I.
Ezra Klein uses my analysis of race and justice as a starting point to offer a thoughtful and intelligent discussion of what exactly it means to control for something in a study.
I’m not really going to call it a critique of my piece, because it only applies to two of the six areas I looked at, and in those two areas Klein’s thoughts were already carefully integrated into my conclusion – I described both as showing “ambiguity over the level of racial bias, depending on…how strictly you define racial bias.” The Vox article repeated and expanded on that conclusion rather than contradicting it.
But it’s still an important issue and I’m glad it’s come up since I didn’t have time to deal with at enough length in the original post.
The argument is: any study worth its salt is going to control for things like income level. Therefore, a study that concludes “blacks and whites get arrested at about the same rates” may only mean “blacks and whites of the same income level get arrested at about the same rates”. If blacks on average have lower incomes, then in the real world blacks might still be arrested much more. Blacks being poor and therefore getting uniquely poor treatment from the criminal justice system (Klein says) sounds like exactly the sort of thing we would call “racial discrimination” or “racial bias” or “racism”, but it would be totally missed by the standard methodology of controlling for income.
The solution is terminological rigor, which I foolishly forgot to have. What I should have said at the beginning of my post was “I want to know whether there is any direct bias against black people caused by racist attitudes among police and other officials.” By this definition, all of my conclusions stand.
Klein wants to know whether there is any factor at all that causes disproportionate impact of the criminal justice system on any race. By this definition, my conclusions are only a tiny part of the picture, although at the end I recommend the book Malign Neglect which provides much of the rest.
As long as we keep these two meanings of “racial bias” or “racism” or whatever separate, there’s no problem. Once we start conflating them, we’re going to become very confused in one direction or another.
Ezra Klein and I don’t disagree about any point of statistics. What I think we do disagree about is the terminology.
If we find that much of the overrepresentation of blacks in the criminal justice system is because black people are often poor and poor people often get sucked into the system, should we describe this as “the problem isn’t racism in the criminal justice system, it’s poverty” or as “the problem is racism in the criminal justice system, as manifested through poverty”?
II.
Consider a town with 1000 black people and 1000 white people. 750 black people are poor, and 250 are rich. 750 white people are rich, and 250 are poor. Everyone commits crimes at the same rate – let’s say 10% per year. Rich people have lots of connections and can bribe their way out of trouble in a pinch, so only 50% of rich criminals get arrested. Poor people don’t have any strings they can pull, so 100% of poor criminals get arrested.
We can do the calculations and determine that the black arrest rate will be 8.75% and the white arrest rate 6.25%, a pretty significant difference. The people in the town can do the calculations as well. They correctly observe that in their town, everyone commits crimes at the same rate, so there must be some bias in their system. Using Klein’s definition, they determine that since the system in their town disproportionately affects blacks, their criminal justice system is racist.
The problem is, upon learning that your criminal justice system is racist, what solutions come to mind? The ones I think of include things like increasing the diversity of the officer pool, sending police to diversity training, ferreting out racist attitudes and comments among members of the force, urging officers to consume media that is more positive towards black people, et cetera.
But all of these are unrelated to the problem and will accomplish nothing. We specified the decision algorithm these officers use, and we know it has nothing to do with race and everything to do with class. The townspeople should be attacking the culture of bribery, nepotism, and corruption, not throwing away resources on curing racist attitudes that don’t affect police behavior in the slightest.
Note that this is true even if the poverty is caused by racism. Suppose the town college unfairly admits whites and turns down blacks, which is why the white people in this town are so much richer. I have no problem with saying “the town college is racist”. This suggests the appropriate solutions – educating and/or punishing the people at the college. I have a lot of problems with saying “the town police are racist” as a shortcut for “the town police take bribes, and due to racism somewhere else the people with the cash are all white” because this obfuscates the correct solution.
You can’t just cut links out of a causal chain and preserve meaning. “Blacks are arrested disproportionately often because of gravity” is true, insofar as without the formation of the Earth from the gravitational coalescence of a primordial gas cloud humans and therefore racism wouldn’t exist. But if the natural reaction to hearing the phrase is to solve the problem by attaching hundreds of helium balloons to black people, then say something less misleading.
III.
Klein goes on to say:
An example is research around the gender wage gap, which tries to control for so many things that it ends up controlling for the thing it’s trying to measure. As my colleague Matt Yglesias wrote, the commonly cited statistic that American women suffer from a 23 percent wage gap through which they make just 77 cents for every dollar a man earns is much too simplistic. On the other hand, the frequently heard conservative counterargument that we should subject this raw wage gap to a massive list of statistical controls until it nearly vanishes is an enormous oversimplification in the opposite direction. After all, for many purposes gender is itself a standard demographic control to add to studies — and when you control for gender the wage gap disappears entirely!The question to ask about the various statistical controls that can be applied to shrink the gender gap is what are they actually telling us,” he continued. “The answer, I think, is that it’s telling how the wage gap works.
