Law, Africana-studies, and sociology professor Dorothy Roberts describes race as a “political category that has been disguised as a biological one.” It’s a hard concept for many to grasp. Physical features associated with race, such as skin and hair color, are inherited through our genes. So how is race not biological? The long answer is found in Roberts’s book Fatal Invention: How Science, Politics, and Big Business Re-create Race in the Twenty-First Century, but the short answer is that there’s no clear biological basis to divide humanity into five or six or seven races, any more than there is to divide it into twenty or a hundred or a thousand. The lines are drawn by social and political imperatives, not nature.

Roberts’s thinking about race originated in growing up in an interracial family in Chicago. Her white father, Robert, was an anthropology and sociology professor, and her Jamaican mother, Iris, had been his research assistant at the university and became a public-school teacher. Both of Roberts’s parents taught her that “there is only one human race.” As an adolescent in the 1960s Roberts became interested in the antiwar and civil-rights movements, and her undergraduate years at Yale University introduced her to the women’s liberation movement.

After graduating from Harvard Law School in 1980, Roberts clerked for federal judge and civil-rights champion Constance Baker Motley and practiced law in New York City. In the late 1980s she began to read about women — especially black women — being prosecuted, on charges ranging from neglect to attempted murder, for using illegal drugs while pregnant. Why, she wondered, was a threat to the health of mothers and their babies being treated as a criminal-justice matter?

It was the beginning of her research into the stark differences between white and black women’s reproductive freedom in the U.S., culminating in her 1997 book Killing the Black Body: Race, Reproduction, and the Meaning of Liberty. From there Roberts explored racial injustice in the child-welfare system, and more recently, the fields of genomics and medicine, where she finds that misguided ideas about race are leading to different treatments for patients of different races and false genetic explanations for racial health gaps actually caused by social inequities.

In 2012 Roberts joined the University of Pennsylvania in Philadelphia as its fourteenth “Penn Integrates Knowledge” professor and became the founding director of the Penn Program on Race, Science, and Society. I met with her one winter morning in her cheerful, cluttered university office. She had expressive eyes and an easy laugh. As I tried to find a place for my recorder amid the piles of books, she chuckled and said, “I like stacks.”


Leviton: What was it like growing up in a biracial family?

Roberts: My father began studying interracial marriage as a twenty-two-year-old grad student at the University of Chicago. His parents were immigrants from England and Germany, and he grew up in a European-immigrant Chicago neighborhood. He interviewed interracial couples as early as 1937, and in 1940 he wrote his master’s thesis on interracial marriage and the racial order in the United States. It’s amazing to me that he even found these couples, some of whom had gotten married in the late 1800s. There were barriers to race-mixing in Chicago then, though there wasn’t an outright ban on interracial marriage in the state of Illinois. In the 1950s my mother, who was originally from Jamaica, was one of my father’s students. They fell in love and got married while working on this black-white-marriage project. I was born a year later.

My parents were devoted to the principle that there is only one human race. It was a mantra in our house. My father was doing his research not just because of his personal relationship to the subject, but because he believed that interracial marriage was the answer to this country’s race problem. He believed that if blacks and whites got to know each other — preferably intimately [laughs] — racism would wither. He saw black Americans and white Americans as incredibly similar. My father would say that if you compared a black American, a white American, and someone living in a tribal village in Liberia, where he also conducted research, there was no doubt that the two Americans had a lot in common.

From an early age I believed strongly in the equal humanity of all people. I also was very aware that my mother was black and my father was white. A number of interracial couples were friends of our family. My piano teacher was in an interracial marriage. So was our plumber. When I was young, I was proud to be part of an interracial family. I remember walking down the street holding my mother’s hand on one side and my father’s on the other and wanting people to see us: Look at us! Black and white people can live together! I thought of myself as being both black and white.

By the time I was in seventh grade, I’d become interested in the civil-rights and antiwar movements, and I began to think of myself as a black person living in a racist country. I also felt a duty to do something about racism.

Leviton: Do you feel the civil-rights movement made you “choose a side”?

Roberts: Maybe. Also, my mother had dark skin, and even though I was intellectually close to my father and would talk to him about ideas all the time, when it came to racial identity, I was closer to my mother. I don’t think I’ve quite figured out myself exactly when I decided to identify as black and not biracial! [Laughs.]

As I became more and more aware of racial injustice, I started to question my father’s view that an increase in intimate relationships between blacks and whites by itself could overcome racism and white supremacy.

Leviton: You argue that race is not a biological condition but a “political invention.”

Roberts: Right. I’m not saying that race is a natural division of human beings that can lead to unjust hierarchies. I’m saying that the very concept of race was invented to create and enforce such hierarchies.

