What We Know and What We Don’t Know: Human Genetic Variation and the Social Construction of Race
Published on: Jun 07, 2006

Joseph L. Graves, Jr. is University Core Director and Professor of Biological Sciences at Fairleigh Dickinson University. His research concerns the evolutionary genetics of postponed aging and biological concepts of race in humans. He is the author of The Emperor's New Clothes: Biological Theories of Race at the Millennium, and The Race Myth: Why We Pretend Race Exists in America. He was elected a Fellow of the American Association for the Advancement of Science in 1994.

Even the devil can quote scripture: genetics for the human race

“A Family Tree in Every Gene” by Professor Armand Leroi makes much of the Nature Genetics supplement entitled: “Genetics for the Human Race.” This volume originated in a symposium entitled Human Genome Variation and Race, held at Howard University in the spring of 2003. The conference was the brain child of the National Human Genome Center at Howard University, a historically African American institution, led by scientists who hail from different parts of the African Diaspora. At this time I was a member of its external advisory board and a participant at the conference. Leroi describes its results as signaling the end of race as a social construct: “In the supplement . . . [b]eneath the jargon, cautious phrases, and academic courtesies, one thing was clear: the consensus about social constructs was unraveling. Some even argued that, looked at the right way, genetic data clearly shows that races exist.” As an attendee of the conference, I can state that his characterization of both the conference and the supplement is not accurate. The Howard conference was anything but “courteous” or politically correct. Scientists presented state-of-the-art papers on genetic variation and the concept of race. The discussion of the papers was intellectually rigorous, many times heated and contentious, and in no way governed by attempts to avoid controversy.

“Genetics for the Human Race” began with a commentary imploring researchers to change from socially constructed race-based to genetic population-based thinking. In fact, every one of the papers published in this volume discussed the limitations of present genetic data to reconstruct 19th century typological racial schemes. All recognized that geographic variation exists, and that this variation may play a role in specific diseases, but none expressed that “race” as classically defined was an appropriate way to classify individuals.

The baby with the bathwater

Leroi relies on a 2003 paper published in Bioessays by Cambridge statistician A.W.F. Edwards.1 Edwards shows that a single genetic locus is insufficient to classify the ancestry of individuals. If one looks at many loci it is possible to unambiguously identify an individual’s geographic ancestry. This result is hardly new. So, why should anyone choose this time to marshal this fact as an assault on social construction theory? The impact of social construction theory in anthropology or sociology did not begin or end with Richard Lewontin in 1972. Charles Darwin raised similar issues in The Descent of Man (1871).2 He pointed out the difficulty inherent in human racial classification, showing that naturalists had failed to agree upon the most important taxonomic characters. Thus, the racial schemes of his time varied from 2-63 named races. The morphological theories of race employed in the Western world by pre-Darwinian naturalists were deeply influenced by their social views.3 Social construction theory is certainly evident in the thinking of Franz Boas and his student Ruth Benedict in the late 1930’s. However, the first full and clear articulation of this was by Boas’s student Ashley Montagu. He was a physical anthropologist and wrote Mankind’s Most Dangerous Myth: The Fallacy of Race (1942). Montagu pointed out that the physical features found in populations were not consistently correlated with each other (the principle of discordance.) For example, sub-Saharan Africans, East Indians, and Australian aborigines have dark skin, but differ in other anatomical traits, such as body proportions, skull proportions, hair type, or ear wax consistency. Indeed if one attempts to take multiple physical characters to define racial groups, you arrive at categorizations that are not indicative of their evolutionary history. Montagu wrote a series of articles for both scientific and popular journals between 1939 and 1942 outlining this concept.4 In fact, Edwards himself published a paper showing that using 63 physical traits you would classify Eskimos closer to Swedes and French populations than Eskimos are to North American Indians, with North American Indians closer to Swedes, French, and Eskimos than they are to South American Indians. Similar errors were observed with other populations, such as linking Australian aborigines with sub-Saharan Africans. In the same paper published in 1964, he and Cavalli-Sforza showed a tree based on 20 genes did not match the tree based on physical featutres!5 Neither is the discordance between physical features and genetic variation a phenomenon only found in humans.6 C. Loring Brace, Curator of Biological Anthropology at the University of Michigan’s Museum of Anthropology, has shown that while physical features can be used to demonstrate the likely origin of an individual skeleton, these features do not allow unambiguous classification of races. Geneticist Sewall Wright made the same point concerning discordance in his discussion of the genetic differentiation of the races of mankind.7 Thus Leroi is mistaken when he claims that human physical characteristics are correlated with each other in ways that reflect genetic relatedness.

