Framing For Light Instead Of Heat

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.)

This entry was posted in Uncategorized and tagged , . Bookmark the permalink.

427 Responses to Framing For Light Instead Of Heat

  1. Anonymous says:

    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.

  2. llamathatducks says:

    I think there’s an important part of Ezra Klein’s piece that you’re not addressing.

    Specifically this:

    In other words, we police black communities more heavily and we are more aggressive about enforcing drug laws against drugs that black people use more frequently. Controlling for those facts isn’t helping you isolate the role racial discrimination plays in drug enforcement. Those facts are the role that racial discrimination plays in drug enforcement.

    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.

    • Anonymous says:

      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.

      • Llamathatducks point is well-made. I personally have a strong bias against anything Ezra Klein writes, now that he’s producing tripe media at Vox. That is unwise of me, as Ezra Klein is no fool, despite Vox. However, that’s as far as one should go, in the direction of “because of racism”, prior to further consideration and analysis.

        Racial divisions and wealth are drivers in politics, but so is within-race ethnicity e.g. in Miami now and Boston, NYC during the 19th to mid-20th century.

    • youzicha says:

      And I guess we already discussed an isomorphic problem to death in the Great Intelligence Debate. The point of the Glymour paper is that it only makes sense to do a regression “controlling” for something if you already know the causal structure of the problem, and are controlling for a cause. Then the regression tells you the size of the effect. But if you are controlling on an effect, the numbers you get are junk, they do not let you infer the direction of the causality.

    • youzicha says:

      I remember talking to some guy giving tours in Harlem, who said it was well known that NYC police officers has some kind of quota of arrests they should make, and if they are not making enough they will drive a detour through some predominantly black area and arrest someone on flimsy evidence. Of course, just because it is believed to happen does not necessarily mean that it does happen. But this seems like a good example of an overtly racist practice which would get missed by controlling for neighborhood.

      • Anonymous says:

        http://en.m.wikipedia.org/wiki/Adrian_Schoolcraft

        There is fairly good evidence of this phenomenon occurring, and almost always in neighborhoods like Bed-Stuy rather than somewhere in, say, Midtown.

        The more I read and think about this issue, the less value I feel willing to attribute to any light that Scott may have shed. The micro data seems trvival next to the macro picture. Remember this is a country that went about locking up record proportions of young black man in the middle of a precipitous drop in the crime rate. Meanwhile, countries the world over experienced similar levels of decrease in crime without a similar increase in incarceration. There is strong evidence, much to a biodeterminst like Scott’s liking, that biological/environmental factors drove the change, implying that US levels of incarceration are in some large part exorbitant.

        In sum, Scott’s micro evidence paints a picture of mild discrimination against blacks, but the macro evidence strongly suggests that the criminal justice system is at least indifferent to the health of the black community in America.

        • haishan says:

          I think a better way to interpret Scott’s evidence is that people at the end of the justice system, like police, judges, and juries, exhibit mild discrimination against blacks, on top of the huge structural discrimination blacks face.

          Police look for arrests in Bed-Stuy rather than Midtown because there are very few likely costs to them for arresting a young black man living in Bed-Stuy. He probably won’t sue the department if he’s handled roughly or for false arrest, or if he does, he probably won’t be successful. He’s considerably more likely to be breaking the law by chance (drug possession, firearm possession, jaywalking) than a white dude in Midtown is, which turns a quota arrest into a real one. These facts are not the fault of our police officer; she’s just responding to her incentives. (As is the black guy with the gun, presumably.)

      • John Schilling says:

        “Detour to a predominantly black area and arrest someone”, will not show up if you control for neighborhood. “Detour to a predominantly black area and arrest someone black”, will. What are your priors on these two possibilities?

        • llamathatducks says:

          If the area is overwhelmingly black, your chance of randomly picking a white person to arrest is pretty small, so I don’t think it would be likely that we could distinguish between your two possibilities with the existing data.

      • RCF says:

        If cops have a quota, and are arresting people on flimsy charges, why in the world would the issue of racism be the most important issue? It seems to me that all this talk of racism is creating a lot of minddeath. Slavery was bad because some people were being forced to work in horrible conditions with no pay, not because there was a “racial disparity” in the rates at which people were being forced to work in horrible conditions with no pay. It would still have been evil even if black and white people had been forced to work in horrible conditions with no pay at the same rates.

        If cops have quotas and are arresting people on flimsy charges, then the question of how those arrests are distributed among different races should not be the primary issue; the primary issue should be the fact that cops have quotas and are making arrests on flimsy charges.

        • llamathatducks says:

          I agree with you in part. I think that the solution to racial disparities in poverty is not to distribute poverty more equally but to fight poverty. But it’s possible that terrible practices like arresting people on flimsy charges are enabled by the social vulnerability of certain classes of people, like black people and poor people. In that case, addressing that bigotry might help.

          Plus there’s the possibilty of things like unconscious bias, which could also be counteracted by bringing it to light.

          • Anonymous says:

            In conversations like this I usually find it useful to remind people that Martin Luther King’s final creation wasn’t the “Campaign for Colored People.” It was the multiracial “Poor People’s Campaign.”

            Few groups get strawmanned like racial minorities.

        • Liskantope says:

          Your point about the unjust acts themselves being the issue rather than the races of the people involved… this is a major idea, and should probably be the subject of its own blog posts with their own lengthy comment threads.

          I don’t think anybody here will argue with the fact that such unjust acts are still evil even without a racial component; the point of contention is your implication that they are no worse because of the racial component. I’m inclined to think that the racial component makes them far worse, but the argument for that would be fairly deep, and at the moment I’m not sure how best to assemble it.

          I think that a large part of my objection is that if people are marked for discrimination by something as obvious and unambiguous as the color of their skin, this will lead to further decreases in utility for those with the disadvantageous skin color. When the determining characteristic associated to certain inequalities is so overt, those inequalities seem to self-perpetuate.

          Perhaps (in the absence of not-well-developed points such as in the paragraph above) the following would be an interesting thought experiment. Compare the following two societies: one in which the founding members declare “Things will run more smoothly if a certain portion of the population is forced to work for the rest with no pay and horrible conditions, so let’s invent an algorithm that randomly selects this portion of the population”; or “Things will run more smoothly if a certain portion of the population is forced to work for the rest with no pay and horrible conditions, so let’s decide that this will be the subset consisting of individuals of such-and-such race”.

          • Omar says:

            Why restrict it to race? Assuming this isn’t a system in which everyone spends part of their time being a slave, but rather one where random individuals of the population are slaves for the rest of their lives, the slaves will end up forming a minority group on the same level as races/genders/sexual orientations/classes/etc., won’t they? If that’s the case, shouldn’t we expect the negative consequences of targeting particular races to occur anyway, just to another slice of the population? I realize it’s less visible than race, but it could be comparable to sexual orientation.

            Back the initial topic: if blacks use drugs as heavily as whites, and if the police is not racist, then a group is still disproportionately harmed by drug prohibition: heavy drug users.

        • 27chaos says:

          I agree it is not the primary issue, but note that if such problems are concentrated in one group they’re often worse than if evenly distributed throughout the population.

          For example, assuming for simplification purposes that employability is equal, if the least productive 20% of black people go unemployed due to unfair reasons, this is worse than if the least productive 5% of total people go unemployed due to unfair reasons, because the black people at the 10% mark are more likely to deserve a job more than the general person at the 5% mark.

          Employability isn’t equal, of course. But this form of analysis lets us arrive at good decisions after doing our best to control for employability.

          Similarly, if you’re arresting 100 suspicious looking people, it’s better that you arrest the most suspicious looking people there are than that you arrest 100/150 black people in a neighborhood. Arresting the highly suspicious people is more justifiable than arresting the moderately suspicious people.

          You can probably also look at risk dispersal. If risk is dispersed throughout a population, people will be able to rely on their luckier relatives for help. If risk is targeted at a population, this form of insurance goes away.

          (I wonder if this interacts with social capital/social trust.)

        • K says:

          You’re assuming that the “horrible conditions” of slavery would have been just as bad in a world where there was no racial disparity in who was enslaved.

          But we know that’s not the case, because there have been plenty of times and places like that in history – and while slavery always sucks, it’s nowhere near as awful in societies where it’s not racialized. Greek and Roman slaves were frequently paid salaries as motivation, allowed to purchase their own freedom, manumitted at much higher rates than American slaves, and educated to perform skilled professions like medicine. The Greeks and Romans could do those things because their cultural justification for slavery was not “these people are biologically inferior and must be treated like animals for everyone’s good.” In the absence of that justification, slaveholders have an incentive to treat slaves more like the Greeks and Romans did. In the presence of it, slaveholders have a strong incentive to treat slaves as if they were animals and never allow them any chance to prove that they weren’t.

          I suppose the statement that hypothetically, if slavery in 19th century America wasn’t racialized but was still exactly as awful as it was in our world, that would be no improvement on what happened in our world. But we can be pretty confident from looking at history that situation wouldn’t have happened. Racial disparity makes institutions like slavery worse, which is a pretty good reason to treat racial disparity as a bad thing in and of itself.

          • Anonymous says:

            History very clearly answers your question. However, it give exactly the opposite answer you give. Roman agricultural slavery was quite comparable to American agricultural slavery. Sugar plantation slavery was the worst in the history of the world. That was true in the old world, where it was not racial.

    • Scott wrote in his first essay that choice of drug-crime enforcement, and sentencing by judges, both appeared racially biased against blacks but everything else apparently isn’t. The “in other words” sentence is in fact exactly the point he got across. As to the claim that black communities are policed more heavily [than they should be]: what evidence? I’m sure many poor communities with prevalent crime of the sort that could be interrupted by a patrol wish they were better-policed. And isn’t human nature that cops would prefer to patrol where it’s *not* needed? I know I’d rather police the rich neighborhood.

      • llamathatducks says:

        Scott wrote in his first essay that choice of drug-crime enforcement, and sentencing by judges, both appeared racially biased against blacks but everything else apparently isn’t. The “in other words” sentence is in fact exactly the point he got across.

        Sort of but not really. Directly addressing the issue Klein addresses in the “in other words” quote, Scott writes in the original post:

        Likewise, whites are more likely to use low-penalty drugs like hallucinogens, and blacks are more likely to use high-penalty drugs like crack cocaine. Further, blacks are more likely to live in the cities, where there is a heavy police shadow, and whites in the suburbs or country, where there is a lower one.

        and then

        Adjusting for known confounds reduces their rate to twice that of whites.

        Klein’s point is that these two controls aren’t “known confounds”, they’re ways that racial discrimination can happen. It’s possible that [race of most users of drug X] and [penalty associated with drug X] (or [rate at which police officers bust people for drug X]) are actually causally linked. Scott sort of addresses this when he talks about crack cocaine sentencing, but that’s just about sentencing, not arrest rates.

        I don’t think the original Race and Justice post brings up type of driver stop at all.

        But mainly my point is that this post, in its summary of (and rebuttal to) Klein’s argument, leaves out a key part of that point. Here’s the summary:

        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.

        This makes it sound as though Klein says causality goes like this:
        – Societal racism → black people are disproportionately poor.
        – [independently] Police officers disproportionately target poor people (whether because of classism, higher crime, or a mix of both).
        – Therefore, police officers disproportionately target black people [despite not directly trying to]
        – Therefore, law enforcement is actually racist.

        Or like this:
        – For whatever reason, black people mostly do drug A and white people mostly do drug B.
        – [independently] Law enforcement targets users of drug A more than users of drug B.
        – Therefore, law enforcement ends up targeting black drug users more than white drug users.
        – Therefore, law enforcement is actually racist.

        Against that argument, Scott’s response makes sense.

        But the causality proposed by Ezra Klein is actually mostly like this:
        – For whatever reason, black people mostly do drug A and white people mostly do drug B.
        – [independently] Police officers have some degree of racial bias.
        – Therefore, police officers target black drug users more than white drug users.
        – Therefore, police officers end up targeting users of drug A more than users of drug B.

        The part about racism → wealth differences → disparate impact of law enforcement is mostly a side note introduced by Pollack in the final quote of Klein’s piece.

        Ezra Klein’s model of racism in law enforcement includes the kind of racism-in-law-enforcement that Scott agrees deserves to be called that, but this post makes it sound like it doesn’t. I think this is problematic even if you’re right that all communities are policed about the right amount (I won’t wade into that question now because I need to sleep).

    • Thursday says:

      The gold standard for crime is murder, because its easy to count corpses, and so murder isn’t subject to the same questions about whether police are ignoring certain crimes in certain places. If there is a dead body, attention must be paid.

      So, we should at least compare arrest rates for other crimes with arrest rates for murder. Not perfect, but it does provide some controls.

      • Nornagest says:

        I actually did that a while back. Clearance rates for murder cases in the US vary quite a bit between cities but tend to hover around 60%; for other violent crimes they’re generally similar but lower. Property crimes are cleared much less often, rarely more than 20% of the time; 10% isn’t unusual in some categories.

        • Anonymous says:

          That isn’t the exercise Thursday was asking for. Remember, the context is whether the police are racist. If your exercise doesn’t look at race, it is unlikely to be relevant.

          Also, the thread was about drug arrests, a victimless crime with no complaints to clear. Llama complains that Scott ignored Ezra’s claim that aggressive policing of drugs is racist, and that controlling for neighborhood is ridiculous. Whereas, Thursday is suggesting that heavy policing is due to the color-blind reason of a high murder rate. He proposes a test: is the number of drug arrests in a neighborhood due to the race of the inhabitants or the murder rate?

          • Thursday says:

            “Thursday is suggesting that heavy policing is due to the color-blind reason of a high murder rate.”

            Or simply that murder and drug crimes correlate.

            BTW even if a murder case isn’t solved, there’s still a body to say that a crime actually has taken place.

  3. Darcey Riley says:

    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.

  4. elstir says:

    Great post. I think this may be the most Scott paragraph ever:

    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.

    • I love this rhetorical penchant of his too. Take something we all know is retarded, and articulate *why* it’s retarded, and imply that this is how retarded the people we disagree with are. I find myself unable to disagree.

      • chaosmage says:

        The first sentence in that paragraph does all the intellectual work. The rest is an illustration funny enough to make that first sentence stick.

  5. Deiseach says:

    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):

    I also want to see Higher Level Mathematics in the Leaving Certificate becoming part of the minimum entry requirements for Initial Teacher Education and I’ll tell you why – to a highly feminised audience and profession – our research shows that young women who do the Junior Certificate and take Higher Level Mathematics comfortably in the Junior Certificate exam, drop Higher Level Mathematics when they do their Leaving Certificate because it is not a requirement. This is evidence-based research, and that’s why we want to see it happen.

    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.

    • Vaniver says:

      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.

      Right. This is because you expected words to mean what they normally mean, which is part of Scott’s point. The wage gap statistics of the “77 cents on the dollar” sort just compare the median (not even average!) salaries of full-time male and female workers. See wiki:

      In the United States, the gender pay gap is measured as the ratio of female to male median yearly earnings among full-time, year-round (FTYR) workers.

      When you turn it into hourly wage for the same sort of work, the gap is somewhere in the 5 cents range- but then there are still things unaccounted for. Height explains a few cents per dollar of wages, but is highly dependent on sex; would adding height to the list of confounding variables drop it even lower? If it did, would that be meaningful?

      • Multiheaded says:

        Height explains a few cents per dollar of wages, but is highly dependent on sex; would adding height to the list of confounding variables drop it even lower?

        Shorter people getting lower pay due to being seen as less socially dominant and women getting lower pay due to being seen as less socially dominant is, very broadly, part of the same systemic thing! It’s just easier to force qualitative inequality (like silently doing most of society’s day-to-day emotional work) on women as a class!

    • Anonymous says:

      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.

      I am shooting from the hip a little bit, but it is my understanding that the “74 cents on the dollar” claim was arrived at by more or less just dividing the average wage for men into the average wage for women. When looking attempting to look at “same field, same experience” the pay gap shrinks to 90-odd cents on the dollar.

      The usual explanation for this is either “women are acculturated in such a way as to make them bad at wage negotiations” or “overt sexism.”

      The factor that I would be interested in exploring is that equivalent CV’s are not necessarily equivalent. If, for the sake of argument, women on average work 80% of the hours as men then an employer, presented with two CVs that both show 10 years of experience in the field where one of the applicants is a woman and the other not, may “reasonably” conclude that she only has only 80% of the experience. This might lead to, going off of one of your examples, a woman being hired in an acting capacity because “her CV looks great but she might not have the depth of experience we are really looking for.”

      It seems that this sort of thing would contribute to the perception of bias.

      In general I would assume that little ol’ me can’t think of stunning new avenues of research with a half a second of thought, so I further assume that this factor has been accounted for in the literature (though judging by the quality of some of the research Scott reviews I am far from certain), so it may not budge the 90-odd cents on the dollar figure at all.

    • Princess Stargirl says:

      “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.”

      You give the wage gap people too much credit :(.

      The way the major “wage gap” statistics are computed is this. Find the average salary of all women working “full time”* in the USA. Find the average of all men working full time in the USA. Compare the numbers. This is actually where the number the Obama campaign and such used. (warning I think the democrats are a more competent party than the republicans overall).

      So basically the popular wage gap statistics do not even account for hours worked. This basically makes them propaganda imo. Though there is still a significant wage gap when controls for hours worked. The gap even remains if you control for the fact that salary is highly non-linear in weekly hours. For example a person working average 60 hours makes over double (as I recall) the salary of someone working 40. So I am not saying they fabricated the wage gap by not controlling for hours. Just that their failure to do so implies either dishonesty or extreme incompetence.

      I will also note many feminists do report reasonable wage gap figures based on studies with controls. I am not criticizing them.

      *They now use 40 hours a week as full time. Though the first popular wage gap numbers actually used a cut off of 35 hours. Which exaggerates the gap even more.

      • Anonymous says:

        Controlling for hours worked reduces the wage gap, but does not eliminate it.
        I recall that one study that controlled for as many factors as possible found a 9% wage gap, and that convinced me that controlling for more stuff wouldn’t produce a qualitative difference and that arguing about the exact size of the difference was pointless.

      • RCF says:

        “Though there is still a significant wage gap when controls for hours worked.”

        The results vary from study to study. Some even find that men make less than women.

        • Anonymous says:

          If you only control for the objectively measurable hours worked, most of the gap remains. To get to near parity, you have to also control for comparable jobs, but this is a much more subjective step.

          • Richard Gadsden says:

            As well as the subjectivity of measuring job comparability, you also have the problem that the direction of causality between “Job A pays better than Job B” and “Job A is mostly male; Job B is mostly female” is not obviously from the first to the second.

            As teaching has become a more female job, so it has fallen in pay relative to other jobs (ie classroom teachers are at a lower income percentile).

            Now, extending that result to a wider selection of jobs is going to be very hard; but if it proves to be true, then controlling for occupation doesn’t exclude sexism. It just means that the sexism is institutional rather than personal.