Leviton: How old is this political invention?

Roberts: Certainly hundreds of years old. The term “race” came into use to distinguish human groups in the sixteenth century when Europeans began to conquer other peoples and enslave them. To justify capturing Africans and turning them into property, Europeans came to describe them as a separate kind of human being — or even not human at all.

As soon as people invent the concept of race, they rank races into a hierarchy. Some people think it’s harmless to believe in biological differences between races as long as we don’t value one over another, but the whole point of dividing humans into races is to value some more than others. The inventors of the biological concept of race said that Africans were naturally meant to be enslaved, that it was for their own good, that they were better off being slaves! These ideas were written into law in the United States during the slavery era.

Medicine played a big part in promoting racial thinking. Doctors claimed that people of different races had peculiar diseases, or experienced common diseases differently, because they had innately different bodies. In 1851 Dr. Samuel Cartwright argued before the Medical Association of Louisiana that black people had lower lung capacity than white people, and forcing them to work was therefore good for their health. It would “vitalize” their blood and “free them from barbarism.”

Leviton: But is it fair to say that, before colonialism, people did notice differences in skin color, eye color, and hair texture?

Roberts: Yes, and those differences were seen as significant. For centuries groups of people thought that other groups — often distinguished geographically and by physical appearance — were inferior to them. Aristotle thought the Greeks were superior to the barbarians. But those differences in status could be erased. Barbarians could become civilized. Tribal membership could change through marriage. Race is seen as universal and immutable: from the moment of birth you belong to one group and cannot change your nature or status. Even if some black people can “pass” as white, they still possess some essence that makes them naturally black.

First comes the desire to conquer another people and take their land or enslave them. Then follows classifying human beings into races to justify and manage it. Racism isn’t a product of race. Race is a product of racism.

Leviton: Doesn’t our skin color come from our genes?

Roberts: There is genetic variation in the human species. In Africa alone we could divide people into a thousand different “races” if we wanted to, based on various genetic differences. But there would be no point in having a thousand races. If you divide humans into just a few groups, however, then you can build a social hierarchy around those divisions. Besides, skin color varies within races and is consistent between some people of different races.

The biological concept of race has been refuted by evolutionary biologists and geneticists and genomicists for decades. The scientists who led the Human Genome Project made a point of saying human genetic variation isn’t divided into races. There’s no such thing as black genes or white genes. The amount of genetic variation among people of the same so-called race is greater than the amount of genetic variation between races. You might have genes that can be traced to a certain population somewhere on the globe, but there’s no point at which you can draw a boundary line and identify one race on one side and a different race on the other.

All humans originated in Africa and then migrated outward in groups, each carrying a subset of the genetic variation in Africa. No one has identified a point in human history at which these migrating groups evolved into discrete and homogeneous “races.”

Leviton: You’ve written that during U.S. slavery there were many laws and tests to determine who belonged to what race, and these laws changed over time.

Roberts: Mixtures had to be dealt with. Human beings, regardless of race, can have children together. That’s always been a problem for the biological argument. At first some European naturalists believed that men and women of separate races couldn’t produce progeny. When it turned out they could, the naturalists claimed that the offspring would be sterile, like mules. That’s where the word mulatto comes from, by the way. And when that turned out not to be true, the definition of race had to be written into law. The colonists didn’t really care about biological categories except as a means to justify enslaving other people and not people like themselves. The colonists in Virginia needed to determine the race of the children white men fathered with enslaved African women — by sexual assault, since the women had no right to refuse. They also didn’t want these children to have any claim on inheritance from their fathers. Colonial law literally begins with a ruling about “whether children got by any Englishman upon a Negro woman should be slave or free.” British common law said children inherited their social status from the father. In 1662 the white settlers passed a law that, in the Virginia Colony, the social status of children born to black women was inherited from the mother, so they could be enslaved.

By the twentieth century the law of racial inheritance becomes the “one-drop” rule, which says that if you have any discernible African ancestry, from your mother or father, you’re black. You can have seven white great-grandparents and one black great-grandparent, and you’re still black. In American culture white means “pure white.” Blackness can be a mixture, but whiteness cannot. Privilege and entitlement are given exclusively to people who can legally claim to be white. Jewish, Irish, Italian, and Slavic people weren’t always classified as completely white but were eventually able to attain that status. This system is clearly a political invention to establish white supremacy and white dominance.

Leviton: The Naturalization Act of 1790 prohibited non-white immigrants from becoming American citizens. In 1923 an Indian Sikh named Bhagat Singh Thind attempted to have himself categorized as Caucasian, but the U.S. Supreme Court decided his skin color disqualified him.

