Race Unmasked: Biology and Race in the Twentieth Century

In 1998 the public health psychiatrist Mindy Fullilove published an article in the American Journal of Public Health challenging the utility of race as a variable in health research, pointing to the risks of relying on an archaic and imprecise way to organize human diversity and calling upon the field “to abandon race as a variable in public health research.” The race concept “is an arbitrary system of visual classification that does not demarcate distinct subspecies of the human population,” Fullilove asserted, arguing that the concept could not “provide the information we need to resolve the health problems of populations.” In its place, she proposed that public health professionals “invent a new science that embodies the human rights and civil rights essential to the health of human populations.” Although Fullilove acknowledged that “‘race’ is an ingrained part of personal identity,” she believed that abandoning “racial classification schemes is to challenge deeply held and socially endorsed ways of seeing one’s place in the world.” Fullilove’s proposition was as much a call to science to find new methods to study the relationship between health outcomes and human diversity as it was to remind her fellow scientists that “following the illusion of race cannot provide the information we need to resolve the health problems of populations.”¹

Fullilove was not, of course, the first scientist or scholar to call for rethinking the scientific methodologies that measure human diversity. At the turn of the twentieth century, W. E. B. Du Bois called into question the scientific legitimacy of the biological race concept in The Health and Physique of the Negro American.2 In the early 1940s the anthropologist Ashley Montagu called for an abandonment of the race concept in his classic work Man’s Most Dangerous Myth.³ The evolutionary biologist Theodosius Dobzhansky, who had in the 1930s and 1940s been largely responsible for reimagining and preserving the race concept in biology in his work on the evolutionary synthesis, came to believe by the 1960s that “the problem that now faces the science of man is how to devise better methods for further observations that will give more meaningful results.”⁴ And in 1972, the evolutionary biologist Richard Lewontin wrote that human racial classifications have “virtually no genetic or taxonomic significance.”⁵ Despite these calls, the population genetics–based race concept as first articulated by Dobzhansky is alive and well in the biological sciences.

Part of the challenge of situating this history in present-day debates about race is that, in general, unlike pre-twenty-first-century arguments about the race concept in biological thought, today’s debates generally lack the clear ideological and political antipodes of racist and nonracist, reactionary and liberal. Instead, today’s scientific debates take place (with some exceptions, of course) in a way that seeks to avoid the ideological and racist baggage of the past. Claims for race-based medicine in the genomic era are allegedly about drug safety and efficacy, reducing health disparities, and ultimately bettering human health and saving lives.⁶ Most of the discussion today about race in biology concerns the noble goal of utilizing an individual’s self-identified ancestry to help determine the best course of medical treatments as we move closer to personalized medicine. Although there is a legitimate fear that this type of research will serve only to reify racial categories and harden racial prejudices, that is not, in most cases, its purpose. The question as it is often framed in today’s debates about race is not whether racial classifications are harmful—and this is with an acknowledgment that race is a social construction—but whether race is a valuable biological variable and whether we can best improve the health of individuals and populations by continuing to use this variable.

In public health, the attention to health disparities has further deepened the interest of public health professionals in the race concept. Some in public health have chosen to integrate a socially constructed race concept into novel methodologies designed to study health disparities, while others have drawn on quasi-typological ideas of race to understand the relationship between biology and health. The work of Nancy Krieger at Harvard, for example, rejects the biological race concept but utilizes race in a novel methodology that she calls the “Public Health Disparities Geocoding Project.” Her method seeks to understand the underpinning of racial disparities in health by examining the underlying socioeconomic causes of disparities. In order to overcome a virtual “absence of socioeconomic data in most U.S. public health surveillance systems,” she has developed a method that utilizes geocoded residential addresses and area-based socioeconomic measures.7 Others have made alternative proposals. Ichiro Kawachi, Norman Daniels, and Dean Robinson argue that “historical, political, and ideological obstacles” have “hindered the analysis of race and class as codeterminants of disparities in health.” They propose that “racial and class identities are mutually constitutive,” as race is “neither a biologically meaningful category nor a proxy for class, but is a separate construct from class, more akin to caste.” They call for a strengthening of the U.S. data infrastructure “to improve the measurement of race/ethnicity as well as class/socioeconomic status” so that policy makers can better “monitor the effects of their policies on health disparities.”⁸ Still others have called for more clarity in the use of race in health research. “At a minimum,” R. Dawn Comstock, Edward Castillo, and Suzanne Lindsay argue, “researchers should clearly state the context in which these valuable epidemiologic and public health research variables are being used, describe the method used to assess these variables, and discuss all scientific findings.” Doing so, they believe, “will ensure continued constructive dialog about the interpretation of findings regarding race or ethnicity.”⁹