The 1950 UNESCO statement on race was clearly influenced by the thinking of Ashley Montagu and embodies social construction theory. The UNESCO conference was reported on by The New York Times, on July 17, 1950. The article stated that a world panel of experts had concluded that “race was less a biological fact, than a social myth.”8 The panel included a number of pre-eminent geneticists and anthropologists. The 1951 UNESCO statement clarified and refined the definition of race, pointing out that race was a classificatory device that allowed various groups of mankind to be arranged and evolutionary processes studied. It also stated that there were various populations that could not be easily fitted into a racial classification.9

Lewontin in context

Population and human genetics had already begun to cast doubt on the validity of prior racial classification schemes well before the 1970’s. In fact, Edwards’ 1964 co-author, the eminent biological anthropologist Luca Cavalli-Sforza, wrote a section in History and Geography of Human Genes (1994) entitled: “Scientific Failure of the Concept of Human Races” (section 1.6). It stated that individual genetic variation accumulated over long periods of time, that most polymorphisms antedate the separation of humans into continents, that the same polymorphisms are found in all populations within the species, and that the difference between groups is small when compared with that within major groups. It also stated that no single gene was sufficient for classifying humans into systematic categories. Yet, using the very same cluster analysis Edwards claims is required to discriminate humans into racial groups, they rejected the notion of their existence.

In addition Edwards and Leroi unfairly single out Richard Lewontin as the architect of the statistical fallacy at the center of “the dominance of social construction.” In the same year that Lewontin published his paper, Masatoshi Nei and Arun Roychoudhury published similar results.10 the general measurement of natural genetic variation, specifically to test classical theories of natural selection. For this reason, a series of papers were published all returning the same results and with similar analysis. It therefore seems problematic to focus a discussion of the fallacy of dismissing racial classification solely on Lewontin’s work. Nei and Roychoudhury were not considered radical scientists, thus it seems the singling out of Lewontin is more associated with a criticism of his political ideas, as opposed to any statistical fallacy he might have committed.

Finally, and most importantly, the legitimacy of social construction theory is entirely unrelated to the mathematical issues that Leroi raises in his discussion. Saying that legitimate ways to structure human populations exist does not say that the history of racial thinking applied to humans in biology and anthropology applied those same methods. Most galling was his claim of the ubiquity of the “social awareness” displayed by scientists studying human variation. Scientists differ in motive and ability just like people in any profession. Historically, scientists studying human racial variation have varied in political motivation from outright fascists to socialist revolutionaries. No scientist at any part of the spectrum was immune from socialized racial ideology. In the main, these scientists were from the socially dominant populations in their respective countries. The role that racism has played in American systems of social dominance need not be documented here. However, to think that social agendas and individual’s socialized racial ideologies have not and are not presently influencing research on human racial variation is at best hopelessly naïve, at worst dishonest. Ample past examples of this exist, such as the activities of the scientists associated with the Eugenics Record Office in the 1920’s and 30’s. To understand how social dominance is impacting modern research requires more sophistication and will be discussed below.

