Jason Steorts argues that we should not focus exclusively on “statistics on police shootings and crime rates by race” in the current, post–George Floyd debate on race and police use of force. This is partly true, but we should not understate the relevance of those statistics to the particular debate we are having.
Where I agree with Jason is on two points. One, shootings by police are not the only form of deadly force used by the police, as the Floyd, Eric Garner, and Freddie Gray cases illustrate. While comprehensive national statistics are harder to come by, however, there is no evidence I’m aware of showing that deaths of Americans at the hands of police by means other than shootings are some order of magnitude larger than shootings, to the point where they would lead to different conclusions. Quite the contrary. For example, a NIH study of 812 deaths from 2009-2012 across 17 states found that gunshots were the cause of death for 760 out of 797 of those for whom there was an autopsy — 95 percent. If there is an elephant out there in the snow, you’d expect it to leave bigger tracks than that. Better collection of such data on a national, systematic basis is a legitimate function for the Justice Department in shining a light on how law is enforced in America. It is unlikely, however, to give us a different answer.
Two, the debate about racial bias in various aspects of policing and law enforcement does, of course, extend to questions other than the use of deadly force, and we should not give short shrift to the extent that people’s experience with, say, being hassled in stops on the street colors their view of other aspects of policing.
Giving those points their due, however, the data we have are still highly relevant and important. The rhetoric surrounding the current protests is steeped in explicit or implicit claims that the killing by police of unarmed black Americans is (1) epidemic in scale, (2) more common than the killing by police of unarmed whites, (3) a larger problem than the killing of cops, (4) a sufficiently large problem compared to the overall homicide rate to be worth radical revisions of policing, and (5) not getting better. If none of those claims are true — and all of them are contradicted by the available data — that matters.
I noted some of those statistics in a prior article. Rafael Mangual offers additional data on the long-term trends:
Police use of force declined sharply over the past 50 years. In 1971 the New York City Police Department reported 810 firearms discharges by officers, which wounded 220 people and killed 93. In 2016 those numbers were down to 72 shootings, 23 wounded and nine killed.
Consider why we are even talking about the slogan “Black Lives Matter.” It could be simply an aspirational slogan: that we should treat all black lives as mattering. But if that was the main point, there would be no objection to replacing it with “All Lives Matter,” because presumably nobody is arguing that non-black lives should not also matter. The reason why the particular slogan is used is, instead, to argue that black lives do not currently matter, at least not to the police. This is an assertion of fact, and it is an assertion about deadly force in particular. Given the ubiquity of this argument, it is worth asking whether it is actually true. We can’t answer that question by changing the subject to other areas of the criminal-justice system that do not deal in life-and-death decisions.
Coleman Hughes explains how looking more closely at both the stories and the data changed his mind:
[T]he basic premise of Black Lives Matter — that racist cops are killing unarmed black people — is false. There was a time when I believed it. . . . For every black person killed by the police, there is at least one white person (usually many) killed in a similar way. . . . you must do what all good social scientists do: control for confounding variables to isolate the effect that one variable has upon another (in this case, the effect of a suspect’s race on a cop’s decision to pull the trigger). At least four careful studies have done this — one by Harvard economist Roland Fryer, one by a group of public-health researchers, one by economist Sendhil Mullainathan, and one by David Johnson, et al. None of these studies has found a racial bias in deadly shootings.
As Hughes notes of the historical trendlines, “we know progress through normal democratic means is possible because we have already done it.” The relative rarity of such deaths argues in favor of more targeted solutions and fewer sweeping generalizations that treat as a single organism the nation’s 800,000 cops (some 100,000 of them black) across over 15,000 police departments. In fact, not all police departments are the same. As Robert VerBruggen has noted:
New Mexico has the highest rate of fatal police shootings in the country at nearly one per 100,000, New York the lowest at about one-tenth that. These differences are far too big and idiosyncratic to be driven purely by crime rates or the presence or absence of discriminated-against minority groups. (Alaska, Oklahoma, Arizona, and Colorado round out the top five behind New Mexico, while Connecticut, Massachusetts, Rhode Island, and New Jersey follow New York up from the bottom of the list.)
As conservatives, we understand that data is never the only answer to any question. When people feel that their physical safety is threatened, they cannot be argued out of that belief by “but actually, statistics say.” It’s both legitimate and necessary for our political system to respond to visceral concerns informed by the life experiences of ordinary citizens, whether or not they are supported by data. But it is also always relevant, in considering solutions, to know the truth.