tufts universitytufts magazine issue homepage
contact us back issues related links
features The Ultimate Trip The Big Taboo I Am the Walrus Shooting War Roar of the Greasepaint columns jumbolaya planet tufts newswire wedding album departments

The Big Taboo

We don’t talk about race. Often, we don’t even think about it. But according to Tufts psychologists, unconscious stereotypes and racial biases can have a huge effect on our behavior

At the outset of his 1903 book The Souls of Black Folk, W.E.B. Dubois declared that “the problem of the 20th century is the problem of the color line.” Ensuing decades proved him to be dead on, as tensions over race in America led to burnings and lynchings, internment camps and border patrols, sit-ins and marches, battles over what would be taught in our universities and who could be taught there. But with upheaval came progress. Schools have been integrated, and the Civil Rights movement has led to a promise—at least on paper—of equal rights before the law. Jackie Robinson has become Shaquille O’Neal; Paul Robeson has become Denzel Washington; Booker T. Washington has become Condoleezza Rice.

The scholars Henry Louis Gates Jr. and Cornel West—Dubois’ self-described intellectual heirs—have declared the 20th century “the African-American Century.” In their book of that title, published in 2000, Gates and West wrote: “It was the century in which African-American life was transformed—and the century in which African Americans changed America.” They went on: “At the dawn of the 21st century . . . we cannot imagine an American culture that has not, in profound ways, been shaped by the contributions of African Americans.”

And it’s not just black and white. Social commentators have begun to debate whether we are in fact a “postracial” society—or as the essayist Leon Wynter put it in his 2002 book American Skin: Pop Culture, Big Business, and the End of White America, a “transracial” society, in which our dominant culture is a fusion of minority influences, with kids of all races wearing baggy shorts and backwards ball caps, listening to rap, and dropping black slang. Mixed-race marriage has made hyphenated categories useless (some kids now call themselves Blaxican, Mexipino, and other newfangled coinages), and identity politics may be growing less relevant as the class line replaces the color line.

Then came Katrina. As the floodwaters rose in New Orleans in the wake of last year’s hurricane, the images of black people waving on rooftops and floating dead in the water left an indelible reminder of the link between race and poverty in America. During the disaster, black residents were described in news stories as “looting” grocery stores while whites merely “found” food, and hyped reports of violence in the city seemed to confirm the worst stereotypes of black people as criminals. If we needed any reminder that we still live in “two Americas,” a Pew poll in the wake of the disaster found 66 percent of blacks believed the government would have responded more quickly if most of the victims had been white, compared to only 17 percent of whites. In the same poll, 71 percent of blacks said the hurricane shows racial inequality is still a “major problem,” compared to just 32 percent of whites.

Like the O.J. Simpson trial before it, Katrina is one of those events where race rears its head out of the floodwaters of ignorance to show that it is still a grave and lingering problem in America. Even as the most outrageous forms of racism have been eradicated, we as a society are still plagued by stereotypes that cannot be argued away by differences in class or poverty. “You have all of these widespread reports of gruesome violence at the convention center, about babies being raped, and apparently none of these things happened,” says Sam Sommers, a professor of social psychology at Tufts, who researched media portrayals of the hurricane. “At some point, you have to ask, Does the evidence add up to say race might have played a role here?”

Sommers is one of a trio of social psychologists at Tufts who are pushing the bounds of research into perceptions of race. Their research codifies the powerful, but largely subliminal, effects of race on how we see others and even ourselves. “It fascinates me how race is potentially relevant to so many situations in life, but it is this giant elephant in the room that no one wants to talk about,” says Sommers, who is white, and first became interested in the topic after the widely divergent reactions on the part of blacks and whites to the O.J. Simpson verdict. The problem is pinpointed in an episode of Seinfeld, in which George and Elaine are discussing Elaine’s suspicion that her new boyfriend might be black and George nervously insists, “I really don’t think we’re supposed to be talkin’ about this.” In fact, it may be our very reluctance to admit to ourselves that we harbor secret stereotypes that perpetuate prejudice and injustice.

Of course, stereotyping is not just a weakness of whites. Psychology professor Keith Maddox studies the ways that members of racial minority groups discriminate within their group based on subtle distinctions like skin tone. Another Tufts psychology professor, Nalini Ambady, has looked at how stereotypes affect people’s conceptions of themselves.