Population subdivision and biological races

One way to ask if there are significant genetic divisions within the human species is to ask whether our total genetic variation has substructure.11 Sewall Wright developed his hierarchical population statistics to measure genetic variation within a species (called F statistics.) These can examine variation at the level of breeding populations (demes), within regions (DR), regions within primary subdivisions (RS), and primary subdivisions within the total species (ST). Wright felt the latter, measured by Fst was equivalent to the subspecies used by taxonomists (also called biological or geographical race.) Population subdivision can be calculated at individual genetic loci or for numerous genetic loci simultaneously. Wright’s statistic can range between 0 and 1.00. He arbitrarily suggested that the minimal threshold for the existence of great variation was Fst = 0.250 and moderate variation Fst = 0.15 to 0.250. He examined individual loci derived from protein electrophoresis from a variety of species, finding a range of differentiation from 0.023 to 0.501 (average Fst = 0.168).

Subsequent studies of multiple loci, including whole genome analyses, have generally shown human Fst at much less than Wright’s critical value. They also show that variation within regions is smaller than variation between regions.12 Given this fact, it is hard to understand how anyone would claim that no genetic structure exists in the human species, yet at the same time, these data don’t say that there are natural divisions in our species, equivalent to biological races, or corresponding to our socially defined notions of race. Neither is quantification of the amount of genetic difference alone indicative of populations acting in ways associated with the evolutionary meaning of race. For example, if assortative mating is strong enough, populations with very little genetic difference can act as biological races, even within the same geographical region. This has appeared to happen in some insects. Corn borers have adapted over the last 500 years with the introduction of corn in Europe to form populations that feed on different plants and predominantly mate with members of their same ecological race. Fst values within races were 0.004 and between them was 0.132. The latter values are less than Wright's threshold. In this case, population subdivision is less because of less genetic polymorphism in insects, and the mating preferences result from strong selection on a few genes of major multiple effects related to host plant preference.13 Thus one can argue that a wide range of genetic differentiation is consistent with identifying biological races, depending upon the action of the specific loci involved, in the context of a given species biology. The task, of course, is showing that the genetic variation associated with human geographical distributions fits the bill.

Theoretical studies have further examined the limitations of Wright’s Fst as a measure of population divergence. One study showed the statistic is biased toward smaller values due to a failure of certain core assumptions utilized in its formulation, such as its assumption that the effective population sizes of all the subpopulations analyzed are equivalent and that subpopulations are evolving independently of each other (significant gene flow between regions violates this assumption.) Relaxing those assumptions allows Fst values to become larger. Yet, this doesn’t alter the fact that all human populations derive from a common ancestral group, have great genetic diversity, with a complex pattern of variation and no major discontinuities.14

It is instructive to compare human substructure to that of other closely related species. Other large bodied mammals show much higher population subdivision: white tailed deer (0.600), Grant’s gazelle (0.650) and North American gray wolves (0.750). Our closest relatives, chimpanzees and gorillas have more subdivision between their populations.15 It would be more legitimate to identify geographically based races in these species. These species have large differences between their subpopulations because human activity reduced their population sizes and fractionated their habitats. Conversely, anatomically modern humans have always maintained relatively large amounts of gene flow and are contiguous in habit.

Finally, the sampling schemes used in studies of human genetic variation limit their interpretation. To accurately represent the genetic diversity of the world’s people would require a systematic collection along geographic distance between world regions. In addition, within each region, suitable numbers of individuals would have to be examined, particularly to discover genetic variants that are present in low frequency. For example, studies by American drug companies often recruit people with ancestry from three regions, African Americans (representing sub-Saharan Africa), European Americans (representing various parts of Europe), and various Asian American groups. Sampling in this way ensures that individuals from these specific regions will cluster into three groups, simply because individuals from other portions of the spectrum of human genetic variation have been excluded from the study.

The meaning of social construction, biological variation, and their relevance to health disparity

The new drug BiDil has been hailed as a racial pill.16 It reduced the death rate from congestive heart failure in African Americans 43% compared to those given a placebo. BiDil is a combination of a nitric oxide donor isosorbide dinitrate and the anti-oxidant hydrazaline, which also acts as a vasodilator. Nitric oxide is a gas that plays a role in a variety of neurally mediated events including regulating heart processes, programmed cell death, as an anti-microbial agent, and even assisting penile erection in men. Anti-oxidants protect cells against oxidative damage that result from normal cellular respiration and poisons that accumulate over time. In addition, it has been recently shown that oxidative damage to human cells can be heightened by periods of prolonged stress.17 The African American Heart Failure Trial (A-HeFT) trial was motivated by studies showing that people self-identified as “black” had lower levels of available nitric oxide and greater amounts of oxidative stress than those self-identified as “white.”18