Take hours worked, which is a standard control in some of the more sophisticated wage gap studies. Women tend to work fewer hours than men. If you control for hours worked, then some of the gender wage gap vanishes. As Yglesias wrote, it’s “silly to act like this is just some crazy coincidence. Women work shorter hours because as a society we hold women to a higher standard of housekeeping, and because they tend to be assigned the bulk of childcare responsibilities.”
Controlling for hours worked, in other words, is at least partly controlling for how gender works in our society. It’s controlling for the thing that you’re trying to isolate.
Once again, when someone says “women make seventy seven cents for each dollar a man earns”, the response is almost always “That’s outrageous!” and demands that companies stop being so sexist. I don’t even have to speculate here. Google “gender wage gap”, and just on the first page of results you find statements like:
“While some CEOs have been vocal in their commitment to paying workers fairly, American women can’t wait for trickle-down change. The American Association of University Women urges companies to conduct salary audits to proactively monitor and address gender-based pay differences.”
“Our project on sex and race discrimination in the workplace shows that outright discrimination in pay, hiring, or promotions continues to be a significant feature of working life…the Institute for Women’s Policy Research examined organizational remedies such as sexual harassment training, the introduction of new grievance procedures, supervisory training or revised performance management, and reward schemes.”
“Today marks Equal Pay Day, the date that symbolizes how far into the new year the average American woman would have to work to earn what the average American man did in the previous year. With a new executive order issued today, President Obama and Democrats are hoping to peg the gender wage gap as a major issue ahead of the 2014 elections. This week, Senate Democrats also plan to again bring forward the proposed “Paycheck Fairness Act,” a bill that aims to eliminate the pay gap between female and male employees. Both men and women see a need for moves such as this – 72% of women and 61% of men said “this country needs to continue making changes to give men and women equality in the workplace”
Given that the supposed gender pay gap is being used at this very moment to argue for salary audits, sexual harassment training, grievance procedures, and paycheck fairness acts, isn’t it really important to know that a lot of it is due to upstream factors like how men and women are socialized as children to have different values, which wouldn’t be affected by these things at all?
(Given that the entire issue is probably being used to load the term “feminism” with positive affect, isn’t it important to know that it’s mostly unrelated to what we expect feminists to do with their extra trust and power?)
It might be worthwhile to come at this from an ideologically opposite angle. Suppose I state “Professors who identify as feminist give twice as many As to female students as they do to male students.”
(This is true, by the way.)
It sounds like a big problem. So you dig through mountains of data, and you figure out that most feminist professors tend to be in subjects like the humanities, where twice as many students are female as male, and so naturally twice as many of the As go to women as men. If I just give you my best trollface and say “Yes, that’s certainly the mechanism by which the extra female As occur”, you have every reason to believe I’m deliberately causing trouble. Especially if colleges have already vowed to stop hiring feminist professors in response to the subsequent outrage. Especially especially if you know I am a cultural conservative activist whose goal has always been to make colleges stop hiring feminist professors, by hook or by crook.
If twice as many women as men take English literature classes, that’s compatible with something about gender socialization unfairly making men feel less able to study or less enthusiastic about studying literature. That could be considered biased or discriminatory, I guess. But phrasing it as “feminist professors give twice as many As to women” is calculated to produce maximal damage. It’s the sort of thing you would only do if you wanted to throw a match on a gunpowder keg for s**ts and giggles.
IV.
So I guess I’ve moved on from “poor choice of terminology” to “active misrepresentation”. Let’s stick with that.
This issue makes for the ultimate motte-and-bailey doctrine.
You go around saying “Gender gap! Women making less than men! Discrimination! Sexism!” Everyone puts on their Gricean implicature caps and concludes that they mean what these words mean in everyday speech. The appropriate remedies are trotted out – companies need to raise their female employees’ pay, companies need to hire more discrimination officers, feminists need to talk more about all the ways men talk over women in the workplace and mansplain to them, etc. This is the bailey.
Then someone says “Wait, according to our study, a lot of this is just that women prefer working shorter hours to have time with their families” – and so they retreat to their motte: “Yeah, that’s the mechanism for the gender gap. You mad, bro?”
But the thing about mottes is that nobody actually cares about them when there’s this awesome bailey they can fight over instead. By turning differential socialization into the motte for sexual harassment or something, we’re doing a disservice not only to sexual harassment, but to the principled study of differential socialization.
Anyway, the situation is actually even worse than this. If you hear “The problem with the criminal justice system is disproportionate impact on the poor,” then you’ll probably start coming up with ideas for how to deal with that, and other people will probably start listening. If you hear “the problem with the criminal justice system is racism,” then you will start sharpening your knives.