Finally, epidemiologist Camara Jones argues that racism, not race, may be a “root cause of observed race-associated differences in health outcomes.” Race, according to Jones, “is a social classification in our race-conscious society that conditions most aspects of our daily life experiences and results in profound differences in life chances.” An understanding of what she calls the three levels of racism—institutionalized racism (“differential access to the goods, services, and opportunities of society by race”), personally mediated racism (“prejudice and discrimination, where prejudice means differential assumptions about the abilities, motives, and intentions of others according to their race, and discrimination means differential actions toward others according to their race”), and internalized racism (“acceptance by members of the stigmatized races of negative messages about their own abilities and intrinsic worth”)—may offer a better understanding of the etiology of health disparities as well as concrete strategies to redress them.10

The latest ways to utilize race and begin to rethink how to create a measurement variable from human genetic diversity are, despite their best intentions, overwhelmed in the larger research enterprise by studies that claim that race is, in fact, a methodologically sound classificatory tool in the biological sciences.¹¹ Continued claims that race is, in fact, a useful tool for classification is driven by four factors. First, the reductionist ontology that underlies genomic science has given rise to the belief that biology is destiny, and that both at the individual and population levels, genes hold the ultimate information for understanding our health and understanding who we are. The philosopher of science Sahotra Sarkar worries that such reductionism in genetics “attempts to explain phenotypic properties from a genotypic basis without attributing any particular structure to the genotype.”¹² Claims that race is a meaningful biological classificatory tool do make sense as an extension of reductionism as the line between phenotype and genotype is blurred.

Second, genomic technology has enhanced our ability to examine the 0.1 percent of nucleic acids in the human genome that, on average, vary between individuals. Over the past decade, the Human Genome Project and the technological revolution that spawned it are driving novel research to examine these differences. Many of these 0.1 percent differences will involve phenotypic variations including hair color, body shape and size, skin tone, disease susceptibility, and blood type. The 0.1 percent of variations between people can also be used to study both individual and population-level traits, including disease risk. However, this is still prohibitively expensive to do for everyone, so scientists are instead using population and ethnic groups as well as races as proxies to measure susceptibility and risk. One area where we are seeing some of the earliest manifestations of such an approach is in pharmacogenomic testing—a way to predict, based on an individual’s genotype, how one will react (either negatively or positively) to a particular drug. Will drug A, for example, be an effective and safe treatment for patient Y, or will it be toxic and nonefficacious? But, for the moment, because of the expense of sequencing individual genomes, scientific studies are often relying on an individual’s race to make pharmacogenomic predictions.

The problem with such an approach, as the bioethicist Sandra Soo-Jin Lee points out, is that it inevitably “suggests that the population under study is assumed to be homogeneous with respect to allele frequencies but in fact comprises subpopulations that have different allele frequencies for the candidate gene.”13 In other words, as Dobzhansky tried to teach us many years ago, “racial differences are more commonly due to variations in the relative frequencies of genes in different parts of the species population than to an absolute lack of certain genes in some groups and their complete homozygosis in others.…Individuals carrying or not carrying a certain gene may sometimes be found in many distinct races of a species.”¹⁴ It is therefore not accurate to make assumptions about an individual’s genes based on their race. Race as a proxy predictions will sometimes be accurate if the gene frequencies for, let’s say, a drug’s metabolism are high enough in a particular racial group. But inevitably, many will receive no benefit or may even be harmed based on these assumptions.

The 0.1 percent difference between people is also being used to promote a closer examination of genetic ancestry. A simple Google search reveals companies that claim a pretty simple correlation between your genes and your race. Although it doesn’t use the term “race,” one company allows you to “discover your estimated percentage of ancestry from four different population groups: European, Indigenous American, Sub-Saharan African, and East Asian.”¹⁵ Another company’s promotional video claims that DNA “is the key to discovering where your family is from, and learning about the places and culture that make you who you are.”¹⁶ Critics, however, have pointed out that these tests “cannot pinpoint the place of origin or social affiliation of even one ancestor with exact certainty.”¹⁷ Keith Wailoo, Alondra Nelson, and Catherine Lee also worry that “the mere hint that genetic markers are distributed in different frequencies across populations has led some people to quickly treat such variation as a proxy for racial and ethnic differences, lending renewed authority to biological conceptions of human difference.”¹⁸

Third, the critical task of understanding and reducing health disparities has researchers looking at all possible explanations, including genetic ones, for disparities in health outcomes. The renewed focus on race and genetics suggests, however, that an analysis of the complex relationship between individuals, populations, and health has been surrendered to a simplistic, racialized worldview. An inability to digest (and frequently even to acknowledge) these complexities restricts scientific theory and practice to simplicity when complexity is needed. This underlies the drive to correlate race, genetics, and health disparities.

