Human variation and race: the non-existence of biological races, clines, and adaptation

Intuition Beginner

Race is real as a social category but not as a biological one. Societies have sorted people into groups by appearance and ancestry for centuries, and those groups carry enormous social weight — shaping wealth, health, law, and identity. But modern genetics has demolished the older claim that humanity divides into a small number of discrete biological races.

The genetic evidence is striking. There is more variation within any so-called racial group than between different groups. Roughly 85 percent of human genetic variation exists within local populations, while only about 7 percent separates continental groups. Two randomly chosen Africans are likely to differ from each other more than either differs from a European or an East Asian.

There are no clear genetic boundaries between races. Human variation is clinal: traits change gradually across geographic space without sharp breaks. If you walked from the equator toward the Arctic, you would see skin colour lighten by tiny increments over thousands of kilometres, never by a sudden jump. Clines are the opposite of the neat boxes that racial categories impose.

Consider skin colour. Dark skin near the equator protects folate, a vitamin that ultraviolet radiation would otherwise destroy. Light skin at high latitudes lets enough sunlight through to make vitamin D under a weak sun. The same selective pressure — balancing UV exposure against vitamin production — produced smoothly graded pigmentation across the globe. Skin colour tracks latitude and ancestry, not deep genetic difference.

The traits we use to define race — skin colour, hair texture, facial features — are a tiny fraction of our genetic makeup, and each was shaped by its own local environmental pressure. They do not come as a coordinated package. Grouping people by these superficial traits ignores the vast diversity within every population and exaggerates the differences between them.

Race is a social construction: a way societies organise people into hierarchies — granting privilege to some and denying it to others — rather than a natural biological category. The categories themselves, and the traits used to mark them, have shifted across time and place. What changes is not human biology but the social meanings attached to it.

Visual Beginner

Source of genetic variation	Share of total
Within local populations	~85%
Between populations within races	~8%
Between continental races	~7%

Approach to human variation	View of diversity	Key figures
Typological	Discrete races as fixed types	Blumenbach, Cuvier, Morton
Clinal	Gradual change, no sharp boundaries	Livingstone, Brace
Population / adaptive	Local adaptation, gene flow	Lewontin, Jablonski

Adaptive trait	Environmental pressure	Geographic pattern
Dark skin	Folate protection from UV	Equatorial
Light skin	Vitamin D synthesis	High latitude
Sickle cell trait	Malaria resistance	Endemic regions
Lactase persistence	Adult milk digestion	Pastoral populations
Larger body (Bergmann)	Heat retention	Cold climates
EPAS1 variant (Tibetans)	High-altitude hypoxia	Tibetan plateau

Worked example Beginner

Example 1: Reading Lewontin's 85-7-8 split

Lewontin's 1972 study measured variation at 17 genetic loci across populations grouped by race. He found 85.4 percent of the total within local populations, 8.3 percent between populations within a race, and only 6.3 percent between races. If you pick two people at random from anywhere on Earth, knowing their race tells you almost nothing about how genetically similar they are. Most of what differs between them is variation that exists inside every group, not variation that separates groups. That is why racial categories are biologically uninformative, even though a few visible traits like skin colour do differ between populations.

Example 2: The skin-colour cline

Skin colour follows a smooth gradient. Near the equator, intense sunlight destroys folate, a vitamin essential for embryonic neural development, so dark pigmentation is favoured. Near the poles, weak sunlight cannot penetrate dark skin enough to synthesise vitamin D, so lighter pigmentation is favoured. The transition is gradual: there is no latitude at which "Black" skin suddenly becomes "White." Light skin evolved independently in European and East Asian lineages through different genetic mechanisms (SLC24A5 in Europeans, OCA2 variants in East Asians), a textbook case of convergent evolution under the same selective pressure.

Example 3: Sickle cell is not a racial trait

The sickle-cell allele protects carriers against malaria. It reaches high frequency wherever malaria is endemic — West Africa, the Mediterranean, parts of India and the Middle East — and drops wherever malaria is absent. If sickle cell were a racial trait, it would track racial categories. It does not. It tracks mosquito ecology. Greek, Turkish, and Indian populations carry the allele at frequencies comparable to West African groups, while many African populations in malaria-free zones carry almost none. The example shows that adaptive variation follows environment, not race.