Racism is a uniquely divisive issue. Minorities hear it and think of Klansmen trying to kill them. White people hear it and think of witch-hunters trying to get them fired. A single death in a random Midwestern town has turned half the country into experts on ballistics because it involved race. Bring up race, and people will change their opinion in the opposite direction suggested by the evidence just to spite you for having a different opinion about it than they do.
Once you’ve said words like “racism” or “racial bias”, this dynamic is already in play and you have lost control of the conversation from then on. If you mention the word and then suggest that we should do something about the police bribery or whatever, then ten percent are going to yell “HOW DARE YOU IMPUGN OUR OFFICERS’ HONOR, YOU POLITICALLY CORRECT FASCIST”, another ten percent are going to yell “HOW DARE YOU DERAIL THE CONVERSATION ABOUT RACE, YOU WHITE SUPREMACIST ASSHOLE”, and the other eighty percent are going to be yelling so loud at each other they can’t even hear you. By the time all the fires have been put out and all the rubble cleared, it’s a pretty good bet that nobody is in the mood to hear about policy ideas for reducing the impact of police on lower-income individuals anymore.
Klein ends his piece by interviewing a professor who states that “Liberals sometimes overstate the extent of overt racism as a direct explanation of justice system disparities.” He acts like this is some sort of inexplicable quirk of the liberal mind. I wonder whether it might have more to do with liberals reading things like the recent Vox article, “America’s Criminal Justice System Is Racist”, which declares the thesis “There is no reason to be subtle on this point: the American criminal justice system is racist”, then goes on to repeat the phrase “America’s criminal justice system is racist” five times in the next five paragraphs. It never mentions that possibility that any of this racism is anything but overt.
If, like Robin Hanson, you believe in the metaphor of tugging policy ropes sideways, then I can’t think of any worse way to ensure that everyone will be tugging against you in every direction than trying to focus the discussion about race.
That’s why I limited my review to direct bias within the justice system itself, and why I think other ways of framing the issue are less productive.
(Comment screening is on again, I guess. Comments that will start flame wars or derail the conversation will vanish into the aether. Unrelated: the book review yesterday got popular and this blog might go down every so often because of too much traffic. It’ll be up again shortly.)
On an unrelated note: I’m glad to see your writing is getting more exposure. You’re probably one of the best essayists on the internet right now.
I think there’s an important part of Ezra Klein’s piece that you’re not addressing.
Specifically this:
In other words, it seems like you and Ezra Klein are interpreting the same piece of data differently.
The data says: Controlling for location of drug use and type of drug reduces the racial arrest gap.
You interpret this as: The police aren’t as racist as we thought – it’s just that black people happen to live in more heavily policed areas and they happen to be into more heavily policed drugs.
Ezra Klein interprets this as: The police are racist, and therefore they more heavily police black neighborhoods and more rigorously enforce laws against drugs that black people use.
Same for “controlling for type of vehicle stop reduces or eliminates the racial gap in car searches”. It’s totally possible that racist police officers more often stop black people for non-speeding because, say, they wouldn’t normally stop someone for an obscured license plate unless it was an excuse for racial profiling.
In both these cases it’s very possible that the racism is indeed situated in the law enforcement system. And the proper remedy would in fact be to create some less-racist police officers.
I don’t think it’s possible to determine whether your interpretation or Ezra Klein’s is correct by looking at the data we have. But his is perfectly plausible, and I don’t think it’s fair to totally ignore this half of his argument.
What weight should what we know objectively of the history of law enforcement (ie pretty awful levels of racism) place on a reading of the data? My opinion is that the weight should be very high. Racial divisions are one of the great drivers of American politics. From a bayesian perspective, it seems much harder to be overconfident in any given “racism drives this” explanation than one might think.
Thank you for writing this. This is so underemphasized in the rationalist community: that two different framings, even when they’re both factually correct, lead people to draw different conclusions, and this fact about human inference needs to be taken into account when communicating with one another.
Great post. I think this may be the most Scott paragraph ever:
I thought the wage gap was about rates of wages/salaries/pay, not about hours worked. Of course if John works fifty hours a week for ten dollars an hour, and Sally works thirty for ten dollars an hour, then Sally will come out with less money for her work week than John. But that would be true if we replaced Sally with Tim for the same hours.
What I’ve heard about the wage gap is that if John applies for a job, his hourly rate is likely to be (say) twelve dollars an hour. Sally, with the same qualifcations and experience, is likely to be offered ten dollars an hour, because historically women have been offered lower rates of pay. Even with attempts to equalise this bias, men in general still are ahead because they get promotions or fancier job titles with increased rates attached, while women are more likely to be (as I’ve seen in my job) acting posts with the increased responsibility but the same or stagnant wages.
