The Diversity Delusion

by Heather Mac Donald

  Implicit-bias researchers do not discuss the cognitive skills gap. I asked Greenwald if gaps in academic preparedness should also be considered in explaining socioeconomic disparities. He responded simply by offering up more wellsprings of bias: “There are sources of unintended disparities other than implicit bias (esp. institutional discrimination and in-group favoritism).” But a 2014 study for the Federal Reserve Bank of Chicago by economist Bhashkar Mazumder found that differences in cognitive skills measured by the Armed Forces Qualification Test account for most of the black–white difference in intergenerational mobility. Blacks and whites with the same score on the AFQT have similar rates of upward and downward mobility. The AFQT should overpredict upward mobility for blacks if bias were holding them back; it does not.

  The iron grip of the implicit-bias concept on the corporate world will merely result in a loss of efficiency as workers are again trundled off to this latest iteration of diversity training and are further pressured to take race into account in personnel decisions. Most ominously for productivity, signatories to the CEO Action for Diversity & Inclusion have pledged to encourage more conversations among their employees about race, even though a recent report found that 70 percent of employees are not comfortable discussing race relations at work—understandably, given the potential tensions created by diversity preferences and the oversaturation of race talk in American life. Procter & Gamble is on the steering committee of the CEO Action for Diversity & Inclusion. You would think that its managers would have better things to do than lead bull sessions about racial microaggressions, in light of the company’s lackluster growth over the last decade and the ongoing fight for control of its board.

  But it is in law enforcement that the mania for implicit-bias training exacts its most serious cost. Police officers unquestionably need more hands-on tactical training to avoid ending up in a position that requires the use of force. Officers need tools for keeping their cool in highly charged, hostile encounters. They should practice de-escalating confrontations and gaining voluntary compliance. Some officers pay out of their own pocket for tactical training, since their departments offer too little of it. But now there will be less time and departmental money available for the necessary skills upgrades because precious training resources are being diverted to the implicit-bias industry. And that wasteful training is being carried out in the name of a problem that does not even exist: bias-driven police killings of black men.

  Joshua Correll, a psychologist at the University of Colorado, has been studying police shoot/don’t shoot decisions for years. His experiments require officers to react to rapidly changing images of potential targets on a computer screen. He has found that officers are no more likely to shoot an unarmed black target than an unarmed white one. Officers are slightly quicker to identify an armed black target as armed than an armed white target, and slower to identify an unarmed black target as unarmed than an unarmed white target. But the faster cognitive processing speeds for stereotype-congruent targets (i.e., armed blacks and unarmed whites) do not result in officers shooting unarmed black targets at a higher rate than unarmed white ones.

  Correll’s conclusions were confirmed in 2016 with the release of four studies that found either no antiblack bias in police shootings or a bias that favored blacks. Three of the studies—by Roland Fryer, Ted Miller, and the Center for Policing Equity—reviewed data on actual police use of force; a fourth put officers in a more sophisticated life-size video simulator than the computers that Correll uses.6 That study, led by the University of Washington’s Lois James, found that officers waited significantly longer before shooting an armed black target than an armed white target and were three times less likely to shoot an unarmed black target than an unarmed white target. James hypothesized that officers were second-guessing themselves when confronting black suspects because of the current climate around race and policing.

  Both experimental and data-based research, in other words, dispel the claim that police officers are killing blacks out of implicit bias. That has not stopped the implicit-bias juggernaut, however. Police departments across the country are subjecting their officers to implicit-bias training at considerable cost; any controversial shooting invariably triggers a pledge to bring in the bias consultants. The New York Police Department will now start requiring recruits and officers already on the job to attend a full-day seminar in implicit bias, time that could be better spent practicing tactical and communication skills.

  All the IAT-inspired lecturing cannot change the reality that drives police activity: the incidence of crime. And that is a topic about which implicit-bias trainers have little to say, as I discovered while observing a three-day training program in Chesterfield, Missouri, in May 2016.