Check your understanding Beginner

Formal definition Intermediate+

Typological versus clinal approaches

The typological approach to human variation, dominant in eighteenth- and nineteenth-century physical anthropology, sorts humanity into a small number of discrete, supposedly primordial races. Blumenbach's 1795 five-race scheme (Caucasian, Mongolian, Ethiopian, American, Malay) and Cuvier's three-race scheme (Caucasoid, Negroid, Mongoloid) are canonical examples. These schemes were hierarchical: the Caucasian type, defined against a skull from the Caucasus, sat at the apex, and the others were treated as deviations. The traits used to define the types — skin colour, hair form, cranial shape, facial profile — were assumed to co-vary and to mark deep biological difference.

The clinal or population approach rejects discrete types. Livingstone's 1962 dictum — "there are no races, there are only clines" — captures the shift. A cline is a gradual change in the frequency of a trait or allele across geographic space. Because each trait responds to its own selective pressures and migrates by its own history, traits vary clinally in different directions and at different rates. Skin colour, lactase persistence, and body proportions each follow their own gradient, and the gradients do not align into a small set of packages. No trait distribution carves humanity into three or five or seven discrete groups.

The fixation index F_ST

The fixation index F_ST quantifies how much populations differ genetically. For a locus with total heterozygosity H_T (the expected heterozygosity if all subpopulations were pooled) and average within-population heterozygosity H_S, the fixation index is F_ST = (H_T − H_S) / H_T.

F_ST equals zero when subpopulations are genetically identical and approaches one when they are completely differentiated. Across the human genome, continental F_ST is approximately 0.12 to 0.15. By comparison, subspecies and recognised biological races in other species typically show F_ST above 0.25. Humans are genetically uniform for a species of our global range — a pattern consistent with a recent common origin and high gene flow.

Clinal adaptation

Clinal variation is the signature of local adaptation acting on a mobile species. Skin pigmentation follows the UV gradient (Jablonski and Chaplin): equatorial populations need melanin to protect folate from photodegradation, while high-latitude populations need lighter skin to synthesise vitamin D under weak sunlight. Body size and shape follow climate rules — Bergmann's rule (larger bodies in colder climates, which retain heat better) and Allen's rule (shorter extremities in colder climates, which reduce surface area). Lactase persistence, the ability to digest milk sugar into adulthood, coevolved with dairying in several populations through distinct mutations.

These adaptations do not come as a package. Each trait tracks its own environmental gradient, and the gradients intersect without aligning. A Maasai herder and a Finnish farmer both digest milk as adults, but through different mutations, and they share little else of selective relevance. Adaptation is mosaic, not racial.

Key result: Lewontin's apportionment and the end of biological race Intermediate+

In 1972 Richard Lewontin analysed variation at 17 genetic loci across populations grouped into seven geographic races. He partitioned the total variation into three components: 85.4 percent within local populations, 8.3 percent among populations within races, and 6.3 percent among races. Because the between-race component is small relative to the within-population component, Lewontin concluded that racial classification has "virtually no genetic or taxonomic significance."

Whole-genome studies have confirmed and refined the result. SNP-based and sequence-based estimates of human F_ST cluster near 0.12 to 0.15, consistent with Lewontin's apportionment. Jorde and colleagues, and the HapMap and 1000 Genomes projects, reproduce the pattern: most variation is shared, and allele-frequency differences between continental groups are modest for the overwhelming majority of loci.

Edwards' 2003 critique ("Lewontin's Fallacy") observed that when many loci are analysed jointly, multilocus genotypes classify individuals into continental clusters with high accuracy. The point is correct but does not overturn Lewontin's. The informativeness of multilocus classification depends on accumulating tiny allele-frequency differences across thousands of loci, each of which is individually small. Classification success is not the same as large between-group differentiation. Most variation remains within populations; race captures a thin slice of human diversity, even if that slice is technically detectable.

