What Counts as Evidence of Racism?

I like to think of The Manhattan Institute as the poor person’s AEI. One of MI’s “scholars” is Heather MacDonald, an old time culture warrior (her book The Burden of Bad Ideas is about the damage caused to America by the 1960s, an obsession of theirs). In 2003 she collected some of her City Journal columns into a book called Are Cops Racist?, a collection she apparently believes satisfactorily resolves that either/or question in the negative.

Heather MacDonald seems to have never heard a statement made by a cop that she didn’t take at face value. In these collected columns, she uses two kinds of arguments to back up her claim that racism is not a significant factor in policing, and that instances of racial profiling show evidence merely of “pockets” of racism. One is a critique of statistical studies that purport to prove cops are racist by showing that blacks are stopped on the highway at a higher rate than their numbers in the population would seem to warrant, or that they are arrested at higher rates than whites. The other is quoting police officers about their intentions. The first deserves to be taken seriously; the second does not. One wonders why, while she takes cops’ statements uncritically, she consistently discounts blacks’ claims that they are stopped over and over for no reason when they are driving. I’ve seen her on television call-in shows taking phone calls from one exasperated black man after another recounting their experiences with racism. She simply insists, with a sunny smile on her face, that they must have been doing something wrong. This obvious arbitrariness makes it hard to believe that she does not have an agenda.

Nevertheless, she raises a legitimate question. Her claim is that “systematic racial profiling by police does not exist.” As mentioned above, one of two main kinds of evidence she produces to try to support this claim are statements by police of their intentions. Unsurprisingly, not many officers cop to approaching people on the basis simply of their race, unless police are specifically looking for a suspect who has been described as black. The other kind of evidence she relies on is data on “highway stops.”

MacDonald cites a study done of police stops of drivers on the New Jersey Turnpike as evidence that police do not use race as a reason to stop drivers. The study, unlike research on racial profiling that uses population measures for benchmarking, establishes a “violator benchmark” by which to measure whether officers stop drivers arbitrarily. Researchers at the Public Service Research Institute in Maryland used high-speed cameras to take photos of almost forty thousand drivers. Then they showed the photos to evaluators who determined the race of each driver. The photos were correlated with data collected from a radar gun, and the researchers determined that “black drivers speed twice as much as white drivers, and speed at reckless levels even more.”

MacDonald is right that to make sense of data on police stops of drivers, or of any kind of arrest data, one needs to compare such data to a relevant benchmark; otherwise there is no way to avoid arguing in a circle, or assuming what you are trying to prove. She argues that population is not an adequate benchmark because using population assumes that blacks commit crimes at the same rate as other racial groups—thereby assuming the proposition that is to be proved. She’s right about this. That doesn’t mean, however, that using population as a benchmark doesn’t tell us anything useful about racism. If blacks commit more crimes or, more accurately, more of the kinds of crimes for which arrest data is available, it doesn’t necessarily disprove the existence of racism. What such data means depends on how you think the world works.

The problem with relying on “stop data” to try to prove a contention as generalized as “systematic racial profiling by police does not exist” is that the phenomenon she is trying to disprove is more complicated than what can be captured in “stop data.” Her focus on this kind of data, as well as explicit statements she makes, gives an indication of her view of how the world works, which in turn determines what she accepts as evidence in the first place. She does not accept any arguments about the existence of racism that rely on “end results,” such as higher unemployment rates for blacks than for whites, or higher incarceration rates, or higher poverty rates. She even puts the phrase “root causes” in sneer quotes when criticizing civil rights-minded people for looking past the surface of things for explanations of such differentials. She says that

…“looking at the end result” is the hallmark of bogus civil rights analysis, designed to shift attention from individual deficits onto the resultant disparate outcomes, which are then attributed to racism. We can argue about whether society is, in some structural way, still somehow to blame for Cincinnati’s idle, functionally illiterate young dealers, but let’s not brand employers as racist.