  About three dozen officers and supervisors had come to this green suburb of St. Louis from as far away as Montana, Virginia, North Carolina, Michigan, and Kentucky for a “train-the-trainer” session offered by the premier antibias outfit in the field. Lori Fridell has been lecturing to police departments about bias-based policing since the “driving while black” notion emerged in the 1990s. But the implicit-bias idea has boosted her business enormously, as has the Black Lives Matter movement. In 2016, Fridell was fielding a call a day from police departments, courts, and other parts of the criminal-justice system. The Obama Justice Department funded her organization’s implicit-bias trainings for police departments that it considered particularly troubled. Other agencies pay their own way.

  A day and a half into the three-day Chesterfield training, the attendees had been informed that the Michael Brown shooting was a function of implicit bias (even though Brown had tried to grab the officer’s gun and had assaulted him) and that the overrepresentation of blacks in prison was because blacks get longer sentences than whites for the same crime (in fact, sentences are equal, once criminal history is taken into account). The attendees had learned about the IAT; they had watched a video of singer Susan Boyle’s victory in the television show Britain’s Got Talent; they had viewed photos of a hot babe on a motorcycle and a female executive with a briefcase; they had written down stereotypes about the “unhoused”—not activities directly related, say, to serving a felony warrant safely.

  The theme of these exercises was that everyone carries around stereotypes, and that to be human is to be biased. In the case of police officers, the two trainers explained, those biases could put an officer’s life in jeopardy if he discounts a potential threat from a white female or a senior citizen because it is counter-stereotypical. But those implicit biases are also killing black men, said trainer Sandra Brown, a retired Palo Alto police public-affairs lieutenant.

  Brown described a study by Stanford psychologist Jennifer Eberhardt in which Stanford students in a psych lab were shown a blurry object on a computer screen. The students were quicker to identify it correctly as a gun if they had been shown an image of a black face right beforehand. (Greenwald and Banaji also invoke this study.) “Black men are dying because we see the gun too quickly,” Brown said—never mind that the aforementioned research on police shootings shows that black men are not dying because police officers “see the gun too quickly.” Why might such a priming function occur? Eberhardt and her coauthors, of course, attributed it to irrational stereotype. But another explanation comes to mind: Blacks are objectively more associated with crime. The Chesterfield training only tiptoed up to this topic.

  It is “partially factual,” Brown said, that “people of color” are disproportionately involved in street crime. Actually, it is fully factual; street crime today is almost exclusively the province of “people of color.” In New York City, for example, blacks and Hispanics committed 98 percent of all shootings in 2016; whites, who, at 34 percent of the population, are the city’s largest racial group, committed less than 2 percent of all shootings.7 Those figures come from the victims of, and witnesses to, those shootings. These disparities are repeated in cities across the country. If you’re hit in a drive-by shooting, the odds are ov
erwhelming that your assailant will be black or Hispanic—and that you will be, too, since blacks and Hispanics are usually the victims of such crimes. If the public associates blacks with violent street crime, it is the tragic facts that lead to that association.

  Yes, a police action should not be based on a “stereotype,” as Brown rightly admonished. But crime is the overwhelming determinant of policing today, and to pretend that implicit bias drives policing distracts from the challenges that officers face. By day two, the audience was interjecting some social and political reality back into the training. “Are there any studies about black and white officer shootings?” asked a black officer. “No one’s outraged if I shoot a black, but if a white officer does, it will be pandemonium.” Another local officer said that he worried about the violence in the black community: “It’s so disproportionate. When black people are shot by other blacks, it doesn’t make the news. There were over a dozen people shot in a theater the other day. I worry about that disparity.”