The American Association of Physical Anthropologists' 2019 Statement on Race and Racism distils the consensus: biological race does not exist in humans; race is a social and historical construction; and racism — not race — produces real biological consequences through unequal exposure to stress, pollution, deprivation, and violence.

Exercises Intermediate+

Advanced results Master

Population structure and the reality of clusters

Genome-wide studies confirm Lewontin's apportionment while complicating the interpretation. Rosenberg et al. (2002) applied the STRUCTURE algorithm to the CEPH diversity panel and found that individuals cluster into approximately five continental groups when K is set to five. The result is widely cited as evidence that genetic structure recapitulates continental race.

Bolnick (2008) and others caution that the number of clusters is sensitive to sampling: rare or geographically intermediate populations shift the optimal K, and the assumption that K = 5 corresponds to natural races confuses an analytical choice with a biological fact. Witherspoon et al. (2007) showed that multilocus genotypes classify individuals accurately only when many loci are used, and that pairs of individuals from different continents are sometimes more genetically similar than pairs from the same continent when few loci are examined. Classification power grows with the number of loci but does not imply large between-group differences at any single locus.

The clinal turn in physical anthropology

Franz Boas's 1912 immigrant study measured cranial form in children of European immigrants to New York and found that head shape changed within a single generation in response to environment — refuting the typological claim that cranial morphology was a fixed racial marker. Gravlee, Bernard, and Leonard (2003) reanalysed Boas's data with modern statistics and confirmed that the plasticity signal is real, although its magnitude is smaller than Boas reported.

Sherwood Washburn's 1951 address "The New Physical Anthropology" called for replacing typology with population thinking and evolutionary process. Frank Livingstone extended the logic to clinal variation; C. Loring Brace argued for thinking about race without the race concept; and Stanley Garn developed the idea of geographical races as fluid breeding populations rather than fixed types. The cumulative effect was to dissolve the typological frame from within the discipline.

Race in medicine and pharmacogenomics

Race enters medicine in two contradictory ways. As a proxy for genetic ancestry, it sometimes correlates with allele frequencies of clinical importance — CYP variants, HLA types, warfarin dosing — but the correlation is crude and population-specific rather than racial. The drug BiDil (isosorbide dinitrate with hydralazine), approved in 2005 for heart failure in self-identified Black patients, became the first race-specific drug and a flashpoint.

Critics (Kahn, Braun) argued that the trial was designed around a regulatory convenience rather than a biological mechanism, and that marketing it as racial essentialised a heterogeneous population. Wilson et al. showed that genetic clusters align poorly with self-identified race in many cases, and that ancestry-informative markers outperform racial labels for pharmacogenomic prediction. Braun's work stresses the distinction between pragmatic clinical use of race and its reification as biology. The contemporary direction is to replace racial categories with direct genotyping or ancestry estimates wherever possible.

Race in forensic anthropology

Sauer's 1992 paper posed the central puzzle: if races do not exist, why are forensic anthropologists so good at identifying them? His resolution: forensic anthropologists classify skeletons by social race because racialised environments leave biological traces. Diet, occupational stress, disease exposure, and healthcare access shape bone density, cranial form, dental wear, and stature. The forensic classifier reads the embodied record of a racialised life, not a racial genotype. The finding supports the social-construction view: social categories become biological through embodiment, without being biological in origin.

Admixture, ancestry, and population history

Tishkoff and colleagues' work on African genetic diversity established that the continent contains more genetic variation than the rest of the world combined, consistent with its longer habitation and larger effective population size. African-Americans carry roughly 16 to 20 percent European ancestry on average, with wide variance reflecting regional and individual history. Latin American populations show varying degrees of European, Indigenous, and African ancestry shaped by colonial caste systems and subsequent mixture.

The Siddi of South Asia are an African diaspora population whose genetic ancestry diverges from their cultural context. Alondra Nelson's The Social Life of DNA traces how genetic genealogy became entangled with African American identity politics and claims of ethnic belonging. Kim TallBear's Native American DNA examines how tribal citizenship and genetic testing collide — genetic ancestry cannot resolve questions of political belonging that tribes define by their own criteria.