Despite her grudging concession to the possibility that structural problems exist, other statements she makes about idle, illiterate do-rag clad drug dealers who hang out in front of places of business shooting holes in the walls show that she believes people end up where they are according to their individual moral choices. She already believes that the “end results” often cited as evidence of systemic problems are primarily the result of individual choices. What she accepts as evidence for any kind of racial differential must by definition fit into this individualistic perspective. Taking at face value what police say to her about their intentions when they stop drivers when she is interviewing them for articles on how unfairly the police are treated might be naïve, but it is consistent with her view of how the world works. Her reliance on this kind of surface analysis is comparable to claiming that corporations don’t pay taxes merely because they take advantage of legal tax shelters, while ignoring corporations’ role in influencing the legislation that creates the tax shelters in the first place.

What if blacks do speed more than whites on the New Jersey Turnpike? That fact, whether substantiated or not, doesn’t generalize to an assertion that “systematic racial profiling by police does not exist.” It doesn’t explain the over-representation of blacks in the prison system, since nobody goes to prison for traffic violations. MacDonald claims blacks are not “over-represented” because blacks commit more crime than whites. She makes the assertion that “some officers inevitably started playing the odds” to explain why they stop blacks more than whites on known drug trafficking routes into New York City. The problem of circularity is especially difficult with this issue. To make any valid claims about profiling there needs to be some independent source of data on crime rates for comparison purposes, besides the rates generated by police making arrests and courts handing out sentences. MacDonald makes the facile claim that police “go where the [drug] deals are.” There is evidence, however, according to Jennifer Wynn, that “a high proportion of people from all walks of life have at some time or another committed serious crimes.” In her book Inside Rikers: Stories from the World’s Largest Penal Colony, Wynn cites a study of 1,700 relatively well-off New Yorkers who had never been arrested that showed

91 percent had committed either a felony or a serious misdemeanor. Sixty-four percent of the men and 27 percent of the women had committed at least one felony, for which they would have been sent to state prison had they been caught.

People commit the kinds of crimes they have the opportunity to commit. Wealthy people commit “white collar crime” because they are in a position to do so. There are systematic biases in the likelihood of getting caught, depending on the crime and the criminal. Some kinds of crimes are committed outdoors, and in neighborhoods where police patrol and harassment is common; it certainly makes sense that such crimes will come to the attention of patrol officers at a higher rate than crimes committed in high rise bathrooms or inside large homes in wealthy neighborhoods. Benchmark data on highway stops doesn’t get at this kind of problem of race, class and crime.

A main anchor for much of MacDonald’s argument, a point she begins with and goes back to over and over, is the dramatic drop in crime in New York City since the mid-nineties, which many argue was brought about by improved police tactics, such as a crackdown on nuisance offenses, holding precinct commanders accountable for producing results, and the use of Compstat, which uses computers to map crime patterns in the city. She claims that since the adoption of Compstat, policing is now data-driven and, therefore, cannot possibly be racist (She said this on Ben Wattenberg’s show “Think Tank” on the subject “Is Racial Profiling Real?”). She is probably right to some extent. But I’m not sure there is anyone who claims police are always racist. She argues that because more punitive police tactics in New York City are responsible for the dramatic drop in crime since the 1990s, those police tactics are justified whether or not they happen to target blacks more than whites.