Roberts: The Thind case is a perfect example of how the decision of who gets to be white is not a biological one. Thind brought to court scientific evidence that his ancestors were of the same Aryan blood as Englishmen and white Americans. In fact, he argued that his superior caste position in India made him of purer white descent than white people in the U.S. The Supreme Court basically dismissed his data and said, in effect, “No white man in America is going to think you’re white. Therefore you can’t be.” The legal category of “Asian” was created to identify those who would be excluded. But where’s the border? Is Russia less “Asian” than China? U.S. law considers Russians to be white but Chinese people to be Asian — because historically Russians were more welcome as immigrants, though not as welcome as the English.

First comes the desire to conquer another people and take their land or enslave them. Then follows classifying human beings into races to justify and manage it. Racism isn’t a product of race. Race is a product of racism. People think it’s OK to categorize people by race as long as they’re not racist, but any division of people into supposedly natural races promotes a racist agenda, whether we intend it to or not.

Leviton: You have written about how early efforts to gain female reproductive freedom — through the work of Margaret Sanger, Planned Parenthood, and others — became entangled with the eugenics movement, which advocated “selective breeding” in humans. What were the limitations of Sanger’s thinking?

Roberts: The movement led by Sanger for women to gain greater control over their reproductive lives and bodies began at the same time that eugenics was gaining popularity, in the 1920s. Contraception gave women more autonomy and removed the burden of constant childbearing, but there was also brutal suppression of reproduction among those who were deemed “unfit” to have children, including mandatory sterilization of people confined to mental institutions.

In Buck v. Bell in 1927 the U.S. Supreme Court upheld Virginia’s 1924 compulsory-sterilization statute, which was meant to prevent reproduction by “potential parents of socially inadequate offspring.” Justice Oliver Wendell Holmes Jr., an ardent eugenicist, wrote, “It is better for all the world if, instead of waiting to execute degenerate offspring for crime or to let them starve for their imbecility, society can prevent those who are manifestly unfit from continuing their kind.” Some people were being locked up in asylums in order to sterilize them.

Sanger may have started her campaign hoping to emancipate women, but it was quickly distorted by the desire to control certain populations. She became involved with eugenicists, espoused their principles in writing and speeches, and founded the American Birth Control League, which was firmly eugenicist. Racist organizations like the Ku Klux Klan agreed with her that birth control could eliminate “undesirables.”

The eugenics movement targeted those who were socially devalued — poor people, people with disabilities, black people, Mexicans, Puerto Ricans, and Native Americans. To justify this, eugenicists tried to pass off social hierarchies as natural. If certain groups of people are at a disadvantage, the thinking went, it must be because of their biological inferiority, not due to state violence and structural inequalities. In the U.S. eugenics became especially brutal, producing public policies and statutes that would later be adopted by the Nazis in Germany.

Such thinking is still used to explain social inequality in the present: If we believe biology produces these unequal social and economic conditions, then how can they be immoral and in need of change? They are “natural.” The situation can’t be changed.

Leviton: Politically it seems there’s always a need to define some group as a burden on society, undeserving of the resources and advantages upstanding citizens have earned.

Roberts: This is a familiar discourse in the U.S. Black women on welfare. Mexican immigrants. They are costing taxpayers too much. We need government policies to control them: welfare rules that deny benefits to women who have another child; immigration laws that keep Mexicans from giving birth on American soil. To me this is the language of eugenics, even if the people espousing these policies don’t mention sterilization or genetics. It’s still focused on the threat to society posed by certain people’s childbearing.

And you hear the argument, both in the past and today, that these policies are for the benefit of the disadvantaged people themselves. That 1924 Virginia law mandating sterilization was passed in a package that included the Racial Integrity Act, which criminalized marriages between blacks and whites. That law wasn’t overturned until 1967, in Loving v. Virginia.

One of the creators of the Racial Integrity Act, Earnest Cox, said, “The sane and educated Negro does not want social equality. They do not want intermarriage or social mingling any more than the average American white man wants it. They have race pride as well as we. They want racial purity as much as we want it.” In other words, We’re doing this for their own good!

Likewise, Holmes argued in Buck v. Bell that if the government sterilized the mentally ill, it could release them from asylums and let them live freely. Sterilization benefited them! So Holmes framed a violent act as benevolent.

Samuel Cartwright, the nineteenth-century physician I mentioned earlier, argued in a medical journal that blacks suffered from “dysaesthesia aethiopica,” a mental condition that caused laziness and could be cured by forced hard labor. Slavery “unchained the mind” of black people from barbarism and ignorance. These attitudes persist today in the way legislators talk about how to “cure” the problems of the black community. In 1965 Democratic senator Daniel Patrick Moynihan issued a report titled The Negro Family: The Case for National Action. It concluded that poverty in black communities was caused by “matriarchal” black single mothers, the poor character of black men, and ghetto culture — not forced residential segregation, substandard schools, and employment discrimination.

