths about human biodiversity and behavior. Embracing reality is the only way to harness it for good ends.

Conclusion: The investigation presented here, grounded in extensive peer-reviewed evidence, demonstrates that human populations exhibit significant genetic differentiation, even by conservative measures like \(F_{ST}\)​ or more refined ones like Jost’s D—at levels comparable to those used to classify subspecies in other mammals​ ^{[1] [1]}. Under the Phylogenetic Species Concept (PSC), multiple human lineages could be recognized, yet this notion is resisted for ideological rather than scientific reasons.

In behavioral genetics, lines of evidence ranging from classical twin/adoption studies to modern polygenic research consistently point to a substantial genetic basis for both individual and group differences in intelligence and other traits ^{[16] [17]}. Environmental interventions have repeatedly shown only limited and temporary effects, reinforcing the conclusion that biological factors play a dominant role ^[19]. Despite this, political and moral concerns have driven resistance to these findings, with notable scientists like Lewontin and Gould obscuring or outright denying evidence out of ideological commitment ^{[28] [29]}. However, their arguments—such as Lewontin’s claim that 85% of variation is within groups—fail under deeper scrutiny and have been refuted by later analyses​ ^[1].

History has repeatedly shown that ideological interference in science—whether creationism, Lysenkoism, or enforced taboos around human genetic differences—is ultimately unsustainable in the face of determined empirical inquiry. It is time to bring human taxonomy and behavioral genetics fully into the realm of objective science, free from euphemism, distortion, or fear. A rational, truth-seeking society must ground its policies and ethics in reality, not in ideological wishful thinking.

Let us, then, continue this exploration with curiosity and honesty, nihil humani a me alienum puto – “nothing human is alien to me.” In studying human variation, we are studying ourselves, and only through truth can we truly know ourselves and improve the human condition.

Sources:

• Woodley, M.A. (2010). Is Homo sapiens polytypic? Human taxonomic diversity and its implications. Medical Hypotheses, 74(2), 195-201. lesacreduprintemps19.files.wordpress.com​lesacreduprintemps19.files.wordpress.com​lesacreduprintemps19.files.wordpress.com​lesacreduprintemps19.files.wordpress.com​lesacreduprintemps19.files.wordpress.com​lesacreduprintemps19.files.wordpress.com
• Edwards, A.W.F. (2003). Human genetic diversity: Lewontin’s fallacy. BioEssays, 25(8), 798–801. humanbiologicaldiversity.com​humanbiologicaldiversity.com
• Rosenberg, N.A. et al. (2002). Genetic structure of human populations. Science, 298(5602), 2381-2385. pmc.ncbi.nlm.nih.gov
• Witherspoon, D.J. et al. (2007). Genetic similarities and differences within and between human populations. Genetics, 176(1), 351–359. pmc.ncbi.nlm.nih.gov​rosenberglab.stanford.edu
• Jost, L. (2008). GST and its relatives do not measure differentiation. Molecular Ecology, 17, 4015–4026. (Explains limitations of \(F_{ST}\) and introduces D)​molecularecologist.com​molecularecologist.com​molecularecologist.com
• Bouchard, T.J. & McGue, M. (1981). Familial studies of intelligence: A review. Science, 212, 1055-1059. (Heritability from twin/adoption studies)​www1.udel.edu
• Rushton, J.P. & Jensen, A.R. (2005). Thirty years of research on race differences in cognitive ability. Psychology, Public Policy, and Law, 11(2), 235-294. www1.udel.edu​www1.udel.edu​www1.udel.edu
• Plomin, R. & von Stumm, S. (2018). The new genetics of intelligence. Nature Reviews Genetics, 19(3), 148-159. pmc.ncbi.nlm.nih.gov​pmc.ncbi.nlm.nih.gov
• Protzko, J. (2015). Effects of early interventions on intelligence: A meta-analysis. Intelligence, 53, 78-92. sciencedaily.com​sciencedaily.com
• Lewis, J.E. et al. (2011). The mismeasure of science: Gould vs. Morton on skulls and bias. PLoS Biology, 9(6): e1001071. journals.plos.org
• Discover Magazine (2012). The Mismeasure of Stephen Jay Gould (Jason Lewis et al. commentary on re-evaluation of Gould’s claims)​discovermagazine.com​discovermagazine.com
• Pew Research Center (2019). Darwin in America: The evolution debate. pewresearch.org​pewresearch.org
• Depew, D. (2018). Richard Lewontin’s ideology and rhetoric in science (Poroi, 13(2)). pubs.lib.uiowa.edu​pubs.lib.uiowa.edu
• Feldman, M.W. (2014). Echoes of the Past: Hereditarianism and A Troublesome Inheritance. PLOS Genetics, 10(11): e1004818. (Provides historical context on Jensen and the debate)​pmc.ncbi.nlm.nih.gov​pmc.ncbi.nlm.nih.gov
• Kanazawa, S. (2017). Beyond the Blank Slate: Evolutionary Psychology and Social Sciences (mention of politically motivated resistance by Lewontin, Gould, Kamin). academic.oup.com (Various sources on Marxist influence)