If ever we are going to get past race as a problem in the 21st century, the research suggests, then it is not enough to look at legislative and judicial solutions to racial inequity. We are going to have to understand the psychology behind how we discriminate based on the color of people’s skin.

Not-So-Blind Justice

Much of Sommers’s research concerns the composition of juries—one place in American society where colorblindness is supposedly sacrosanct. At least since the days of black Supreme Court Justice Thurgood Marshall, legal scholars have pushed for diversity on the jury panel. “When any large and identifiable segment of the community is excluded from jury service, the effect is to remove from the jury room qualities of human nature and varieties of experience . . . that might have unsuspected importance,” wrote Marshall in a 1972 ruling. In fact, through a process known as “peremptory challenge,” lawyers are allowed to remove potential jurors from a panel for any reason they choose—except for race. A 1986 legal decision required lawyers to justify challenges suspected of being based on race.

A study by Sommers and Harvard Business School professor Michael Norton, however, suggests that lawyers haven’t stopped removing jurors from juries because of their race; they’ve just gotten better at hiding their motives. For the experiment, Sommers and Norton constructed two juror profiles that would be equally distasteful to a prosecutor—a journalist who had written about police misconduct, and an advertising exec who said he was skeptical about statistics. They then asked groups of college students, law students, and actual trial lawyers which of these two individuals they would remove from their jury if they could. When told the first juror was black and the second juror white, the great majority (77 percent) of study participants chose to exclude the first juror. When told that the first juror was white and the second juror was black, however, the majority (54 percent) again chose to remove the black juror. Clearly race played a role. Yet less than 10 percent of participants in all three groups mentioned race as a factor in their decision—suggesting they were either lying to the investigators or lying to themselves.

“In this day and age, race still influences our judgments even when we think it is least likely to occur,” says Sommers. “This is a serious decision that a prosecutor makes. You want to think that race never factors into those judgments.”

In a study published this year, Sommers took this experiment one step further to look at how the racial composition of juries affects justice. Before his study, there was little hard data to back up Justice Marshall’s statement that diverse juries were fairer than homogeneous (read: all-white) ones. But in a study of mock juries drawn from members of actual jury pools, Sommers and Norton showed that a mixed jury was less apt to convict a black defendant—and not because of the black people on the jury, but because of the white people.

In fact, Sommers’s and Norton’s study found that, on average, 33 percent of whites on a racially mixed jury voted guilty, compared to 50 percent of whites on an all-white jury. More surprisingly, he found that on the mixed jury, whites expressed more skepticism about the prosecutor’s case than blacks did. “You would expect that it would be the black people raising novel ideas,” says Sommers, “but what’s interesting is that many of the differences come from white people behaving differently.”

The all-white juries appeared to be governed by the Seinfeld rule (George’s “we’re not supposed to be talkin’ about this”). The few times jurors brought up the issue of race, other jurors cut them off. The net effect was to take race off the table as a possible motivation for the allegations, leading to a higher probability that a defendant would be unjustly convicted.

True colorblindness in jury selection, says Sommers, wouldn’t rely on lawyers to justify their own decisions, but would be better achieved by selecting juries behind a screen, as is done at symphony auditions—or just by getting rid of the idea of peremptory challenges altogether, an idea suggested by Thurgood Marshall 30 years ago. “Lawyers would hate that,” he adds. “But why is it important for them to know the juror’s race? It shouldn’t be.”

Fear of Darkness

As insidious as stereotyping is between whites and minorities, it can be just as pernicious—and seldom talked about—within racial groups, says Sommers’s colleague, Keith Maddox, director of Tufts’ Social Cognition Lab. His work has focused on what he calls “skin-tone bias,” or prejudices based on the relative lightness or darkness of someone’s skin.

Maddox has theorized that people classify each other in two ways: one, by using multiple physical characteristics to sort people into broad groups such as black, white, or Latino, and coming up with judgments based on those groups; and two, by focusing on certain physical features and applying judgments within those groups based on those features. Maddox has spent most of his work investigating the second form of classification, looking into stereotypes based on skin tone—“colorism,” as it’s often called.

As far back as slavery, it has been an open secret within the black community that those with lighter skin receive favorable treatment from whites and fellow blacks alike. Slaves with lighter features would find easier work in the house, while dark-skinned slaves performed backbreaking labor in the fields. After abolition, lighter-skinned blacks were more successful in pursuing education and finding employment. Even today the wage disparity between the two groups is almost as high as that between blacks and whites.