Anthropometric history

Richard Steckel and John Komlos developed anthropometric history as a window onto the biological standard of living. Adult height reflects net nutritional status — diet minus disease load and physical labour — during the growing years. Steckel's work on slave stature, Komlos's on European height cycles, and the broader literature on secular trends in growth and puberty timing use bodies as records of economic and environmental history.

The Boas 1912 immigrant study, revisited by Gravlee and Brace, fits the same framework: cranial plasticity reflects the biological impact of migration and changing conditions. Anthropometric history shows that "biological" outcomes are deeply historical, mediated by institutions and inequality, and cannot be read off genetics alone.

Skin pigmentation genetics and recent selection

Light skin in Europeans is driven substantially by the SLC24A5 variant, which rose to near-fixation within roughly the last 10,000 years under strong selection, alongside SLC45A2 and KITLG variants. East Asian light skin evolved independently through OCA2 and other loci — convergent evolution under the same selective pressure. MC1R variants regulate melanin production and show signatures of relaxation of constraint in depigmented populations.

Williamson et al. (2007) documented widespread recent and ongoing positive selection across the human genome, much of it acting on diet, immunity, and pigmentation. The implication is that many traits used to mark "race" are evolutionarily recent and were acquired separately in different lineages — the opposite of ancient, deep racial divergence. Chaplin and Jablonski's revision of the vitamin D hypothesis accommodates these findings, treating skin colour as a fast-evolving adaptation rather than a fixed racial attribute.

Race statements and curriculum

The AAPA's 1996 statement on race affirmed that pure races do not exist, that human variation is continuous, and that racism has real social and health consequences. The American Anthropological Association's 1998 statement went further in stressing the social construction of race and was criticised by Brace and others for understating the biological dimensions of population differences.

The AAPA's 2019 statement is the strongest and most current professional position: it states plainly that biological race does not exist, that race is a social construct, and that racism produces biological harm. Edgar and Hunt, and John Relethford's Human Population Genetics textbook, address how to teach the topic. The consensus is to present clinal variation, F_ST, and Lewontin's apportionment together, while foregrounding the social and medical consequences of racism.

Structural racism as a biological force

Clarence Gravlee's 2009 paper "How Race Becomes Biology" argued that race becomes biology through embodiment: racism shapes exposure to stress, environmental toxins, poor housing, limited healthcare, and chronic discrimination, and these exposures leave measurable biological traces — in cortisol patterns, blood pressure, inflammatory markers, and birth outcomes.

Nancy Krieger's ecosocial theory formalises embodiment and the cumulative biological impact of social inequality. Arline Geronimus's "weathering" hypothesis documents accelerated aging and earlier health deterioration in African American populations, particularly women. Collins and David found that African American middle-class women still have worse birth outcomes than White women who did not finish high school — a finding incompatible with simple socioeconomic explanations and pointing to the chronic stress of racism itself. Epigenetic inheritance of trauma remains debated: Rachel Yehuda's studies of Holocaust survivor descendants are suggestive but methodologically contested.

Ancestry-informative markers and consumer genomics

Ancestry-informative markers (AIMs) are loci chosen for large frequency differences between reference populations. Consumer genomics companies (23andMe, Ancestry.com) use them to estimate ancestry proportions, but the estimates depend on the composition of reference databases, which are heavily biased toward European populations and sparse for African and Indigenous groups.

Bolnick and colleagues documented that Native American ancestry estimates are particularly unreliable, both because reference panels are thin and because tribal belonging is political rather than genetic. TallBear's Native American DNA argues that genetic ancestry testing threatens indigenous sovereignty by redefining belonging in genomic terms. Nelson's work on African American genetic genealogy traces how consumer genomics intersects with questions of identity, reparation, and belonging that genetics cannot resolve.

Race in primatology and the history of display

Sarah Hrdy's primatology of infanticide provoked a feminist reckoning with how human categories — gender, race, family — are projected onto primate behaviour. Donna Haraway's Primate Visions traced how race, gender, and colonial power shaped primatology from its origins, treating the discipline as a site where human hierarchies were naturalised through the study of apes.