  Then an officer from Chesterfield raised the most pressing concern in the Black Lives Matter era: depolicing. Seventy-five percent of the apprehended shoplifters in the Chesterfield mall were black, he said. (Chesterfield’s black population was 2.6 percent in 2010.) “We struggle with depolicing; it’s difficult to tell officers to enforce the shoplifting laws when they will be confronted with the implicit bias issue.” That is the dilemma facing officers today: If they enforce the law, they will generate the racially disproportionate stop-and-arrest statistics that fuel specious implicit-bias charges. But it is the reality of crime, not bias, which results in those disproportions.

  The trainers had nothing to offer to resolve this problem. “It’s hard to answer these tough questions,” Brown said. Her partner, Scott Wong, also from the Palo Alto Police Department, gamely tried to bring the discussion back to the official topic. “You need a passion for this; you have to believe in implicit bias and how it affects officers.” But while many officers could do with a courtesy tune-up, they are overwhelmingly not making bad decisions based on invidious stereotypes. What they are doing, on a daily basis, is trying to deal with the breakdown of family and bourgeois norms in inner-city areas that leads to so many young black men gang-banging in the streets. Joshua Correll has found that officers’ neurological threat response is more pronounced when confronting black suspects. Might that be because black males have made up 42 percent of all cop-killers over the last decade,8 though they are only 6 percent of the national population? Or because the individuals involved in the daily drive-by shootings in American cities are overwhelmingly black? Until those realities of crime change, any allegedly “stereotypical” associations between blacks and crime in the public mind will remain justified and psychologically unavoidable. Those crime rates will also affect the pool of job candidates without a criminal record, further reducing the likelihood of proportional representation in the workplace.

  The Chesterfield training did offer several profound pieces of advice: “Make every day the day you try to change someone’s perceptions” of the police, Brown said. She urged officers to get out of their cars and talk to civilians: “They need to know us; people are afraid to talk to us as human beings.” However sage this message, though, it should not be necessary to contract with a pricey implicit-bias trainer to convey it.

  The implicit-bias crusade is agenda-driven social science. Banaji seems to see herself on a crusade. In an email to New York’s Jesse Singal, she attacked both the credentials and the motives of the academics who have subjected the IAT narrative to critical scrutiny: “I don’t read commentaries from non-experts,” she wrote (those “non-experts” are overwhelmingly credentialed psychologists, like herself). “It scares people (fortunately, a negligible minority) that learning about our minds may lead people to change their behavior so that their behavior may be more in line with their ideals and aspirations.” The critics should explore with their “psychotherapists or church leaders” their alleged obsession with the race IAT, she suggested. Kang has accused critics of holding a “tournament of merit” vision of society and of having financial reasons for IAT skepticism. (Of course, the fact that Banaji and Kang hire themselves out as IB trainers, for “non-trivial … fees,” as Kang puts it about himself, and that Greenwald serves as a paid expert witness in discrimination lawsuits, does not lead Kang to impute financial reasons for such pro-IAT advocacy.)

  * * *

  A thought experiment is in order: If American blacks acted en masse like Asian Americans for ten years in all things relevant to economic success—if they had similar rates of school attendance, paying attention in class, doing homework and studying for exams, staying away from crime, persisting in a job, and avoiding out-of-wedlock childbearing—and we still saw racial differences in income, professional status, and incarceration rates, then it would be well justified to seek an explanation in unconscious prejudice. But as long as the behavioral disparities remain so great, the minute distinctions of the IAT are a sideshow. America has an appalling history of racism and brutal subjugation, and we should always be vigilant against any recurrence of that history. But the most influential sectors of our economy today practice preferences in favor of blacks. The main obstacles to racial equality at present lie not in implicit bias but in culture and behavior.

  * * *

  CONJURING DISRESPECT

  * * *

  The attempt to find implicit bias in policing has come to this: the difference between an officer saying “uh” and saying “that, that’s.” According to a team of Stanford University researchers headed by psychology professor Jennifer Eberhardt, police officers in Oakland, California, use one of those verbal tics more often with white drivers and the other more often with black drivers. If you can guess which tic conveys “respect” and which “disrespect,” you may have a career ahead of you in the exploding field of bias psychology.