Actually, these results do not indicate that BiDil is a “racial” pill. What we know about the mechanism supports that assertion. Nitric oxide is synthesized by individual cells and this is catalyzed by an enzyme known as endothelial nitric oxide synthase (eNOS). Genetic variation at position G894T in this enzyme influences arterial stiffness (after controlling for sex, age, body mass index, insulin, heart rate, and mean arterial pressure).19 African Americans that had the T allele had less elasticity than those with the G allele. European Americans showed no significant difference between T and G, but the trend was similar. However, they also found that the frequency of T was 0.131 in African Americans v. 0.321 in European Americans, respectively. This of course means, if all other factors were equal, that more “whites” should have less elastic arterioles than “blacks.” If so, BiDil should help whites more than blacks, yet present data do not support this, meaning that other factors must be at play.

Social dominance creates different environmental conditions for the socially constructed races in America. A number of studies exist documenting the relationship of stress to lowered health outcomes. A recent study experimentally demonstrated a mechanism by which emotional stress could actually cause cellular damage.20 It found psychological stress was significantly associated with oxidative stress, lower activity of the enzyme telomerase, and shorter telomere length. Telomere lengths of women giving care to chronically ill children were significantly shorter than women who gave care to healthy children. Finally, telomere length in chronically stressed groups was proportional to the years of care giving and to the perceived amount of psychological stress.

This result could revolutionize research into chronic illness. Individuals experiencing chronic racialized stress should also show shorter telomere lengths. Racialized stress increases the probability of pre-term and low birth weight deliveries and negatively affects the mental health of pre-school children.21 BiDil may work for African American patients because they have greater oxidative damage in their cells, due to chronic stress. This would mean that the drug is acting on an environmentally induced difference, not a genetically based one. If the drug were used in Western Africa, where Africans face less racialized stress and a variety of environmental factors differ, we may not observe any “race-specific” effect.


Human genetic variation is real. It is best described by isolation by distance, meaning that individuals who have ancestry in particular geographic regions are more likely to share genes than those from disparate regions. The overall amount of measured human genetic variation, however, is very small, yet this does not mean that it cannot be categorized. This is facilitated for individuals by using multiple loci particularly when they are examined at the level of DNA sequence variation. This greater “signal,” while allowing the ancestry of individuals to be readily determined, may be discordant with any particular phenotypic trait (physical features) of interest, especially since much of the classification salience originates from DNA that does not influence the phenotype.


1 Edwards, A.W.F., “Human Genetic Diversity: Lewontin’s Fallacy,” Bioessays 25: 798-801.

2 Graves, J.L., The Emperor’s New Clothes: Biological Theories of Race at the Millennium (New Brunswick, NJ: Rutgers University Press, 2001), pp. 64 – 71.

3 Ibid., pp. 37 – 51.

4 Montagu, A., Man’s Most Dangerous Myth: The Fallacy of Race, 3rd Edition, (New York: Harper and Brothers Publishers, 1952), p. xvi.

5 Cavalli-Sforza, L.L. and Edwards, A.W.F., “Analysis of human evolution,” Proc. 11th International Congress of Genet. 2: 923-933, 1964. The two figures I refer to in the text are reprinted in Cavalli-Sforza, Menozzi, and Piazza, The History and Geography of Human Genes (Princeton, N.J.: Princeton University Press, 1994) on pp. 68 and 71.

6 Podolsky, R.H. and Holtsford, T.P., “Population structure of morphological traits in Clarkia dudleyana I. Comparison of Fst between allozymes and morphological traits,” Genetics 140: 733 – 744, 1995.

7 Brace, C. Loring, “Region does not mean ‘race’--reality versus convention in forensic anthropology,” J Forensic Sciences 40(2): 171-5, 1995. Wright, S., Evolution and the Genetics of Populations, Volume 4: Variability within and among natural populations (Chicago: U. Chicago Press, 1978), pp. 449-50.

