Race Bibliography

What does race have to do with genetics? Can humans be categorized according to the color of their skin and shape of their face? Though for much of human history, race has been considered a topic of social problems, blamed for ills ranging from crime to illegitimate birth, the argument that race represents a tangible biological divide between humans has been used as an excuse to defend both the privileged status of whites and the mistreatment of those in minority groups. About a century ago suspicions about innate differences between human races developed into a movement, known as eugenics, that spawned sweeping and destructive government policies like those implemented in Nazi Germany, during World War II. The science of eugenics was deeply flawed and later discredited, and today we benefit from a wealth of knowledge about human origins that has put the concept of race as a biological construct to rest. Well, sort of. Today, researchers and policy makers are leading a march back in time on a topic that has been both reviled and celebrated as an answer to the vast economic and social disparities present in our society. Recent articles in the popular press have reported an increase in the number of studies examining whether innate differences exist between “racial” groups in areas as disparate as drug responsiveness and intellectual ability. When the arguments are examined closely, however, it is clear that confusion tends to rule the discourse. 

In a March 2005 Op-Ed piece in the New York Times entitled “A Family Tree in Every Gene.” Armand Marie Leroi, the author and a British biologist, defends the use of race by medical researchers as a proxy for genetic identity. In his discussion, however, he mixes up the term race with references to local tribes or populations, managing only to confuse his audience. Demonstrating how antiquated notions about race continue to belie real scientific understanding Leroi speaks in one sentence of the “continental races” and in another of how Hispanics are made up of an “evolving blend of European, American Indian and African genes.”[1] Which is it?

Much of the confusion surrounds the term itself. What is race and how is it defined? What relevance does race have to science, if any? These are questions that help to frame a debate that is often confusing, where words such as ancestry, population, and ethnicity are bandied about and superimposed upon one another often inappropriately by researchers and journalists alike. Members of the public are faced with conflicting headlines such as “DNA Test in LA. Killings Is Said to Have Indicated Attacker’s Race,”[2] or, “Race is the biggest reason America treats its poor more harshly than any other advanced country,”[3] which highlight the debate between nature and nurture, but offer no guideposts to discerning it. Within communities of scientists there seems to be little agreement on the semantic platform as to a common definition. For example, some researchers use the term “Latino,” to signify a group of individuals who share a common racial background. The designation, however, references groups originating from several continents each of which has a different history of colonization, immigration, and groups of indigenous residents. Are these last criteria sufficient grounds from which to extrapolate a common genetic lineage?

If the debate were restricted to a few quibbling biologists continuously re-drawing lines in the sand, the issue would be far less provocative and dangerous. However, extreme perspectives are re-emerging in concert with speculations like those made by Leroi. The Chairman of Britain's Commission for Racial Equality, Trevor Philips, was quoted in theEconomist recently suggesting performance disparities between black and white children in British schools might be solved by segregation.[4] Though tremendous sociological evidence collected over the past several decades alludes to multiple complex environmental factors implicated in such disparities, statements such as Phillips’ represent a faction of contemporary opinion focused solely on innate biological causes for racial differences the roots of which are quite old.

What is Race? Definitions From Yesterday and Today

The eugenics movement, spawned by Sir Francis Galton, a 19th century scientist who conducted research aimed at the improvement of the human race through controlled breeding practices, was aimed at denying criminals, “barbarous” individuals, and those with less than ideal characteristics from reproducing. It is clear that such a movement in the guise of science would open the floodgates for racist, sexist, and other discriminatory practices. Galton himself referred to individuals with darker skin color, “negroes” as having a stubborn tendency to continue reproducing even after being exposed (through slavery) to civilized society, unlike other animals, who in captivity tend to greatly reduce their reproductive practices.[5] Clearly in comparing those of African descent to captive animals in a zoo, he was not so subtly revealing his concept of where such individuals were ranked in an imagined taxonomy of the human species. Indeed, Galton’s conceptual framework became a springboard for the primary movements of racial and ethnic cleansing implemented in the early twentieth century in Germany, Scandinavia, and the United States.