And it’s complicated by the fact that traditionally ‘women’s work’ has been lower-paid than men’s work. Heavy industry was the high-earning employment for the blue-collar worker, and women were debarred by custom and opportunity from that. It’s not much better for white-collar work; professions traditionally majority male, when women become employed in significant numbers, undergo wage decline and decline in perception of desirability and status. Our now-former Minister for Education made a misstep when speaking to a conference of primary school teachers about encouraging pupils to take higher level mathematics in exams; requiring teachers to have higher level maths themselves would attract more men into teaching because teaching was a feminised profession (which sounded to his audience as if he was saying that women were too stupid to teach, or incapable of teaching, their pupils at a higher level):
Our out-and-admitted atheist and progressive and the divil an’ all about taking the Irish education system out of the grip of the Church and being all modern and STEM-oriented and non-sexist, non-racist, non-homophobic, etc. etc. Minister still managed to insult a majority-female teaching audience despite all his (champagne) Socialist credentials!
And on top of this, it may well be connected to hours worked – John regularly works fifty hours a week so he’s seen as a better employee and he gets promoted. Sally is seen as less committed to the company. But John may be facilitated in working those extra hours precisely because he’s married to Sally, who takes up the slack in “bring the kids to the doctor/take time off to be there when the plumber needs to call/pick up my prescription medicines/keep the house running/visit my sick mother in the hospital”. Women have gone for part-time/flexble hours work precisely because they’re juggling two lots of responsibilities.
It’s not just childcare, it’s the whole range of the domestic sphere that women take on, which includes looking after their husbands’/partners’ health – I’ve regularly seen in all my working experience women making phone calls setting up doctor’s appointments, arranging hospital visits, ringing around looking for help for obscure conditions, etc. on behalf of men as well as other family members.
If women put as much time into work, and let society take up the slack of the unpaid social support they do, it would have a huge impact that is not measured but would cost €€€€€€ more in educational, charitable, medical and other kinds of support just to keep life outside of work ticking over.
Pollack says:
I don’t know which society he lives in, but American society is obsessively interested in addressing those conditions! Our domestic political conversation often seems to be about little else; both major parties are deeply invested in the problem, and have tried their own solutions (incorrectly or insufficiently, depending on whom you ask).
Anytime someone frames an obviously controversial issue as “people just don’t care enough”, I’m suspicious. It feels like a back-door ad hominem: “My opponents may act like they care, but they really don’t, because they’re terrible people.” Talk about prioritizing heat over light…
I’ve seen that argument made even more strongly than you suggested: “My opponent’s suggestion of doing X to alleviate problem Y proves that he actually wants problem Y to continue.” This, of course, is a version of ESR’s kafkatrap; I have proposed calling it the Model N after Miss Nevada 2014, who was attacked with this pseudo-argument.
Good response! As a practical measure, my suggestion would be to use these terms around here, in contexts where we need a shorthand:
“gender bias” and “racial bias” for individuals’ irrational beliefs and aliefs about gender/race;
“gender stereotyping” and “racial stereotyping” (or ‘profiling’, ‘generalization’, ‘inference’, etc.) for inferring non-gender/race features of an individual based on their gender/race (which may be accurate or inaccurate, helpful or harmful, etc.); and
“gender disprivilege” and “racial disprivilege” for the general fact (or claim) that people of a certain gender/race are worse off (for whatever reason) than people of another gender/race. (Or use ‘disadvantage’ or something if you don’t like talking about ‘privilege’.)
If you want to build in causal claims about why groups are disadvantaged, I’d suggest speaking in full phrases, or using language like ‘gender-bias-caused disprivilege’ and ‘racial-stereotyping-caused disprivilege’. A system like this lets us sidestep the ambiguity of terms like ‘racism’, ‘racial discrimination’, ‘oppression’…
If we get used to drawing distinctions like these, it may make it easier for us to communicate with outside groups as well — not because we can yell at them for using language differently, but because we can give clearer options when we ask ‘When you say X do you have in mind Y?’ and can pick up on indicators of meaning.
Another way this comes up in a bad way is if the most inflammatory metrics lag actual discrimination by decades. Like, even if there were no sex differences and everybody had turned perfectly non-sexist 10 years ago, even somehow to the extent that they socialize their children regarding gender, there would still be way more male CEOs. Climbing that ladder takes a career. Even tech start-up CEOs would still be disproportionately male because the girls fully socialized under gender equality would just be turning into teenagers. Society isn’t like that of course but it is still true to a lesser extent that wage gaps and whatever else you want to use will reflect the state of society over the past several decades and so solutions based on them without consideration for time will severely overshoot, if there has indeed been “progress” during that time.
Also from the Vox article, “They show that racism exists even in our control society — the one with equality of income, and education, and neighborhood, and car choices. The one where we’ve wiped out most every difference but pigment. The one where we’ve left ourselves no excuses for our prejudice. It is remarkable how much discrimination can survive.”