8 UNESCO, Race and Science: The Race Question in Modern Science (New York: Columbia University Press, 1961) and “No scientific basis for race bias found by world panel of experts,” The New York Times, July 17, 1950. The list of notable geneticists who reviewed the 1950 UNESCO statement before its publication included E.G. Conklin, Th. Dobzhansky, L.C. Dunn, Julian S. Huxley, and Hermann J. Muller.

9 See discussion of this in Graves 2001, pp. 149 – 51.

10 Nei, M. and Roychoudhury, A., “Gene differences between Caucasian, Negro, and Japanese Populations,” Science 177: 434-435, 1972; with a follow up paper, “Genic variation within and between three major races of man, Caucasoids, Negroids, and Mongoloids,” Am. J. Human Genetics, 26:421-443, 1974.

11 Wright, S., Evolution and the Genetics of Populations, Volume 4: Variability within and among natural populations, (Chicago: U. Chicago Press, 1978).

12 Many studies have returned this result, including: Bowcock, A.M., Kidd, J.R., Mountain, J.L., Hebert, J.M., Carotenuto, L., Kidd, K.K., and Cavalli-Sforza, L.L., “Drift, admixture, and selection in human evolution: A study with DNA polymorphisms,” Proc. Natl. Acad. Sci. 88: 839–843, 1991; Tishkoff, S.A., Pakstis, A.J., Stoneking, M., Kidd, J.R., Destro-Bisol, G., Sanjantila, A., Lu, R.B., Deinard, A.S., Sirugo, G., Jenkins, T., et al., “Short tandem-repeat polymorphism/alu haplotype variation at the PLAT locus: Implications for modern human origins,” Am. J. Hum. Genet. 67: 901–925, 2000; Akey, J.M., Zhang, G., Zhang K., Jin L., and Shriver, M., “Interrogating a High-Density SNP Map for Signatures of Natural Selection,” Genome Research 12:1805 – 1814, 2002; Tishkoff, S. and Kidd, K., “Implications of biogeography of human populations for ‘race’ and medicine,” Nature Genetics Supplement, No. 11, November 2004, pp. S21-S27.

13 Malausa, T. et al., “Assortative mating in sympatric host races of the European corn borer,” Science vol. 308, 8 April 2005: 258-260.

14 Long, J.C. and Kittles, R., “Human genetic diversity and the non-existence of biological races,” Human Biology 74(4): 449 – 471, 2003.

15 Kaessman, H., Wiebe, V., and Paabo, S., “Extensive nuclear DNA sequence diversity amongst chimpanzees,” Science 5 November 1999, Vol. 286: 1159-62.

16 Marchione, M., “Heart failure drug for blacks expected to become first pill sold for a specific race,” Associated Press, November 9th, 2004.

17 Taylor, A. et al., “Combination of isosorbide dinitrate and hydralazine in Blacks with heart failure,” The New England Journal of Medicine 351 (20): 2049-57.

18 Ibid.

19 “Nitric oxide synthase gene polymorphism (G894T) influences arterial stiffness in adults: The Bogalusa Heart Study,” American Journal Hypertension 2004 17(7): 553-9.

20 Epel, E.S. et al., “Accelerated telomere shortening in response to life stress,” Proceedings of the National Academy of Sciences, 101(49): 17312-15, 2004.

21 Caughy, M.O., O’Campo, P.J., and Muntaner, C., “Experiences of racism among African American parents and the mental health of their pre-school aged children,” American Journal of Public Health, 94(12): 2118-24, 2004; Mustillo, S. et al., “Self-reported experiences of racial discrimination and Black-White differences in preterm and low-birthweight deliveries: The CARDIA Study,” American Journal of Public Health,; 94(12): 2125-31, 2004; Collins, J.W. et al., “Very low birthweight in African American infants: The role of maternal exposure to interpersonal racial discrimination,” American Journal of Public Health 94(12): 2139-48, 2004.