    • Multiheaded says:

      Yes to all of this! The real issue is conveniently avoided by conservatives and neoliberal “lean in” feminists both.

      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.

      YES YES YES! And one more thing – emotional labour; society really pervasively and insistently outsources it to women as a class, and then covers it up with romantic notions. A mother worried sick about her children, a wife listening to her husband went – that’s work, taxing, poorly and irregularly compensated work; it’s very important for the society that disregard it. This is where the real feminist revolution has hardly even begun yet!

      • Multiheaded says:

        s/went/vent

      • Andrew says:

        What men do for their spouses and children is also “disregarded” by society in the same way, though.

        However, at least the spouses won’t (always) disregard it.

        Also, if there’s a divorce, society will suddenly stop disregarding it, as well.

      • nydwracu says:

        A mother worried sick about her children, a wife listening to her husband vent – that’s work, taxing, poorly and irregularly compensated work; it’s very important for the society that disregard it. This is where the real feminist revolution has hardly even begun yet!

        What follows from the classification of all that as work? Is it to be compensated? By whom? The state? Or outsourced to people like therapists? What effect would that have on the social fabric? Wouldn’t that be atomizing?

        Perhaps a less atomizing solution could be found. Could the people who do the ‘work’ be paid by the people they ‘work’ for? Not on a salary, of course; that would undermine it. The payment plan would have to be something along the lines of support. And they couldn’t do it for multiple people, for the same reason — it would just be ineffective that way.

        Oh wait.

      • William O. B'Livion says:

        What is your evidence that society forces this on women rather than women preferring these sorts of roles?

        http://www.forbes.com/sites/meghancasserly/2012/09/12/is-opting-out-the-new-american-dream-for-working-women/

        If you want a world where people can do what they want, you have to let them want what they want, not what you want them to want.

      • Tracy W says:

        I don’t follow your logic here. Are women not part of society?

        As for worrying about your children, or listening to your husband being poorly compensated, surely that depends on how much you like your children and your husband?

        • llamathatducks says:

          Re: your last sentence: that’s kind of like saying that if you enjoy your software engineering job you shouldn’t be paid for it.

          Though I’m not saying that women should be paid for emotional labor (unless they’re teachers/therapists, then of course they should be). Rather, emotional labor should be recognized as labor, and attempts to achieve equal division of labor should include emotional labor as part of the “calculation”. (Which really means that fathers should in general do as much parenting as mothers, and spouses should support each other equally.)

          • Andrew says:

            OK, but why do you think it isn’t recognized? What exactly is the lack of recognition?

            Fathers don’t need to “do as much parenting” in order to recognize parenting as contribution, do they? Might they provide compensation in some other form?

            Also, have you given any thought to the “sexual marketplace”? Maybe the spouse who does more of the labor in the relationship can obtain a higher quality spouse. Maybe if women want men to do all the work, they can just choose men who have no options for “better” women? But maybe they’ve chosen instead to have “better” men, even though such men don’t have to (and won’t) put up with that?

    • I’ve also heard that part of men putting in more hours is that they take longer commutes– this gives them a larger pool of potential jobs.

    • Tracy W says:

      Just a minor point: society pays regardless. If women do this sort of work instead of making widgets for pay, then society pays in the sense of having less widgets than it otherwise would. If women did none of this work and thus men produced less doodackeys then society would pay in the sense of less doodackeys. If the work comes out of leisure time instead of work time then obviously people are poorer (the whole point of production is to consume). If the government set up a bureaucracy to do this sort of work then obviously society would have to pay the taxes and associated costs for that bureaucracy, including the opportunity costs of the wackadoodles the bureaucrats would have made instead.
      The same argument is why con scripting men into the military didn’t save money over paying them to join.
      Obviously there are more and less efficient ways of providing all these services, and it probably is true that many bureaucracies don’t account for the costs they put onto their users and could be organised to lower these costs. But society pays regardless, and society pays even more the more inefficient the system is.

      • Tarrou says:

        Monetizing unpaid work brings it into the taxable arena too, don’t forget.

        You don’t tax mothers, you tax day care! And the mothers work so they can pay day care costs, and you tax them too!

        It’s not clear to me that this brings an increase in net income to anyone, or that living standards (tech adjusted) or happiness levels have increased as a result. But it’s about status, not money, in the end. That’s the part people forget.

        • RCF says:

          Childcare is tax deductible.

          • Luke Somers says:

            Up to a limit, but the child care workers do pay income taxes they wouldn’t if they were just taking care of their own kids.

          • RCF says:

            It’s not clear what alternative universe you’re comparing this to. If we look at a particular dollar spent on child care, it would be taxed if it weren’t spent on child care, and it would be taxed at the parent’s tax rate, rather than the child care worker’s tax rate, and the latter is probably lower on average than the former.

          • roystgnr says:

            Childcare is only effectively tax deductible if you itemize, and most tax filers (two thirds?) take the standard deduction rather than itemize.

            Is that incorrect? Or is it a moot point, because child care deductions are so large that people who are eligible for them are certain to itemize?

          • Luke Somers says:

            > It’s not clear what alternative universe you’re comparing this to.

            I’m not making a between-worlds comparison. I was making a statement about this world. It will apply to any comparison of this world to some other world you wish to make.

            About other tax issues: Child-care benefits can be given pre-tax up to a substantial amount (of course this only applies to actual benefits). Also, there is a tax credit you don’t need to itemize to get, but that’s small.

    • RCF says:

      But, at least in a community property state, anything that John is paid is part of Sally’s income.

      • Tracy W says:

        If John & Sally are married then if Sally has a lower income John’s consumption is probably lower.

  6. stillnotking says:

    Pollack says:

    Our society seems remarkably uninterested in addressing the conditions that lead to such disparities in criminal offending.

    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.

      • Ryan says:

        That’s great. Another form is “my opponent is opposed to proposal X to alleviate problem Y, proving he wants problem Y to continue.”

  7. 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.

  8. AR+ says:

    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.

    • Deiseach says:

      I dislike the term “socialisation” because, although it may not be meant that way, it sounds as if it’s some harmless process: boys get to play with tipper trucks and girls get to play with dolls, so when they get old enough to go to work, girls go for part-time work or ‘softer’ jobs and that’s just how it is!

      The point I want to make is that this doesn’t “just happen”; that there’s a huge and uncounted reliance on “women’s work” by society as a whole and by people in general, even the most liberated and egalitarian ones.

      I want to conduct a quick and unscientific mini-poll here. Out of the males/male identifying/leaning towards that identity people on here, how many of you are the ones who (using examples I’ve heard today from my female colleagues):
      (a) take time off work to go clean the house from top-to-bottom for a friend/family member who is coming home from the hospital?
      (b) use your lunch hour to make phone calls on behalf of children/parents/other family members to banks, doctors, utility companies, etc.?
      (c) nip out during the afternoon to go to a car boot sale selling legal text books on behalf of your college-attending children? (Again, for those of you not old enough to have kids going to college/childless, imagine you’re doing something like this on behalf of a friend/other family member).
      (e) will leave after a work meeting and not attend the this-time-of-year social gathering afterwards but collect a child from the train station to drive them to a interview preparation coach?
      (d) those of you single or not in relationships – or even if you are – who of your parents is the one who keeps track of anniversaries, birthdays, significant events? Is it your father or your mother who rings up to remind you Cousin Sally is getting married, send her a card? Are you the one who keeps in touch with friends by remembering to send e.g. birthday cards, Secular Solstice Commemoration Cards for this time of year, or do you go “yeah they were a great bunch of guys but now I’ve moved halfway cross the country, I’m letting the friendship lapse by not getting in touch”?

      Society relies on women being available to do all sorts of work in the domestic sphere that, if women worked at their main careers the way men do, would need to be paid for in the form of health support, education, cultural/sporting activities, childcare, elder care and the like – either by public provision by the state, or by employers having to pay wages to women high enough to pay for the childcare etc. (a perennial complaint for working women; most of their wages go on childcare expenses if – once again! – there isn’t a grandmother available to take care of the baby/younger children during the working day).

      There’s a huge invisible cost to “women and men are socialised differently” that is not, I propose, counted in when asking “Well, why do women choose such-and-such professions that don’t pay as highly as men? Or why don’t women just ‘lean in’?”

      • Joyously says:

        Three points:

        1. While it is true that women are more likely to do unpaid “women’s work,” men also do unpaid “men’s work.” For example, just over this Thanksgiving holiday my father: constructed a handrail on the front porch steps at my feeble grandmother’s request, put up the Christmas tree/lights, and drove for most of a day while I relaxed/napped/played computer games. (I’m a woman. If one of my brothers had been there instead, I suspect he’d have had to take a turn driving.) These are all “masculine” duties, and I’m sure you can think of more. I suspect that women do *more* unpaid women-work (since they do less paid work), though I’d need to look up studies to confirm this. Doing this sort of work is essential to most household economies, and by extension to “society” as you word it. (A society that pays my father for driving on road trips and my mother for sending out Christmas cards would be a *bad* society too entangled with private life.)

        2. Is this “invisible cost” you speak of a cost in money or freedom? Because if you’re talking about cost of *money* that doesn’t make sense to me. A woman who works 8 hours a day taking care of her own children and a woman who works 8 hours a day making money which she uses to pay some other woman to take care of her children are, economically speaking, exactly equivalent. If your entire paycheck goes to your nanny, that only means you’re contributing exactly as much to the household economy by holding a job as you would be if you stayed home. This doesn’t change if the child care is paid for by the government–that just means the cost has been shifted partly from the childed to the childless. The *cost* is still the same–8-woman-hours worth of work. The argument for subsidizing childcare should not be different than the argument for subsidizing households that need resources for any other reason.

        If what you mean is that there is a cost of freedom, that makes more sense. If a woman with children wants to work a paid job, she has to either pay for childcare or have a husband/partner who stays home. She has less freedom to make that choice. Though that just boils down to socialization again–her husband has less freedom to make the choice to stay home.

        3. Sweden, Denmark, and France wealthy countries with famously generous work-family policies including “up to 16 months of paid leave after the birth of a newborn, extra tax credits to defray the cost of child-rearing, plus access to regulated, subsidized day care facilities that stay open from 6:30 in the morning until 6:30 at night” (http://www.newrepublic.com/article/118294/us-should-copy-sweden-and-denmarks-work-family-policies) have unadjusted pay gaps in line with the rest of Europe and only somewhat higher than the Us’s. (http://epp.eurostat.ec.europa.eu/statistics_explained/index.php/File:Figure1_The_unadjusted_Gender_pay_gap,2012.jpg)

        • nydwracu says:

          When my divorced mother’s car broke down, she called me and my uncle to fix it.

          When she needs her gutters cleaned, she calls me and/or my uncle to clean them.

          When a wall cracks and needs patching, she calls me to do it.

          When a lightbulb burns out, she waits until a man is visiting and gets him to replace it. (I am not making this up.)

          Men’s work probably takes less time than women’s work, but it requires more skill (except in the case of cooking, and a lot of men can cook these days) and more physical exertion. (Possibly more risk, but I’m not sure how climbing / cutting down trees, crawling around on ladders, dicking about with power tools, and so on stacks up against driving.)

          • Nornagest says:

            If you’re looking for skilled “women’s work”, I’d point to sewing (and, in a premodern context, weaving and spinning) before cooking. Cooking at least well enough to feed yourself isn’t hard; there are people that can’t do it but I think that’s more of a learned helplessness thing than an actual skill gap. But doing anything more complicated with fabric than sewing a button back on? That takes real skill, and in a domestic context it’s historically been highly gendered.

            If we look back far enough it’s been even more dramatic. Before the Industrial Revolution, most fabric was handmade, and that takes both skill and a lot of time: we use distaffs as a female symbol (“the distaff side of the family”) because, at one point, all the adult women in your average family would have been walking around with a distaff and drop spindle, spinning thread or yarn, for a good portion of most days.

          • Anonymous says:

            …it’s almost as if men and women seem to contribute different, probably incalculable, efforts. You almost had me thinking that we should hesitate to overthrow a long-negotiated balance of effort based on a random measure which shows a gender disparity. ALMOST! I like my random measures of disparity too much… especially when they show that my gender gets the short straw.

          • Nornagest says:

            Are you talking to me or Nydwracu?

          • Multiheaded says:

            Time, skill, exertion? Okay, but what about worry? What about actually caring about the place? Like bothering to put things where they belong every damn time when it’s just such a small thing and could always be done later? Actively keeping even the appearance of entropy away?

            I wouldn’t say it qualifies as a full-time job alone – there are other factors – but in some contexts people get paid a half-decent wage to do essentially that.

          • Nornagest says:

            Okay, but what about worry? What about actually caring about the place? Like bothering to put things where they belong every damn time when it’s just such a small thing and could always be done later?

            I think you’re conflating two mostly distinct issues (cleanliness and emotional attachment) here, I think casting the latter in terms of labor is extremely sketchy (worry is unpleasant, but I also don’t think it’s very valuable; emotional attachment more generally comes with its own rewards), and I think they’re both less gendered than you’re pitching them as.

            I can only speak for myself, but I’m better at keeping a neat house than every girlfriend I’ve ever had. It’s not a skill that gets emphasized much in my generation, male or female.

          • Keeping a place clean isn’t necessarily about worry, but it does take thought to keep track of all the tasks.

          • Joyously says:

            Nornagest’s mention of sewing reminds me–among people who aren’t trying to tear down the gender structure, I think doing “gendered” work for each other is often a form of flirting. I had a roommate who would offer to sew on cute guys’ buttons for them. Another roommate’s boyfriend replaced the two-prong plugs in our living room with three-prongs. Maybe this is just a specific form of doing-nice-things-for-the-person-you-like, but I suspect the gender element is significant.

            The guys could of course learn to sew on their own buttons (and many do) and we could learn to replace our own plugs (as many women do) but then we’d need different mating rituals.

      • lunatic says:

        As a male reader who does more of the tasks you list than my partner, I want to point out that the reason I seem to wind up doing a lot of these is that I have a less demanding job than my partner. That is, many of them are not so much about gender roles (house cleaning seems to be an exception) as a pragmatic compromise regarding who has the time to do it.

        I’m claiming its always this way, but that the causation can go in both directions.

      • Tracy W says:

        Childcare is expensive because it’s labour intensive. Eg in my part of the world the legal maximum ratio of staff to children under three is 1 to 3. If every child care worker is paid the median wage then that means they must be paid 1/3 of the median wage per child. A rough rule of thumb is that the total marginal costs of hiring someone is twice their salary (eg holiday leave, sick cover, benefits, admin costs) so each child costs 2/3 of the median wage.
        Now each childcare worker might be paid less than the median wage but then there’s all the other costs of childcare like the building and heating the building and buying food and nappies and vast amounts of paint and glitter. So 2/3 of the median wage looks about roughly right. And thus once you’ve paid taxes as well, it’s not surprising most parents have little money left over.

        Raising wages generally will raise the wages of the childcare workers (as they would otherwise quit to take the newly high paying job) and thus raise childcare costs.

      • Shenpen says:

        I don’t understand the take time off work part. I am a European. Is it common for Americans to have flexible hours, not the 40 hours a week schedule? (This also why I don’t understand “hard work” : everybody works just about the same hard. Just not the same smart.)

        a) Never do such a thing, that friend or family member will pay a cleaner. (And yes they are usually women, cash under the table means you get one often under or near the min. wage).

        b) I do e-mails, and yes I do that, not wife, as they find it harder to ignore me. She does the phone calls later on, because if my emails don’t get answered my phone calls would be overly aggressive.

        c) nobody does stuff like this as far as I can tell, people buy them at school or online

        d) We have no social gatherings. We work then go home. I never understood pretending to be friends with coworkers. It is a job. It pays bills.

        e) hardly anyone gives a fuck about anyone not in the immediate family. Why on Earth would I care about a cousin’s birthdate or anniversary or even a friends. This is their thing. If they want to invite us to a birthday party they will tell.

        Wow, it seems our lives are really different. We seem to do everything less socially?

        Now it sounds like your description of an American woman is some kind of Informal Social Life Organizer. But is that work or a kind of hobby? Why care about a cousins birthday if it is not a hobby to care?

        Unpaid, unaccountend for womens work here seems to be the usual: childcare, cleaning, cooking…

        Now for fun, unpaid / unaccountend for mens work in Central Europe:

        1) assembling all the furniture
        2) household repair, DIY
        3) take care of the family car, if having two cars, take care of the wifes car as well, know when the next oil change is due, wash it on the weekends
        4) keep the garage and tools in order
        5) usually everything in the garden that does not involve flowers, rake the leaves etc.
        6) fix all the computers all the possible ways
        7) research and decide about all the electronics purchases, then set them up
        8) “do something” where the Internet is down

        It is not equal – cooking is a bitch of a timesink, and infants take a lot of looking after, it is about 30-70. I am just saying it often adds to quite a lot sometimes.

        • llamathatducks says:

          Is it common for Americans to have flexible hours, not the 40 hours a week schedule?

          In my understanding, yes. Many office jobs, especially “creative” rather than administrative ones (like in software, finance, etc.) don’t have defined hours, you just work some amount of time that lets you be productive enough for your boss to be happy with you. (And in some places the culture is to work as much as possible – I’ve seen lots of articles about how absurdly workaholic the culture in big financial firms is, for instance.) Also e.g. doctors and lawyers have absolutely absurd working hours. Basically anyone who’s on a salary rather than an hourly wage is likely to have undefined and ever-expandable working hours.

          On the lower-wage end of the scale, it’s often the other way around: hours are “flexible” because they often don’t add up to a full 40 hours (because then the employer would be on the hook for healthcare and such) and they can also be unpredictable as companies do complicated things to determine when they’ll need more people working, but they can’t do that very far in advance.

    • Karmakin says:

      This is a very important point. Statistically, the seeds we plant now will take decades to take full bloom. That’s not to say that planting the seeds isn’t important, but people really need to understand how this sort of statistics play out.

      Unless you want to pass a strict equal pay for equal work legislation, which is something I support…most people don’t however because they think they’re “better” than their co-workers and as such deserve to be paid more. I personally think measuring productivity in today’s interconnected world in most cases is a fool’s errand.

      • Emily says:

        Labor markets are highly gender-segregated. The idea that you have a man in one office or cube or section of the assembly line doing the same work for the same hours as the woman next to him, but earning more money, would not be an accurate representation of what’s driving wage gaps. (And, of course, it is already illegal to do that.) Men and women do different kinds of work. If you want them to get paid the same, and you want to legislate that, you would have to say that people in job x must be paid the same as people in job y. Calling that “equal pay for equal work” is begging the question, since the dispute here is whether job x is actually equal to job y.

      • Anonymous says:

        “I personally think measuring productivity in today’s interconnected world in most cases is a fool’s errand.”