Fourth and finally, the history of the biological race concept demonstrates that race is deeply embedded in scientific and social thought, and that racialized thinking was an integral part of genetics in the twentieth century.19 This history has shaped scientific thinking about human difference as well as popular thinking about that difference. That Craig Venter’s criticisms of the race concept at the outset of the twenty-first century are similar to those made by W. E. B. Du Bois at the beginning of the twentieth—calling into question both the utility of racial classifications and the claim that health disparities can be explained by elucidating the relationship between genetics and race—suggests that we are having frustratingly similar arguments about race and human difference despite the benefit of one hundred years of knowing better.²⁰

In light of this history, we should not underestimate the tenacity of the race concept. As scientists continue to struggle with the meanings of race, they should remember Dobzhansky’s paradoxical thinking about the race concept: on the one hand, Dobzhansky believed the race concept to be a critical methodological tool for biologists to make sense of genetic diversity within species.21 On the other hand, Dobzhansky understood the imprecise nature of the term, its limited utility, and its potential for abuse. But despite these acknowledged contradictions, Dobzhansky held fast to and publically defended the concept for much of his career. Toward the end of his career he struggled to disown this paradox. He did this most probably because of the ways in which he saw the biological race concept being appropriated in the early years of the civil rights movement by those fanatically opposed to African American civil rights, because of disputes with other scientists who he believed were misusing the race concept in their work, and ultimately because he seemed to realize that he was fighting a Sisyphean battle in trying to separate what he had formerly considered the separate domains of the scientific and social meanings of the biological race concept. Nevertheless, genetics continues to function within Dobzhansky’s paradox. Despite his brilliance as an evolutionary theorist, Dobzhansky left us with only a warning and not a new method to study human diversity. History has shown that even acknowledging that race has a social meaning cannot disconnect the concept from its typological and racist past (or present). Despite the best intentions of scientists and scholars, race will always remain what Ashley Montagu once called a “trigger word.…Utter it and a whole series of emotionally conditioned responses follow.”²²

And so while the race concept can provide us with some information about health and human evolutionary history, ultimately that information is incomplete. What we know from Dobzhansky, his sometimes collaborator L. C. Dunn, the evolutionary geneticist Richard Lewontin, and, most recently, Craig Venter, is that race is not a particularly useful measure of human genetic diversity. We also know from Du Bois and other historians and social scientists that the biological race concept is inseparable from its social history, and that race can be harmful to an understanding of both human health and evolutionary history.

We are a genetically diverse species, and there is meaning in that diversity. But as a species we seem thus far unable to reliably distinguish between the scientific ramifications and the social meanings of human difference. “Race” is a historical, not a scientific, term. Yet until the scourge of racism is eliminated from our lives and institutions, developing scientific methods unburdened by racial ideology to study human difference will be limited by the historical nature of the race concept itself.

In 1940, toward the end of his career, W. E. B. Du Bois penned his autobiography, Dusk of Dawn: An Essay Toward an Autobiography of a Race Concept. David Levering Lewis, Du Bois’s biographer, believes that the book was intended to be “not so much the story of his own life, then, but the autobiography of the twentieth century as it had lived through him.”23 The book is thus, in large part, both a personal account of the meaning of race in Du Bois’s own life and an evaluation of its political, sociological, and scientific meanings over the course of his lifetime. Du Bois understood that scientific concepts of race had played a critical role in both providing support for racist claims and in buttressing the racial order, writing in Dusk of Dawn that he “had too often seen science made the slave of caste and race hate.” Yet Du Bois also understood that the world and the race concept had changed, in some ways for the better and others for the worse, over his lifetime. “Life has its pain and evil—its bitter disappointments,” Du Bois wrote at the book’s close, “but I like a good novel, and in healthful length of days, there is infinite joy in seeing the World, the most interesting of continued stories, unfold, even though one misses THE END.”²⁴

In our own time, we continue to struggle with the power of the race concept. Biologists may today develop new ways of studying human populations that, to whatever degree, distance themselves from the race concept and its historical baggage, and those new methods may, in fact, be an improvement over where we stood before. Yet we should not be waiting for or expecting science and scientists to change our thinking about race. Science may have helped bring us to this point, but it is unlikely to extract us from it.