1. Tweetstorm on the Heather Mac Donald (@HMDatMI) op-ed being widely cited by critics of #BlackLivesMatter. At a @UChicagoLaw webinar, I referred dismissively to her yearslong campaign to debunk the “myth” of police racism. Some audience members wanted more engagement. Can do.
2. Mac Donald is a big-time conservative crime pundit. Jeff Sessions called her “the greatest thinker on criminal justice in America today.” She wrote a book called “The War on Cops.” Here’s the op-ed: wsj.com/articles/the-m…
3.A basic problem with Mac Donald’s work is that she fails to mention the *hundreds* of empirical studies of disparities in policing and other criminal justice stages. Almost all find anti-black disparities/evidence of bias. Mac Donald cites only a few exceptions.
4. Indeed, even when she cites Roland Fryer’s study “finding no evidence of racial discrimination in shootings” (her words), Mac Donald fails to mention the parts of Fryer’s study that DO find (w other datasets) that police disproportionately subject blacks to use of force.
5.And Fryer acknowledged that even if there is no racial disparity in whether a police interaction results in a shooting (as he found w a Houston dataset), that doesn’t rule out “important racial differences in whether or not these police-civilian interactions occur at all.”
6.That’s a key concession. Black Americans indisputably bear the disproportionate brunt of police violence in this country. The only debate is as to why. Mac Donald and her allies want to blame crime differences, #BlackLivesMatter emphasizes racism.
7.Importantly, the role of racial discrimination need not be limited to the way in which racial differences influence the split-second decision to pull a weapon. Disparities also could result from systematic overpolicing of black individuals and neighborhoods.
8. Racial disparities in stops, arrests, etc., produce a lot more interactions that can turn deadly. And many studies find that crim justice disparities are substantial *after* controlling for crime measures. See this @radleybalko piece compiling research: washingtonpost.com/graphics/2020/…
9.E.g., research on NYC’s stop-and-frisk policy by @JFagan46 found enormous disparities unexplainable by factors like neighborhood crime rates. Note the vast majority stopped under that policy were innocent. Law-abiding Black New Yorkers are far more likely to be stopped.
10.But besides cherrypicking a few study findings, Mac Donald herself tends to stick to a much more rudimentary argument. She likes to compare the percentage of police killings that are of black people to the percentage of homicide or robbery arrests that are of black people.
11.This comparison might possibly make sense if all police killings consisted of police interrupting a homicide or robbery in progress. But they obviously don’t. Many have arisen from routine traffic stops, misdemeanor arrests, etc.
12.So why does Mac Donald use homicide and robbery as her baseline? Of the major crime categories, these two have the highest black share of arrests by far. Look here: ojjdp.gov/ojstatbb/crime…. She picks the metric that best serves her narrative.
13.Also, using arrests for any crime as the baseline is absurdly circular. Arrests are POLICE CONDUCT, subject to the same biases Mac Donald is trying to debunk. When trying to disprove police bias you can’t just assume that arrests are a neutral measure of who commits crime!
14.BTW, I wrote a paper explaining some of the challenges of estimating disparities, including the fact that we can’t directly measure crime. I also show in more detail why comparisons like Mac Donald’s mislead. papers.ssrn.com/sol3/papers.cf…
15. One more super-dodgy move made by Mac Donald and others: Using the @washingtonpost database, she states that police “fatally shot nine unarmed blacks and 19 unarmed whites in 2019.” Then she proceeds to discuss those cases alone, as though they are all that matter.
16. But wait: there were 1000+ fatal police shootings in 2019 (plus killings by other means). What about all the others? By ignoring them and focusing only on the “unarmed” subset, Mac Donald implies if the victim is armed, a police killing requires no explanation.
16. This is breathtaking. ”Armed,” in this database, means having any weapon (including toy weapons) and does not mean you are brandishing it, much less that all the requirements for self-defense are met. Many cases involve police shooting a fleeing suspect in the back!
17. Don’t forget: Carrying a weapon is, subject to some limitations, a *constitutional right* in this country. It wouldn’t be if I ran the show, but I don’t. That’s the law, and I’d guess that most of the people sharing Mac Donald’s article support it.
18. Yet the Second Amendment ceases to exist when many talk about the legitimacy of police killings of black men. Possession of a weapon = justification for a government agent to kill you; no need for further discussion.
19.The last point I want to make is that none of this is really what matters (so, sorry for all the preceding tweets!). The obsessive focus of Mac Donald and others on whether police killings result from intentional discrimination misses the point of what “systemic racism” is.
20. There are many problems that the current movement is focused on: excessive force, excessive police presence generally, overcriminalization of problems like addiction and poverty, mass incarceration, militarization, a gang-like code of mutual protection among officers, etc.
21. All those problems are unquestionably racially disparate in their consequences. And they are deeply embedded in a society in which everything *else* also disfavors black communities
22. And those problems have led to a situation in which black people throughout the country widely fear the police rather than seeing them as a source of protection. That is a terrible and unsustainable situation, and does not help police to actually protect communities.
23. One way to think about systemic racism here: Society wouldn’t tolerate these problems if white people suffered them at the rates that black people do-if white people widely lived in fear of the police, or if 1 in 15 white men were behind bars. That’d be a national emergency.
24. None of that depends on whether Mac Donald is right or wrong in her claims about what the data show. But she’s wrong (or at least misleading and highly selective), and a lot of people do listen to her, so it’s worth saying so. And that’s it. Finally.
16.5 (skipped one, sorry!): The dismissal of "armed" cases is breathtaking. ”Armed,” in the WP database, means having any weapon (including toys!) and does not mean brandishing it, much less that requirements for self-defense are met. Many involve shooting a fleeing suspect!