Alternate interpretation that somebody has to say explicitly because it obviously doesn’t go without saying: whatever is left after successfully controlling for all the effects of racism which also cannot be shown to be the result of overt discrimination might be actual racial difference.
Ezra Klein, now. That’s going to produce some volume.
I feel you’re approaching the point in most blogs’ growth cycle at which the comments become a net liability, as a larger and larger fraction of posts are by people making their first comment. Was there a plan for dealing with it?
I haven’t read the studies, so I don’t know how they corrected for the various factors, but , depending on the technique, correcting for factors might take out some of the race variable. For example, if you corrected for neighbourhoods by subtracting the average amount of stops per person in the neighbourhood, then you wouldn’t just be removing the effects *caused* by the neighbourhood, but also allocating some of the effects due to race to the neighbourhood factor. But perhaps they are using more sophisticated techniques
There is another thing that can happen in your statistically symmetrical town though. The police see the conviction rates for each race and know they have to get a decent conviction rate to be seen as good at policing, so they want to stop, search and arrest people that are more likely to be convicted. But it’s easier to see straight off whether someone is black or white than whether they are rich or poor, and if this is especially than when you’re looking at someone of another race. If the police are mostly white, they will arrest more rich black people than rich white people. They may go as far as assuming that a rich black person entering their own house is probably a burglar, or that a black person in a flashy car is more likely to be a drug dealer than a lecturer…
“If the police are mostly white, they will [in their mathematically optimized pursuit of enhanced conviction rates] arrest more rich black people than rich white people.”
Is this not equally true if the police are mostly black, or green-skinned space alien mercenaries we hired to avoid the whole issue? The police increase their conviction rate if they focus on black people regardless of the race of the police themselves. This is something that we should try to control for, something that is likely very hard to control for, and something that is not helped by bringing the racial demographics of the police into the discussion when you are posulating a mechanism where that doesn’t matter.
OTOH, if it’s just random data-mining in hope of insight we are talking about, then maybe it would be enlightening to try measuring a difference in the rate at which white and black cops stop/search/arrest rich white vs. black suspects. But it would probably be best to do that without postulating a mechanism in advance.
The mirror-image problem is also worth thinking about. I.e., you don’t want to fix problems far upstream of their source. Troll example: ending pork-barrel politics by ending democracy, which most people who aren’t neoreactionaries think is silly. Real example: dealing with social costs of drug use via prohibition.
In my opinion, this discussion is operating on the wrong level. What you have to first solve is an (analytic) philosophical problem of defining what you are after.
For example, there is a bit in Pearl’s book where he (formally) defines discrimination by gender as a direct effect of gender on hiring (importantly, indirect effects of gender are intuitively ok). “Mediation analysis” studies direct and indirect effects.
If you agree with that definition (this is where the bulk of the discussion should be, do you agree with the formalization of discrimination/racism/etc) then it becomes a math problem, and you can ask people who know about that kind of math to tell you what you should do to your data to get that target.
I think starting with some data and thinking of ad hoc ways of massaging it to possibly get what we want is putting the cart before the horse.
Causal inference (which is pretty much what we want, right?) is super hard though, especially in the social sciences where you can’t easily do controlled experiments. In some cases it may be impossible to do meaningfully better than controlling for factors X, Y, and Z, and then using what that tells you to gesture in the general direction of causation.
Also, a criminal justice system which is biased against poor people will amplify poverty.
I’ve only started reading The New Jim Crow, but one of the points is that the criminal justice system in the US does huge reputational damage to convicts– they face legal and social costs because they’ve been permanently labeled as felons.
A general point which I haven’t seen addressed elsewhere: what are the consequences of a legal system which the majority of the people can’t afford to use?
As an aside, I am fairly sure that Scott would prefer comment and criticism to slavish praise, but for someone who avoids race and gender, he handles these topics with the care and precision they deserve. I think our conversation is far better for these last couple postings.
I largely agree with your analysis here, but I’ll pick on one minor point in the way you frame the divisiveness of race. You say:
Minorities hear it and think of Klansmen trying to kill them. White people hear it and think of witch-hunters trying to get them fired.
This frames racial issues, naturally enough, as pitting white people against non-white people. But I don’t think that’s actually how racial issues function in our society. It’s not minorities who associate such extreme negative affect to any mention of racism; it’s progressives. And it’s not white people who worry about witch-hunts; it’s conservatives. As you have observed (http://slatestarcodex.com/2014/09/30/i-can-tolerate-anything-except-the-outgroup/), “white” is often a code word for conservatives in these discussions.
I think this is an especially important point for conservatives to make, because otherwise we’re both misidentifying the problems (from our perspective) caused by political correctness and being unfair to some of those affected by them. For example, I want to make very clear that I do not feel at all discriminated against or attacked in my field (academia) as a straight white male. People often use that language, but I know that’s not really what they mean; there are plenty of straight white males — the majority, actually — who can feel perfectly comfortable here, because they fit in with the prevailing progressive orthodoxy.
