Possibly the only thing worse than racism itself is the pseudo-intellectual way in which some seek to justify it. For instance, consider the standard conservative response to those of us who argue that the criminal justice system is the site of significant racialized unfairness. Whether the subject is racial profiling, stop-and-frisk rates, arrest rates, rates of incarceration, or the rates at which blacks are shot by police, those on the right are quick to dismiss disparities in these areas by claiming that because rates of criminal offending are higher in black communities, disparities in enforcement of the law are only to be expected.
This line of reasoning has been the default position, for instance, of conservative scholar Heather MacDonald, whose new book, The War On Cops, is but the latest in her years-long attempt to rationalize away any and all disparities in the justice system. According to MacDonald — who previously made this case so as to defend the NYPD’s stop-and-frisk policies, and who now uses the same logic to justify disproportionate use of force against blacks by police — if rates of arrest, incarceration, and the rates at which blacks experience police force are consistent with rates of criminal offending, there is no evidence of racism.
But there are several problems, both theoretical and concrete, with these arguments.
First, although black arrest and incarceration rates for crimes like murder, aggravated assault, rape and robbery, do roughly mirror the rates at which blacks commit those crimes, arrest and incarceration rates for other offenses (especially drug offenses) suggest a significant disproportionality, above and beyond rates of black offending. In other words, even using the standard of analysis preferred by the right, there is still evidence of bias in the justice system.
As I have explained elsewhere, once one compares the best estimates we have of drug usage rates, with the rates at which whites and blacks are arrested for drug possession, there are roughly 160,000 blacks each year who are arrested, above and beyond what their rates of offending would otherwise justify; and there are 160,000 whites who are not arrested, who would be, if arrest rates mirrored rates of drug possession violations. That is no small degree of disparity. Over a decade, it means 1.6 million more blacks and 1.6 million fewer whites with drug records than would be the case in a system where law enforcement treated all equitably.
But even for more serious crimes, “controlling” for offending rates as a way to debunk the possibility of racial bias is an inherently flawed manner in which to analyze the subject matter. Think for a minute about what is being said when people make this argument. They are saying that if blacks as a group commit x percent of crime in a community or nation, there can be no racial bias operating so long as the share of persons stopped and frisked, arrested, prosecuted or incarcerated for those crimes is equal to x, or below that level. Only if rates of arrest and punishment exceed x by some appreciable amount can we say that racism could be the culprit. But how is this true at the level of actual personal experiences and cases? How does the fact of general offending by blacks at x percent mean that racism can’t be operating in given cases, even many of them, involving not abstract black people but actual black people who may or may not be guilty of anything?
For instance, let’s say that in a given community, 90 percent of the violent crime is committed by black people, perhaps because the community is pretty much all black. And let’s say 90 percent of the people stopped by police in this community, on suspicion of having done something wrong are black, as are 90 percent of the people arrested, prosecuted and sent to jail from that community. According to conservative theory, these aggregate numbers would prove there had been no racism operating. But actually, those numbers could look exactly the same, whether there was no racism or complete racism. How? Simple.
Just for the sake of argument — and to test this hypothesis — imagine that every cop in this neighborhood’s precinct was a bigot who targeted black people randomly for stops and frisks, without much or any reason to do so, based solely on their own bigotry, which they “justified” in their own minds by reference to aggregate data about crime rates. Obviously, at the individual, experiential level, those stops would involve racism. They would involve cops singling people out because they don’t like African Americans and/or perceive them all as interchangeable criminals. And those black people they single out would be innocent in almost every case. So the fact that other black people in the neighborhood were not innocent, would not change the fact that those who were stopped would have experienced a racialized injustice.
In a slightly less blatant fashion, this is roughly what blacks experienced for years under stop-and-frisk in New York. Only six percent of persons stopped ended up being arrested, and less than 1.5 percent were found with drugs or weapons. Overwhelmingly the black people who were stopped — who comprised 52 percent of all persons stopped — had not done anything. But according to MacDonald, to the extent blacks commit a disproportionate share of crime in New York — and over 85 percent of all shootings — there was no racism operating. If anything, blacks were stopped less often than they should have been, given the crime data. But this argument is almost stunning for its logical and factual ignorance.
First, only about 15 percent of stops under stop-and-frisk were made by officers who were investigating violent crime—this, according to the police department’s own records. So the fact that blacks commit a disproportionate share of violent crime — and nearly 9 in 10 shootings — has little bearing on something like stop rates. If I’m not stopping you on suspicion of having committed one of those crimes, what relevance does that data have? None, by definition. Stops were overwhelmingly for subjective causes like “furtive movements,” suspicion of trespassing, or simply because the person stopped was in a “high crime neighborhood,” the last of which justifications could obviously be used to rationalize stopping every single person who lives there, every single day, regardless of actual behavior, let alone criminal guilt. To deny that such a thing as that constitutes racism (and almost by definition), suggests that in the mouth of Heather MacDonald, words have no meaning and language is dead.
Secondly, many of the stops were for suspicion of drug activity, but since whites use drugs at roughly the same rate as blacks (and actually deal them at similar rates as well), as noted previously, disproportionate stops for drugs cannot be justified with reference to rates of infraction.
