Imagine I have a jar of marbles, 70% of which are yellow. If I pull out ten marbles at random, would you think it odd if seven of them were yellow? What if it was thirteen out of twenty, or seventy-two out of a hundred? I think most people would intuitively expect a number of items pulled randomly from a mixed population would approximately match the proportions of the original, and that expectation is mostly right. Randomness can generate some odd results, but over time those should converge on their probabilities.
But pundits are not most people. A major media talking about mass shootings over the last few weeks has been to note that the majority of perpetrators are white men. “White men have committed more mass shootings than any other group,” notes Newsweek. “Las Vegas is actually the norm in the United States,” reports Think Progress. “[W]hite men are disproportionately responsible for mass shootings.” “White men commit mass shootings more than any other group — why aren’t we talking about that?” asks Bustle, apparently unaware that a lot of people are talking about exactly that. George Ciccariello-Maher, the Drexel professor famous for satirically calling for white genocide, argued, “[W]hiteness is never seen as a cause, in and of itself, of these kinds of massacres, of other forms of violence, despite the fact that whiteness is a structure of privilege and it’s a structure of power, and a structure that, when it feels threatened, you know, lashes out.”
All three publications rely on data from Mother Jones — which in my opinion is the best available mass shootings data set — which shows that white men are indeed the majority of such shooters. The number these articles produce, 54%, is out of date: organizing the data this morning put the number at 56.4%.
But whites make up 61.3% of the population. “You can’t talk about risk without talking about population size,” notes Northeastern University criminologist James Alan Fox. With respect to population demographics, whites are actually underrepresented among mass shooters.
But it’s about what you’d expect to find: the racial distribution of mass shooters is pretty close to the racial distribution of Americans. “In a narrow sense, these stories are correct: The plurality of mass killers are white,” Daniel Engber wrote in Slate. “But the notion that white men of privilege are disproportionately represented among mass shooters—indeed, that they make up ‘nearly all’ of them—is a myth.” There’s simply no evidence that race is a valid predictor of mass shooters in America beyond demographic representation. Pulling more yellow marbles out of the aforementioned jar doesn’t say anything about the marbles per se, but it says something about the composition of the jar on the whole.
Gender, however, is a valid predictor of mass shooters: of the 95 shooters in the Mother Jones data set, 93 of them have been men — 98%. These shootings are not a race issue, but they are a gender issue. “This is not a phenomenon that’s unique to mass shootings—most killers of all types are men,” observes Engber. “[M]ale rage can be deadly at any time or any place, and at every level of analysis.” Whether there is a meaningful differentiation between why men are more violent in general and why men are more likely to commit mass shootings in particular is, as far as I can tell, an open question.
(I went through the song and dance of creating the binomial distributions for mass shooters by race. They are at the bottom of the post for anyone interested. I used the same approach as when I analyzed police shootings by state racial demographics.)
Why does this matter? Well, there are a few reasons. First of all, we should demand accuracy from those advancing policy goals, no matter what their ideological alignments happen to be. Secondly, we cannot hope to solve a problem that we do not properly understand.
But perhaps most importantly, I have no interest in conceding to racists and white supremacists that drawing major conclusions based on race is a valid lens through which to see the world. “This may be an attractive, tidy take for explaining mass shootings — one guaranteed to generate controversy and clicks,” wrote Jesse Singal in NYMag. “It’s also an extremely inaccurate, overly reductive, and misleading approach to a serious societal problem. People should really stop making such a glib and substance-free argument.” Singal continues, “When Ciccariello-Maher and others posit whiteness as an explanation for violence or other bad behavior on the part of whites, they are, in part, offering a rejoinder to the pernicious idea that blackness can explain such behavior on the part of blacks.”
Thomas Chatterton Williams notes the dangers of trying to use white supremacy as an all-encompassing explanatory tool: “Though it is not at all morally equivalent, it is nonetheless in sync with the toxic premises of white supremacism. Both sides eagerly reduce people to abstract color categories, all the while feeding off of and legitimizing each other, while those of us searching for gray areas and common ground get devoured twice. Both sides mystify racial identity, interpreting it as something fixed, determinative and almost supernatural.”
A blogger you may know made a similar point when discussing the issue of white terrorism. “When we turn wickedness into a competition, we adopt the frames that perpetuate that wickedness,” I argued. “[B]y keeping score, we adopt the simplistic logic of racists: (the idea) that reference to identity in abstract – whether Christian or Muslim, white, black, Hispanic or Asian – is a useful way to differentiate between good and evil, human and inhuman.”
“What all these ideas about black crime have in common is that they essentialize,” Singal adds. “[T]hey claim that acts of violence can be explained, in a meaningful way, by the race of the perpetrator: Black people are like this, and white people are not, which is why blacks commit more crimes. Ciccariello-Maher is making an argument of this same form, but swapping out one race for another: White people are entitled [or whatever] in a way members of other races are not, which can explain why they shoot up churches and concerts.”
Chatterton Williams went on to describe an exchange he had with the prominent white supremacist Richard Spencer.
This summer, I spent an hour on the phone with Richard Spencer. It was an exchange that left me feeling physically sickened. Toward the end of the interview, he said one thing that I still think about often. He referred to the all-encompassing sense of white power so many liberals now also attribute to whiteness as a profound opportunity. “This is the photographic negative of a white supremacist,” he told me gleefully. “This is why I’m actually very confident, because maybe those leftists will be the easiest ones to flip.”
However far-fetched that may sound, what identitarians like Mr. Spencer have grasped, and what ostensibly anti-racist thinkers like Mr. Coates have lost sight of, is the fact that so long as we fetishize race, we ensure that we will never be rid of the hierarchies it imposes. We will all be doomed to stalk our separate paths.
Either race is a useful heuristic by which we can predict criminals or it’s not. Singal closed out his piece by saying, “The way to atone for yesterday’s pernicious, racist narratives about crime is to abandon them in favor of smarter and more informed accounts, not to simply tweak them so they apply to majority rather than minority groups.” I couldn’t agree more.
Here are the binomial distribution graphs of mass shooters by race.
Asian perpetrators (n=94, k=7, p =.057):
Black perpetrators (n=94, k=16, p=.133):
Latino perpetrators (n=94, k=7, p=.178):
Native American perpetrators (n=84, k=3, p=.013):
And White perpetrators (n=94, k=53, p=.613):
All of these lands squarely on the heart of their distributions.