Eugenicists identified individuals from different groups based on physical features such as skin color, nose shape, and hair texture. Anthropologists also used a system based on outward appearance when they grouped humans into the five racial categories, Negroid, Caucasoid, Mongoloid, Capoid, and Australoid. Today we distinguish between human groups in much the same way, through outward physical characteristics. This suggests that not much has changed regarding concepts of human race and current science supports the notion that variation between groups is mostly skin-deep. However, race is also socially constructed and designations can be used to describe an orientation to religion, culture, socio-economic status or historical event. The first definition may overlap with the second or not at all, depending on the context.

Taking the above definitions into account, race has traditionally been viewed by biologists as having no strong genetic basis. This position has arisen from over thirty-five years of research on human genetic variation and the human genome project. These analyses have shown that race is too broad a category with which to classify humans biologically. 

R.C. Lewontin, a Harvard scientist, explains in his essay “Confusions About Human Races,” that 85% of human genetic variation occurs within any given population, such as those of European descent. About half of the remaining 15% is attributable to variation between local populations, such as the French and the Ukrainians, whereas an unclear 6-10% defines variation between the more “classically defined geographical races.” (e.g., African versus European). Lewontin maintains that the exact proportion of this variation is unclear due to the inherent problems in attributing groups to racial categories.[6] For example citizens of the country of Turkey look “European” but speak an Asiatic language. How should they be categorized? Information about these geographical races is found by utilizing a type of non-functional DNA know as microsatellites which are markers that define the small and obvious differences between populations such as skin color, hair color and texture, and facial characteristics.

L.B. Jorde & S.P. Wooding of the University of Utah Department of Human Genetics, assert in a November 2004 article published in Nature Genetics that “the average nucleotide differences between a randomly chosen pair of humans is consistently estimated to lie between 1 in 1,000 and 1 in 1,500. This proportion is low compared to many other species…reflecting the recent origin of our species from a small founding population.”[7] There is no evidence to suggest that the small number of markers isolated in some groups implicates for differences beyond mutations for a handful of diseases or outward physical appearance, yet the use of race in genetics is often stretched to areas where it does not belong.

But even within the parameters that define the tiny biological differences between groups, race is not useful as a proxy to judge these differences. This is true for a couple of reasons. First, a person’s apparent race often does not reflect the genetic diversity represented in their biology. For example, though they are of one race, many African-Americans have descendants from Europe and the Americas as well as Central and West Africa. Their “race” is considered to be African-American, but their ancestry is mixed. Ancestry refers to an individual’s line of descent that can be traced through ancestors back to a geographic region(s) of origin. A given individual may have ancestors who all originated from the same region, but this is rare. Further, each individual’s ancestry is different and it is possible for individuals from the same “race” to have very different ancestries. What you see isn’t always what you get. Second, it is possible, even likely, that a given individual will not know the extent to which they are genetically mixed. Medical advice and clinical studies based on information about “race” from self-reports or by judgments about “race” made by doctors and researchers, therefore, can only lead to erroneous conclusions.

Can Ancestry Information Be Used to Assess Disease-Risk?

The majority of studies have found that there is some evidence that genetic mutations implicated in a few disease clusters within populations have been geographically isolated under certain environmental stressors. An example of this is the higher incidence of sickle cell anemia within individuals of Sub-Saharan African and Mediterranean descent.[8] Those who are homozygous for this mutation (having alleles from both parents) have an increased susceptibility to the disease. Those that are heterozygous (having an allele from only one parent) are considerably less susceptible to the disease but have an increased resistance to malaria. Selective pressures over long periods of time within isolated populations can allow for small mutations such as this.