        When isn’t it? People seem to think that “productivity” as economists use the term literally exists as real world “productivity” in the English word sense, which betrays a complete lack of understanding of how supply chains work. Every step in the supply chain is necessary, from the chemical company that provided fertilizer to the farmer to the farmer doing farm labor to the truck driver who drove the produce to the store to the manager of the store to the store employees…. None of these people have meaningful “productivity” besides being a step in that chain – all of the steps in the chain are completely valueless without the others.

        How much each step in the chain is “worth” is determined by supply and demand. People are fond of calling that “productivity,” but it has to be stressed that you might as well call it “cheese factor” or something – how much you get has nothing to do with what you actually do. It has to do with how many other people can do what you do and how many companies need people who can do what you do, and the various protectionist measures taken by labor or capital to try and get as good a deal as possible.

        A nurse helps sick people. That’s what they do. That’s the only “productivity” in a literal sense that they have. They also can get a certain price for their services. This has nothing to do with how much their services are “worth” to society. In a world where there aren’t enough nurses and people think of it as something like garbage collection, the price of a nurse is high, like the price of a waste management engineer is high. In a world where tons and tons and tons of people want to get into nursing, the price of a nurse is lower. This despite there being zero change in what they do or how much their services are “worth” to society.

        If anything, the more a service is “worth” to society, the less “productivity” (price) they have, because services that are prestigious will have more people trying to get in, all else being equal, than services that aren’t. Another factor is that the more something is “worth,” the more likely it is to be simple – you don’t have to be a brain surgeon to grow food or drive it to market, but it’s just about the most important thing for society. You also don’t have to be a genius to build houses, but that’s probably the second most important thing for society. And taking care of sick people is mostly straightforward but dirty work, and reliably keeping sick people clean, fed and hydrated is incredibly important for society.

        You get the point. If you use the economic term “productivity” to mean actual productivity in the sense of how much “value” is being provided, you’re going to have a very very very bad time. “Productivity,” meaning how much someone is paid, is a measure of their class. Upper class, intelligent people are who push the boundaries of human capability, so in that sense there is value to be found among those with higher “productivity.” But make no mistake, the biggest value is in the people who grow the food, build the shelter, keep law and order, and nurse the sick, and they’re around the bottom of the “productivity” numbers.

        So what does this imply? Should people be paid in proportion to their value to society? Of course not – the most valuable tasks are things that pretty much anyone can do, and there’s no sense in promoting a huge oversupply of farmers or whatever. No, what it implies is that you should not ascribe any moral weight to economic “productivity.” There is no reason that someone “deserves” more money than someone doing a different task – it’s just a question of how the amoral market values their skill.

        This has big implications, among them that it’s not some sort of horrific injustice to flatten the wage curve through interventions. The market should be allowed to communicate shortages and surpluses. We don’t need it to tell us what people are “worth” in terms of how much luxury or poverty they should live in. When we tax people who get paid lots and lots and give it to people who don’t get paid as much, we aren’t committing some kind of sin, and there is no sense in which the person being taxed “deserved” that money. It is useful to have them be paid more so that the price system can communicate what needs to be communicated, but there is no reason, moral or practical, to have the take-home income be the same as the market price.

        They didn’t “produce more” and get it taken away and given to people who “produce less” – they got a better deal in the amoral market, which morally entitles them to nothing and practically entitles them to whatever percent of the gain is necessary to maintain incentives. For people who manage to get hugely good deals in the market, that’s going to be a very small percent of the gain.

        • Chris H says:

          Doesn’t this miss a potential use of the term “productivity “? Specifically as a means of comparing how well two people in similar jobs do relative to each other? You’ve made a case for not using the idea of productivity between jobs, but if one nurse can effectively help five patients an hour and another can help seven with the same effectiveness, can’t we call the latter more productive? Admittedly, most aspects of inequality, whether it be prestige or income, arises between not within most professions (though some exceptions may exist such as writers) but the idea of productivity still seems to have some usefulness.

          • Karmakin says:

            Using your nurse analogy, here’s why I don’t like the measure of productivity. What if the 7 patient nurse chooses easier patients than the 5 patient nurse? What if the 7 patient does things in a way that increases costs down the line? And to be honest, more than likely they’re both working with the same pool of patients, so if say the 7 patient nurse doesn’t do something that the 5 patient nurse then takes care of, then that’s a problem.

            It’s those types of things that are the reason that determining productivity is so difficult.

      • Mary says:

        The last thing we need is some irresponsible government bureaucrat with the authority to decide what is equal work.

      • Tracy W says:

        Also how do you define what is equal work? The factors that go into that are immense, eg the hours, how unpleasant the job is, where the job is (including how many people of the opposite sex will be around), who the co-workers are, what sort of training is needed, etc. And what if for some reason you need more people working in a particular area suddenly, or less people?

        • Karmakin says:

          I agree, defining equal work is tricky, although I’d take hours out of the equation by saying everybody has to be paid hourly (but that’s another kettle of fish). Past that, generally speaking I’d use the concept of equivalent tasks. But again, that’s hard to define.

          This isn’t the hill I’d like to die on, so to speak. While in theory I support the concept of equal work legislation, I understand it’s pretty much impossible. More so, it irritates me that people who should support that sort of thing don’t.

          • Murphy says:

            Also, should you have the right to negotiate? if you apply for a job and they offer you a figure, you say “no, I think I’m worth more than that because…”

            Which also leads to a gender pay gap as women are less willing to negotiate.

          • Tracy W says:

            although I’d take hours out of the equation by saying everybody has to be paid hourly

            Even if everyone is paid hourly that doesn’t mean all hours are of equal value. Eg if you’re a midwife then your hours are tied to when the babies decide to come. If you’re a farmer and a flood’s coming and you need to get your stock to safety you’re working until that’s done. Or you’re working all hours when the weather’s right for harvesting. And different people have different preferences for weird hours, many students wait tables for that reason.

    • llamathatducks says:

      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.

      That pretty clearly doesn’t apply to the particular example of being hassled by police, which is primarily what Klein was talking about there.

      • William O. B'Livion says:

        s/race/culture/ where race is used as a visible proxy.

        • llamathatducks says:

          I’m not sure I understand. What I’m saying is that if police officers treat white people and black people in the same situation differently, that can’t be ascribed to racial difference, it has to be ascribed to racism. When police officers saw student Cory Booker with suspicion when he was a non-criminal Stanford student, that was definitely due to the police officers’ prejudices rather than any of Cory Booker’s attributes.

  9. Toggle says:

    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?

    • Hanyo Ilthium says:

      I feel the ideal solution would be for Scott to disable comments in concert with LessWrong allowing link posts to the main.

    • Scott Alexander says:

      You would think so, but actually that article got me fewer clicks than a moderately popular LW post that links to me does.

      I am pretty sure that even for Vox readers, an article on the ambiguities of statistical controls isn’t really clickbait.

      • Randy M says:

        “Even for Vox readers”
        Is there some study that shows readers of Klein’s site are more interested in mathematical precision than readers of Instapundit, NRO, or whatever?

      • Anonymous says:

        Huh, that’s an interesting statistic in itself. Sometimes I worry that the average person isn’t very enthusiastic about the ethics of rigorous sampling in multivariate analysis.

      • Karmakin says:

        It also might be that Vox readers are not really that interested in reading up on primary sources.

  10. Chris Leong says:

    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

    • Scott Alexander says:

      I would think at the very least they would use something based on comparing white people to black people within the same neighborhood, or a white person in one neighborhood to a white person in another neighborhood.

  11. Rowan Santry says:

    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…

    • John Schilling says:

      “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.

      • llamathatducks says:

        My interpretation of what Rowan said is that black police officers might be better at telling rich black people from poor black people, either because (a) white officers are too racially biased to tell or (b) white class markers are somehow different from black class markers, and each person is better at distinguishing the class markers of one’s own race.

        I’m not sure I agree, but I think that’s the proposed mechanism. (And anyway I don’t like police work that specifically targets poverty.)

  12. haishan says:

    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.

    • DES3264 says:

      Scott’s hairdryer story is a great example of fixing problems downstream.

      I always understood “affirmative action” to refer to people saying “look, we know we’re not really addressing the causes of racial/gender disparities in hiring/school admissions, but we think it will be worthwhile to solve the problem downstream.”

      • Jacob Schmidt says:

        That’s largely how I see it. Affirmative action is a sloppy bandage covering a complicated problem. Its done partially because its easy to measure. You get complaints about equal outcome vs. equal opportunity, but the latter is really difficult to measure, particularly when you include subtle things like being gently dissuaded from certain fields at a formative age. Done perfectly, the affirmative action measure would mirror the systemic biases. That’s not remotely feasible, though, so we get things like “30% of your new hires need to be women” or similar.

        Affirmative action does have the benefit of making minorities more visible, which in turn might help us in disassembling biases, so its not entirely a downstream process. The stream has a recycle stream? I dunno, the analogy breaks down a little bit, there.

      • Noah Siegel says:

        Are you stating the moral of the hairdryer story to be, “do what increases net utility, *regardless* of where in the causal stream you put your intervention” or as “downstream solutions are more likely to increase net utility.”? I’m pretty sure it’s the former.

        If your point is that affirmative action might be a good solution *regardless* of being downstream, then I agreee with you.

        • houseboatonstyx says:

          I think net utility will often be increased by looking for the point in the stream where your own power can be most effective.

          Looking too far upstream — well, it breaks the metaphor, for one thing, so here’s a real example. If you’re a business owner or an academic head of whatever and want to recruit more women, yourself doing some form of affirmative action actually gets more women appointed. “Our culture socializes females wrong” just leads to trying to get other people to change their attitudes by flames or petitions etc.

          The wider a target, the further your own power is diluted, the less leverage you have, and the more metaphors you mix.

          Your affirmative action program can actually snowball. The new women appointed can work up to be department heads themselves and hire more women who appoint more women, etc; can make a decisive difference for their women students; and a goodly number of women can jump a rung or two on the pipeline.

          I’d better post this before I get to the glass ceiling.

      • Tarrou says:

        Ok. Has Affirmative Action helped downstream? I think you’ll find teasing a direct benefit out of AA is going to be quite as difficult as nailing down a concrete example of structural racism in the justice system.

    • Ryan says:

      So I think we can look at three somewhat unique categories:

      First off, everyone agrees we have X and that X is a problem.

      Category 1: There are multiple, real viable solutions to the problem. “What are we going to do for dinner?” Well there’s Taco Bell, Chilis, the leftovers in the fridge. And if there is any debate, it is about which is the best option to choose.

      Category 2: There is in fact just one solution to the problem. “The patient is flatlining, do we administer drug A or drug B?” Then the debate is about why the patient is flatlining and which is the correct drug to administer.

      Category 3: There is in fact no solution to the problem. “I love my grandma and I want her to live forever.” “But this closed system would be so much better if we decreased its entropy!”

      I think American (at least) politics tends to do two things which further complicate things. First, it assumes that every problem is a category 2 type problem, when point of fact the problem may have several or zero viable solutions. Second, it makes certain value judgments regarding solutions which can turn a category 1 problem into a category 2, and a category 2 into a category 3.

      Drugs is of course a great illustration. Everyone agrees that heroin addiction is a problem. No one agrees if there are zero, one or multiple solutions. And plenty of people have very narrow lists of what is an acceptable solution to begin with.

  13. Ilya Shpitser says:

    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.

    • haishan says:

      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.

      • Ilya Shpitser says:

        I agree that it’s super hard.

        Actually what Scott is talking about is even harder than ordinary causal inference, because even if you could do an RCT you may not be able to do mediation analysis (which is what questions like discrimination are about). In addition to controlling treatment assignment, you need to be able to split treatments into components (e.g. nicotine patch vs nicotine-free cigs, if you think it is nicotine that is effecting one pathway and smoke another pathway).

        But you can at least say: “given that this your target [this is the important bit, agreeing on the target], if you make these assumptions, here are some bounds.” And so on. But importantly, you start with the target, not with the data.

      • Vaniver says:

        In case you’re not aware, Ilya Shpitser is a technical expert in the subject of causal inference.

        • haishan says:

          This is what I get for not reading peoples’ names, I guess. At least I didn’t say anything too flagrantly wrong.

          • Ilya Shpitser says:

            I really worry about self-censoring due to that, actually.

            Presumably, saying something stupid and then adjusting (silently or otherwise) is more virtuous behavior than holding your (biased) peace.

            I have a list somewhere of stupid shit I thought/said about causality at one point.

      • Totient says:

        You’re talking to Ilya Shpitser; believe me, he knows how hard causal inference is:
        http://ftp.cs.ucla.edu/pub/stat_ser/shpitser-thesis.pdf

        Controlled experiments are still the (only) gold standard for inferring causality. But if you’re willing to make additional causal assumptions (e.g. the existence/non-existence of causal connections) you can make additional causal inferences (e.g. how strong is this causal effect).

        So what happens when people disagree on which assumptions to make? You get more debate. But it’s debate ratcheted up another level of abstraction which I’m pretty sure would be a huge win, especially for this issue.

        I’m all in favor of turning problems into math.

    • The Do-Operator says:

      Direct and indirect effects are only defined relative to a mediator. It seems to me that whether the direct effect of gender should be considered to be “discrimination” will depend critically on what mediator we are considering.

      For example, if you are interested in the effect of sex on income, and the mediator is occupation preference, it seems most rationalists will consider the indirect effect to be morally neutral

      However, if the mediator is a hiring decision made by gender-biased companies, the indirect effect may no longer be morally neutral.

      Even more confusing when trying to make sense of ethics in a causal inference framework, it can plausibly be argued that confounding pathways may have moral relevance. I think this is at the core of the disagreement between Scott and Ezra. I tend to agree with Scott on this: If we are discussing ethics for the purpose of making fair decisions, confounding pathways will lead to misleading conclusions.

      It is also important to note that to capture all the ethically relevant features of the situation, it may be necessary to consider between-unit interference.

      • Ilya Shpitser says:

        I agree that the mediator is important (or more precisely, we have to settle on a graph first, and then settle on what pathway we want in that graph).

        I also agree that we may have to keep interference in mind.

        What I was trying to call for with the GP post was moving from “what confounders do we adjust for in this dataset?” to “what do we actually want, anyways?”

  14. 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?

  15. Tarrou says:

    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.

    • Not THAT anonymous commenter says:

      I agree – I disagree with Scott often (from the left), but pretty much always think he’s being careful and nuanced, which is great.
      Since this subthread is giving Scott cookies, I’d also say that so far I’m impressed with the comments that have gone through moderation. I’m sure it’s a lot of work to do that kind of weeding. Nice job, man. 🙂

  16. Troy says:

    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 (https://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.

    • John Schilling says:

      I don’t think white progressives are terribly worried about themselves being knifed by the Klan. That was a factor once, but not now and not in any plausible near future.

      White progressives may be worried about black people being knifed by the Klan, but their actions look to me like they are optimized for being on the winning side of the “All wrong-thinkers should be fired” witch hunt rather than for actually preventing black people from being knifed by the Klan – and it’s very hard to be both, because every wrong-thinking racist who gets fired for his views is one more strong candidate for the Klan with maybe nothing left to lose. Meanwhile, the stuff that actually might protect black people from the Klan, is mostly being ignored by the progressives.

      I am also skeptical that it is just conservatives who are afraid of witch-hunts, but that’s another matter and I want to think about it some more.

      • William O. B'Livion says:

        I am also skeptical that it is just conservatives who are afraid of witch-hunts, but that’s another matter and I want to think about it some more.

        It’s entirely dependent on the dominate culture. I’ve spent about 1/2 the last 6 years in the Defense Contracting space, and about 1/2 contracting in the Internet Sector.

        In the DoD space people are still very conscious about race, but people are very anti-communist and largely anti-progressive. They generally view themselves as “Conservative” even if they’re not really clear on what that means[1].

        I know that among the contractor/civilian side of things people who strongly support gay marriage and are in favor of gun control (as examples) generally were very quiet about their positions, and I have never met one (employee, not spouse. Mix marriages are odd!) who actively advocated more socialism. I’ve met soldiers, sailors, marines and Airmen who did (never met a Coastie who did, but they’re rare birds anyway) but not on the civilian side.

        I can’t speak to whether they are afraid for their jobs (probably not) or just don’t want to make waves.

        I do know that it is generally the case that conservative politics are inherently reactionary (duh, it’s built in) and that it is much more generally assumed that, for example a conservative position on race relations (today, not in 1950) looks like “blaming the victim”, but since #9 on Kirks list is “…perceives the need for prudent restraints upon power and upon human passions” we take the position that *generally* you won’t get arrested and convicted of a crime you did not commit.

        But taking this position is perceived as being racist (or at least bigoted, and it arguably is) and *does* open you up for a “hostile work place” complaint. Ditto if you’re a committed Christian in one of the sects that hold that homosexual behavior is a sin in the eyes of god. Whether you and I believe that is irrelevant, there are people for whom this is a deeply held worldview, and they are entitled to not only *hold* that opinion, but to voice it. But when you do, even if it’s to remind a co-worker that not everyone shares their beliefs and maybe the work place isn’t the best place for political signaling/arguments then it’s your fault and YOU are creating the hostile work environment with YOUR intransigence.

        Keep in mind that todays corporate climate is to advocate *for* more gender, race and sexual orientation diversity, even in DoD/IC environments. It is *not* acceptable to question whether that is a supportable long term goal.

        Now, as a hard-leftist if one were to argue that those of us on the Right (or Christians, or Republicans) should be purged from the corporate ranks and forced out of polite society then that would also open you up for a EEOC complaint, and it certainly would be creating a hostile work environment.

        It was my experience in a couple of Sillycon Valley startups and more established companies that the default assumption in that environment was that one was Liberal to Progressive, and just saying “no, not all of us here are like that, you should tolerate a little more diversity” really upset people, and it was generally my fault for not shutting up and going along with the crowd.

        [1] I tend to go by Russell Kirks 10 principles as a general test/guideline for what American conservatism is, since it’s the stuff that the conservative intellectuals generally wind up referring back to at some point. Him or Burke, but Burke is Irish/English and Things Are Different There).

    • Vaniver says:

      Witch-hunts go after women and minorities too when they express the wrong opinions.

      Ferguson has been particularly obvious for this; any black person who looks at the forensic evidence and gets convinced that Wilson’s story was mostly accurate immediately gets called an Uncle Tom, because racial slurs are okay when they’re about conservatives.

      • Toggle says:

        An important difference between ‘Uncle Tom’ and most racially coded slurs is that it doesn’t implicitly treat the race as an insult. If anything, it accuses the target of being insufficiently black, for some social value of blackness. The race itself is treated with pride. Most other slurs that I can think of are simply a racial description with negative connotations.

        • Anonymous says:

          To me, it’s even worse than that, it’s, “You aren’t sufficiently racist in our favor.” Let’s call pathological social norms like they are.

    • William O. B'Livion says:

      Another thing to consider is that people are just flat out *odd*, and it’s not “minorities” that are always the problem. I’ve known quite a few people who were fine with one race, or even attracted to it but didn’t care for another race.