But I do feel attacked as a conservative. I know that I have to keep my conservative opinions to myself or risk, at best, alienating my colleagues and, at worst, losing my job. And while being a straight white male doesn’t help me here, I’m confident that conservative minorities have it worse off than progressive straight white males. Witch-hunts go after women and minorities too when they express the wrong opinions.
I would have assumed that one of the elements of this where racism applies is that if poor people are treated badly and poor people are often black, the general public will care less about the poor treatment.
If police brutality, corruption, high incarceration rates, high repeat offending levels with little ex offender support or rehabilitation is being done to some other group of people (black people), everyone else will not have to worry too much about it happening to them.
So one of the main functions of a democracy – a just criminal justice system – can dwindle away without anybody noticing too much if most of the people going through that system are already an out group.
But any police force will become corrupt if there isn’t a great deal of political will pushing in the other direction. It will become corrupt even when staffed by kind, compassionate decent people if they are expected to work under poor quality policies. A lot of people will have had the experience that they have worked in places where bad decisions started to be made about vulnerable people because of poor policy and management, and they as an individual felt powerless to change that. It is just the same for the police.
As for the gender gap in pay… Every country making high levels of progress in reducing this does it through paid maternity leave, high quality childcare subsidised or free for those on low incomes, promotion opportunities for part time workers, respect for people returning to work after career breaks spent caring for young children or other vulnerable people, extra support for carers of people with disabilities, extra social support for single mothers, creating jobs with responsibility where possible that are within school hours, financial support for students with children, classing maternity and pregnancy as protected characteristics under equality law and so on.
That’s because the point is never to fix the problem, the point is to paint a dialectic of oppressor and oppressed to use the oppressed as political allies.
From Klein’s perspective, it seems worthwhile to separate out those root causes in order to communicate with people. Here’s my rough observation:
1) If a progressive says: everyone should understand the economic structure in this country is racist, a lot of undecided people will think: “Hey, I try my best to treat everyone equally when I hire. If anything, I probably would prefer to hire a minority applicant over a similarly qualified majority applicant because I feel bad about how disproportionately white my team is, etc.”
2) If a progressive says: “Hey, the system is unfair because minorities have a higher hill to climb because they start from a disproportionately worse position, disproportionately have more cultural adjustment required to interact in majority culture, are spending more of their grit dealing with microaggression, etc.”, I think they’ll have an easier time preaching beyond the choir, even if their ultimate solution is something at the hiring level to try to make up for things.
Wait, wait, you mean to tell me that correlation is not causation?
I’m impressed, actually, that you got through this post without once talking about causation and correlation. Confusing the two is basically what’s going on here, but shouting FALLACY! at people usually doesn’t do very much good.
Although I notice that the light-heat metaphor might just be another way of talking about correlation-causation. When two parts of the machine move in tandem, and we want one to change its behavior, we sometimes try to achieve this by pushing on the other. If they’re actually causally related in the right way, the machine’s behavior will alter in the way desired; if not, it just creates waste heat. Heat is what happens when you try to produce an effect through a mechanism not actually causally related to that effect.
(Also, first-time commenter; hello SSC world.)
I guess John Stuart Mill was overly optimistic when his main argument against censorship was that in an open debate, on an open marketplace of ideas, the best arguments and thus truth tends to win. This blog, and a lot of stuff I’ve read on lesswrong.com is excellent at documenting how it is not so.
I know some old-fashioned Catholic conservatives who think censorship is a good idea, but based on NOT whether ideas are true, NOT whether ideas are dangerous or morally repugnant, but simply based on how seductive they are. The idea that by adding pain to ideas that feel good, all ideas roughly feel neutral, and in that case truth may win.
I don’t really know how that could work in practice. It sounds fairly incredible to tell someone “you are going to pay for saying this, not because it is wrong, but because saying this feels too good, and I want you to feel roughly the same way about all ideas so that you can decide rationally” but there is a certain kind of sense in it.
The problem with censorship is not whether it’s ever justified — it’s easy to come up with thought experiments justifying it in certain circumstances, perhaps even very broad ones — but the same old “quis custodiet” problem that democracy itself was designed to address. The primary reason to have a free and open marketplace of ideas is to optimize for fairness and stability, not happiness or perfect decision-making.
The liberal-democratic thesis is that greater happiness and better decision-making tend to manifest in the long run from ideological freedom, as compared to even well-intentioned ideological policing. IOW, a system that is vulnerable to being taken over by a small, powerful cabal of (for example) evil Catholic theologians eventually will be.
I’m not a statistician, but: Isn’t this sort of the point? When we go looking for the cause behind an effect, aren’t we searching for those things which, when we control for them, make the effect go away?