Should researchers develop drugs tailored to a groups’ disease-risk? Possibly, the most well-known attempt at this type of medicine can be found in the recent approval by the U.S. Food and Drug Administration of a heart failure medication for African-American patients. BiDil, the first drug ever developed for a specific race, is a combination of two generics already on the market that have been repackaged and marketed by the Massachusetts-based pharmaceutical company NitroMed. BiDil’s success came after a failure in its initial trial to show any effect on a broad sample of heart failure patients several years ago.

In 1999, University of Minnesota researchers reexamined the data from this trial and found that the effectiveness of the medication was considerably greater for those individuals in the sample who self-identified as black.

Thus a second study was launched, only this time the researchers required study participants to identify their racial background. Participants claiming to be African-American were invited to join the trial. Though the data from this second study showed a significant effect for the group, it is interesting that participants from other racial groups were not included. In the November 2004 issue of Nature Genetics, biologist D.B. Goldstein of the University College, London, and his colleague S.K. Tate stated "Many differences in drug response associated with race or ethnicity are due to environmental factors [such as diet] rather than population genetic differences…In the case of BiDil, it is not currently known whether it works differently in African Americans and European Americans because of genetics, environment, or both.”[9]

Tracking disease risk utilizing ancestry information from patients presents serious problems. Though the probability of determining risks for certain diseases based on ancestry information provided by a patient may increase, the probability that risks for other diseases will not be accurately calculated or that they will be missed altogether also increases. This is due to the fact that most individuals do not know their full ancestral background. Therefore, it remains very important that physicians continue to press firmly for lifestyle, family history, and other background information from patients rather than screening for ancestry.

In choosing candidates for its study based on self-identified race, NitroMed proved nothing about a genetic predisposition to certain types of heart disease in this population. Would the participants be able to accurately identify their ancestry across the board? It is not clear whether the study participants were asked on what basis they self-identified as being of African descent. Due to the fact that most individuals are unaware of or mistaken about the nature of their varied and complex genetic lineages, (in the US, many self-identified African-Americans have as much or more “European” ancestry), it seems unlikely that this study sample represented clear information from a genetically isolated “race.”

Adding to the confusion are the unknown effects of discrimination and racism on an individual’s health. A recent study published in the New England Journal of Medicine [10] suggests that they may be quite severe. A.K. Jha, a Harvard School of Public Health professor and lead researcher for the study, cites findings which suggest that even when access to health care is controlled for, Black patients consistently receive different treatment than White patients: “We’re not talking about small differences; we’re talking about substantial differences that have a very profound impact on whether people live or die, what quality of life they have. And in the year 2005 [these inequities] really should not be something we should be willing to live with,” states Jha in an interview on National Public Radio. [11] Sometimes these differences come in the form of fewer life-saving surgeries such as heart bypasses. Other times access and advocacy are an issue. Appearance serves as a proxy for different treatment by physicians, the study showed.

New York University professor T. Duster, points out in his essay, “Race and Reification in Science,” that in heart failure patients, Americans of African descent show much higher rates of hypertension than Americans of European descent. Additionally, he points out, darker skinned blacks show higher rates of hypertension than lighter skinned blacks. [12] It is not currently known what the reasons for this difference are or if drug treatments such as BiDil are more or less effective based on this etiology. Notably, black individuals of Caribbean descent do not show a significant response to BiDil. Mixed pieces of evidence, such as these, combine to form a muddy picture in terms of racial differences, disease, and response to drugs. Questions such as whether BiDil works only for African-Americans have not been answered.

DNA Identification of Criminals and Commercial Tests of Ancestry

Databases put together by the Federal Bureau of Investigation currently match forensic DNA to samples based on race. Prosecuters attempt to determine the likelihood that a given suspect committed a crime by calculating the frequency that a particular set of markers present in the forensic sample would occur within the suspect’s assigned “racial” group. This frequency is calculated against samples catalogued in the FBI database according to “race.” For example, if a witness to a particular crime judges an unknown suspect to be of “South Asian” descent, law enforcement officials would choose to match the DNA evidence collected from that crime scene to a database made up of samples collected from “South Asian” individuals. From here they would calculate the frequency of a particular set of markers thought to be “South Asian” to verify the race of the suspect.