Getting back to the main point: racism cannot be proved or disproved based solely on aggregate statistical comparisons; but rather, real-world facts involving actual interactions. As a more obvious example, let’s imagine a community where all the violent crime was committed by black people because the entire community was black. Under conservative theory, nothing the police might do in that community to actual black residents could be considered racism. Nothing. Because no matter how many people they stop, frisk, beat, kill, or arrest, the percentage of those who experienced these things and were black could not exceed 100 percent, by definition! So even if there were a group of police in this community who were secretly members of some Neo-Nazi gang (and thus, by anyone’s standards, racists), nothing they proceeded to do to the residents there could qualify as racism to hear Heather MacDonald tell it, at least not so far as data would allow us to see. Even if these cops conspired to single out black people in the community to murder, solely to satisfy their racial animus against blacks as a group, statistical inference would exculpate them of any racist injustice.
So let’s imagine these officers went out and each murdered one black person a week, targeting them and then fabricating evidence that would allow them to claim that the blacks in question had pulled weapons on them, thereby justifying the shootings. And let’s assume that only black people were shot by police in this community. Since 100 percent of the crime had been done by blacks and 100 percent of the people shot were black, everything would be considered fine, presumably. Obviously, I’m not suggesting that this is what happens — that cops just go randomly murder people as some standard practice — but the point is, even if they were, the model of statistical inference upon which the right relies to debunk racism would acquit the police of any wrongdoing, so long as their misdeeds remained undiscovered. Their racism would be obscured by the aggregate data.
By the same token, using conservative logic, virtually nothing Bull Connor’s police might have done to black folks in terms of disparate law enforcement in early 1960s Birmingham would have been racist either. Since neighborhoods were segregated, for instance, it goes without saying that virtually all the crime in black neighborhoods (save, perhaps, Klan bombings) would have been committed by black people, since that’s who lived there. So if Bull Connor’s cops regularly stopped and harassed black people for no reason other than to satiate their bigotry, and framed certain black folks for crimes they didn’t commit — and surely even the Heather MacDonalds of the world can’t deny this might well have been common at the time — data would not indicate anything untoward. After all, 100 percent of the crime in the neighborhood would have been done by blacks, and thus we should “expect” 100 percent of the stops, arrests, beatings, shootings and incarcerations from those neighborhoods to be of blacks as well. The fact that the blacks stopped, arrested, beaten, shot or incarcerated could theoretically have all been innocent, and the victims of racist policing — even as other blacks in the community really were committing those crimes — is unaccounted for and unrecognized by the logic of conservative denialists. To them, in effect, black people are interchangeable — whether innocent or guilty — and so long as the percentage unjustly harassed or mistreated does not exceed the percentage who actually did something wrong, everything is chill.
In short, racism could be operating in every case, theoretically, and the numbers might look the same as if there were no racism operating. I’m not saying that’s the case, but one cannot determine based solely on aggregate data and statistical inference, what is happening on the ground in particular cases. Although the above hypotheticals may seem extreme, to the extent they demonstrate the weakness of relying on aggregate data to rationalize inequality, they torpedo the conservative methodology for debunking racism. But using more realistic possible hypotheticals can make the case just as well.
So, for instance, and putting aside the prospect of Neo-Nazi police gangs for a second, what if police are quicker to presume guilt for a particular suspect because of racial bias — perhaps subconscious and subtle? Or what if prosecutors are? Or what if juries are quicker to presume guilt and discount exculpatory evidence when the defendant is black? In those cases, which certainly seem within the realm of the possible, racism could be implicated in particular arrests, prosecutions or incarcerations, even if the larger data suggested nothing was wrong and even if the crimes in question actually werecommitted by black people.
So too, if half of murders are committed by blacks, and half of arrests for murder are of blacks, this doesn’t disprove racism in any given case. What if police are making these arrests on the basis of weaker evidence than they might require were the suspect white? What if prosecutors are quicker to bring the case to trial than they would be for a white person given the same fact pattern? What if jurors are quicker to presume guilt and convict? In other words, the fact that some black people are committing murder – and are indeed half of all murderers – does not necessarily mean racism is not operating in the given arrest and prosecution of a particular black person, or even all black people.
Theoretically it would be possible for every single black defendant arrested for murder to be innocent, and to have been the target of racist police, and still have the aggregate data look as it does. Although I am not suggesting this is the case, let’s just pretend that police were all so incredibly racist that whenever there was a murder committed by a black person they just went and grabbed random black people off the street, planted evidence on them and called in the D.A. In other words, lets pretend that in every single case involving a black defendant, the actual black person being charged and prosecuted was innocent and was only facing trial because of the bigotry of the cops. At the end of the year, so long as the aggregate number of blacks arrested, tried and incarcerated for murder didn’t exceed 50 percent – even if every single actual black person who was arrested, tried and convicted was innocent and had been racially targeted – people like Heather MacDonald would look at the data and say, no harm, no foul.
The test of racism then, is not whether stop and frisk rates, arrest rates, or incarceration rates mirror offending rates, but whether individual persons stopped, arrested or incarcerated are experiencing those things because of racial bias. And that determination requires quite a bit more nuance and willingness to listen to communities of color than the likes of Heather MacDonald can apparently countenance.
The point is, black people are experiencing policing differently, and in a more hostile way, than whites. Innocent black people who have committed no crime are being stopped, frisked, profiled, and detained, in ways that innocent white people are not. Unarmed black folks are more likely to be shot by police than unarmed white folks — about 3.5 times more likely — even when they are not behaving in such a way as to make the shooting understandable. The fact that other people who look like these innocent black people happen to be guilty of something does not justify what is happening to the innocent. In truth, it often can’t justify even that which is done to the guilty, since due process is still a thing that theoretically disallows extrajudicial execution. And anyone who fails to understand that, or in this case to call that racism, is not worthy of being taken seriously as a scholar or commentator on issues of such great social importance. They are merely hiding behind loaded footnotes to justify systemic injustice. But racism, no matter how highbrow, is still racism.
