fbpx
Politics Foreign Affairs Culture Fellows Program

Lies, Damn Lies, And Racist Statistics

I understand why Matt Yglesias wants to argue this, but I don’t think it’s going to get him where he wants to go: Trolling the universe this morning, Richard Cohen wrote a column arguing that it wasn’t racist of George Zimmerman to suspect Trayvon Martin of being a criminal because everyone knows that a disproportionate share of […]

I understand why Matt Yglesias wants to argue this, but I don’t think it’s going to get him where he wants to go:

Trolling the universe this morning, Richard Cohen wrote a column arguing that it wasn’t racist of George Zimmerman to suspect Trayvon Martin of being a criminal because everyone knows that a disproportionate share of violent crimes are in fact committed by young black men.

I think what Cohen really means to be arguing isn’t so much that neither he nor Zimmerman are racists, but that racism is the correct social and political posture. That white people have good reason to fear black men, and that therefore all black men should be put in a subordinate position. But as a logical argument, Cohen here is falling afoul of very poor statistical inference. For example, the vast majority of newspaper op-ed columnists in America are white men just like Richard Cohen. But that doesn’t mean it’s reasonable to see a white man walking down the street and assume he’s a newspaper columnist. If you look specifically at Jewish men, you’ll see the stereotype that we are disproportionately represented in the field of political commentary is absolutely accurate. And yet it is still not reasonable to assume that some randomly selected Jewish man is a professional political writer. Even right here on the mean streets of Washington, D.C.—a city that’s legendary for its high rate of punditry—a clear majority of Jewish men are not pundits. It’s just a very rare occupation.

By the same token, the fact that young black men are disproportionately likely to be involved in violent crime in no way licenses the inference that you should stop random black men on the street and begin treating them like criminals.

Let me translate this argument: if crime is extremely rare, then it is wrong to assume that anybody is a criminal, even if most of the (extremely rare) criminals you’ve ever heard of look somewhat like the person you happen to be wondering about.

That’s true enough. But what if crime isn’t extremely rare? Or, since “extremely rare” isn’t a technical term, let’s just say that crime is high enough to have political and economic consequences – that it is materially driving the behavior of large numbers of people who are afraid of being victimized.

A situation like that could be perfectly consistent with the proposition that the overwhelming majority of any readily-definable demographic group is not involved in crime.

Let’s assuming our population is divided into two groups, X and Y. Say that no more than 1% of group X is involved in violent crime, but that no more than 0.01% of group Y is involved in violent crime. Yglesias is correct that, in this case, it would be irrational to assume that any given member of group X is a violent criminal – it’s too unlikely. But, by the same token, if you were to say “we need to catch more violent criminals before they commit a crime” and resorted to something resembling the Bloomberg “stop and frisk” dragnet to try to catch those criminal fish, obviously part of the profile for somebody to be stopped would be “is the person a member of group X.” Undoubtedly the profile wouldn’t stop with that one factor – but excluding that term from the profile would be very foolish if you were trying to be efficient in allocating resources.

So, again, what Yglesias’s argument boils down to is “crime isn’t high enough to warrant subjecting any group to the disparate treatment implied by racial profiling,” not that if one were to profile it’s foolish to include race as a factor.

But that isn’t the only argument against that kind of profiling. There’s a political and moral argument, one that doesn’t depend on arguing that profiling is statistically illogical.

Cue Ta-Nehisi Coates:

[W]e should take a moment to appreciate the import of Cohen’s words. They hold that neither I, nor my twelve year old son, nor any of my nephews, nor any of my male family members deserve to be judged as individuals by the state. Instead we must be seen as members of a class more inclined to criminality. It does not matter that the vast, vast majority of black men commit no violent crime at all. Cohen argues that that majority should unduly bear the burden of police invasion, because of a minority who happens to live among us.

Richard Cohen concedes that this is a violation, but it is one he believes black people, for the good of their country, must learn to live with. Effectively he is arguing for a kind of racist public safety tax. The tax may, or may not, end with a frisking. More contact with the police, and people who want to be police, necessarily means more deadly tragedy. Thus Cohen is not simply calling for my son and I to bear the brunt of “violation,” he is calling for us to run a higher risk of death and serious injury at the hands of the state. Effectively he is calling for Sean Bell’s fianceé, Trayvon Martin’s parents, Amadou Diallo’s mother, Prince Jones’ daughter, the relatives of Kathryn Johnston to accept the deaths of their love ones as the price of doing business in America.

The unspoken premise here is chilling — the annihilation of the black individual.

“Annihilation” is a hyperbolic word, but I think his point is a valid one. It’s not just that the policy may have political consequences in terms of resentment and such. It’s that it implies a very different view of the citizenship and individuality of different groups within our society.

Coates continues:

Perhaps the standards should be different when it comes to public safety and violence. But New York City’s murder rate is as low as it has been in 50 years. How long should a racist public-safety tax last? Until black people no longer constitute a disproportionate share of our violent criminals, one assumes.

Coates goes on to talk about the legacy of slavery and racism as a primary cause of criminality in the black community, which is fair but I can already imagine the manifold objections – but, more to the point, I think it is a distraction from the truly central moral and political question. Abraham Lincoln once responded to the argument that slavery was justified because Africans were (according to the argument) intellectually inferior by saying that, by this logic, if a white man should encounter someone more intelligent than he, he should accept being enslaved by him. And of course no white man would.

If legal equality depends on factual equality – equality of talent, of temperament, of taste – then there is no hope for the idea of legal equality.

I live in New York City, and I am well aware of just how massively the city has benefitted from the extraordinary drop in crime, particularly violent crime. But we can’t simply ignore the costs incurred in that process. I wouldn’t put the humiliations of “stop and frisk” at the top of that list – I’d put the manifold horrible effects of mass incarceration at the top – but it’s certainly on the list. We – folks like me and Richard Cohen – are only able to ignore those costs, or discount them, because they are born disproportionately by another group of citizens.

Advertisement

Comments

The American Conservative Memberships
Become a Member today for a growing stake in the conservative movement.
Join here!
Join here