There are a couple of problems with this method. First, the assumption that an eyewitness can correctly identify a suspect’s race is erroneous, because many citizens of the United States possess mixed ancestries. Notably, African-Americans, the most frequently incarcerated group, do not have a “pure” genetic lineage. R. Schwartz in an editorial for the May, 2001 edition of the New England Journal of Medicine states, “After 400 years of social disruption, geographic dispersion, and genetic intermingling, there are no alleles that define the black people of North America as a unique population or race.”[13] Any black individual in the United States might be identified by an eyewitness as African-American however, or catalogued by law enforcement in the same way.

But this concern applies to whites as well. R. Hubbard points out in Exploding the Gene Myth, that Caucasians have immigrated to the U.S. from all over Europe at different times over the last few hundred years. At any given time, this “racial” group may be represented by those whose ancestors came from small villages in different parts of the continent which remained geographically isolated, or from large industrial cities boasting populations which were considerably more mixed. Inter-breeding “racially” has occurred in these centers or in the U.S. and may have been happening for centuries or a few years. [14]

This highlights the second major problem with this forensic identification method. If the FBI is assigning race to perpetrators and database samples based on appearance or even self-identification they are not taking into consideration the fact that these individuals may be mixed or from an entirely different set of ancestral groups altogether. Therefore, samples from other “races” or categories which might be represented in the perpetrator’s and the forensic DNA are not considered. This makes the calculations of frequency inaccurate and could easily lead to the incarceration of an innocent suspect.

There are newer methods utilizing polymorphisms, microsatellites, and chromosomal markers to identify the ancestry of a perpetrator that work in the reverse direction. Through a process called admixture mapping, scientists search for very small portions of DNA, known as “Ancestry Informative Markers” (AIMs)[15], that vary between populations. These markers can provide a sketch of the different population percentages inherited by an individual from his or her descendants. Law enforcement officials recently used admixture mapping in 2003, to convict Louisiana serial killer, Derrick Todd Lee, a self-identified African-American truck driver from Baton Rouge. Based on eyewitness reports, police were searching for a white man in connection with the crime. When biological material collected from the crime scenes was analyzed by DNA Print Genomics Inc. [16], they were told that their suspect was actually an individual with 85% African and 15% Native American ancestry. Based on this information, police began to look for suspects with darker skin color. Fortunately, Lee had voluntarily submitted a sample that later proved a match. [17] Though, in this case, ancestry analysis of DNA proved a useful guidepost, it should not be confused as a proxy for identifying a perpetrator’s “race”. Though the test in this case was used to determine the suspect’s ancestry, the word “race” was used liberally in the news coverage of the case as if the two terms are interchangeable. One article included the headline, “DNA Test in Louisiana Killings Is Said to Have Indicated Attacker’s Race.” [18] Confusing the terms in this way only adds to the false notion that race is biological. Additionally, there is the possibility of making incorrect assumptions regarding physical appearance

Tests for determining one’s “hereditary profile” are now commercially available to consumers. Companies such as Ancestry By DNA, [19], [20] and Family Tree DNA, [21] offer services to assist individuals in tracing their ancestors’ geographic origins. The New York Times ran an article recently entitled, “Blacks Pin Hope on DNA to fill Slavery’s Gaps in Family Trees,” [22] published in the Science section which demonstrated why the tests are so popular. One woman who was interviewed said that her test showed that her African roots were quite old. Her light skin color gave her lower status among members of her community, but the test, she said, “showed underneath I’m deepest Africa.” However the tests do present some problems. For example, Ancestry by DNA claims that they are able to map a person’s ethnicity to among four anthropologically defined groups: Native American, East Asian, Sub-Saharan African, and European. Europeans are traditionally thought of as white, but “European” in this instance refers to individuals with continental European, Middle Eastern, Eurasian, Central Asian, and South Asian descent. For additional money, the “EuroDNA” test will break down the percentages of different ethnicities within that group to four other categories: Northern European, Southeastern European, Middle Eastern, or South Asian. It is easy to see how this would be confusing to clients. Further review of the websites found that fine distinctions such as these are not made clearly enough. It is important for consumers to understand that these tests cannot identify race. Confusion between the categories race, ethnicity, and population is something that professionals and scientists in the field of genetics should be trying hard to combat.