      There was an Australian (this story was relayed to me, I didn’t witness it) who was waiting for a bus to work with an American Black, and the Australian started using “The N Word” in a diatribe about Aboriginals. The American was *deeply* offended and the Australian didn’t get it.

      A guy I was in the Marines with (20+ years ago) really did not care for Black folks, but thought the world of Chinese, especially the women.

  17. Audrey says:

    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.

    • Deiseach says:

      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.

      Which is excellent and badly needed, but I think indirectly supports my point about socialisation: which gender is getting supported here? And why?

      Yes, women (so far) are the only ones who have babies (excuse me, those of you who want to point out that trans men can also and do also have babies, that’s not the particular axe I want to grind right now) so pregnancy leave etc. is going to be tilted towards them.

      But this also assumes and presumes women are going to be predominantly the ones who need part-time/flexi time to take care of family duties, because women have always done that and are now taking on paid work outside the home on top of that traditional role.

      Make it paternity or merely parental leave; let men be the ones care-taking elderly relatives or children and who need to work part-time/flexi time; childcare for low-paid male workers with children – make it as normal that a man will be the one needing to leave at three o’clock in the afternoon to bring little Johnny to the dentist, as it is now normal for women to be the ones doing this, and we’ll finally see real social progress.

      Otherwise, we’re only going to make the cul-de-sac of “women’s careers aren’t as valuable as men’s because women don’t put the time into building their careers” a little nicer, more comfortable and easier but we won’t have changed expectations that the daughters/sisters/wives/girlfriends/female partners/mothers are the ones still juggling the domestic with the paid work.

      • Audrey says:

        I think both things have to happen together (And indeed more. The Fawcett Society also covers the points you raised in your other points). There are various social benefits to men doing more unpaid work, but the consequences to pay per hour and promotion of working part time that two adults in a household doing it increases poverty compared to one full time worker.

        I would not want to see a situation where the gender of the people doing lots of unpaid work changed but those people were still needlessly disadvantaged by it. We need to both increase men’s unpaid work and create paid working environments that fit with the needs of society.

      • Audrey says:

        http://www.fawcettsociety.org.uk/2013/11/equal-pay/

        Sorry, no edit function. The Fawcett society covers many of the points you raised in your previous post:

    • llamathatducks says:

      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.

      Yes, this. I really wonder if there’s any way to test this, though; it seems pretty untestable.

      If this is true, though, then “black lives matter” is extremely relevant regardless of whether there is blatant racial discrimination going on. Black lives matter, so we should be a lot more careful than we are in our police work, because lives are at stake.

      • houseboatonstyx says:

        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 the majority thinks “It can never happen to me, it only happens to Blacks or poor people,” they won’t get as involved, and the abuse will continue. So it seems important to show that, for instance, police shooting does happen to all demographics.

        In promoting local measures such as requiring all police to wear cameras, for many white people “Black lives matter” will be less effective than “Yes, it can happen to you.”

        • llamathatducks says:

          You might be right, which is kind of depressing.

          On the other hand, I wonder if “black lives matter” can also be effective in bringing out people’s desire to feel like a good person. As in, most people wouldn’t want to disagree with “black lives matter” because then they’d be racist assholes.

    • Anthony says:

      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 …

      Not true. The United States went from a “gender pay gap” of 59c to 71c from 1981 to 1989 – 1.5c/year. While I happen to think that the policies (and the President) during that period were rather good, I wouldn’t argue that they actually constituted what you recommend.

    • William O. B'Livion says:

      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.

      The Irish in the 1600s through the early 1900s. The Italians in the late 1800 and early 1900s. The Chinese during mid to late 1800s. In all of these cases the dominate culture thought those in the dominated culture “deserved” it for some reason. And in both these cases and todays inner city black communities there are folks who want to change the culture. Except today they want to change the culture without changing the culture or admitting they want to change the culture.

      It’s not the *race*, it’s the perception their behaviors and choices are keeping them poor.

      Look at the Appalachian communities for the last at least 30 years (I’ve been hearing about how poor they are since I was in late high school). Freaken NO ONE but a few sociologists[1] cares about them AT ALL and they’re mostly white. They’ve got mostly the same pathologies as the inner city blacks, but they don’t ride the subway next to CEOs and Doctors. They also tend to solve their problems within their communities (probably because the CEOs and Doctors don’t see them hanging out on the street corners and send the cops after them) so they don’t get on CNN.

  18. Arthur B. says:

    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.

    • Ryan says:

      Malcolm X said this in a speech in 1963:

      The Honorable Elijah Muhammad warns us daily: The American government is trying to trick her twenty-two million ex-slaves with promises that she never intends to keep. the Crooked politicians in the government are working with the Negro civil rights leaders, but not to solve the race problem. The greedy politicians who run this government give lip-service to the civil rights struggle only to further their own selfish interests. And their main interest as politicians is to stay in power.

      In this deceitful American game of power politics, the Negroes (i.e., the race problem, the integration and civil rights issues) are nothing but tools, used by one group of whites called Liberals against another group of whites called Conservatives, either to get into power or to remain in power. Among whites here in America, the political teams are no longer divided into Democrats and Republicans. The whites who are now struggling for control of the American political throne are divided into “liberal” and “conservative” camps. The white liberals from both parties cross party lines to work together toward the same goal, and white conservatives from both parties do likewise.

      The white liberal differs from the white conservative only in one way: the liberal is more deceitful than the conservative. The liberal is more hypocritical than the conservative. Both want power, but the white liberal is the one who has perfected the art of posing as the Negro’s friend and benefactor; and by winning the friendship, allegiance, and support of the Negro, the white liberal is able to use the Negro as a pawn or tool in this political “football game” that is constantly raging between the white liberals and white conservatives.

      Politically the American Negro is nothing but a football and the white liberals control this mentally dead ball through tricks of tokenism: false promises of integration and civil rights. In this profitable game of deceiving and exploiting the political politician of the American Negro, those white liberals have the willing cooperation of the Negro civil rights leaders. These “leaders” sell out our people for just a few crumbs of token recognition and token gains. These “leaders” are satisfied with token victories and token progress because they themselves are nothing but token leaders.

  19. Jos says:

    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.

  20. Joseph S says:

    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.)

    • anodognosic says:

      I’m wary of the phrase “correlation does not imply causation,” because it’s something of a blunt instrument and conversation stopper–a gotcha! that too often leads to no further inquiry. Instead, I would suggest replacing “correlation does not imply causation” with the more sophisticated “causation may imply naive causation, a more complex causal mechanism or mere coincidence, and the issue requires further investigation to figure out which it is.” If you already do this in your head, great! But I see it too often as a way to incuriously dismiss correlational evidence or score points in a debate.

      • Ilya Shpitser says:

        Your replacement is basically like “up to 15% or more” here: http://xkcd.com/870/

        I agree that we should not treat “correlation does not imply causation” like a lexical token in arguments. I also agree that people think correlational evidence has policy implications all the time, and that has to stop.

      • Exfernal says:

        Why not use instead: “correlation does not guarantee causation” and “correlation does not preclude causation”?

  21. Shenpen says:

    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.

    • stillnotking says:

      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.

    • koreindian says:

      “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.”

      Do you have a linkable references for this position? I’m asking because it sounds very interesting.

    • Ryan says:

      I think a similar principle underlies the rules of evidence used in US courts. Two examples. First, you cannot offer character evidence to prove actions consistent with that character. The reason why isn’t that it’s not a logical argument, the reason is that it’s way too logical. How many people saw the convenience store footage of M Brown grabbing the clerk and shoving him aside and immediately decided the cop shot in self defense? Admitting evidence like that does not for a fair trial make.

      Another example is what’s called “the golden rule.” You cannot ask a jury to put themselves in the shoes of anyone involved in the trial. So you can’t say “imagine you were the one held down and raped, or imagine it was your daughter.” The issue here is perhaps the “seductiveness” you refer to. If a person really does imagine themselves in the shoes of another, that is going to drastically change how they think.

    • Anonymous says:

      This is very interesting. I’ll have to think about it. At this moment I think that thinking like this is probably counterproductive, but at the moment I can’t pinpoint the exact reason why.

    • Multiheaded says:

      As repeatedly seen in SSC comments (and with some LW users previously), the enormous issue this line of thought opens the door to is countersignalling. It drives me pretty mad, to be honest, but not so mad that I cannot describe the problem.

      Like, the ever-popular thing: person 1 goes “Women are people! Yay! Applause light!”; person 2 gets socially rewarded for going, “Tsk tsk, you silly youngster, evidence strongly suggests that they are not” *butchers the evidence and disregards plainly observable reality* *person 1 gets disheartened and doesn’t want to argue*

      • That’s countersignaling only in the sense that #2 is showing fearlessness/willingness-to-look-stupid. In other words, when that happens, I assume that #2 actually believes awful-thing, not that he’s trying to show off his intellectual courage, because courage in saying poorly-supported gets-you-fired things is not (to me) impressive. If it’s well argued and convincing, then sure, I’m impressed (and convinced).

    • RCF says:

      I don’t think Mill’s position was so much that in a free debate, the truth is guaranteed to win, so much as the people who, in a free marketplace are engaging in fallacies and demagogy, would in a regime of censorship be simply banning other points of view; as useful as censorship would be for those promoting the truth, it would be even more useful for those opposed to the truth. Truth is not guaranteed to win in a free marketplace, but at least it has an advantage. In a world where what people say is based on what is allowed to be said, rather than what people believe, what advantage does truth have over error?

  22. Error says:

    which tries to control for so many things that it ends up controlling for the thing it’s trying to measure.

    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.

    • Yeah, showing what caused something is hard.

      But if the effect is erased and the list of controls consists only of things we consider morally right, then we would rightly conclude that there’s no morally wrong effect.

  23. Koken says:

    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?

    • Troy says:

      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.

      Controlling for kind of drug still clarifies the debate by letting us ask: should our legal system treat these kinds of drugs differently? Perhaps the answer is “yes,” and the police are not being racist in targeting the one drug more heavily. Perhaps the answer is “no,” and the police are being racist in targeting that drug more heavily. Or perhaps the answer is “no,” but the police target the drug more heavily for non-race related reasons — e.g., because sellers of the drug tend to do so more openly, and so are easier to catch.

      Clarifying the causal mechanism is important because it tells us what questions we have to answer in order to see whether there’s racial discrimination going on or not.

      • Koken says:

        True, but Scott’s initial objective was to evaluate the level of racism in law enforcement. The point is simply that excluding this is at least plausibly excluding not a confounder to discovering the level of police racism, but some actual police racism.

  24. theLaplaceDemon says:

    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.

    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.

    • Deiseach says:

      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.

      Well, look at the tilting on here towards STEM subjects. How many are people working in fields of the humanities versus the ever-popular debates about AI, statistical analysis, and how to engineer a better-model human via chemical, biological or technological re-design?

      I definitely think I’m the stupidest person daring to comment on here, both from assumption re: raw IQ and the fact that I’m dyscalculic. I can’t say whether the audience here skews majority male or not; I have the impression it just about does, but I could be very wrong about that. I have no doubt that it skews majority STEM fields (with medicine being a ‘soft’ science) rather than arts/humanities.

    • Tracy W says:

      most measure to divorce the influence of parents’ socioeconomic status from their kids success would probably be seen as massively unfair by most.

      Also very costly. If A is a better surgeon than B because of A’s parents’ background, she’s still a better surgeon. Issac Newton and Charles Darwin were from wealthy backgrounds which made it far easier for them to make the discoveries they did, but that doesn’t mean that society didn’t benefit immensely from their work.

    • Scott Alexander says:

      Yes, I agree with that. It’s a great question to ask, just not a question to ask under the guise of “Why do feminist professors give more As to women?”

      I feel like the real world equivalent, “Why is Google disproportionately male?” or whatever, is very much the feminist-professor phrasing, and the structural-raising-people for-certain-interests framing would be more profitable.

      • llamathatducks says:

        I feel like the real world equivalent, “Why is Google disproportionately male?” or whatever, is very much the feminist-professor phrasing, and the structural-raising-people for-certain-interests framing would be more profitable.

        People often do argue that there are barriers to diversity within tech companies.

        Here’s one post that explicitly rejects the premise that the gender imbalance in engineering can be completely explained by the pipeline.

        And here is one black female engineer’s personal story of how the lack of diversity at Google has been harmful to her (including some obvious racism that was not addressed).

        I agree with you that [Google’s engineers are disproportionately male and white+Asian] –> [Google is sexist and racist] doesn’t work by itself. But people do flesh out that argument in valid ways.

  25. Fnord says:

    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.

    • gattsuru says:

      Mr. Alexander’s previous piece pointed that blacks *are* more likely to be stopped for non-speeding violations, that this effect persists even after reasonable controls (although some relevant ones like income do reduce the magnitude of the effect). That’s actually the summary of heading A in the previous thread, and the only defense police officers can come up with involves whether these stops find a reasonable amount of contraband. It’s actually the biggest certain source of racism-qua-racism demonstrable, closely followed by arrest rates for minor crimes.

      • Fnord says:

        It’s still a valid criticism of Scott’s point in this post. Klein’s point is not merely framing for heat rather than light; at least in some cases, the “control away the actual effect” problem can apply to Alexander-definition, “cops need diversity training”, racism too.

    • Also, at least some non-speeding offenses (like a missing light or lack of registration) could be the result of poverty.

    • Scott Alexander says:

      This seems like a pretty simple methodological problem. If the problem is that cops stop white people for speeding offenses but black people for drug offenses, then that’s your finding. If you didn’t think it was your finding, then you should probably do a study intentionally looking for that effect. Then you won’t have to call it “a thing you controlled for”

      • Fnord says:

        I’m not saying it’s impossible to do social science research, if you do it carefully. I’m saying that one of the things to be careful about is not controlling away the effect you’re looking for.

  26. Randy M says:

    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.

  27. eqdw says:

    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

    • Karmakin says:

      I suspect you’re right for a lot of people here, including me. Where we go wrong is that whole idea of “What can we do to fix the problem?” That’s not a very popular question to be honest. Because generally it starts to require sacrifice all around, and people start getting squeamish at this point.

    • llamathatducks says:

      This isn’t addressed at your main point, but how does one distinguish between “it’s actually that cops shoot poor people way more often, and black people tend to be poor in America relative to baseline” and “it’s actually that cops shoot black people more often, and black people are disproportionately poor”? If poverty and blackness travel together, how do you decide whether police officers are shooting people because they’re poor or because they’re black?

      • Alex Godofsky says:

        By looking at how often cops shoot poor white people vs. rich white people, and by looking at how often cops shoot poor black people vs. rich black people.

        • Matthew says:

          This is actually more complicated than it seems. It’s not clear how law enforcement would operationalize “discriminate against poor people;” poverty is not always obvious in the way that skin color is obvious. In practice I think it’s more likely to be “discriminate against poor neighborhoods.”

          But while most middle-income whites live in middle-class neighborhoods, most middle-income blacks still live in lower-income neighborhoods.

      • 27chaos says:

        One useful measure is what happens as black people’s (in)equality of income changes over time. Potential for confounders due to the causes for changes in income equality also altering shootings, of course.

        Probably there aren’t any perfect methods.

    • haishan says:

      Strategically, I think the correct thing to do (in some cases!) may be to say “Yes, I may be a racist, but not the bad kind of racist.” At least, this is what I do, admittedly usually pseudonymously.

    • Anthony says:

      Conservatives do this about government finance issues. They claim to want to reduce the budget deficit, and some are willing to actually propose real spending cuts, but mostly they’re not serious about it.

  28. social justice warlock says:

    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.

    • Vaniver says:

      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.

      I think this misses the point; racism is useful as a club because it’s so amorphous. See Scott’s post on the Non-Central Fallacy.

      • Ryan says:

        Wait Yvain is SA? No wonder every time someone brings up the Mote and Bailey technique I always think “isn’t this more the noncentral fallacy?”

          • ryan says:

            Elsewhere, especially on Less Wrong, I blog under the nickname “Yvain”. I mention this mainly because of the surprising number of people who read things in both places and never manage to figure out that we are the same person. Don’t be the person who tells me I should talk to that Yvain guy because we seem to have a lot in common.

            Bah ha ha ha ha. I love it.

        • Anonymous says:

          The World is a small place, there are actually much less people than you think there are 🙂
          Sometimes two people are actually one person 🙂
          Just look around you and what you’ll see might surprise you 😀

    • Thad says:

      I’ve seen some people use “white supremacy” where I thought the word “racist” was a much better fit. My guess is that changing terminology will just push the problem around, with side effect like confusion and acting as a new vector for signaling.

      • Susebron says:

        It could be useful to separate racial bigotry and structural racism, though.

        • Tarrou says:

          Sure! “Structural racism” is what is blamed when the data clearly shows no racism. It’s hidden racism, so cleverly devious and pervasive you can never see it, but trust me, it’s there.

          One can think of “Structural Racism” kind of like the Illuminati or International Jewry. Once you accept that it’s a thing and actually runs the world, it’s easy to see it everywhere!

          • Randy M says:

            As Scott pointed out in the last post, when the effects of racism diminish, it’s not “less racism”, it’s “more insidious racism.”

          • Susebron says:

            Structural racism is more about things like, for example, greater penalties for drug use when it comes to drugs which are more heavily used by black people. It’s also used for unconscious things like people favoring identical resumes with white names. It needs a new name, but it’s not the Illuminati.

          • Ockraz says:

            @Susebron – I don’t think usage is consistent enough for that to definitively be what it -is-. I’ve seen it used to mean persisting negative effects to the African American community that are caused by racism that occurred to prior generations.

          • Susebron says:

            Well, it’s certainly used badly, which is why we need new terms that are more clear. Lumping everything into “racism” only causes problems.

          • Illuminati Initiate says:

            Hey! The Illuminati runs the world for your benefit! How dare you compare us to racism!

            … (joking obviously)

      • stillnotking says:

        Seems like a straightforward appearance of the euphemism treadmill. People use “white supremacy” because “racism” has started to lose its emotional punch from overuse.

        It’s unintentionally hilarious at times — describing some random, slightly insensitive dude as a “white supremacist” is like describing the school bully as “Dark Lord of the Sith”.

    • Tarrou says:

      Sure! Just give me a definition for any of those and I can tell you whether it’s a thing or not.

  29. 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.

    • anodognosic says:

      I’m going to assume that fairness and concern for others is a terminal value.

      Race is a relevant category because it’s deeply grooved in society. It used to be that race determined your legal status in the United States, including, among other things, the possibility of being a slave. This difference was enshrined into law until the middle of the twentieth century (well, ostensibly equal separation, but you know how that worked out). White supremacy–people acting consciously to privilege white people over black people–is still a significant force in many places. And unconscious bias against black people is amply documented in social psychology.

      To borrow a legal term, all this makes black people a *suspect class*–a group that is likely to be discriminated against. It’s a group that has had consistent disparities throughout US history. Thus, even if you question the causal mechanism by which black people are disadvantaged, slicing up the population in this way is likely to yield significant explanatory power.

      • Explanatory power is a *positive* claim. I’m asking a *normative* question.