If we control for everything under the sun and the effect is still there, we don’t know any more than “the cause is not within the list of things we controlled for.” We’ve ruled possibilities out rather than in. Presumably we would prefer a finite list of things-ruled-in rather than an infinite list of things-not-ruled-out.
This response functions fairly straightforwardly with a variable like poverty, but there are definitely some much more ambiguous controls described in the Klein piece. For instance, regarding the question of which kinds of drug use see the most intensive policing: that is presumably something that is in large part within the control of law enforcement officers, whose decisions could be influenced by their awareness of the racial differences in the use of various drugs. Hence a control that separated out arrests for different kinds of drugs would also risk obscuring actual racism in law enforcement.
Presumably the problem arises when certain controls de facto remove the possibility of meaningful comparisons. E.g. If I control for, say the level of deprivation in the neighbourhood where arrests happen, and either neighbourhoods are generally mixed or there are plenty of examples of both black and white (for simplicity’s sake, no other races exist in this example) neighbourhoods with similar levels of poverty, then that is valid – it is a perfectly good way to test whether the police are treating black and white people differently. This however stops working when my controls become such a good proxy for race in themselves that there are hardly any white neighbourhoods in my study that haven’t already been ruled out of comparison with the black neighbourhoods by the controls.
Does that make sense?
I agree with your overall point about causal vs non-causal factors for race/gender/etc gaps, but I do think “Why aren’t there more male students in the humanities?” is a legitimate question to ask. If, say, men are being turned off from the humanities because of rigid gender stereotypes and gender policing, that’s a thing worth knowing and combating. Not The Most Important Problem We Face, but not a silly dismissible problem either.
More broadly, I think that there is a big problem in some circles with overestimating overt racism and underestimating the tremendous effect of structural factors have in perpetuating inequities. It’s easy to see why – personal racism has a conceptually simple solution (get people to stop being racist) whereas structural/institutional racism is much more complex to figure out a solution for, and gets down to the core of how we think society should be organized. For example, it feels unfair to punish children for their parents socioeconomic statuses, but most measure to divorce the influence of parents’ socioeconomic status from their kids success would probably be seen as massively unfair by most.
It’s not just that in Klein’s critique, though. Some of it is indeed the stuff like the controlling for income.
But there’s also the thing with controlling for the kind of traffic stop, and finding that 1) there are more searches during stops for non-speeding violations than stops for speeding and 2) blacks are more likely than whites to be non-speeding violations. Maybe it’s just that blacks are more likely to have broken tail lights and failing to signal as they exit the high way (and police are more likely to search vehicles stopped that way for some facially neutral reason). But it’s also possible that police are more likely to use non-speeding offenses to make pretextual stops, and they’re more likely to make pretextual stops on blacks.
This is a good piece that makes important points that liberals (among others!) need to keep in mind in order to be efficacious in their declared goals, so I suppose that excuses a few easy cases of oversimplifications, such as:
“If we find that much of the overrepresentation of blacks in the criminal justice system is because black people are often poor and poor people often get sucked into the system, should we describe this as “the problem isn’t racism in the criminal justice system, it’s poverty” or as “the problem is racism in the criminal justice system, as manifested through poverty”?”
There may be a third causal factor that influences both higher rates of poverty and lower priors for positive interactions.
[from Klein, not Scott] ” As Yglesias wrote, it’s “silly to act like this is just some crazy coincidence. Women work shorter hours because as a society we hold women to a higher standard of housekeeping, and because they tend to be assigned the bulk of childcare responsibilities.””
Who’s this we, kemo sabe? It tends to be other women who set the standards of housekeeping (or perhaps advertising industry…), and women play a large part in assuming the opportunity for childcare responsibilities–let’s not completely strip them of agency now.
” isn’t it really important to know that a lot of it is due to upstream factors like how men and women are socialized as children to have different values, which wouldn’t be affected by these things at all? ”
Socialization is the primary factor? Aren’t you the biodeterminist guy? Politics, religion, personality, that’s genetic, but gender traits are about how children are treated?
I know you will say, of course not, clearly it is a mix, but I think it is a bit… noteworthy that you allow the impression that all gender differences are due to sexist parents to stand. I think I understand why you do so–a spoonful of sugar for the medicine you want the liberals to take to be more accurate with their studies. I’m not complaining, just think it is noteworthy.
I think you’ve put into words something that I’ve been feeling for a long time, but never really known how to explain.
This is a pretty common pattern in my life. I will hear about something terrible (say, “The Legal System Is Racist” or “Employers Pay Women Less”) and thing “something must be done!”. So I will start researching an issue, trying to understand how it works, why it happens, and what I might do to resolve it.
Along the process, I will start to find serious holes in the “marketing slogans” for given issues. On the benign level, it may be something as simple as an obscuring oversimplification. Often times, I find a more benign level that points to some kind of corruption or personal interest on the part of whatever issue’s leaders. And this never ends well.