The disadvantages suffered by black people today are not due to innate flaws in black people themselves but stem from the institutional racism that’s been practiced over centuries in this country. Our society is unequal because it is engineered to be that way.

Medical professionals, geneticists, criminologists, sociologists, and psychologists who don’t think of themselves as racist still find it hard to conceive of doing their work without categorizing people into natural races. Doctors ask, “How can I treat patients without taking their race into account?” Researchers ask, “How can I do this study without having subjects tick the box to indicate race?” By treating race as if it were an essential, innate part of who we are, they are upholding institutions and ideas that perpetuate inequality.

Avowed white supremacists are aware of their investment in the system. They are dedicated to propping up institutions that favor white people based on a belief in white superiority. But most white people don’t realize the depth of institutional racism. It’s painful for them to confront how they have benefited from these systems. They want to believe they deserve what they have. Because they don’t understand or don’t want to acknowledge how being white confers privileges, they do nothing to dismantle these systems where racism is embedded, and they even support policies and ways of thinking that maintain them.

Leviton: You mentioned having research subjects tick the box indicating race. How do researchers misuse that information?

Roberts: There’s a whole research industry built around studying the problems of black children. White researchers’ work often veers into biological explanations: Something’s wrong with the kids’ brains, their genes. They have low cognitive capacity, which leads to a lack of self-control. This is why black children don’t do well in school; why they have higher rates of juvenile detention; why they are more often placed in foster care than white kids. When I meet with these researchers, I ask them, “Have you studied the amount of police presence in predominantly black schools? Have you looked into the fact that black children get expelled and arrested at far higher rates than white children for the same behaviors, like missing school, talking back to a teacher, or roughhousing? Have you looked for any explanations other than within the gray matter of their brains?”

The researchers often get defensive, as if I am questioning their qualifications or motives. They don’t want to consider that maybe white kids aren’t smarter or don’t have more self-control than black kids but instead have protections that black children don’t have.

The myth of biological race keeps many people from seeing that there’s nothing wrong with people who suffer from social disadvantage. What’s wrong is a society that puts people at such social disadvantage because of their race.

Leviton: You’ve said that government policies that do not overtly specify race still have unequal outcomes for different racial groups.

Roberts: Government statutes that appear to be race-neutral often have racial assumptions embedded in them, or they have a different impact on people of color because other institutions have already put those people at a disadvantage. “Stop-and-frisk” police policies don’t mention race at all, but when practiced in a country where racial segregation is a fact, stop-and-frisk can discriminate by race. It enables officers to go into black neighborhoods and routinely stop people for suspected infractions that are rarely addressed in white neighborhoods, even though they take place there, too.

Vagrancy laws, which also don’t mention race, have traditionally been used to arrest black people simply for their presence in public. A Chicago “gang loitering” ordinance gave police officers license to arrest groups of people who had “no apparent purpose” on city streets. Thousands were arrested for not walking away fast enough when told to move. Ordinances like this don’t mention race, but they allow police to target those who look like they belong to a gang, and we all know there’s a racist stereotype of a gang member: it’s a black or Latino person. As some commentators noticed last year, when mostly white Philadelphia Eagles fans smashed windows, looted, and burned cars after their team’s Super Bowl victory, they were “celebrating.” There were few arrests, and the police mostly watched. When black people in Ferguson, Missouri, protested police brutality in the wake of the killing of Michael Brown in 2014, they were greeted with a militarized police force, tear gas, and arrests. This is how “race-neutral” laws operate in our country.

I certainly support universal, race-neutral policies that promote equal access to resources — for instance, Medicare for All. But even if we had that, we would need to recognize that it won’t benefit everyone equally. Residential segregation has put black Americans in neighborhoods with fewer high-quality hospitals, cutting-edge technologies, and medical caregivers. Black people have to travel farther to get care, often using substandard transportation systems. Then there are the living conditions that put black people at greater risk of poor health in the first place, like the proximity of toxic-waste sites, air pollution, lead in the water, police violence, and higher levels of stress. Universal access to single-payer health insurance is urgently needed, but it won’t solve those problems. People who live in better-resourced areas, who can buy more-nutritious food and have less exposure to toxic chemicals and other unhealthy conditions, are still going to have a health advantage.

Most of my work looks at institutionalized disadvantage, not individual bias, but that doesn’t mean there isn’t plenty of individual bias on the part of police, social workers, employers, landlords, teachers, and health-care providers. Studies show that many doctors falsely believe in biological differences between races, and they incorporate these beliefs — especially negative myths about black people’s bodies — into their medical practices.