        You’ve brought up the positive fact that the law previously discriminated against blacks. What is the normative principle you can use to say that that fact implies contemporary blacks deserve some sort of compensatory treatment (to choose one example of a conclusion people draw that slices by race)?

        For example, a normative principle might be “sins of the grandfather pass to the grandson.” That principle is insufficient (slicing by race includes immigrants who’s ancestors were not in the US in racist times) but it’s getting closer.

    • Jadagul says:

      Because we have historical reasons to have concerns about this particular division.

      It’s pretty transparently clear that, say, 200 years ago black people were discriminated against in a really deep-going and pathological way. It’s reasonably clear that the same was true sixty years ago. Asking “to what extent is this still true?” seems pretty reasonable.

      In contrast, I have no reason to expect that there is prejudice/discrimination/structural bias against people with names beginning with B, L, R, and T. So if you generated a study showing that people whose names began with those letters performed worse than people whose names did not, that looks a lot like data mining.

      • llamathatducks says:

        This.

        And indeed racism is probably more likely than other kinds of bias (except classism) to persist economically, because people generally inherit wealth (if any) from people of their own race. Whereas women get their wealth from people of any gender, and LGBT people get their wealth from people of any sexual orientation/gender identity. (It’s still worthwhile to look at potential gaps along lines of gender, sexual orientation, and gender identity, because current discrimination/bias can play a role, but in the case of race it’s especially worthwhile because the past plays such a strong role.)

      • stillnotking says:

        But is it easier/fairer/better to dismantle discrimination based on slices, or the slices themselves? I’m old enough to remember a time when the rallying cry of the left was a post-racial society; these days, advocating a post-racial society will get you labeled a racist.

        We have actual, historical examples of the slices changing. In the not-so-distant past, “Irish” was a race in America. Now Irish people are just white. The same seems to be happening currently with Asian-Americans, especially East Asians — they’re becoming “white people with epicanthic folds” in much the same way Irish are “white people with funny names who sunburn easily”. I’m willing to bet that in another ten years or so, overt racism against Asians will be seen as not so much wrong as quaint, absurd even.

        I have no policy proposals around this, and I don’t think it’s possible or desirable to try to force people’s opinions by fiat, but I find it interesting.

      • eqdw says:

        I have an interesting edge case for you to consider then: What if we slice and dice humans by height?

        Women currently hold 5.2 percent of Fortune 500 CEO positions and 5.4 percent of Fortune 1000 CEO positions.

        Surveys have uncovered that less than 3% of CEOs were below 5 ft 7 in (1.70 m) in height. Ninety percent of CEOs are of above average height.[12]

        As a descriptive answer to Chris Stuccio’s question, I think you’re bang on: This one matters because of historical reasons. But I’m also sympathetic to an implication I’m reading in his post: There are certain other ways to partition humanity that show disproportionate impact, that the majority of society doesn’t care about. Are these not also important?

        Take this case: statistically, if the low rate of women CEOs of American companies is both a) evidence of discrimination; and b) normatively a moral problem in need of correcting, then the same logic should apply to shorter people. And yet, whenever I talk about being politically disadvantaged at work for being 5’5″, people look at me like I’m crazy.

        Thoughts?

        • Anonymous says:

          Or maybe women aren’t CEOs because they are short.

        • A dude says:

          Of course this is important. Ultimately I think that it is interesting to look at Status-indicating-attributes and determine as a whole their impact on outcomes. Height would certainly fall into that category, race, and gender too. An interesting study I am too lazy found that racial discrimination was significantly larger in jobs involving customer relations and almost nonexistant in noncustomer-facing jobs.

      • There is known prejudice/discrimination/structural bias against people who are short and ugly. I see no reason to believe this bias hasn’t happened historically. No one seems to care or even oppose this.

        Further, various other groups (Irish, Asians) were also discriminated against, yet slicing by race explicitly puts them into the “white” group.

        So this explanation both a) isn’t a normative principle and b) doesn’t work even if you assume some normative principle based on historical bias.

        • llamathatducks says:

          Further, various other groups (Irish, Asians) were also discriminated against, yet slicing by race explicitly puts them into the “white” group.

          First off, while (East) Asians may be moving towards whiteness, they’re not there yet. People around me routinely juxtapose “white” and “Asian” in ways that obviously refer to different groups (and I know a lot of white and Asian people).

          Second, I think it’s absolutely valid and important to ask question about how these groups are doing in our society. All of the following questions are interesting:
          (0) Can most people tell this group apart from other groups?
          (1) Are there currently stereotypes about these groups? If so, what are they?
          (2) What effects do these stereotypes (if any) have?
          (3) Can we observe discrimination against these groups? If so, how prevalent is it?
          (4) How are these groups doing in terms of wealth and other indicators of wellbeing?
          (5) If there are differences between wealth/health/etc. levels of these groups and others, why?
          (6) (especially in the case of Asians) Within this group, are there differences in wealth/wellbeing between descendants of immigrants from a long time ago (when overt discrimination was more prevalent) and more recent immigrants?

          My guess is that there is little to no remaining prejudice against Irish people in the U.S. (but it’s possible that descendants of Irish immigrants from a long time ago are still poorer than other white people) and that there is still prejudice against Asian people, but of a different sort than there used to be and of a different sort than black people encounter.

          Third, the answers to questions (0) and (1) matter a lot. (Btw, I only used zero-based numbering because I thought of question (0) last and didn’t want to renumber all the things…) Because black people are a recognizable group in the U.S. that has stereotypes attached to it, it’s extra-likely that we as a society don’t treat them the same way we treat non-racialized (white) people. (Same with Asians, but the stereotypes are different.) Therefore it’s important to examine that.

          Fourth, I think it’s important to note that while Irish and Asian people faced exclusion and discrimination, they nevertheless weren’t enslaved, and even post-slavery the nature of their oppression was different (and possibly less severe). This doesn’t mean that the weight of history of anti-Irish and anti-Asian racism isn’t worth studying and considering. It just means that there is extra weight of history in the case of black people.

          There is known prejudice/discrimination/structural bias against people who are short and ugly. I see no reason to believe this bias hasn’t happened historically. No one seems to care or even oppose this.

          I agree about prejudice and discrimination and disagree about structural bias. I think the prejudice and discrimination are obviously unfair, that it’s interesting and important to study them, and that it’s even worth including these things in the list of things people should know they might be biased about. (e.g. in an “unconscious bias” workplace training) But it’s also important that we have a history of explicitly, intentionally, systematically, and drastically oppressing/disadvantaging black people in a way that’s just not true of ugly and short people. The attractiveness bias has probably always existed, but it’s probably always existed in the sort of low-key, unacknowledged way it exists now, whereas racism used to be explicitly the law of the land. This makes potential racism a particularly important thing to examine.

    • Intrism says:

      You seem to be missing the key insight in what makes discriminating against a group X-ist or not. Specifically, whether affiliation with that group was neither chosen nor changeable. The rationale for this is that it’s not fair to judge people for such characteristics.

      Race is the “big one” in X-ism because race is not chosen (it is instead assigned by birth) and cannot be changed. National origin, gender, and ethnicity are protected classes for the same reason. It’s not entirely clear how sexual orientation is assigned, but it is clear that it’s neither chosen nor changeable, so that’s a protected class as well.

      Poverty, incidentally, is not considered such a class, at least not in the United States. While this isn’t actually true, it’s still a generally accepted belief in American politics that, if one is willing to work hard, one can easily lift oneself out of poverty. Therefore, it is changeable.

      The only major exception to this rule is religion. It’s not generally chosen (like race, religion is usually assigned by birth), but conversion is generally very easy (well, providing you’re not Islamic in the Middle East). It’s treated, however, more in line with race. This probably has a lot to do with history.

      (And then, of course, things like “are you a racist” or “what political party do you support” or “what football team do you support” are both chosen and changeable, so people are accordingly free to discriminate on those categories.)

      • anon1234 says:

        You haven’t answered his question. Poverty, ugliness and a potential infinity of other things you’re arbitrarily choosing not to divide humanity by are all heretible. Why the groups you have chosen?

        • Intrism says:

          I covered poverty. Clearly you didn’t bother to actually read my comment, I’m not sure why I should bother responding…

          Eh, whatever. Ugliness is largely a result of how much effort one puts into one’s appearance, therefore “changeable,” but when it isn’t people are indeed concerned about that. (see: “fat-shaming.”)

          The potential infinity of other things either don’t fit the “unchosen and unchangeable” criterion, don’t have any significant amount of discrimination associated with them, or are already considered protected classes.

          • Anonymous says:

            I don’t understand, you claim ugliness is mostly changeable and cite weight as an exception?

    • Scott Alexander says:

      Er, I might be misunderstanding you, but I would expect most slices to either have no significant differences (ie people whose names begin with “A” vs. “M”), many slices to have significant differences that turn out to be proxies for other important things (ie favorite TV show being a proxy for class or race), a few slices to have tautological or obviously fair differences (ie poor people have less money than rich people, or lazy people get less money than hard-working people) and a few slices to have very significant differences that are not otherwise explainable (ie race).

      I would expect most people to consistently say disparities in this last group should be addressed, although of course some people might place more emphasis and activism power on one than another.

      I agree that some disparities get less attention than others – for example height vs. race. I would argue that the height disparity is less strong than the race disparity, that it tends not to concentrate and feed in on itself because short people don’t all live in the same areas and only know other short people, and it has less historical baggage behind it. I think these are good reasons it gets less attention – but I also think most people would agree in principle that height disparities should be addressed.

      I might be confused, though. Can you give an example of what you mean?

      • Contrarian says:

        “a few slices to have tautological or obviously fair differences (ie poor people have less money than rich people, or lazy people get less money than hard-working people)”

        Is it even true that lazy people get less money than hard-working people? I’m really skeptical of that. I guess if you control for a tremendous number of variables, then you’re likely to find a significant correlation. Still, putting that up as an example of a fajr difference next to your example of a tautology (which is unambiguously tautological), seems like a bad example. I’d say that’s especially true given that one of the two things examined in your essay is how to understand ‘women get less money than men.’

        Maybe I’m being nitpicky. (?)

      • anon1234 says:

        ” obviously fair differences”

        Why obviously fair? Poor people, stupid people, short men, fat people etc. didn’t choose their genes. Just like with race or gender.

        “I agree that some disparities get less attention than others – for example height vs. race. I would argue that the height disparity is less strong than the race disparity,”

        How are you coming to this conclusion? What if we split up the height group into a binary of the bottom 20% of men vs everyone else? I bet this difference will be huge and is just as valid as the split between “men” vs “women” or between the races. Easily noticeable and measurable too. Why not the bottom 10% instead? Another arbitrary objectively measurable grouping like gender and race where you will find big inequalities. What about ugly AND fat people? etc.

        Anyway it seems Chris Stucchio is trying to ask for some objective reason to accept some normative claims like “you should strive for racial pay equality” or whatever. You respond with reasons like historical baggage etc. This is obviously just a nonsensical discussion because morality is subjective.

      • Azathoth123 says:

        > Er, I might be misunderstanding you, but I would expect most slices to either have no significant differences (ie people whose names begin with “A” vs. “M”), many slices to have significant differences that turn out to be proxies for other important things (ie favorite TV show being a proxy for class or race),

        Not quite. You’re not imagining what the typical slice looks like. A slice is simply a (very long) list of people. Most slices, like you said have no significant difference. Among the ones that do have significant differences most aren’t “black” or “ugly”. Most are simply an arbitrary list of about 3 billion people most of whom happen to be doing worse then average.

        The question is why is something like “black” any different from one of these arbitrary lists?

        • Scott Alexander says:

          First of all, I think you’re wrong. The chance that two randomly selected lists of 3 billion people each show the same disparity as two specifically selected lists of black and white people is probably less than one in 10^10^10^… If you disagree, I can try to justify this statistically, but basically think of how low the p value for the null hypothesis would be on the 3 billion person “sample size” study of whether race affects income or whatever.

          That suggests that this isn’t a coincidence and something is causing the disparity. If the thing causing the disparity is bad, we should figure out what it is and remove it.

          (for example, most pairs of individuals will be of unequal health and happiness. If I punch you in the face and break your nose, that just means that one of a billion billion pairs has changed its inequality level. But I should still stop punching people in the face)

          • Anonymous says:

            2 lists of 3 billion are the same list with probability 1 in 10^10^10, not 10^10^10^… The probability that 2 lists are similar is much higher.

    • Your literal ‘why race?’ question is easy to answer re: why – because people tend to identify as black or white, whether or not there’s any difference in population statistics beyond that. Perhaps it could be less so, but for now, it is. And it hurts most (I speculate) to imagine you’re persecuted because of something you visibly are (and think of yourself as), than because of something that only shows when you choose to show it. An even more immediate “why” is because people are especially sensitive about those classes by habit and battle lines already drawn (but I gather you were asking why should it be so, in which case refer to my first answer: people don’t consider themselves [permanently] poor, but they do black/white/etc. Identity.)

      • Matthew says:

        I suspect the causality here is backwards. (Some) people see race as a very salient aspect of their identity because racism has been such a pervasive influence in society in the past.

        • Richard Gadsden says:

          Which came first, racism or racial identity?

          There seems to be a pretty solid case that the answer is racism.

          That is, having an easy physiological marker for “slave” made plantations much easier to operate.

    • 27chaos says:

      Evenly dispersing bad events is better than having them targeted at a particular area, in many situations. Probably almost all of them, in social science. The black people slice is an arbitrary target, but not a random one. It’s a Schelling point and they’re hit with injustices over and over again.

  30. Flavio says:

    Ezra Klein mentioned this blog? Sorry, but I don’t think SSC is cool anymore.

  31. Anonymous says:

    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.

    • Jaskologist says:

      If we have to be constantly vigilant no matter what the data says, why bother looking at the data at all?

      Isn’t the whole idea behind Bayesian reasoning that when data comes back saying there’s not much evidence of discrimination, you should adjust your view that policing is “ultimate system 1” or “quite/uniquely susceptible to racial implicit bias” or something like that?

      • Or that exactly the right amount of don’t-get-caught-being-racist paranoia is keeping the two forces roughly in check – see the tendency of many white cops to look for non-black collars, which on average balances out the “look where the crime is” intuitive profilers (and the higher % of black arrests by black cops etc)

        • Richard Gadsden says:

          Connectedly, there’s been quite a bit of research into the causes of political candidates being mostly male in the UK.

          The evidence is that women who put their names forward have approximately the same probability of being chosen by their party as men who put their names forward, but that, at each stage in the process, a smaller proportion of those putting their names forward are women. This is a classic pipeline problem (cf other postings about the STEM pipeline).

          The hypothesised mechanism is that there are a small proportion of sexists and a similar proportion of anti-sexists, but the sexists tend to make female candidates’ (or potential candidates) experiences unpleasant, while the anti-sexists vote, or otherwise influence the process in favour of the women, counterbalancing the sexists, but do not counterbalance the unpleasantness of the experience for the women (in part because of biases in favour of remembering unpleasant experiences, and in part because they are more likely to operate behind the scenes, and in part because “I’m voting for you because you’re a woman” is not a particularly pleasant or enheartening experience for a female candidate).

          This suggests that there are two possible solutions; one is for anti-sexists to be deliberately hostile to male candidates and make their experiences crappy – perhaps we could get some tumblr SJWs to help with that? The alternative is some affirmative action so that the women who don’t drop out are able to take up the places that should have gone to those who did drop out.

          If people are experiencing bias but there is no statistical evidence, then the hypothesis that there are biased people who are visible and counter-biased people who are invisible seems like a sensible place to go rather than telling people that their lived experience is invalid.

  32. Anonymous says:

    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 🙁

    • Anonymous says:

      Also, whatever happened to the edit button?

      • Vaniver says:

        I suspect it’s disabled if comment screening is enabled, which may or may not be intentional. It has the benefit that you can’t post something innocuous, and then edit it to something malign once it’s been approved.

      • Nornagest says:

        The report button seems to be gone too. I think something broke Scott’s CSS magic — although the comment counter still seems to be working fine.

        • Anonymous says:

          Edit and Report use server resources. He probably removed the plugins because the server was falling over under load yesterday. Whereas the comment history (and any CSS) is client-side. Actually, the real difference is probably that the comment history isn’t a wordpress plugin, so Scott doesn’t know how to remove it.

        • Jiro says:

          One of my comments which already was moderated as acceptable seems to have gone missing (and I am puzzled why my other comment was not moderated as acceptable, and I suspect it got lost too.) Something is odd.

    • Scott Alexander says:

      Right now there’s no way to do that. The WordPress forums specifically say that there isn’t. If someone knows of a way or wants to make a way, I’ll do it.

      A lot of features were turned off yesterday to help the server deal with a big bump of traffic from Reddit. They should all be back on now.

  33. Ryan says:

    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.

    • Randy M says:

      I like your comment until the end, which implies that it is impossible to honestly determine the relative weights of various parts of the process by which a difference appears, or unimportant to do so. Maybe the former is right, but I don’t agree that the latter is.

      • Ryan says:

        Yeah actually reading the last part again I think it’s wrong.

        If I might retreat back to my bailey, when one side says we’re controlling for hours worked and the other side says you’re controlling for sexism in action, I think there’s not an objective correct answer, that we’re simply seeing how two competing world views see the situation.

        But as you say that shouldn’t stop anyone from trying to honestly determine the relative weights of various parts of the process by which a difference appears.

    • Jos says:

      There are a bunch of interesting ways to slice it.

      1) One is, once you have identified the mechanism for a disparate result, do you still think the disparate result is unfair at all? For example, if it turns out that 2% of people with hair commit crimes at hair care salons and only 1% of bald people do, an otherwise undiscriminatory policy of convicting guilty people is going to have a disparate impact on hirsute people. We might look at that and say that while we oppose discrimination in the general case, in this specific case, it’s appropriate.

      2) Another is, once you have identified a mechanism for a disparate result, is it still appropriate to address the result at the end of the process? If we learn that due to a since corrected chemical impurity in hair gel, 5% of hirsute people are now irretreivably criminally insane versus only 1% of bald people, we can expect to see a dispirate number of hirsute people arrested/jailed/committed, etc. On the other hand, now that we understand the mechanism, I think most people would think it’s not appropriate to let some guilty hirsute people go free or convict some innocent bald people just to even the odds.

      3) A third is whether identifying the mechanism may be more helpful for achieving our social goals. It might be that it does more good to treat the mechanism, or that it’s easier to convince people to treat the committed humanely if you say “they can’t help it, and for many of them, it was the hair gel” instead of “a system of convicting guilty people discriminates against the hirsute.”

      • Ryan says:

        I’m not sure how on topic this is, but I guess it relates to your hair gel analogy. Criminal defendants in capital cases especially are starting to use genetic predisposition to crime offensively as a sort of excuse for their behavior. “Look your honor, my client’s brain simply doesn’t break down neurotransmitters fast enough to effectively control compulsive behavior.” I don’t think they have much traction yet, but it raises an interesting moral conundrum about personal responsibility.