Anecdotal evidence of colorism abounds. When Henry Louis Gates arrived at Yale as an undergrad in the 1960s, he was subjected to the then common practice known as the “paper-bag test,” in which an African-American fraternity holding a party would deny entrance to anyone darker than a brown paper bag. Spike Lee satirized the practice in his 1988 film School Daze, in which two black sororities, one with light-skinned women and the other with dark-skinned women, battle it out over “good” hair and “bad” hair, and deride each other as Jigaboos or Wannabes. He returned to the subject of colorism in his 1991 film Jungle Fever, which took on the thorny issue of sexual preference for lighter-skinned women.

As well known as these stereotypes are in the black community, however, there is a taboo against talking about them, says Maddox, since the idea that blacks could be seen as prejudiced in any way vitiates the struggle for black racial equality. “The extent to which you acknowledge that blacks pay attention to racial features and use them to determine the ways they interact with each other undermines the beef you have against other people who do the same kind of thing,” he says. It’s not just blacks who employ stereotypes based on ethnic variation. “It’s almost as pernicious among Latin Americans,” says Maddox. “There is a real fondness for people who look more Eurocentric.” The same goes for places like India, where the caste system has favored the pale. “It’s not even under the surface in India,” says Maddox. “You look at the personals in Indian newspapers, and they will say they want someone with lighter skin.”

Maddox, who is black, became interested in the general topic of racial stereotyping while growing up in a white neighborhood in metropolitan Detroit. “I am a fair athlete, but I know people had expectations I would be a better athlete,” he says. Similarly, he was seen as a poor dancer among black relatives, but a great dancer among his white friends. “I had the same skills, but the expectations were different, and they would look at my performance through a different lens.”

It wasn’t until grad school, however, that Maddox encountered the concept of “skin-tone bias” and decided to look into how such perceptions are formed. In doing so, though, he ran up against a resistance to talking about the topic within the psychological community itself. A 1956 study found no difference in the ways that whites perceived light- and dark-skinned blacks. Thereafter, most social psychologists ignored skin tone. “That study was the gold standard,” Maddox says. “People assumed that variation didn’t matter, but there was so much evidence that it did. Maybe I am naïve, but my belief about this is you won’t see diversity in perspectives in social psychology until you see diversity in the researchers.”

Maddox is one of the first researchers to show that skin tone really does matter—and not just to blacks but to whites as well. In an early experiment, he asked black and white students to list their beliefs, as well as the cultural stereotypes they held, about different racial groups, including “light-skinned blacks” and “dark-skinned blacks.” The last category was much more apt to be categorized with labels like “aggressive,” “criminal,” or “poor.” In 2005, Maddox and a colleague, Professor Travis L. Dixon of the University of Illinois, went on to look at how those stereotypes played out in the news media.

Many studies have found that blacks are overrepresented as criminals on evening newscasts. For example, one study in 2000 by Dixon and Daniel Linz of the University of California at Santa Barbara found that 37 percent of the lawbreakers shown on the nightly news in Los Angeles were black, even though blacks accounted for only 23 percent of arrests during the same period.

For their experiment in 2005, Maddox and Dixon expanded those findings to look at the implications of skin tone, showing participants a news program in which they embedded a short story about a murdered police officer. During the story, the program flashed a brief picture of the suspect in the murder, which Maddox and Dixon used computer manipulation to present as a white person or a light-, medium-, or dark-skinned black.

Participants shown a darker-skinned perpetrator were one-third more likely to say the story made them feel concerned about crime than those shown a white perpetrator, while the reactions of those shown a lighter-skinned black suspect were not statistically different from those shown a white suspect. Participants who deemed themselves heavy news watchers were two-thirds more likely to be concerned about the dark-skinned perpetrator. Their response may actually have been “primed” by the overrepresentation of black criminal suspects on newscasts.

“People looked at the feature”—dark skin—says Maddox. “Even though they all belonged to the category”—black—“the extent to which they had that feature made them a better fit” for the negative stereotypes associated with blacks. Since launching the study of skin-tone bias in 2002, Maddox has expanded his hypothesis to include other racial characteristics—for example, the broad nose and full lips of blacks or distinctive eyes of East Asians. He theorizes that these differences will gain importance, relative to racial group membership, as our society grows more integrated and interracial marriages become more common.