One atrocious example is the undertreatment of black patients for pain. Black patients receive less pain medication than white patients for the same injuries, and black people are less likely to get any pain medication at all. A study of children coming into emergency rooms with severe appendicitis found that white patients were more than twice as likely to be given opioids to relieve their suffering than black patients. In some emergency rooms, because of a myth that black people exaggerate their pain, nurses are instructed to give black patients less pain medication than is indicated by how they describe their pain.

A 2016 study of medical residents and students at the University of Virginia showed that a significant number had false beliefs about black people, such as that they have thicker skin and less-sensitive nerve endings than white people. These beliefs could be folklore the students brought with them, but they are affirmed by a medical-school education that focuses on race-based biological differences.

Some doctors have tried to explain away the difference in treating pain by saying that black children are taught not to express their pain. So it’s not the doctor’s fault for undertreating them. Yet in a study of long-bone fractures — which are very painful — even when black patients rated their pain level the same as white patients, they were less likely to get the same treatment.

Leviton: Nobody in these situations is filling out a detailed family history. Two people checking the box for “black” on the intake form could have very different heritages.

Roberts: You shouldn’t treat a social category as if it were biological — that’s the whole point. In its directives to the researchers it funds, the National Institutes of Health (NIH) uses the census categories, which the U.S. government has said are social. These categories change over time. There is discussion of adding a new category — Middle Eastern or North African — to the 2020 census. Right now people from those regions are placed in the “white” category.

Medical researchers mostly use self-identification to determine study participants’ race: if the participant says she or he is white, that’s the box that gets checked. Even if the researchers were doing genetic tests to determine race, their studies would still be flawed and inconsistent, because there’s no valid or agreed-upon genetic test for race.

A few years ago Stanford sociologists Aliya Saperstein and Andrew M. Penner found that, over the course of an important long-term study of children, a significant number of the children changed race. This was because new researchers coming into the study made different determinations of the children’s race. What’s also interesting is that the race of the subject was more likely to change from white to black if something negative had happened to the child in the interim. For instance, if the child’s parent had become unemployed or gone to jail, it was more likely that the researcher would identify the subject as black. This shows how the identifiers of race are not just physical. There are social traits associated with each race — stereotypes, expectations of how each person will be, think, act, and achieve in the world based on race.

Leviton: You cite one study in which subjects were categorized as “Hispanic” based on nothing more than their last name. It’s incredible to me that such bad methodology doesn’t prevent publication in a peer-reviewed professional journal.

Roberts: How can a name indicate a supposedly biological category? It’s absurd. And remember, these individual researchers often don’t decide that they want to include race because it’s essential to answer their research questions; they are simply following NIH guidelines. In an effort to diversify clinical research, the NIH required that race be applied in recruiting research subjects. The problem is that the racial categories are often maintained throughout the study, whether they are relevant or not, and are often treated as biological variables. What started as an effort toward inclusion became a way of encouraging bad science.

Black people have historically been coerced into participating in studies that have been harmful to them — most infamously the Tuskegee Study of Untreated Syphilis in the Negro Male, where government-funded researchers told black men they were getting free health care, then allowed them to go untreated for syphilis and later autopsied their bodies to discover the supposedly different way the venereal disease affected them.

There are innumerable examples of brutal experimentation on black people going back to the days of slavery: experiments that treated them inhumanely because they weren’t considered to be fully human. J. Marion Sims, the so-called father of gynecology, used enslaved women, without anesthesia, in his surgical experiments. So it’s no wonder people of color are often skeptical about participating in medical studies.

Leviton: But isn’t there any validity to race as a factor in medical studies? For instance, aren’t blacks more likely to get sickle-cell anemia [a condition in which the red blood cells grow rigid and contract into crescent shapes resembling a sickle — Ed.]?

Roberts: We shouldn’t confuse a higher prevalence of a particular genetic variant in a population with a racial disease. It is true that the genetic mutation that causes sickle-cell disease is more prevalent in certain parts of Africa, as well as in other parts of the world where malaria is endemic, because the sickle-cell trait provides a protective advantage against malaria. But the trait doesn’t exist only in black people. Nor are individual black people likely to have it just because they are classified as black. There’s also a reliable test to detect the sickle-cell gene. Why not use it instead of depending on flawed race classification?

The association between sickle-cell and being black was intensified with a strategy that was supposed to improve the health of black people. In 1972 Congress passed the National Sickle Cell Anemia Control Act. By 1975 there were more than 250 screening programs around the nation. Because screening programs often provided no counseling, however, there was confusion about who was a carrier and who actually had the disease. Even the preamble to the federal law erroneously stated that 2 million Americans had sickle-cell disease rather than just carrying the trait. Fourteen states made sickle-cell tests mandatory for any blacks getting a marriage license, enrolling in school, or entering prisons or mental institutions. Hysteria about the disease led to widespread discrimination. Blacks were denied entrance to the Air Force Academy because of the perceived possibility that their red blood cells might “sickle” at high altitudes. Commercial airlines fired or grounded black pilots and flight attendants who merely carried the trait. Carriers were also sometimes denied insurance coverage or charged higher premiums. Women who were carriers were counseled not to have children at all.