The deeper history includes human zoos — colonial exhibitions that displayed African, Asian, and Pacific Islander peoples as "primitive" spectacles for European and American audiences — and Samuel Morton's nineteenth-century collection of skulls, ranked by race and capacity, later reanalysed by Stephen Jay Gould in The Mismeasure of Man as a case of bias shaping measurement. Recent decades have seen repatriation of Morton's collection and broader reckoning with the colonial foundations of physical anthropology.

Connections Master

This unit presupposes the evolutionary framework established in 31.04.01, which introduces natural selection, population genetics mechanisms, and the distinction between genotype and phenotype. The fixation index F_ST, balancing selection, and gene flow used here are developed there for the general case.

The hominin fossil record and Out of Africa dispersal treated in 31.04.02 pending supply the demographic substrate against which clinal variation is read: the founder effects along dispersal routes, the bottleneck reducing non-African diversity, and the adaptive introgression from Neanderthals and Denisovans all shape the present-day distribution of human variation discussed here.

The social construction of race connects directly to cultural anthropology 31.02.01 and to anthropological theory 31.01.02 pending, where the Boasian critique of typology and the history of racial thought are treated in depth. The embodied biological consequences of racism bridge to medical anthropology and to the applied anthropology curriculum in section 31.06, where the ethics of research and intervention in racialised health disparities are examined.

The F_ST apportionment and population-structure analyses connect to evolutionary biology and population genetics in section 19, where the mathematics of heterozygosity, fixation indices, and the coalescent are developed for general diploid populations rather than the human case alone.

Historical and philosophical context Master

The typological race concept has deep roots. Carl Linnaeus's 1735 Systema Naturae classified humans alongside the great apes and divided Homo sapiens into geographic varieties — europaeus, afer, asiaticus, americanus — each with attached moral and temperamental stereotypes. Johann Friedrich Blumenbach's 1795 De generis humani varietate nativa introduced the five-race scheme (Caucasian, Mongolian, Ethiopian, American, Malay) and the term "Caucasian," drawn from a skull Blumenbach considered beautifully formed. Georges Cuvier reduced the scheme to three and treated the races as hierarchically ordered. The typologies were not innocent classifications: they hardened into justifications for slavery, colonial rule, and scientific claims of European superiority.

Samuel George Morton's nineteenth-century collection of hundreds of human skulls, measured for cranial capacity and ranked by race, supplied American physical anthropology with its supposedly empirical core. Stephen Jay Gould's 1981 The Mismeasure of Man reanalysed Morton's data and argued that systematic bias — selective sampling, inconsistent averaging, and plausible mismeasurement — produced the racial ranking Morton expected. Lewis et al. (2011) partially rebutted Gould by remeasuring the skulls and finding some accusations overstated, while confirming that Morton's racial hierarchy was unsupportable. The episode remains a methodological touchstone for how prior belief shapes measurement.

Franz Boas's 1912 immigrant study marks the empirical turn against typology. By measuring cranial form in children of European immigrants to New York and showing that head shape changed within a generation, Boas undermined the claim that cranial morphology was a fixed racial marker. Sherwood Washburn's 1951 address "The New Physical Anthropology" called for replacing typology with population thinking and evolutionary process. Frank Livingstone's 1962 dictum — "there are no races, there are only clines" — captured the new consensus: human variation is gradual, and the traits used to define race respond to different pressures and vary clinally in different directions.

Richard Lewontin's 1972 paper "The Apportionment of Human Diversity" supplied the genetic argument. Analysing 17 polymorphic loci across populations grouped into seven races, Lewontin found that 85.4 percent of variation lay within local populations, 8.3 percent between populations within races, and 6.3 percent between races. He concluded that racial classification has "virtually no genetic or taxonomic significance." The result was confirmed by the SNP and whole-genome era: F_ST for humans is approximately 0.12 to 0.15, well below the threshold typical of subspecies in other species.

The institutional reckoning has been slow. UNESCO's 1950 statement on race, drafted by a committee including Ashley Montagu, declared humanity a single species and rejected biological race — but was widely criticised as politically motivated. The American Anthropological Association's 1998 statement stressed social construction and was faulted by some biological anthropologists for overstating the case. The AAPA's 2019 statement on race and racism is the current professional position: biological race does not exist in Homo sapiens; race is a social and historical construction; and racism, not race, produces real biological harm through the embodiment of inequality.