Self-Fulfilling Prophecies

As much as racial lines might be blurring, we are still a long way from the time when people will stop identifying themselves as black, white, Asian, or another commonly recognized race. For the first time in 2000, the U.S. Census allowed respondents to check boxes for two or more races on their forms. Only 3 percent of the population did (smaller portions of the populace checked three, four, five, and even six races). According to the research of Tufts social psychologist Nalini Ambady, the box we check on that form might have as big an effect on how we see ourselves as our conception of race has on the way we see other people. Ambady has spent years testing the power of students’ group identity to affect their academic performance.

In a study published in 2001, for example, she and other researchers from Harvard University and the University of Michigan administered a difficult math test to a group of Asian-American girls. For some of them, the researchers very subtly “activated” the girls’ female identity by giving them a questionnaire on the topic of coeducation. For others, they activated their Asian identity by giving a questionnaire about the languages and foods at home. For a third group, they asked “neutral” questions about animals and seasons. With nothing more than that, the girls performed dramatically differently on the test.

The girls who were reminded of their Asian identity scored up to 30 percentage points higher (depending on their grade level) than those reminded of their gender. Subsequent studies by Ambady and her colleagues with Asian boys found that boys’ scores improved when they were reminded of either their race or their gender. With black girls, by contrast, scores decreased when reminded of either category.

“The very fact that we could move around their performance really, really surprised us,” says Ambady, who joined the Tufts faculty from Harvard in 2004. “When we asked them at the end if they had been aware of any sense of their identity, they said no.” The findings suggest that we can internalize stereotypes so deeply that we subconsciously allow them to mold our behavior depending on the images that society feeds us—such as the impression that Asians are good at math, or that blacks aren’t.

To learn how that might work, Ambady and her graduate students at Tufts recently hooked subjects up to EEG sensors to measure the electrical activity in different regions of the brain. “We find that women whose gender is being activated are allocating less attentional resources towards math and more towards verbal,” she says. “They identify and decide to devote more energy to something else.” The good news, at least for gender, is that Ambady’s experiments show a marked change in the upper elementary school years, between 8 and 12, when girls actually do better on math tests if they are reminded of their gender. “This is the age that they are more chauvinistic and compete with the boys,” she says. “To me this is very strong evidence that girls’ underperformance is sociocultural, and not genetic.” For older students facing racial stereotypes, however, the effects can be harder to counter. “It’s the groups that”—academically speaking—“have no positive stereotypes that are the most difficult part,” she says. “We really have to work to develop role models to counter the effect of negative stereotypes.”

That, of course, is a tall order. Difficult as it was to change laws and overthrow discriminatory institutions—the achievements of the Civil Rights era—it is even harder to change the way people think. If the work of these three researchers teaches us anything, it’s that stereotypes are more ingrained than we’d like to admit—and that trying to ignore them, even with the best intentions (à la George Costanza), only makes them worse. “Not that overt forms of racism don’t exist,” says Sommers. “But it’s more fascinating to look at the subtle ways that race influences perceptions and judgments of people who see themselves as fair-minded.”

Maddox agrees that the kind of biases people have shown in his studies don’t necessarily mean they are bigots. “I would stay away from labeling people as racist,” he says. “The ways we interact with each other are a consequence of a broader racism—a systemic racism.” A first step in tackling that deeply embedded societal discrimination is to keep looking at how it operates. Sommers and Ambady are currently collaborating on a study to determine at what age children form race-based stereotypes. Maddox and Sommers are jointly studying the influence of skin tone in jury judgments.

“Knowing is a big part of the battle,” says Maddox. “Even if you are motivated to control prejudice, unless you have an understanding of how it works, you won’t be able to do anything about it. But if you know you might be acting according to false perceptions, maybe it becomes easier to do something about it.”

That sounds like a challenge for the 21st century.

MICHAEL BLANDING is an award-winning magazine writer whose work has appeared in The Nation, The New Republic, Boston Magazine, and the Boston Globe Magazine. He previously taught a class in magazine production at Tufts’ Experimental College.

  © 2006 Tufts University Tufts Publications, 200 Boston Ave., Suite 4600, Medford, MA 02155