  In June 2017, the team of nine psychologists, linguists, and computer scientists released a paper purporting to show that Oakland police treat black drivers less respectfully than white ones. The study, published in the Proceedings of the National Academy of Sciences, elicited a huzzah from the press.

  The Washington Post, The New York Times, and Science, among many other outlets, gave it prominent play. “Police officers are significantly less respectful and consistently ruder toward black motorists during routine traffic stops than they are toward white drivers,” gloated The New York Times.

  Reading the coverage, one expected reports of cops cursing at black drivers, say, or peremptorily ordering them around, or using the N-word. Instead, the most “disrespectful” officer utterance that the researchers presented was: “Steve, can I see that driver’s license again? It, it’s showing suspended. Is that—that’s you?” The second most “disrespectful” was: “All right, my man. Do me a favor. Just keep your hands on the steering wheel real quick.”

  The researchers themselves undoubtedly expected more dramatic results. Undaunted by the lackluster findings, they packaged them in the conventional bias narrative anyway, opening their study by invoking the “onslaught of incidents” involving officers’ use of force with black suspects that have “rocked” the nation. A cofounder of the Black Lives Matter movement helpfully commented in the San Francisco Chronicle that the study goes beyond individual racism to highlight a “systemic set of practices that has impacts on people’s lives.”

  The study is worth examining in some detail as an example of the enormous scientific machinery being brought to bear on a problem of ever-diminishing scope, whether in police departments, on campus, or in American society generally. The most cutting-edge research designs, computer algorithms, and statistical tools, such as Fisher’s exact tests, Cronbach’s alpha, and Kernel density estimation, are now deployed in the hunt for crippling white racism, while a more pressing problem—inner-city dysfunction—gets minimal academic attention.

  The Oakland Police Department gave Eberhardt virtually unlimited access
to its policing data as part of a federal consent decree governing the department’s operations. Her first study of the department—on racial profiling in police stops—managed to run nearly four hundred pages without ever disclosing black and white crime rates in Oakland. (Hint: They are vastly disparate.)

  This latest study analyzed officer body-camera footage from 981 car stops that Oakland officers made during April 2014. Blacks were 682 of the drivers in those stops, whites 299. The resulting officer-driver conversations yielded 36,738 discrete officer utterances. In the first phase of the study, college students rated 414 of those officer utterances (1.1 percent of the total) for levels of respect. The students were shown what, if anything, the driver said immediately preceding each officer statement but were not shown any more of the earlier interaction between officer and driver. They were not told the race of the driver or officer or anything else about the stop. The students rated police utterances to white drivers as somewhat more respectful than those to black drivers, though the officers were equally “formal,” as the researchers defined it, with drivers of both races.

  In the second phase of the study, the linguisticians tried to tease out which features of the 414 officer utterances had generated the student ratings. They came up with twenty-two categories of speech that seemed most determinative. On the positive scale were, inter alia, officer apologies, the use of surnames, the use of “um” and “uh” (known in linguistics as “filled pauses”), use of the word “just,” and what is referred to as “giving agency” (saying “you can,” “you may,” or “you could”). The eight negative categories included asking a question, “asking for agency” (phrases such as “do me a favor,” “allow me,” “may I,” “should I”), “disfluency” (a repeated word such as “that, that”), informal titles (“bro,” “my man”), first names, and, most disrespectful, the phrase “hands on the wheel.” If some of those distinctions seem arbitrary—“could I” is disrespectful, “you could” is respectful; “um” is respectful,” a word repetition is not—they are. More important, they are minute and innocuous. The twenty-two categories each received a score allegedly capturing their degree of respect or disrespect, with apologizing at the top of the respect scale and “hands on the wheel” at the bottom. There were no categories for swear words or even for unsoftened commands, presumably because officers never engaged in those forms of speech.