Blacks are not the only ethnic group associated with a genetic disorder. In the seventies at least fifty genetic disorders could be identified, but none of them resulted in the institutionalized abuse that sickle-cell did.

Leviton: What about the prevalence of Tay-Sachs disease among Jews?

Roberts: According to sociologist Shelley Z. Reuter, Tay-Sachs became associated exclusively with Jewish people largely due to anti-Semitism in the nineteenth and twentieth centuries. There was a desire to prevent Jews from immigrating into the country, and stereotypes about their bringing illnesses played into that. Tay-Sachs is not just a Jewish disease. Other groups — including French Canadians and Cajuns in Louisiana — also have a higher prevalence of it. And even within Jewish communities, it’s mostly European Jews, rather than Middle Eastern Jews, who were affected. And are Jews a biological race? Is the world “naturally” divided into Jews and gentiles? Noticing the resurgence of anti-Semitism following Trump’s election, Emma Green wrote a piece in The Atlantic titled “Are Jews White?” She points out that white-nationalist groups define white people as “non-Jews of European descent.”

Donald Trump refers to his “good genes” as the reason he’s so smart. He believes people are like racehorses: their success depends on how they’re bred. He thinks his cabinet has the highest IQ ever. Note that it’s made up mostly of white men.

Leviton: Let’s talk more about how medical treatment is affected by racial categories. In Fatal Invention you tell a story about Richard Garcia and his childhood friend. What did you learn from him?

Roberts: Dr. Garcia is a physician who wrote an article in Pediatrics about how notions of race can lead to misdiagnoses. He used the example of a friend of his, an African American girl named Lela, who from a very young age had suffered from respiratory problems. She kept coming back to the emergency room, and doctors could not figure out what was wrong with her. Her medical chart contained doctors’ notes identifying her illness as “another pneumonia” and a “fever and cough.” When Lela was eight years old, an X-ray technician looked at her X-ray without knowing her race and immediately asked, “Who’s the kid with cystic fibrosis?” She’d had the symptoms from the beginning, but the doctors thought only white people carried the gene for cystic fibrosis.

There is no such thing as a black, white, Asian, or Native American gene. There’s genetic variation among humans, all of whom descended from Homo sapiens on the continent of Africa. Cystic fibrosis may occur more often in a certain population of European descent, but that doesn’t mean that it cannot be carried by someone who is not of European descent or who is not identified as white. And this African American girl could have had European ancestry — in fact, she could have had mostly European ancestry — but doctors still would have identified her as African American because her skin was a certain shade, or her hair was a certain texture, or because her parents ticked the “black” box. In the U.S., “black” means having any discernible amount of African ancestry.

Doctors use race as a proxy for genetic differences because it’s convenient, or because they don’t have time to do genetic tests. But testing for cystic fibrosis is better medical practice than continually admitting a patient like Lela to the hospital. Her race wasn’t considered just a factor; it was considered a categorical exclusion. It meant she couldn’t have cystic fibrosis, even though she had the symptoms of the disease.

Race overwhelms everything else. If you’re a doctor, even when your patient has the clinical indicators and the symptoms are right in front of you — everything you need to make a diagnosis — you might discount it all because of the patient’s race. That makes race a dangerous category for doctors to rely on.

Race is also embedded in the technology doctors use. Let me give an example from my own life. My daughter sent me a copy of a blood test she’d gotten at a university medical center. At the very top, beside a measure for kidney function, it said, “African American 86 / Non-African American 89.” In other words, if the doctor looks at my daughter and identifies her as African American, the test gives one reading. If she’s not, it gives another. This modern clinical test treats African Americans as if their bodies functioned differently than those of other human beings. Why isn’t the American Medical Association advocating to change this? That number is an important indicator of whether the patient is at risk for kidney failure, but since black people’s bodies are presumed to be different, the level of concern for them is adjusted. Could this presumption contribute to the undertreatment of black people for kidney disease?

So where does this adjusted number come from? It appears it’s related to the idea that black people as a group have greater muscle mass than other human beings — and muscle does have an impact on the blood-filtering process. But why doesn’t the doctor look at the patient to determine the amount of muscle mass? The doctor can adjust his or her concern based on that and any other relevant factors presented. Family history, for instance, has been shown to be a better predictor of susceptibility to certain diseases than any specific gene or set of genes, and certainly better than an association between genes and race. There’s no need to “correct for race.”