    • Emily says:

      It’s not that I don’t think women being primary caregivers is the result of social forces, it’s that I don’t care whether it is or isn’t for the purposes of policy. “Is the result of social forces” is just not a meaningful category. Social forces can be impossible to overcome, or we might not want to overcome them because they have positive effects. And there are innate differences that absolutely we should fight and can be fought effectively (like, poor eyesight with glasses.) In this example, women being the primary caregivers seems pretty intractable, regardless of where it’s coming from. What we’d have to do in order to change that would be monumental and not worth it. That might change in the future: maybe we’ll get exceptional child-taking-care-of robots. (Also, your good sociologist, regardless of their values, is going to present multiple models where they condition on various different things in various different ways.)

      • Ryan says:

        Maybe call that the gray tribe interpretation? Don’t discriminate between upstream and downstream solutions? I personally suspect the leading cause of myopia is excessive reading in childhood (I have supporting data if it means the plural of anecdote). Assuming I’m right, crafting glasses for the myopic is a much better idea than slapping books out of the hands of children.

        If we think it’s a problem that women earn X% less money because of various consequences of being principle child caregivers, why do we seem to immediately shut out solutions like cut their taxes to the point that take home pay is the same?

        • Matthew says:

          I personally suspect the leading cause of myopia is excessive reading in childhood

          I’m pretty sure I’ve seen this reported as fact in a phenotype = genotype + environment article somewhere. (Nearsightedness should be strongly selected against in proto-humans and hunter-gatherers, yet clearly has a significant heritable component and is widespread. Literacy appears to be the novel environmental stimulus.)

        • Will says:

          100% of the people I know with 20/20 vision were voracious readers in childhood.

          • Matthew says:

            You need the myopia gene(s) and voracious reading to actually end up with myopia.

          • haishan says:

            One of my parents is nearsighted and the other has astigmatism; I read like a maniac as a child; I had 20/20 vision throughout my childhood and up until my cornea started doing weird things a couple of years ago. I can only conclude that phenotypes are weird black magic that make no sense.

          • Richard Gadsden says:

            “phenotypes are weird black magic that make no sense” seems to be a remarkably accurate summary of a lot of complex research.

            So far.

          • Wulfrickson says:

            These people hypothesize that increases in myopia are related to lack of exposure to sunlight, but not to time spent on reading or other near work.

      • Jos says:

        IMHO, what portion of the pay gap is attributible to what issues is informative to (a) do we feel obligated to correct the disparity and (b) if so, what is the preferred method of addressing it?

        a) Should We Address It: If it turns out that due to cultural forces, men prefer opera and women prefer watching football, and opera costs more to produce per viewer/hour and therefore women get fewer hours of entertainment per dollar, I don’t personally see a societal imperative either to adjust the relative consumer cost of opera and football or try to address the social forces that cause this disparity. On the other hand, if it turns out that due to social forces, men are socialized not to make eye contact when speaking, and that the dominant culture views that as sign of under confidence and dishonesty, which is holding them back from full participation in society, then I’d like to address that.

        b) How Can We Effectively Address It: Again, knowing how the disparity arose gives us some information about how to address it. (Can we educate both women and men about eye contact so they meet in the middle somewhere? Can we educate one group or the other? Can we just require that companies promote representative numbers of men and women? What would be the benefits and costs of each proposal? Etc.)

    • Nornagest says:

      That sounds a little glib to me. From where I’m standing, it looks more like both sides are happy to assign all sorts of powers to socialization iff it happens to point toward policy goals they like. A rightist might place most of that socialization pressure in the hands of traditional families and therefore decry gay marriage and no-fault divorce; a leftist might assign it to “microaggressions” and therefore go to great lengths to police discriminatory speech and behavior. Either, if we’re having that moral panic today, might attempt to censor violent or sexual media.

      • Ryan says:

        I think you’ve made a very good point. Maybe I just way overcomplicated things and the actual difference is the blue tribe sees equal income for men and women as a morally important outcome and the red tribe simply doesn’t?

        • John Schilling says:

          More generally, blue tribe sees outcomes as important whereas red tribe values opportunities.

          In this case, if the opportunities, “Earn $100K/yr via total commitment” and, “Earn $50K/yr with plenty of time left over to spend with the kids”, are equally available to both genders, red tribe is satisfied. Blue tribe is only satisfied if both genders make the same choice.

          The idiot segment of blue tribe seems to believe this can be accomplished by making the options, “$75K plus time for the kids” and “$100K minus an extra $25K in taxes”.

          • Ryan says:

            I have to wonder if the extra $25k in taxes imposes a lower overall cost than the regulatory burdens of discrimination lawsuits, diversity officers in every HR department, etc.

          • Luke Somers says:

            Hrrmmmm, no, I don’t think that you’ve characterized the blue tribe well, here.

    • Anonymous says:

      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.

      I think many red tribesmen would admit that this disparity is due to social forces, but would likely claim some combination of the following:

      1) Social forces did not arise in a vacuum or via conspiracy. They may have resulted from a very long timescale negotiation between abstracted groups.

      2) Computing the ‘actual worth’ of the tradeoffs involved is an impossible task. How much do people actually value the ability to stay home with their children? Is this ‘worth it’ when compared to some of the tradeoffs?

      3) Social forces may have arisen from biological facts, anyway. Given the chemical cocktail floating around in a brain due to pregnancy/childbirth, the ‘actual worth’ of staying at home with the children may strongly correlate with gender.

      4) Given (1)-(3), it may be Very Hard (TM) or even unintentionally counterproductive to mount a campaign against all social forces which lead to an identifiable disparity in some particular measure… even if these social forces are as real as my receding hairline.

      I feel doubly strongly about this, because I keep hearing red tribesmen saying (and getting mocked for saying) that the cause of such-and-such disparity is “culture”. That is social forces, even if they have no idea how to go about ‘fixing’ them… because again, they think it’s Very Hard. Many red tribesmen seem to care a lot about social forces and how they can support the norms that they care about.

      • Ryan says:

        “Given (1)-(3), it may be Very Hard (TM) or even unintentionally counterproductive to mount a campaign against all social forces which lead to an identifiable disparity in some particular measure… even if these social forces are as real as my receding hairline.”

        I think that’s a really good point. I wonder if the most cynical of the blue tribe see this and think “Excelent! Permanent Revolution!”

    • llamathatducks says:

      I am, I suppose, a “blue tribe” person, and I’m primarily concerned with equality of opportunity rather than equality of outcome. It’s just that I define “opportunity” differently than most “red tribe” people do.

      In the case of the gender wage gap, I think it hurts equality of opportunity that we have a societal expectation that (a) each family has a caregiver and a breadwinner (b) the breadwinner doesn’t need to worry about caregiving much, so they have to work super hard, and (c) typically the breadwinner is male and the caregiver is female.

      Because jobs are typically designed in a way that doesn’t let working people also do a lot of meaningful parenting work, and we give maternity leave not very much but still more than paternity leave (which perpetuates the gendered expectations), and we don’t provide great support for working parents in the form of accessible and good childcare. These factors take away opportunity from people who want to both parent well and work well.

      • Bugmaster says:

        While I generally agree with you, I am compelled to point out that the need for maternity leave is only partially social (breadwinner vs caregiver). The other part is purely biological, since, given our current level of technology, humans are a sexually dimorphic species who reproduce by live birth. Thus, I would argue that women (or, more precisely, female humans) do have a greater need for maternity leave, since there are many tasks they (unlike male humans) are physically unable to perform shortly before as well as shortly after giving birth.

        • llamathatducks says:

          Yes, that’s true. However, there’s also a social component. The parent who gave birth needs to physically recuperate for a time, but the baby needs someone to be around constantly for a longer time. Historically, the way to deal with this has been for the baby’s mother to quit her job (if she had one) and just take care of the baby for as long as needed. Now, if a mother’s job provides reasonable maternity leave, she might exhaust that and then have to face the choice of going back to work and leaving the baby to some kind of carer (which could be scary with a really young kid) or quitting her job. But another option (assuming a male-parent-and-female-parent household) is for the father to take additional leave to take care of the kid after the mother has rested. (This is also the most fair in terms of eliminating incentives not to hire women who might get pregnant.)

          And given different physical needs for people who get pregnant and those who don’t, lack of maternity leave impedes equality of opportunity too.

          • John Schilling says:

            Insofar as there are heavy transaction costs in taking substantial time off most jobs, to both employer and employee, the economically efficient solution is for the mother to cover both child-birth and child-raising, thus not double-paying the transaction costs.

            If you try to impose an economically inefficient solution in the name of fairness, expect the economy to push back. And expect a whole lot of people, even people who would benefit from the fairer solution, to not want to be collateral damage in that fight and just quietly go along with the more efficient solution. I think if your metric for success is, “Men handle 50+% of the child-raising, thus everyone works equal hours for equal pay”, you’re going to be waiting at least until the uterine replicator and probably a generation beyond.

      • Ryan says:

        The normal response I see to this line of argument is that it goes against some foundational notions of economics, in particular that specialization and division of labor are efficient. The emergence of the caregiver and breadwinner roles should be seen as rational actors maximizing utility.

        That doesn’t explain why the usual situation is women as caregivers and men as breadwinners. The simple explanation is women have a uterus and lactate. I’ve always found that a perfectly satisfactory explanation, never seen the need to speculate about effects of societal expectations.

  34. Ryan says:

    In regard to the issue of controlled variable actually hiding the affects, prior record is one that continues to bother me. Assuming a “racist police” world where whites are systematically under-policed and blacks over-policed, we’d expect a given arrested black person more likely to have a prior record than an arrested white person even if they’ve committed an equal number and severity of crimes. Therefore in this world, a regression that controls for prior record would seem to wash out the over-arresting of blacks. Or am I missing something about how regression works?

    • That’s true, and the identification of “real criminal activity” can be similarly impugned (that’s why we love stats about killings – they’re pretty much all reported and often solved). But over time your hypothesized circularity will decay out of the system like a convergent geometric series. I suppose your theory is that the true difference (which must exist) is only in past bias, and it just hasn’t decayed yet.

    • Richard says:

      I understood it to mean that control for prior records were only applied to sentence length where it has a direct effect because of mandatory sentencing and three strikes and whatnot (my understanding of the exact principles is rather vague)

      • Anonymous says:

        There are a lot of studies that control for a lot of things. The very first study Scott mentioned in his original post found that blacks’ cars were searched 2x as often as whites, but when you control for a bunch of things, including prior record, it drops to 1.5x.

  35. Ilya Shpitser says:

    [ meta: is there a way to get the software to allow edits in comments awaiting moderation? Usually by the time a comment makes it, and I check back, it is past the edit timeout. ]

  36. John Schilling says:

    A related issue may be the arbitrariness of “root cause”. This comes up a lot in my work because when e.g. a rocket fails, the immediate cause (part X failed) may be interesting but useless (the whole rocket is now junk at the bottom of the ocean) whereas the root cause may be something we can fix to prevent problems in the future. And it is always expressed as THE root cause, singular, because that’s how monkey brains work when it comes to this sort of thing and it’s usually not worth the trouble to fight it.

    So (and this example is really not related to any exploding rocket you have read about in the news lately):

    The rocket failed because the fuel/oxidizer mixture ratio was wrong.

    The fuel/oxidizer mixture ratio was wrong because one of the lines was blocked.

    The line was blocked because someone put a plug in it during a repair and forgot to remove it.

    The technician forgot to remove the plug because there wasn’t a “sponge count” in the checklist for the repair procedure.

    There wasn’t a sponge count in the procedure because the people writing the procedure didn’t understand what was so important about it.

    …because the repair work was done at a facility that was excluded from both the design and the operation of the rocket, which is where that information comes from.

    …because the contractor’s management determined that consolidating that sort of work would save costs and didn’t adequately consider risks

    …[skip a dozen steps] because capitalism.

    …[skip a hundred steps] because Let There Be Light, or maybe Original Sin.

    So, which one of these is the singular Root Cause of the rocket not working? Where do we apply corrective action?

    The answer, of course, is that root cause is wherever you want it to be. Maybe just because you’re tired of asking “But why…?” like a hyperinquisitive five-year-old and this is as good a place as any to stop, but maybe because you’ve found something you really feel good about blaming. In rocketry, lots of people really like to blame FOD (Foreign Objects & Debris), mostly because it’s easy and unfalsifiable and the corrective action is just “try to be a bit cleaner”. And an awful lot of causal chains go through a step that includes some piece of crap where it shouldn’t be, so root cause is FOD and we can go home now.

    With social problems, an awful lot of causal chains go through race or something correlated to race if you follow them far enough, and an awful lot of people seem to really want the root cause to be racism.

    • Anonymous says:

      I think this is a hilarious and relevant expansion of Scott’s joke about gravity (…both of which I’m very partial to as a ‘rocket scientist’, myself). Picking a useful level of abstraction is a hard thing that often involves assumptions and flat out being wrong. It’s also very closely linked to what it is you’re trying to do… and what you’re capable of doing.

      If you’re just a theoretical control guy who occasionally programs a controller for fuel delivery, then you just chalk up the failure to an idiot with a wrench. Maybe you take the ambition to develop a fault-tolerant controller that could detect this type of issue.

      If you’re a program manager trying to make sure your delivered rockets hit X% success rate, and you see a surprising number of similar failures coming from a facility, you possibly change facilities or demand more information concerning possible solutions to the failure rate.

      If you’re the President… or voting for the President… or participating in large-scale politics… perhaps you vote against capitalism.

      If you’re God, you take back the dust.

      Sometimes, you can jump categories (I can develop a model-predictive controller and vote for president!), but it’s always dependent upon ‘what am I doing right now?’ Am I just trying to do cool work and keep my job… or am I in line at the polling both? Am I doing something that I know I’m an expert in (control theory) or am I relying on my understanding of some other expert’s choice of abstraction (capitalism)?

    • Anonymous says:

      I find this reasoning particularly amusing since racism is both demonstrably historically powerful (thus it almost has be a partial cause of many current social phenomena) and perhaps the mother of all confirmation biases, greater than any anti-racist excess imaginable. It seems very likely that throughout most of post-Columbian, western history the group of people desiring the cause of social ills to be blacks or immigrants with weird languages, or jews has been the majority and the most politically powerful. It is entirely possible that anti-racists suffer from a serious confirmation bias problem, but if you think there’s more value in tackling them over the almost certainly larger biases of racists, I can only presume you spend much less time among elderly and evangelical whites (ie the most powerful voting block in the country) than I do.

      • Personally, I fear unchecked anti-racism/sexism. That there are some old people who haven’t been reformed doesn’t worry me; they’re on the way out, though I’m definitely as afraid of sincere religious conservatism as anything and am happy to point out their poor scientific methodology if anyone feels that point has been insufficiently driven.

        I don’t think it’s wise or necessary to befriend every enemy of my enemy.

      • John Schilling says:

        It isn’t just a matter of which bias is larger, but of which bias is more tractable and which bias is causing more harm.

        The racist bias of elderly and evangelical whites is intractable save by the passage of time, and I do not think it causes significant harm in the specific area under discussion – it isn’t elderly evangelical whites who are arresting young black men in poor black neighborhoods, and it mostly isn’t elderly evangelical whites who are setting up police patrol patterns and the like. So, for this purpose, I’m not going to waste my time on it.

        The racist confirmation bias of young white police officers, and the anti-racist confirmation bias of liberal politicians and political activists, those might be easier to recitify. And almost certainly more influential in this particular context, so that’s where I would like to focus.

        But which one is the dominant effect, is not obvious. That’s where work like Scott’s is so valuable, and I’m glad he’s trying to tackle it because just getting rocket-builders to do sponge counts is aggravating enough for me.

    • Bugmaster says:

      I agree; instead of asking, “what is the root cause of this rocket failure ?”, we should be asking, “what is the most effective way we can prevent rocket failure in the future ?”. As it turns out “try to be a bit cleaner” is a cheap but ineffective solution; “add sponges to the checklist” is more expensive but vastly more effective; and “overthrow capitalism” is so vastly expensive that it’s not even worth considering. So, once you’ve followed the chain of cause and effect to that point, it’s time to stop and look at the proposals you’ve got so far, then pick the best one.

      This approach probably works for rockets; I know that it more or less works for software engineering; but, from what I’ve seen, it can never work for social issues because they are driven by emotion, not pragmatism. But perhaps I’m being too pessimistic…

  37. Emily says:

    Since your blog is getting so much attention, I’d suggest thinking about what would happen if someone with a big platform said extremely critical (and perhaps race-related) things about you and that person (or their followers) connected it with your real name and tried to get you fired, to what extent you can/wish to avoid this by avoiding saying certain things, and whether there are things you can do in your life now to insulate yourself against the consequences of someone doing that. Maybe this is dumb and you’ve already thought about this, I don’t know. But it’s easy to forget there’s a larger context out there and that it might notice you and try to stomp on you. No one has publicly revealed your real name yet because, even though it’s easy to find, everyone who cares to find it likes you. But you can’t count on that lasting.

    • Matt says:

      TL;DR

      Tread lightly with the SJWs.

    • Multiheaded says:

      You’d be surprised as to how little negativity there actually is towards Scott in SJW spaces. I’m not implying that the people there are really fair to him – but “privilege-blind” and “deliberately obtuse” are about the harshest things said.

      • Emily says:

        I’m glad for that. However:
        1. That a group is aware of your deal and basically-apparently-ok with it doesn’t mean that tomorrow they won’t turn around and invoke it as they try to get you fired.
        2. It’s not just or even primarily the SJW people you have to worry about, it’s the gawker piece that misrepresents your views and calls you names.

  38. I’m having a hard time placing Ezra Klein on my tribal spectrum. I knew he read SSC when his essay on reproductively viable entities lifted a lot of examples and narrative from your original posts on politicization. It smelled a lot like a an attempt to preemptively capture subcultural attention of our beloved Gray Tribe.

    Is he a misunderstood pseudorationalist, or the root of all evil? The more I read of him, the more it seems like he has legitimately few moral opinions and is simply an extremely polished and well-versed sociopath type.

    • Karmakin says:

      Ezra Klein is a policy wonk in a world that couldn’t care less about policy wonks. So he doesn’t do as much of it anymore as he used to. His policy wonkdom is really good. His political/cultural wonkdom is really bad.

    • Scott Alexander says:

      I’m totally not getting the sociopath thing. He seems nice and in the 95th percentile of good political writers (I know, I know, damning with faint praise). He’s also taken some genuinely unpopular positions, which is something I would expect a manipulator to avoid.

      • I like his writing a lot. I think he really pissed me off in that linked article about the “polarization of everything” — he runs a freaking Vox Media property and writes things like

        Because for one in five women to report an attempted or completed sexual assault means that everyday sexual practices on college campuses need to be upended, and men need to feel a cold spike of fear when they begin a sexual encounter.