To use this post’s example, I’ll be spending my time trying to understand the racist impacts of the justice system. I will see things like “cops shoot black people way more often than white people” and start researching. I will find some confounders that eventually draw me to either “there is not actually a problem” or “the problem is different from what people are saying (read: “it’s actually that cops shoot poor people way more often, and black people tend to be poor in America relative to baseline”). I will start trying to boost the signal on my findings (“Hey everyone. Turns out what you think is racism is really just classism. We need to focus on helping people of colour to become richer, through education, stabilization of life shocks, extension of credit, encouraging entrepreneurship, etc, and then this problem will be solved!”). And when this happens, I’ll inevitably run afoul of everyone else. It will turn out that the masses don’t care about this, because it sounds dangerously close to “Rabble rabble I am a racist blah blah blah”.
When all the dust settles, I’ll look back and see: “I saw an issue. I researched the issue and attempted to enact a viable solution. I faced strong pushback from the people who claim moral leadership of solving this issue. They must not want to solve this issue. This issue must not actually be that important, if it’s just people doing political jockeying. For that matter, the morally correct thing may be to oppose these people.”
In my process of trying to a) earnestly solve a problem; and b) oppose people trying to twist the problem to their own ends; I end up getting indirectly nerd sniped into taking very socially unacceptable positions. Sometimes it’s “Oh, maybe the Bible is wrong and I should stop taking it so seriously”, and the world becomes a little bit better. Sometimes it’s “Oh, maybe the anti-racists are wrong and I should stop taking them so seriously” and I become a little bit of a worse person.
In short: When I find people being imprecise, obsfucatory, or framing an issue for heat instead of light, in the long run it tends to actively polarize me against whatever the issue at hand is. I doubt I’m the only one. And I wish that widely-listened liberals* didn’t do this all of the time; posts like Ezra’s, with their vague equivocations and quirks of framing, tend to turn me against issues that I believe, rationally, are probably good ideas.
—-
*: I don’t doubt that conservatives also try to do this, but I generally don’t listen to them so if they do, I don’t see it
It strikes me that the ethnic tension (har har) over “racism” has rendered it useless in any speech community that hasn’t settled on the meaning, and that if we replaced it with “racial supremacy” and “racial bigotry” (or whatever) there would be a lot less confusion.
Another place where I see similar lack of careful language on display is when it comes to *normative* questions. Suppose you observe that a system is “racist” in the Ezra Klein sense. So what?
I assert that any system will be X-ist in the sense of Ezra Klein, provided I get to draw the boundaries of my groups. I.e., instead of slicing humans as { h \in Human : black(h) }, I might slice { h \in Human : broke(h) }. In Scott’s example, the broke group will be even more discriminated against than the black group.
Somehow the language about racism also hides an obvious normative question. Specifically, there are 2^{6 billion} subsets of humanity, and a similarly large number of ways of partitioning humanity. Ezra Klein believes that if we slice in one particular way and observe disparities, there is an injustice. Why is his choice of slicing valid?
I’ve asked this question a number of times before – the typical response is merely to cite statistical facts (which are true of many subsets) and then call me racist, e.g.: https://news.ycombinator.com/item?id=8368287#up_8372671
If anyone here has some useful philosophical insight that goes beyond tribalism, I’d be very interested to hear it.
Ezra Klein mentioned this blog? Sorry, but I don’t think SSC is cool anymore.
This is all well and fine. Terminological rigor is always useful. It does strike me, however, that police officers would be uniquely vulnerable to racial biases. Police work seems like one of the ultimate system 1 biased occupations, and thus quite susceptible to known racial implicit biases. To me, this means that even if we can only observe mild discrimination when the data is subjected to rigorous controls, the potential for abuse is constant and requires vigilance.
I also think that given this country’s long history of pernicious law enforcement (eg the ban on blacks testifying in court), the burden of humility rests a little more solidly on the shoulders of anyone trying to explain away racial prejudice.
Scott, is there a way you can turn comment screening on for some posts but not others? I was pretty excited about the book review comment thread
Also, whatever happened to the edit button?
My tribal bias isn’t going to scream out of this analysis.
I think two things explain like 90% of the tribal disparity on a whole host of social issues. Blue tribesmen care about fairness in outcomes and believe in the raw power of socialization. Red tribesmen care about fairness of process and don’t put nearly as much stock in the power of socialization.
So take for example women working fewer hours on average leading to pay disparity. To a blue tribesman that’s an unfair outcome, damming the mechanism creating that outcome, which they identify as social forces making women primary caregivers for children. To a red tribesman they see a fair process: get payed for the work you do. And they’ll agree women being primary caregivers for children makes them work fewer hours, but that’s just women being women, it’s not the result of social forces.
So my opinion would be that what’s really going on when a sociologist “controls” for various factors is that they’re injecting their personal morals and values into the research.