Statistical epidemiologist Jay Kaufman and physician Richard Cooper recently did a study of sickle-cell-anemia testing and concluded it would be more economical and more medically beneficial to test all babies for the trait rather than to assume only black children could have it. It’s the same with BRCA1/2, gene mutations associated with an increased risk of ovarian and breast cancer. Katrina Armstrong, physician-in-chief at Massachusetts General Hospital, coauthored a study that found black women were far less likely to be counseled to get a BRCA1/2 genetic test even when there was a family history of breast cancer and other indicators the patient might be at risk. BRCA1/2 are thought to occur more often in Ashkenazi Jews, but also show up in African Americans, Hispanics, Asians, and so on. In one study, Latina women had relatively high rates of these mutations, and the researchers suggested these individuals must have been descended from Jews in Spain. That could be true, but the genes could also come from other parts of their ancestry. Why was race necessary for any of these predictions?

Leviton: A 2005 study found that white women in Chicago were slightly more likely to get breast cancer than black women, but black women were far likelier to die from it. What happened there?

Roberts: This was a study by epidemiologist Steven Whitman at Mount Sinai Hospital and his colleagues. The research team’s conclusion was that white women’s survival rate increased because they got the benefit of advances in breast-cancer detection and treatment, and black women got no benefit. Dr. Whitman pointed out that Chicago was a very racially segregated city. Black women lived in neighborhoods with higher rates of poverty, and the hospitals there had lower-quality mammography. In one neighborhood the mammogram machine had been broken for many months. Women living in that neighborhood had to travel long distances to get a mammogram — which, of course, reduced the number of them who got tested. They couldn’t afford just to hop in a cab, and they were more likely to have low-wage jobs where they weren’t given time off. Whitman and his colleagues couldn’t study every factor involved, but common sense tells you black women aren’t innately predisposed to die from breast cancer.

Genes are one factor in an individual’s health, but we’ve got to study how social disadvantages produce major differences in health between racial groups. Genes don’t do anything by themselves. They act in relationship to the environment. If you want to save the lives of black women, don’t spend all your time looking at their genes. Look at how to improve their access to medical care, which is certainly an indicator of whether they’re going to survive cancer, as well as their access to other resources that promote better health outcomes. We’ve long known that the best predictor of outcomes for patients is their position in the social hierarchy.

In some emergency rooms, because of a myth that black people exaggerate their pain, nurses are instructed to give black patients less pain medication.

Leviton: You’ve written about how white women can avail themselves of a system that encourages childbearing, while black women are discouraged from having more children.

Roberts: There is a racial caste system in reproduction. Black women in particular have been stigmatized for their childbearing, to the point where it’s seen as a social problem. There’s a long history of punishing black women for having babies, from forced and coerced sterilizations, to the way welfare policy is written, to how black women are prosecuted for drug use during pregnancy. The hysteria over “crack babies” during the Reagan administration was just that: hysteria. More-recent, long-term studies have shown that crack use during pregnancy didn’t produce the dire physical and social harms that were predicted. In any case, drug use during pregnancy should have been treated as a health problem instead of a crime. Racism and the devaluation of black mothers were key to this punitive approach.

Medicaid is paying for disproportionate numbers of black women to be sterilized, at the same time that doctors are discouraging young white women from seeking that option. White women who want tubal ligations sometimes have to wait years to find a doctor who will perform the procedure, but judges have offered women of color the option of getting their tubes tied as a way to get their sentences reduced or even to obtain early release from incarceration. It’s the same idea Oliver Wendell Holmes Jr. had about letting inmates out of the asylum if they consented to sterilization. Black and Native American women are twice as likely as white women to undergo tubal sterilization. Meanwhile we have a multibillion-dollar fertility industry catering primarily to white couples. And when a white woman receiving artificial insemination was mistakenly inseminated with a black man’s sperm in 2014, she brought an unsuccessful lawsuit. The damages sought weren’t just for receiving the wrong sperm, but also because a black or biracial child wouldn’t fit into the family. The child was somehow too different, even though there was the same degree of genetic relation to the mother. Race affects how people think about who belongs in a family.

This is why Founding Father Thomas Jefferson could have biological children with a woman he enslaved, Sally Hemings, and not recognize that kinship. This is one reason I could never adopt my father’s view that intimate relations between blacks and whites could eradicate racism. The political will and work to dismantle racism may lead to more interracial intimacy, but I don’t think it will happen the other way.

Leviton: You’re also concerned that, instead of proving race doesn’t exist, genetics is being used to add biased aspects to existing systems.