        I don’t expect him to agree with him on everything, of course, but I get a strong entryist/apostate reaction when he starts adopting “Gray Tribe” narratives with one hand, while engaging in the exact same polarizing media shittiness with his other. I think Clark from Popehat wrote an article on this, but I’ve invoked enough of the reproductive ant gods for one day.

        This is pure tribal policing right there, and I don’t want to make a habit of it. Perhaps this is my own moral failing.

        • The last thing any man intending to use a condom needs is a spike of fear. Ezra clearly does not use condoms.

        • Whateverfor says:

          Did you read his old blog before he moved to Vox, or his initial statements about what Vox was supposed to be?

          Klein himself is a relatively middle of the road liberal pragmatist who cares more about the nuts and bolts of policy than tribal warfare. He tried to create a new website in this image, but it utterly failed. Packing up shop would hurt too much though, so he created the new Vox that we know and (don’t) love instead.

          • nydwracu says:

            Right. You can run quality posts, but they won’t get clicks, so you need to throw your readers meat in order to get the clicks that can subsidize the quality posts.

            Garbage media outlets like Vox and Gawker do this by throwing the readers meat that appeals to their monkey brains. Thede good, elthede bad! Slightly-less-garbage media outlets like Buzzfeed do this by running ads that look like articles.

            The best media outlets are the ones that subsidize themselves by running not-quite-porn or celebrity shots, like the Daily Caller. But running not-quite-porn or celebrity shots gives the outlet a serious status hit in the eyes of Brahmins, and running vile Streicherite “it is IMPORTANT!!!! to hate the outgroup!!!!” trash does not, and the people who run these things are themselves Brahmins and don’t want to look un-Preftigious in the eyes of their own caste.

            For that matter, what’s the most respected newspaper out there? The Christian Science Monitor. How much money per year do you think that loses?

          • Multiheaded says:

            The Daily Caller appears to be a mix of thoughtful right-libertarian opinion pieces, and… utterly content-free “Obama this, Obama that” garbage? Are you sure that it is not, in fact, the Obama-focused trash that’s driving their clicks?

            And yes, the very idea of Christian Science Monitor having long become a highly respected publication is hilariously Phildickian. I like America a little bit more after having read that.

        • CaptainBooshi says:

          I remember reading that post about college campuses and thinking that it was a pretty clear-cut case of utilitarianism, which is a pretty clear ‘Grey Tribe’ thing. He said that he believes the situation is so bad right now that even if the new rules cause significantly more suffering on the part of men starting sexual encounters, the overall amount of suffering will go so far down that it will be an overall positive. While you may disagree with this, you can’t fairly say it just “pure tribal policing.”

          While I don’t read enough of his stuff to know if what you said is true in general, the specific case you gave is definitely not a good example.

          • Nornagest says:

            You could cast it in utilitarian terms, but the argument as stated is not, in fact, utilitarian.

    • Jaskologist says:

      The dude started Journolist, a special listserv for left-wing (and only left-wing) journalists to come together and discuss (some would say coordinate on) the stories of the day. His tribe should not be hard to identify.

    • Matt says:

      I completely concur. I get that same vibe. Much of his stuff strikes me as deeply cynical.

  39. Steven Herron says:

    Thank you Scott for an excellent treatise on this subject. One that is devoid of emotion, very difficult to accomplish given the topic, filled with substantiating data and quality links. I fear there are those who rely on the ability to stir emotions on this issue for their own economic prosperity rather than objectively discussing it. We will never become a more harmonious society until we choose to debate honestly and logically on any of the issues that strain our society. Sadly our media have a stake in keeping the friction in place.

  40. cheeseandbeans says:

    I love your writing, your posts are always well argued. You are making me question a lot of my beliefs. This is actually somewhat scary for me, because in my social circle I would be ostracized for even bringing up your arguments about race or the wage gap.

    • Anonymous says:

      Inquisitions and heretics are both alive and well in the 21st century, especially since we decided to merely shame, discredit and get the heretics fired from their jobs instead of burning them.

    • Anonymous says:

      You’ll have to learn to talk to your friends in a roundabout way. Do not bring race and gender immediately. Do not claim to be on one side or the other. Simply present things as “I have heard about an interesting study…”. Then your friends might be a little bit less likely to ostracize you. But I think that you’ll have to talk to them. The world where everyone fears to say their honest opinion is much worse place than the world were the statistical average wage of one group of people is a tiny bit smaller than the statistival average wage of another group (individual differences dwarf group differences anyway).

    • Matt says:

      I really don’t mean to be a jerk when I say this: you need a new social circle. That type will find a reason to ostracize you eventually, if you have any thoughts of your own.

      • Anonymous says:

        Better yet, do not keep all your friends in one circle, even if you get ostracized in one circle, you will still have others. I think it is not a good thing to leave one’s social circle entirely as it contributes to creating more and more echo chambers.

  41. Andrew says:

    Eliezer wrote something like this on facebook. There does seem to be a policy proposal to get police nationwide to wear body cameras. It’s getting national attention from the president and congress. You might think it’s a useless policy, you might think it isn’t enough, or you might think it’s attacking the wrong cause, but I find it hard to take the claim seriously that there are no policy proposals coming out of the events in Furguson.

    • Troy says:

      Requiring police to wear cameras is a reform that should be supported by progressives and conservatives alike. Steve Sailer even supports it: http://www.unz.com/isteve/nixon-taped-advocating-cover-up/

    • Not THAT anonymous commenter says:

      The teenage juvenile delinquents (no hyperbole) I teach came up with another policy proposal: reform the programs that give local police departments military hardware without training or guidelines in its use. (One young man also suggested that black Fergusonians would be well served by graduating and going to college at higher rates so they could have more representation in local government and the police force. My jaw dropped at hearing a 17-year-old juvenile offender channeling Bill Cosby.)

      • Multiheaded says:

        My jaw dropped at hearing a 17-year-old juvenile offender channeling Bill Cosby.

        This, uh, has been a staple of African-American politics for the last 150 years? Black people argue for and against this general line all the time behind closed doors?

        (Sources: Glenn C. Lowry, Ta-Nehisi Coates)

    • Scott Alexander says:

      Yeah, supporting cop cameras seems like a no-brainer.

      Part of me wants to argue that the only thing racializing this issue does is get so many white people angry that they’ll oppose cop cameras just because it’s what the social justice faction wants, and if we didn’t racialize the issue everyone could just agree cop cameras are a good idea.

      The other part of me feels like unless we racialize the issue, nobody pays any attention to it and we just get a lot of people who are like “Cop cameras? Sounds like a good idea,” but no pressure for it, and so racializing it turns it from 100% passive supporters to maybe 60% active supporters and 40% active opponents, and maybe the active supporters will get something done.

      This is obviously not a great way to run a country, but I don’t think anyone ever claimed that it was.

      • Ockraz (aka Contrarian) says:

        Actually, the opposition I’ve seen to cop cams has been privacy based. Do you have examples of red tribe knee jerk opposition to them, or is that just speculative. Opposing them on the same basis as opposing CCTVs on every lamppost strikes me as more gray than red. Then again, I’m not certain I fully understand what makes grays gray. (I was actually directed to this blog via an exchange of comments at ruby red site FWIW.)

        I’m also perplexed by the use here of ‘social justice.’ It’s not compatible with either of the two definitions I have been aware of – both of which have substantial scholarship. I only became aware that there’s a term, ‘social justice warrior,’ about a month ago from imgur. Is it pejorative or do people apply the label to themselves? Honestly, I’ve yet to see any indication that it means something more than someone who’s on a crusade based on feminist theory about something that’s especially irking them at that instant.

        • Grumpus says:

          That’s basically it, ergo definitely a pejorative.

          • Ockraz says:

            If someone asked me a year ago who’s the most important author to write about social justice, I’d’ve said the late John Rawls. If asked to name a cultural figure trying to battle for social justice, I’d’ve said Pope Francis. This SJW use of ‘social justice’ is nothing to do with that at all! I find it rather demeaning to the original that the term’s getting abused this way.

          • nydwracu says:

            See also: Father Coughlin.

        • Anonymous says:

          Trigger Warning: very vague assertions ahead.

          I’m fairly sure I’ve seen unironically (that is, as a self descrition) as early as a few years ago. My interpretation was that it was originally a self label that got co-opted as an insult. But I really don’t have any concrete information of the subject (that is, some sort of documentation of the first instances of it being used unironicaly and as a pejorative.

      • Anonymous says:

        Body cameras ARE a no-brainer. They make everybody safer.

        They are not, however, the panacea that some people seem to think. There is a significant practical problem: nobody seems to make one that can hold an entire shift of video. Wilson and all the other officer’s whose interviews I have read were working twelve hour shifts, for example. That is a lot of video.

        Without being able to turn it on at the start of your shift and forget about it, there are going to be a lot of situations were it does not get turned on when things get heavy. An impartial observer (much less a biased one) will be unable to differentiate between oversight/technical problems and malfeasance.

        Suppose that turning on the body camera is item six on the getting-out-of-the-car checklist, and an officer says “I was only on item three on the checklist when this guy started punching me in the face through the window….”

        • John Schilling says:

          “Body cameras ARE a no-brainer. They make everybody safer”

          So, the sole purpose of brains is safety maximization? I must have been misusing mine when I started thinking of unintended adverse consequences to cop-cams that make it less than completely obvious that the net result is positive. I invite you to do the same, considering matters beyond just first-order safety impacts, because I’m actually interested in what you might come up with.

        • youzicha says:

          The lowest quality Youtube 360p videos are coded at 1000 kbps, so 12 hours is 5.4 gigabytes. That should be technically feasible to store, while still providing some useful information afterwards. Though I guess battery life would be a problem if it has to do modern video compression continuously.

      • Anonymous says:

        I think this principle holds for radicalization in general. If an issue isn’t radicalized at all, then oftentimes not enough people will be moved to action. The more an issue is radicalized, the more outspoken voices on both sides will be generated, and the side with the most such voices is likely to win. But if an issue becomes too radicalized, then the opposing camps become so polarized and angry at each other that mutual destruction ensues and nothing actually gets done. Thus, in these cases, the conservative side (the side calling for maintaining status quo) essentially wins.

        So we must hope that the cop-camera issue is radicalized just enough to call people to action, while not becoming so radicalized that everyone just yells at each other and gets nothing done.

        [This is Liskantope, having forgotten to enter my pseudonym.]

      • Avi says:

        For the record, Yudkowsky is against body cameras. https://www.facebook.com/yudkowsky/posts/10152904788444228?comment_id=10152904820329228

        “I think there’s a tremendous problem with body cameras, namely that part of what keeps the current system working at all is the many kindly and just police officers around the country who selectively decline to enforce bad laws. And you would have to be a tremendous optimist to believe that body cameras will help expose these bad laws and change them. But if that’s the policy demand, then I agree and admit that this is a coherent policy demand indeed. Now let’s see that slogan on a set of signs, carried aloft by a coherently organized march to demonstrate political power and capability, directed at whoever is supposed to pass the legislation.”

        • Liskantope says:

          Hmm but suppose that the recordings would only be checked if there is an accusation of abuse of power on the part of the policeman. That is, at least, more or less how I imagined it would be implemented. Maybe that’s naive and/or a misunderstanding of what is being proposed.

          • John Schilling says:

            Selective enforcement of not-bad laws is one of the abuses of power police are frequently accused of. Frequently guilty of, for that matter. And if we all agreed on what the not-bad laws were, they wouldn’t be laws at all.

            So, yes, cop-cams mean FOIA requests for anything that might be a cop letting off some other guy for the thing he arrested today’s designated victim for, and cops tending to arrest them all and let the judge sort them out so as to cover their own ass.

      • houseboatonstyx says:

        The other part of me feels like unless we racialize the issue, nobody pays any attention to it and we just get a lot of people who are like “Cop cameras? Sounds like a good idea,” but no pressure for it, and so racializing it turns it from 100% passive supporters to maybe 60% active supporters and 40% active opponents, and maybe the active supporters will get something done.

        What sort of actions, at which levels, will actually get some police to wear them (without convenient malfunctions)?

        *looks in crystal ball* I see … a small or medium sized Red town that is not especially polarized. A few quiet high status people (probably middle aged white men) talk things over among themselves, individually get input from individual contacts in the police department, do a little research, find out what has happened in similar towns that tried this, talk to the Sheriff, find out what if any approach he, yes, he might consider, work that up. When they are all agreed, then privately talk to the Mayor etc. When the whole clique is agreed, the Mayor makes an executive order, or at most the Commissioners introduce it and rubber stamp it. — Thus a small amount of non-stressful work by a few “Sounds like a good idea” people gets it done.

        I don’t know how many towns might fit this pattern, but I hope they get their measures in place before the issue becomes radicalized.

        I’m probably off tone here. Nearly full moon.

  42. Karmakin says:

    Two points:

    First of all, the concept about working less hours as being a BAD thing is a motte I’d LOVE to destroy, both from a sociological as well as an economic perspective. We’re all (well most of us at least) working too much and we should stop it. At least at our primary job. We have too many hands and too few jobs.

    The idea that the “choice” (no matter how much it’s pressured) to work less is strictly the wrong choice bugs me in so many ways. I’m a guy who would love to work a consistent 30-hour week if it were economically viable. For what it’s worth, people should read about the Temporary Worker program in Canada which is pretty much entirely driven by the desire to get workers who lack community/family ties.

    This is not a world I want to live in.

    Second, about race. Quite frankly, I’ll take the stance every time that the primary form of racism in our society is the assumption that minorities are of a lower economic/social class. I don’t think I’m defending racism by saying that, I feel like I’m actually taking a strong stance against it. But the concept that racism and classism are tightly intertwined, I think is a helpful one in terms of well..fixing problems.

    • nydwracu says:

      First of all, the concept about working less hours as being a BAD thing is a motte I’d LOVE to destroy, both from a sociological as well as an economic perspective. We’re all (well most of us at least) working too much and we should stop it. At least at our primary job. We have too many hands and too few jobs.

      ^^^^^^^

      My father gets back from work at around 10pm, so mentally tired that he can’t do anything but watch old movies on Netflix. I gather that this is not uncommon. And besides, he works for the government — so he’s supposedly better off than his private-sector equivalents.

      Oh well, too bad there will never be anything remotely like a labor movement in this country because the right got co-opted by neoliberal Koch types and the left got co-opted by insane bourgie narcissists who have made it known that they don’t and never will care in the slightest about what happens to white people who have to work a real job for a living.

      Fuck’s sake. I don’t want to have to live like that. And it’s not like anyone with an office job does eight hours of work in the first place.

  43. Blue says:

    On a previous thread someone posted something along the lines of “If you think liberals are leftist you haven’t been hanging out with many leftists”. Glib and almost No True Scotsman here, but it also contains a fundamental point.

    When Harold Pollack says “Liberals sometimes overstate the extent of overt racism as a direct explanation of justice system disparities.” I read liberal there as an _epithet_ , faulting the liberal ideology for focusing too much on the purely evil motivation of racism, and not enough on the structural issues (such as but not limited to the economy) that actually disparately harms groups in our society.

    I understand Scott feeling somewhat misinterpreted by Ezra, but I honestly read Ezra’s piece as saying “our problems are really economic” in a gentle way to his fans and allies. Which is some good progress on motte and bailey issues for the crowd of people who read him. (This is somewhat like the tribes post where Scott said “if you wanted to convince conservatives of climate change, this is how I would phrase it.”) It’s not very critical of his liberal fans, but it may be more convincing.


    Also minor note, it doesn’t escape my notice how Scott focuses more on controlling for income, and Ezra focuses more on controlling for geography. These aren’t really the same thing.

  44. Platypus says:

    In Part IV you give the example of a hypothetical someone saying: “Wait, according to our study, a lot of this is just that women prefer working shorter hours to have time with their families”.

    But I notice the actual text this comes from says: “If you control for hours worked, then some of the gender wage gap vanishes.”

    So in the course of framing an argument about an unrelated topic, we’ve progressed from “some” of the gap to “a lot” of the gap is explained by things which are not sexism.

    I went and found the original Yglesias post, which says: “The pay gap seems to shrink from about 77 percent to around 84 percent when you consider the difference in hours worked.” I would describe this as “thirty percent” of the gap.

    I recognize that the main point of your essay is not about the gender wage gap, and that the motive of talking about that was just to make an analogy to the racism issue from your previous post, so maybe it felt okay to let your hypothetical person fudge the numbers a bit when making their objection. But in fudging those numbers, you’ve now also created the perception, among people who read your essay uncritically, that the gender wage gap is not actually a big deal and we can stop worrying about it.

    This is why we try to use non-controversial topics as our examples when making analogies. 🙂

    • Tarrou says:

      How about the gender death gap? ~96% (depending on the year, sometimes as high as 99%) of all workplace deaths are male.

      As to your post, the serious studies by the US Dept. of Labor have shown consistently that any wage gap is vanishingly small and hard to pin to any sort of “discrimination”.

      From the summary of one such study: “this study leads to the unambiguous
      conclusion that the differences in the compensation of men and women are the result of a multitude of factors and that the raw wage gap should not be used as the basis to justify corrective action. Indeed, there may be nothing to correct. The differences in raw wages may be almost entirely the result of the individual choices being made by both male and female workers. ”

      Link here: http://www.consad.com/content/reports/Gender%20Wage%20Gap%20Final%20Report.pdf

  45. MugaSofer says:

    “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.”

    Ouch.

  46. PS says:

    You win the Internet. Please don’t stop. Maybe someday intellectual honesty will matter.

  47. Barnabas says:

    It seems to be a frequent assumption that minorities or people in poor neighborhoods are harmed more than helped by aggressive policing. Aggressive policing makes for the occasional dramatic episode but it seems to me that a much larger problem in these neighborhoods is lawlessness resulting in a poor quality of life and serving as a disincentive to investment or delaying gratification in general (ie., not using drugs or waiting to have a child,etc). If I live in such a neighborhood and I’m interested in working towards a middle class lifestyle being stopped and frisked frequently or having a certain portion of my co-ethnics in lock up are going to do more to improve my quality of life than harm it. You don’t hear much of this position, particularly in the media, since it would be seen as a defection from the tribe but I do hear it from some of my more conservative minority friends. I have many friends who have emigrated from Central and South America and they speak at length about how lawlessness leads to a cycle of poverty. I think that they would welcome some stop and frisk or profiling if it meant that they could visit home without worry of kidnapped or killed, particularly if it was of the non-corrupt NYC variety where the cops are actually looking for guns and drugs and not bribes.

    • My impression is that poor neighborhoods suffer from random aggressive policing *and* the police not addressing poor-on-poor crime.

      • Barnabas says:

        Perhaps, but the oft heard complaint of the targeting of young black males would be the opposite of random policing.

      • Tarrou says:

        The crime rate decline over the past twenty years begs to differ.

        • Barnabas says:

          I assume by that you mean that lawlessness is a less significant problem than police brutality. The crime rate is still pretty high in my city and I can only imagine that it would be much higher without very high incarceration rates.