The consumer-genomics era has reopened public confusion about race and genetics. Direct-to-consumer ancestry tests promise to tell customers "who they are" through percentage breakdowns, but the underlying reference databases are biased, the categories are imposed, and the political implications — for tribal sovereignty, for reparations claims, for identity politics — exceed what genetics can settle. The curriculum therefore tracks a two-century trajectory from typological classification, through the clinal and population turn, to contemporary genomic confirmation that biological race does not exist while racism, tragically, does.

Bibliography Master

1. Lewontin, R. C. (1972). "The Apportionment of Human Diversity." Evolutionary Biology, 6, 381–398.

2. Livingstone, F. B. (1962). "On the Non-Existence of Human Races." Current Anthropology, 3, 279–281.

3. Boas, F. (1912). "Changes in Bodily Form of Descendants of Immigrants." American Anthropologist, 14, 530–562.

4. Washburn, S. L. (1951). "The New Physical Anthropology." Yearbook of Physical Anthropology, 7, 1–11.

5. Jablonski, N. G. & Chaplin, G. (2000). "The Evolution of Human Skin Coloration." Journal of Human Evolution, 39, 57–106.

6. Edwards, A. W. F. (2003). "Human Genetic Diversity: Lewontin's Fallacy." BioEssays, 25, 798–801.

7. Rosenberg, N. A. et al. (2002). "Genetic Structure of Human Populations." Science, 298, 2381–2385.

8. Witherspoon, D. J. et al. (2007). "Genetic Similarities Within and Between Human Populations." Genetics, 176, 351–359.

9. Sauer, N. J. (1992). "Forensic Anthropology and the Concept of Race: If Races Don't Exist, Why Are Forensic Anthropologists So Good at Identifying Them?" Social Science & Medicine, 34, 107–111.

10. Gravlee, C. C. (2009). "How Race Becomes Biology: Embodiment of Social Inequality." American Journal of Physical Anthropology, 139, 47–57.

11. Tishkoff, S. A. et al. (2009). "The Genetic Structure and History of Africans and African Americans." Science, 324, 1035–1044.

12. Williamson, S. H. et al. (2007). "Localizing Recent Adaptive Evolution in the Human Genome." PLoS Genetics, 3, e90.

13. Bolnick, D. A. (2008). "Individual Ancestry Inference and the Reification of Race as a Biological Phenomenon." In Koenig, B. A., Lee, S. S.-J. & Richardson, S. S. (eds.), Revisiting Race in a Genomic Age. Rutgers University Press, New Brunswick, NJ.

14. Krieger, N. (2005). Embodying Inequality: Epidemiologic Perspectives. Baywood, Amityville, NY.

15. Geronimus, A. T. (1992). "The Weathering Hypothesis and the Health of African-American Women and Infants." Ethnicity & Disease, 2, 207–221.

16. Nelson, A. (2016). The Social Life of DNA: Race, Reparations, and Reconciling the Past. Beacon Press, Boston.

17. TallBear, K. (2013). Native American DNA: Tribal Belonging and the False Promise of Genetic Science. University of Minnesota Press, Minneapolis.

18. Haraway, D. (1989). Primate Visions: Gender, Race, and Nature in the World of Modern Science. Routledge, New York.

19. Gould, S. J. (1981). The Mismeasure of Man. W. W. Norton, New York.

20. Gravlee, C. C., Bernard, H. R. & Leonard, W. R. (2003). "Boas's Changes in Bodily Form: The Immigrant Study, Cranial Plasticity, and Boas's Physical Anthropology." American Anthropologist, 105, 326–332.

21. American Association of Physical Anthropologists. (2019). "AAPA Statement on Race and Racism." American Journal of Physical Anthropology, 169, 400–402.

22. Relethford, J. H. (2013). The Human Species: An Introduction to Biological Anthropology (9th ed.). McGraw-Hill, New York.