Current research in medicine and forensics, and improved techniques in genealogical analysis promise hefty future profits. We are beginning once again to view race as a “holy grail” in the search for truth, hoping that it will yield answers to our biology and ultimately our identity. There is a general sentiment that studying the connection between race and genetics this time around will not produce the sinister effects of the past, but is this true? We may be trudging down an erroneous path, one that will lead to false notions of biological differences and the delay of efforts to remediate inequalities whose origins lie mainly in environmental and social factors. By spending billions in research on race-based approaches to medicine and criminal identification we are furthering the concept that race is real.



1 Leroi, A. M. 2005. A family tree in every gene. New York Times, March 14, New York Edition.

2 Ritter, Malcolm. 2003. DNA test in LA. killings is said to have indicated attacker’s race. Boston Globe, June 5.

3 Krugman, P. 2005. Race is the biggest reason America treats its poor more harshly than any other advanced country. New York Times, September 19, New York Edition.

4 Eds. 2005. Black Marks; race and education.The Economist. March 10. U.S. Edition.

5 Galton, F. 1904. Eugenics: Its definition, scope and aims. The American Journal Of Sociology. X(1):

6 Lewontin R. C.  2005. The fallacy of racial medicine: confusions about human races. Social Science Research Council. (accessed October 4, 2005).

7 Jorde, L. B. and S. P. Wooding. 2004. Genetic variation, classification, and ‘race’. Nature Genetics. 36(11).

8 Lewontin R. C.  2005. The fallacy of racial medicine: confusions about human races. Social Science Research Council. (accessed October 4, 2005).

9 Goldstein, D. B. and S. K. Tate. 2004. Will tomorrow’s medicines work for everyone? Nature Genetics. 36(11).

10 Jha, A. K.; E. S. Fisher; Z. Li; J. Orav; and A. M. Epstein. 2005. Racial trends in the use of major procedures among the elderly. New England Journal of Medicine. 353(7).

11 Ashish K.Jha, interview by Terence Smith for National Public Radio, August 18, 2005. transcript,

12 Duster, T.  2005. Race and Reification in Science. Science. 307(5712).

13 Schwartz, R. S. 2001. Racial Profiling in Medical Research. The New England Journal of Medicine. 344(18).

14 Hubbard, Ruth and Elijah Wald. 1999. Exploding the Gene Myth. Boston: Beacon Press.

15 Yang, N.; H. Li; L. A. Criswell; P. K. Gregersen; M. E. Alarcon-Riquelme; R. Kittles; R. Shigeta; G. Silva; P. I. Patel; J. W. Belmont; and M. F. Seldin. 2005. Examination of ancestry and ethnic affiliation using highly informative diallelic DNA markers: application to diverse and admixed populations and implications for clinical epidemiology and forensic medicine. Human Genetics. (E-published ahead of print September 29.)


17 Ritter, M. 2003. DNA test in LA. killings is said to have indicated attackers race. Boston Globe. June 5.

18 Ritter, M. 2003. DNA test in LA. killings is said to have indicated attackers race. Boston Globe. June 5.




22 Harmon, A. 2005. Blacks pin hope on DNA to fill slavery’s gaps in family trees. New York Times. July 25. New York Editio

GeneWatch: Current Issue
Volume 30, Issue 1: CRISPR & Gene Drives
Remembering the pioneering scientist, ethicist, and founder of the Council for Responsible Genetics.
From a longer letter, "A Call to Conservation With a Conscience," initiated by
the Civil Society Working Group on Gene Drives.
GeneWatch: Archives