That police tactics are the cause of New York City’s crime drop, however, is debatable. Andrew Karmen argues (in an article in Justice Policy Journal v 1 n 1 in 2001) that because the changes in police tactics in New York City were introduced throughout the city, there is no basis for comparison within the city and therefore no basis for a defensible evaluation of the effect of those changes. Moreover, he gives several reasons to doubt the connection. First, the reforms were instituted several years after crime rates began to drop after peaking in 1990. New York City had the highest ratio of resident to officers in the country, but after 1995 when the rate of increase of officers peaked, crime continued to drop. In addition, while murders committed outdoors continued to decline after more officers were hired, murders committed indoors, where activities are not visible to officers on patrol, also declined. Moreover, crime fell in cities that did not hire more police. One reform in New York City was based on the “broken windows” thesis, promoted by James Q. Wilson, according to which treating minor crimes more punitively will forestall more serious crime; but in NYC, where more people were arrested and jailed for minor infractions, the average length of stay in jail decreased, suggesting there was an increase in the numbers of minor offenders who were given merely a “slap on the wrist,”—just the kind of policy Wilson and others claim cause higher crime rates—and still crime rates decreased. Moreover, in cities across the country that experienced similar drops in crime, police did not make numerous minor arrests at all. Finally, Karmen shows that while incarceration rates rose steadily in New York City from 1970 through 1998, then dropped a bit in 2000, the murder rate fluctuated until about 1990, then dropped precipitously thereafter, showing no apparent connection at all to incarceration rates or to police tactics. (See Karmen’s New York Murder Mystery: The True Story Behind the Crime Crash of the 1990s, published 2000.)

Contrast MacDonald’s individualistic perspective to that of an ex- convict who has first-hand knowledge of the circumstances in which many inner city blacks exist. Kenneth took a class called Fresh Start at Rikers, taught by whites. After he was released he became a counselor for Fresh Start inmates at Rikers. He told Jennifer Wynn that

“Sometimes it seems like every black man like me who’s gotten ahead has had a white man in the background,” he says. “I look at my job now—and sometimes I feel like the only reason I got that job is because John recommended me. It’s not what you know, it’s who you know, but the problem is that most black folks don’t know anybody in a position to help them. So we always have to go to white folks to get the jobs, the training, the things we need…. I told a couple of white guys, ‘I don’t mean nothin’ against you, but society accepts you for who you are because of your race. You’ll leave this jail and you’ll be all right because society expects you not to be in jail. Let’s face it—for a white man to go to jail he gotta be a real fuckin’ idiot’….But for us black men,” he says, “society expects us to be in jail, and that’s what we’ve come to believe—like jail doesn’t mean nothin’ to us ‘cause we’re black men, like we rob and we steal because that’s what black men do. We’ve come to believe that’s what we do. Self-destruction is so deeply rooted in us.” (From Inside Rikers, p. 202)

There are several research programs contained in these few sentences. The point is not that individual choice has no impact on the outcome of a person’s life, but that every individual makes choices within a context. The sociological concept Kenneth was using of “social capital” gets at this point. Social capital consists partly in networks that connect people who need resources with those who have them. Those who do well in life have more connections to those who can help them than those who do less well. They have more opportunities to make choices that land them in jobs that can support them, or in college, or in homes in safe neighborhoods.

There are many studies besides those that use “stop data” that show differential treatment of blacks and whites that can’t be explained without reference to racism. Both blacks and whites are more likely to end up on death row if they’ve killed a white person than if they’ve killed a black person. (See the article by Blume, Eisenberg and Wells in 3/04 issue of The Journal of Empirical Legal Studies for some actual evidence.) Blacks serve longer sentences for drug crimes than whites and Hispanics, partly because the penalties for possessing crack cocaine are harsher than for powder cocaine. The differences in physical effect of crack and powder cocaine are small, so the distinction drawn in sentencing can’t be attributed to a concern over those effects.

I think it’s the individualist framework of people like MacDonald that prevents them from getting beyond a polarized, either/or conception of these issues. Looking past the surface of things to discover why people make the choices they do is not to make excuses for behavior. Showing, as MacDonald does in her collection of commentaries, that liberals can grandstand with the best of them about the Cincinnati riots or the Amadou Diallo police shooting doesn’t prove that riots or demonstrations are caused by liberal grandstanding, as she asserts, or that blacks commit more crimes than whites, or that racism occurs only in “pockets” and is therefore an insignificant factor in policing and incarceration in the United States. She has not marshaled anything a thinking person would call “evidence” that might support her assertion that differential outcomes in the criminal justice system are not caused by substantial, meaningful levels of racism.