Whistling Vivaldi: And Other Clues to How Stereotypes Affect Us (Issues of Our Time)by Claude M. Steele, W.W. Norton, 224 pages, $25.95
Two explanations for the persistence of racial inequality predominate in America's ideological battles. The left blames racism, while the right pins the responsibility on minority groups' own culture and behavior. Some on the right insist that racial minorities face no impediments. Just the opposite, they say -- white people now encounter discrimination because of policies such as affirmative action.
The picture of race that the social psychologist Claude Steele offers in his work is different. Racial identity, as he sees it, affects people's thinking and action in countless ways but often without any awareness or malice. People pick up on cues -- a topic of conversation, the language used in an employment brochure, the number of people of their own race or gender in a classroom or office -- and react on the basis of internalized concerns about being stereotyped. While people in every group register these signals, members of racial minorities often experience a threat -- what Steele calls "stereotype" or "identity" threat -- that impairs their performance.
Recently named provost of Columbia University, Steele has spent years studying how self-image and group identity affect individual behavior. The title of his new book, Whistling Vivaldi, comes from an anecdote from Brent Staples, an African American writer at The New York Times, who discovered he could assuage the fears of white people he encountered walking the streets at night by whistling classical music. The music signaled to the people shrinking from him that he was educated, probably middle-class, and in general not the type of black man they should cross the street to avoid.
In Whistling Vivaldi, Steele primarily focuses on underperforming college students, particularly blacks and women, who by reaching college might be thought to have conquered negative stereotypes. But the research that Steele and others have done indicates that when facing identity threat -- that is, when they are reminded of the shortcomings associated with their group -- these students tend to underperform in comparison to their male or white counterparts with similar scores.
The experiments Steele reviews follow a fairly consistent format: Give a test to at least two groups of students at similar skill levels, and with one group, emphasize that the test measures "cognitive ability," while stressing to the other that the test is a "task" for measuring "problem solving" or signaling in some other way that the test is not an intellectual assessment. For example, black subjects who were told that a test measures "cognitive ability" did worse than whites, while blacks who were told the test does not measure ability performed as well as whites did. Similarly, when informed that women always do as well as men on a particular math test, women did better than when they weren't given that cue. And when told that a miniature-golf test would measure their athletic ability, white men performed at a lower level than when they were told the test measured their "sports strategic intelligence."
Other studies have found that test takers under identity threat show physiological signs of anxiety -- increasing heart rate and higher blood pressure -- even when they say they don't feel anxious. Steele refers to this pattern of denial as "trying to slay a ghost in the room." The Herculean effort of trying to disprove a stereotype, Steele argues, "leaves little mental capacity free for anything else we're doing." These cognitive effects, he explains, are as potent a limit on the life chances of minorities as the segregated institutions of his youth in Chicago.
Steele's observations about black students are especially disquieting. Their academic struggles, aren't due to weaker skills or lesser motivation. According to Steele, far from expecting "someone else to do it for them," black students have often internalized the notion that they have to be "twice as good" as anyone else if they are to succeed. Identity threat pushes them to be independent to the point of self-sabotage. Compared to Asian and white students, black students are less likely to study in formal or informal groups, to ask for help from other students or teaching assistants, or to drop classes in which they're doing poorly, and they are more likely to reject feedback from teachers.
In one of the experiments Steele discusses, the black test subjects under stereotype threat opted to do more of the difficult problems on the test than did their white counterparts, who in this situation had nothing to prove. When the black students then did badly, it was a crushing confirmation of group shortcomings -- proof that they didn't "belong there." In fact, the more minority students worried they were being stereotyped, the worse they did. The students who cared more about their education were the ones more likely to underperform. Stereotype threat didn't affect the test subjects who lacked the appropriate skills or didn't care about school to begin with.
Steele also shows how stereotype threat can be mitigated to the point of irrelevance. In an experiment with no identity threat, the black test takers were able to focus on completing the test, and the disparity with whites disappeared. Likewise, Steele points out that giving more constructive feedback, increasing the population of minorities, and encouraging black students to study with one another in groups can entirely eliminate the disparities caused by identity threat -- at least among those students whose deficiencies are caused by identity threat, not by a gap in actual skills.
Steele's experiments show that identity threat can affect a member of any group given the right context, but minorities and women face more of a problem because the stereotypes that work against them are so deeply rooted in our culture. The stress caused by identity threat then becomes an ever-present burden. Steele points to high rates of diseases such as hypertension among African Americans as an example of the physiological effects of coping with identity threat on a constant basis.
Toward the end of the book, Steele explores more deeply the effect of identity threat on race relations. He focuses particularly on the hypothesis that whites avoid interaction with blacks because they fear being seen as racist and don't want to confirm that stereotype. In one experiment, he assembled two different groups of white students, telling the first they would have a conversation about relationships and the second that they would discuss racial profiling. Each subject was to talk with two black men. When the subjects entered the room, they were asked to rearrange the chairs, and the whites who were told they were to discuss racial profiling tended to move the other chairs farther away from their own. To be sure that the white students weren't just reacting to the topic itself, Steele redid the experiment with whites alone, and this time, the subjects who were told they would discuss racial profiling didn't try to place their own seats farther apart from the others.
It's easy to see how the patterns Steele identifies perpetuate social segregation. One side anticipates the other's stereotype that it is inferior, while the latter fears that it will be seen as prejudiced -- and often both sides have no understanding of why they act the way they do. Social segregation can then persist in perpetuity, chugging along with the inertial force of centuries of legalized racist hegemony. The good news is that if deliberate measures can reduce identity threats for students taking a test, there may be ways to reduce those threats more generally.
Steele's research suggests how educational institutions might improve performance among women and minorities. But as he acknowledges, addressing stereotype threat will not affect the underlying problems of poverty, lack of access to good schooling, or estrangement from profitable social networks and other forms of social capital. As Steele notes, "Even if a magic wand waved away all of the prejudice in our society, there would still be pressures keeping us apart."
The political implications of some of the research are also disturbing. We tend to "feel" our identities more acutely when that identity is under siege, but even that minimal identification is enough to make us discriminate against others we see as part of a competing group. That pattern goes a long way toward explaining why Republican white identity politics have been so effective at discrediting liberal attempts to expand the social safety net. Twenty years ago, right-wingers talked about "welfare queens." Now Rush Limbaugh derides the Democrats' health-care reform plan as "reparations" and "a civil-rights bill," coded language calculated to raise the specter of exploitation by racial others.
Whistling Vivaldi conveys an understanding of why race remains such a powerful factor even in a society where racial discrimination is seen as abhorrent and prohibited by law. And while Steele's research gives some measure of hope, I'm not betting that people would change even if they learned about his findings. Dealing with stereotype threat requires a willingness to admit the power that group identities have over our minds and actions, and most of us don't believe that we're susceptible to bias, whether we're the guy whistling classical music or the person sighing with relief and deciding not to cross the street.