Roberts: Yes. If you add a new technology to an already racist system, you’ll get another racist product. Take the criminal-justice system. The police in some states are collecting the DNA of people who’ve been arrested and putting it into a database to then check against DNA evidence in the future. All it takes for the government to get someone’s DNA is for the police to decide to arrest that person, on whatever pretext, and arrests are heavily biased against black people. So you are obviously going to end up with a database in which black people are overrepresented.

Keep in mind that those with DNA in the database haven’t necessarily been convicted of a crime; they’ve only been arrested. Now they are perpetual suspects. And there’s always a chance of error or abuse. Let’s not forget, as we’ve seen in Baltimore and many other cities, police are not above planting evidence at crime scenes.

Predictive analytics based on big data is another technological development that builds on existing racial bias and structural racism. There are proposals to have judges use algorithms to decide who should be paroled from prison. One risk-assessment algorithm, called COMPAS, is being used right now, even though it’s been found to be heavily biased against African Americans. And Julia Dressel and Hany Farid of Dartmouth College recently published a study showing COMPAS predicts recidivism about as well as random volunteers on the Internet.

Leviton: Right-wing pundits often argue that these law-enforcement efforts benefit the black community, which suffers a disproportionate amount of crime.

Roberts: Just because black people are concerned about violence in their neighborhoods doesn’t mean they are willing to waive their constitutional rights. Black people have a nuanced way of thinking about civil liberties. Of course they don’t want violent crime, but they also don’t want the state committing violence against them in the name of protecting them and others from crime. We can hold both ideas in our minds.

If the state is so intent on preventing violence, it should do more to protect women from it in their homes — without relying on police violence and prisons. It wasn’t until 1993 that it became a crime in all fifty states to rape your wife. Mothers have had their children taken from them by child-welfare authorities for “allowing” their children to witness domestic violence against them. I wrote Shattered Bonds: The Color of Child Welfare to highlight how the child-welfare system monitors, regulates, and mistreats poor families, especially poor black families: investigating parents; forcibly removing children from the home; terminating parental rights. That’s how family needs are addressed, instead of by supporting families.

At one point in U.S. history the child-welfare system was used to separate white immigrant children from their families. Charles Loring Brace’s Children’s Aid Society — founded in 1853 — was created to remove Catholic immigrant children from their families and place them with God-fearing Protestant families. That was at a time when child welfare was only for white children.

Leviton: How is it different today?

Roberts: As black families became involved, the system became more punitive. The child-welfare system is run as if black children needed to be protected from their innately inadequate mothers. This leads to black children being removed from their homes with greater ease than white children. There’s a widespread belief that black children are better off in imperfect foster care than in their own homes.

There have always been “undeserving poor,” but black mothers are seen as even more undeserving. The mythical “welfare queen” — an inner-city black woman — became the face of the welfare system, even at a time when there were more white families receiving public aid. According to a widely held myth, white families used government money to care for their children, whereas black families spent it on frivolous luxury items. Black women were actually thought to be having babies just to get a supposedly generous welfare check.

This all led to the ridiculously titled Personal Responsibility and Work Opportunity Act, which President Bill Clinton signed in 1996. The first thing the 1996 law says is “Marriage is the foundation of a successful society.” In other words: You black women need to get married and stop having children out of wedlock. Before 1996 you were entitled to public assistance to care for your children if you had a certain level of need. Since then, Congress has made no bones about the fact that welfare is a behavior-modification system. The law doesn’t examine why welfare recipients need public assistance; it is designed simply to get them off public assistance through marriage, fertility control, low-wage work, or cutting off benefits.

Leviton: So what can be done to change these broken systems?

Roberts: We can start with awareness. Everyone should understand that the reason we have mass incarceration of black men and women isn’t because most of them are violent and dangerous. We don’t have a huge foster-care population because so many black parents are monsters. Researchers are too often trying to figure out how to “fix” black people. My mantra is “There is nothing wrong with black people.” What’s wrong is structural racism. The systems in place ruin black people’s lives and restrict their opportunities to participate equally in society.

I think if more people understood that the problem of inequality is systemic and institutionalized, it would go a long way toward changing the discussion. These issues are interconnected, and we need all progressives, at least, to see the connections. We need people organizing and protesting in the streets and writing op-eds in the newspapers and putting forward legal challenges in the courts and creating more humane alternatives in their communities. I can be pessimistic about this country ever becoming a non-racist society, but I’m heartened when my undergraduate, medical, and law students ask what they can do to bring about such a change.

I don’t believe we should be “color-blind,” that we shouldn’t pay any attention to race. As a political invention, race continues to determine power arrangements and is not going to just go away. We have to dismantle racist institutions to affirm our common humanity. And to do that, we need to understand how the concept of race really functions.