1. lesacreduprintemps19.files.wordpress.com[↑][↑][↑][↑][↑][↑][↑][↑][↑][↑][↑][↑][↑][↑][↑][↑][↑][↑][↑][↑][↑][↑][↑][↑][↑]
2. humanbiologicaldiversity.com[↑][↑][↑][↑][↑][↑][↑][↑][↑][↑][↑][↑]
3. molecularecologist.com[↑]
4. molecularecologist.com[↑]
5. molecularecologist.com[↑]
6. molecularecologist.com[↑]
7. molecularecologist.com[↑][↑]
8. web.natur.cuni.cz[↑]
9. lesacreduprintemps19.files.wordpress.com[↑]
10. en.wikipedia.org[↑]
11. pmc.ncbi.nlm.nih.gov[↑][↑][↑]
12. biology.stackexchange.com[↑]
13. biology.stackexchange.com[↑][↑][↑]
14. lesacreduprintemps19.files.wordpress.com[↑]
15. webspace.ship.edu[↑][↑][↑][↑]
16. www1.udel.edu[↑][↑][↑][↑][↑][↑][↑][↑][↑][↑][↑][↑][↑][↑]
17. pmc.ncbi.nlm.nih.gov[↑][↑][↑][↑]
18. pmc.ncbi.nlm.nih.gov[↑][↑][↑]
19. sciencedaily.com[↑][↑][↑][↑]
20. sciencedaily.com[↑][↑]
21. pmc.ncbi.nlm.nih.gov[↑][↑]
22. pmc.ncbi.nlm.nih.gov[↑]
23. pmc.ncbi.nlm.nih.gov[↑]
24. discovermagazine.com[↑][↑][↑]
25. sciencedaily.com[↑]
26. pubs.lib.uiowa.edu[↑][↑]
27. pubs.lib.uiowa.edu[↑]
28. pubs.lib.uiowa.edu[↑][↑][↑]
29. journals.plos.org[↑][↑][↑]
30. discovermagazine.com[↑]
31. discovermagazine.com[↑]
32. reddit.com[↑][↑]
33. lesacreduprintemps19.files.wordpress.com[↑]
34. lesacreduprintemps19.files.wordpress.com[↑][↑]
35. opencurriculum.org[↑]
36. pewresearch.org[↑]
37. pewresearch.org[↑]
38. pewresearch.org[↑]
39. pewresearch.org[↑]
40. pewresearch.org[↑]
41. ethos.lps.library.cmu.edu[↑]
42. ethos.lps.library.cmu.edu[↑][↑]
43. ethos.lps.library.cmu.edu[↑]
44. lesacreduprintemps19.files.wordpress.com[↑][↑]
45. pmc.ncbi.nlm.nih.gov[↑]
46. rosenberglab.stanford.edu[↑]