    • Multiheaded says:

      1) Do these friends of yours have sons?
      2) Have they ever been stopped-and-frisked in front of their sons?
      3) Have they ever had to explain to their sons who got stopped-and-frisked why that happened?

      Just imagine it, man.

      http://www.theatlantic.com/national/archive/2014/04/what-i-learned-about-stop-and-frisk-from-watching-my-black-son/359962/

      • Barnabas says:

        We don’t have stop and frisk where I live though we do have a terrible amount of violent crime in predominantly black neighborhoods. As far as I can tell stop and frisk was a unique adaptation to a situation where the rich and poor are living in fairly close proximity and simply driving everywhere is not feasible. I understand the argument of injury to racial pride, I’m just not convinced that it’s worth the trade off of having more dangerous neighborhoods.

      • Anonymous says:

        This is exactly the remedy most people need, not in fact reams of confusing studies. A little numeracy is in fact far inferioir, and often dangerous compared to a more humanistic understanding in this case. Too few people are actually interested in the lived experience of poor, minority neighborhoods. To hold forth on exactly what kind of policing is needed for a community that has historically been under total assault, one needs to demonstrate fairly intimate knowledge of what it is like to live and be policed in such an environment.

        I’m learning that self-described rationalists crucially undervalue history.

        • Multiheaded says:

          I’m learning that self-described rationalists crucially undervalue history.

          AMEN!

        • Randy M says:

          “A little numeracy is in fact far inferioir, and often dangerous compared to a more humanistic understanding in this case.”

          Facts vs feelings is a false dichotomy.

        • nydwracu says:

          I’m learning that self-described rationalists crucially undervalue history.

          Not linking my Tumblr but it’s damn near turned into a history blog and also go follow kontextmaschine and bloodandhedonism.

          Giving a rat’s dick about history is the only remaining advantage Tumblr crew has over the rest of the internet now that the trad-Catholic right-Stalinists have all deleted.

    • Lemminkainen says:

      Interestingly, I think that this argument has utility-measurement problems. It’s worth noting that utility-wise, arresting and imprisoning a person is horrible. Like, I would say that being imprisoned even for just a few months is significantly worse than being the victim of any crime that poor people commit other than murder, rape, economically ruinous theft, or an assault which leaves you with permanent injuries or PTSD.

      So, if you’re increasing aggressive policing and the harshness of punishments for petty crimes, you need to actually check whether the utility you gain by reducing the (in just about all of the 20th century, actually quite low) crime rate a little bit is worth the enormous amounts of suffering caused by the aggressive policing and harsh punishments you impose.

      • macrojams says:

        Your post made me think of something that I wasn’t sure was a reply to you or a continuation to my own comment. I edited it on to that one (directly below) but thought you might still be interested.

    • macrojams says:

      I think this is another example of the sort of thing Scott brings up in this post, namely a combination of a high base rate of the condition of interest (criminals in crime-y neighborhoods) and a low sensitivity, low specificity diagnostic (police stoppages).

      Unfortuantely, if the criminals are overrepresented in a particularly race and the police care only about minimizing crime, profiling is going to factor into their Bayes Estimator, upping sensitivity but probably decreasing specificity (this was alluded to in the comment thread on probability matching on the previous post in this sequence). This of course leads to general mistrust and a breakdown of social cohesion, and I don’t know of any good solutions to it; maybe expanding the police’s loss function to match the role of “officer of the peace” more than “crime fighter” so that social cohesion factors in more.

      But that seems pretty hard to pull of considering the financial incentives of police departments and the overall vaguenss/complexity entailed. Plus I think police officers do tend to value that expanded role in far mode, it just gets washed out in confrontations.

      EDIT: Reading Lemminkainen’s comment above brought to my mind the truism that punishments should be frequent, predictable, and minimal, both for the purposes of enforcement and minimizing disutility. This is usually mentioned in the context of child-rearing, but I first encountered it in Ostrom’s Governing the Commons. Turns out she did work on police procedure in the 70’s that fleshes out my vague brainstorming above about policing for social cohseion and throws in the value of local knowlegde. I don’t know much about it, but more can be found here:
      http://en.wikipedia.org/wiki/Coproduction_%28public_services%29
      and here, among other places:
      http://www.indiana.edu/~workshop/publications/materials/volume2.html

  48. Tarrou says:

    I’m not sure if this will make it through the screens, but I wanted to tell a story about racism. Let’s call it “A tale of two racists”.

    I met a man some years back, a solid man who worked as a firefighter in downtown Detroit for thirty-five years. And this man hated black people, wasn’t shy in the least about it. Thought they were lazy, shiftless, criminal, the bunch. But this is a man who lived with them for almost fifty years, and spent most of his professional life pulling poor black folks out of burning buildings. He had a wall of commendations for bravery and pictures with the dozens of blacks he had saved. He had opinions, ones I disagreed with deeply, but only opinions. I never saw him be anything other than professional and polite with people of all races.

    Contrast to me. I’ve got none of those opinions. I support racial legal equality. I’ve dated interracially extensively. I was raised in a totally colorblind community (a cult that cared for theology to the exclusion of any race issues). But I will never, ever, do as much for black americans as that racist firefighter. So think long and hard. I had to. What is really the problem? Is it opinions? Or actions?

    My former cultish training points me to the parable of the Good Samaritan. Who was the neighbor to the man dying in the road? Who of the SJWs can best my racist friend in being a neighbor to inner-city blacks? I know I can’t.

    • Audrey says:

      He was just doing his job! The Fire Service has to provide a service to everyone. Nobody is doing black people a favour by providing them with a service that is supposed to be universal anyway.

      • Tarrou says:

        Your point is taken. The question is, though, why he stayed in Detroit? He could have moved out to the suburbs (where he moved his family, but commuted back to downtown D-town for work for many years). He recently retired as a chief, he had his pick of the firehouses. He stayed in a poor, black ghetto on purpose. Now, I think that purpose was a combination of loyalty to the other lads in the house and adrenaline junkying, and Detroit is the best place in the country if you love actually being in a fire. My point was that opinions often affect actions a lot less than we might think. There’s a raft of solid research on this. Even very racist people often treat individuals of that race well. It’s the amorphous “group” that is hated, not the individuals, usually. Face to face, person to person, racism is a lot harder to maintain.

        • Audrey says:

          I haven’t got a satisfactory answer to your questions, but I have some partial ones.

          Part of this is due to the nature of the fire service increasingly dealing with work that has nothing to do with putting out fires,which isn’t great if that is what you want to do. You’ve addressed that yourself.

          A cynical perspective given in the UK is now that we don’t have an Empire and can’t claim to be ‘helping’ colonised people, socially excluded groups have been turned into an industry which the middle class are employed to ‘help’ in an inefficient way, despite choosing to live elsewhere. If we actually resolved social exclusion, many of the middle class would be out of a job. In that sense, the actions of your firefighter friend is behaving exactly like the social activist people you attempt to contrastwith.

          A third element is simply confidence. People with the confidence to voice unpopular opinions may find it easier to work with different social groups because they don’t get nervous about saying the wrong thing.

    • Multiheaded says:

      1) Pretty sure that the black people would protect that guy in the face of any condemnation; this is why equal access to information and communcations is important.
      2) Pretty sure that the black people would be far less forgiving had he been interacting with them as a cop, a hiring officer or a schoolteacher.
      3) Pretty sure that SJWs would quietly or not-so-quietly suspect that you’re still holding opinions you don’t recognize as racist. I certainly do.

      • Randy M says:

        “Pretty sure that SJWs would quietly or not-so-quietly suspect that you’re still holding opinions you don’t recognize as racist.”
        Of course he hold positions he doesn’t recognize as racist. You mean to imply that they are racist nonetheless?

        Truly, it is easier for a camel to pass through the eye of a needle than a white person to be saved!

        • Jaskologist says:

          Verses from the Social Gospel always tickle me.

        • Multiheaded says:

          Of course he hold positions he doesn’t recognize as racist. You mean to imply that they are racist nonetheless?

          Anecdata (by, IIRC, Belle Waring or some commenter on Crooked Timber) suggests that “racist” was already viewed as an offensive label in the pre-WW2 American South. Wouldn’t you say that maybe the people who didn’t think of themselves as “racists” there and then might still have been very racist?

          If yes, why couldn’t one call Tarrou kinda racist upon being dissatisfied with the details of his opinions?

          (I am not, I repeat NOT, unequivocally calling him racist.)

          • Tarrou says:

            Got any specific complaints other than that you are “dissatisfied” with my opinions? Any evidence of racial animus whatsoever? Or is this just your default setting when you don’t want to think too deeply about an issue?

            You are supposed to be dissatisfied with my opinions. You do realize there are other people whose opinions may differ, don’t you? The connection between that and racism is pretty tenuous. Kind of like the connection between any of your opinions and your direct involvement in the Holocaust. You anti-semite.

      • #1-2 are sharp points but get out of here with that “I think you might be racist” #3.

        • Multiheaded says:

          I have merely strongly suggested that just self-identifying as completely un-racist does not constitute sufficient evidence.

      • Tarrou says:

        Lovely!! Thank you for that. I do love a good accusation of racism! It’s like Kafka, any denial or proof of lack of racism is only actually proof of racism!

        But to be precise, I believe in the perfect moral equality of all humans. No one is “worth” more or less than anyone else except by their actions. I believe in the goal (likely never to be realized) of perfect equality before the law. I believe the science on various minor physical differences between different genetic lines of humans, which often gets me that epithet, but I don’t let the opinions of people who can’t parse a p-value bother me too much.

        If all this be racism, then yes, I harbor those opinions. And your definition of “racism” will have to encompass that.

      • Tarrou says:

        Also, your command of the opinions of the entire black community is pretty solid. Care to answer a raft of questions for them?

        • Multiheaded says:

          I have merely made a guess, based upon what I’ve seen in the American media. Black people talking, rather quietly but publicly, about how Northern liberals are more “enlightened” but less eager to be neighbours then some working-class Southerners, and so on.

          • Tarrou says:

            You know, there are those out there who might claim that your presumption to speak for people of color is racist and embedded with the assumption based on white power structures that you can understand the plight of the underprivileged. Check it.

  49. Barnabas says:

    I think the trend of the USA being a more risk averse society impacts these issues in several ways. (Really, being a risk averse society and at the same time a low trust society.) The most obvious way being that its pretty much accepted that cops respond to resistance with disproportionate force with a tazer or a gun compared to situations decades ago when they might have wrestled with a suspect or used nightsticks. Its hard to object to that while demanding absolute safety in our own jobs and in every other aspect of our lives.

    • Mary says:

      misplaced. . .

    • John Schilling says:

      As noted in Scott’s first essay on this subject, decades ago it was SOP in many jurisdictions to shoot suspects in the back as they ran away from felony crime scenes. Now, the police either use nightsticks if they can catch up with the fleeing suspects or let them go if they don’t.

      I’m generally uneasy with anything that implies using nightsticks is anything less than brutal, bloody, potentially lethal violence. But I think I will call for evidence on your “pretty much accepted” claim that the recent trend is towards a higher bullet:nightstick ratio in law enforcement work.

  50. Ockraz (aka Contrarian) says:

    I’d never heard of Motte-and-bailey argument before, but it appears to mean the same thing as a tactic I know by another name: rhetorical bait and switch.

  51. Ockraz (aka Contrarian) says:

    “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.”

    Odd. Klein may say that’s the sort of thing -we- would call “racial discrimination” or “racial bias” or “racism”, but his we doesn’t include me and I’d guess it doesn’t include more than at the very most 1/8th the folks I know.

    I’d classify that as racially unbiased.

    • If you address unfairness toward the poor in policing by making police be extra lenient to blacks, you do a great injustice to poor non-blacks. So it does matter what the cause.

  52. Will says:

    I think the bigger scandals, having looked at the papers in the review:

    1. how incredibly terrible the data is. The FBI’s uniform crime reporting is awful, and the NIBRS (incident based reporting system) is too small to be really useful. Validating to outside data (like newpaper reports of police shootings) show that violent crimes are under reported in the national data. This data is shockingly bad (and a lot of the review authors note it, but press forward anyway). I’m surprised every criminologist in the country isn’t lobbying for better data.

    2. I see a lot of linear regression models with lots of strongly correlated variables, which makes interpretation very sensitive to the data and the model, and no one seems to be mentioning this potential problem. Instead of calculating standard errors, it would have been nice to see some cross validation of the spot estimates to see what the variance is actually like.

    Basically, this looks like a field where the data is absolutely bad and the methodology is also kind of bad. Why should we conclude anything at all? Isn’t the real scandal how poor the data is?

    • Grumpus says:

      That was my feeling as well.

      • The data collection is sketchy, but some of the analysis is brilliant.

        • Will says:

          Example?

          Also, a brilliant analysis on sketchy data is just garbage in – garbage out.

          Also, I’ve now looked at three well-cited reviews, all of which complain that the methodology of the papers that they are including is very problematic. If its problematic, why include it? It seems like “complain about methodology, use it anyway” is a field quirk.

          • Anonymous says:

            Could you link to the comparison of official data to newspaper coverage?

          • Will says:

            @Anon- the original article is outdated, so here is (better looking) modern data from five thirty eight, which also isn’t paywalled.

            http://fivethirtyeight.com/datalab/reminder-the-fbis-police-homicide-count-is-wrong/

            Here is information about UCR plus the supplementary homicide report:

            http://fivethirtyeight.com/features/how-many-americans-the-police-kill-each-year/

            It is my understanding that these problems plague most of the crime data in general, and aren’t necessarily specific to how many people are killed by police.

          • Anonymous says:

            Thanks!

            Those links are much more positive about UCR than you are. I’ve never heard anyone complain about it.

            The first article links to this, which is better for the comparison with news stories.

            The second link offers three official tallies of police killings and complains that they aren’t the same, but they look awfully close to me. The two articles are by the same guy and he sounds exactly the same when he’s complaining that the three official tallies differ by 10% as when he’s complaining that news coverage is 3x.

            The author goes on and on about how they only count “justified homicides” not “unjustified” ones and how they don’t submit them until the investigation is over. Every
            other source I’ve read on this subject, which is many, says that if they’re going to submit to SHR, they do so immediately, before any investigation, and the label “justified” (and “felon”) doesn’t mean anything.

          • 27chaos says:

            “If it is problematic, why use it?”

            In order to disavow responsibility for one’s conclusions, and also signalling intelligence superior to the authors cited.

    • Nornagest says:

      I’m perfectly willing to believe that the FBI’s crime reporting system is bad, but trying to cross-validate with newspaper reports is insane. That’s like saying that Wikipedia’s not a reliable enough source on some subject, so we’ll back it up by asking five random people on the subway — worse, actually, since random people on the subway don’t have the same incentives to sensationalize.

      • Will says:

        What are you suggesting here? That newspapers are inventing crime to sensationalize things? The newspaper reports should give you bounds on the minimums, at least such is the theory of the people writing the journal papers.

        • Nornagest says:

          I wouldn’t be surprised if there were a few inventions — or at least incidents that don’t meet criminal criteria but get pitched in the media as if they do — but I’m more concerned about selective reporting. I’d also expect getting a reliable incident count out of articles to be hard — incidents are reported out of their jurisdictions, multiple articles are run on an incident, names are very often elided, et cetera — and I don’t really trust your average social science academic to be doing a good job of it.

          If you manage to navigate that minefield it’d give you a lower bound, but unless pages six through fifty of the New York Times were devoted to police blotters when I wasn’t looking, that count’s going to be too low to tell you anything about underreporting. It’s just barely plausible if we’re talking specifically about shootings of civilians by police, but that’s a very media-friendly and very rare subject by comparison with your average murder, let alone your average strongarm robbery.

          The FiveThirtyEight analysis, on the other hand, looks solid to me.

  53. Ockraz says:

    The pay gap argument really irritates me. Advocates sometimes define the problem in as women being paid less because they’re women, strongly implying gender based discrimination (but maintaining deniability since ‘because’ could refer to cultural factors) so that framing this as unjust is absolutely anodyne. However, they in portions of their argument to it’s being women getting paid less on average than men, without qualifying at all. That’s something that might well be fair or just depending on lots of very controversial questions about exactly what you understand ‘just’ and ‘fair’ to connote. Switching back and forth and not acknowledging it, or denying it, or obscuring that there’s a difference – It’s a great example of what I just call bait and switch, but I suppose there’s nothing wrong calling it Mott The Hoople (j/k).

    Here’s a pretty good comic about the difference:
    http://leftycartoons.com/2011/04/08/the-wage-gap-and-womens-choices/

    My view is that it cannot be deemed an injustice, but it’s usually an example of unfairness. That may sound contradictory to some, but it absolutely isn’t. Injustice is much more circumscribed and ethically merits redress whereas unfairness often doesn’t.

  54. RCF says:

    It is not mere implicature that women make less money for the same work. In a State of the Union speech, Obama said “”You know, today, women make up about half our workforce, but they still make 77 cents for every dollar a man earns. That is wrong, and in 2014, it’s an embarrassment. Women deserve equal pay for equal work.” In a bit of quibbling that is hard to not read as liberal bias, PolitiFact says “Ultimately, we decided that Obama’s statement that “women deserve equal pay for equal work” was aspirational rather than a part of his statistical claim, so we’re judging him on his claim that women “make 77 cents for every dollar a man earns.”” http://www.politifact.com/truth-o-meter/statements/2014/jan/29/barack-obama/barack-obama-state-union-says-women-make-77-cents-/ When Obama says “Women make less than men; people should make the same for the same work”, it just disingenuous to pretend that people aren’t going to hear “Women make less for the same work. And in fact PoliFact acknowledges that in a campaign ad, Obama explicitly said that women make less for the same work. http://www.politifact.com/truth-o-meter/statements/2012/jun/21/barack-obama/barack-obama-ad-says-women-are-paid-77-cents-dolla/ Explicitly saying that women make less than men for the same job is quite common; Terry McAuliffe is just one example of someone else who has made the claim. This isn’t just bad framing; it’s outright lying.

    Another statistic that doesn’t get mentioned very much: men die from their jobs at a rate about 11 times the rate for women.

    • Protagoras says:

      The statistic about on the job deaths would be more relevant were it not for the fact that (mysteriously) there isn’t all that much correlation between how dangerous jobs are and how well they pay; the men whose high pay drives the male/female gap aren’t the ones in the dangerous jobs.

  55. Mary says:

    If your last name begins with “A” and you are an author, our alphabetizing schema gives you an advantage over authors whose last name begins with “Z” — or even “V”.

    Edit: Drat. This was supposed to be a response to Scott’s observation “I would expect most slices to either have no significant differences (ie people whose names begin with “A” vs. “M”), “

  56. Andrew says:

    Why do you think that height has less “historical baggage” than race?

    I’d guess that height discrimination has at least 3 orders of magnitude more historical baggage than race discrimination. (It seems to go back millions of years and to be embedded in biology.)

    • social justice warlock says:

      It has a longer history, not but less of what people mean when they say “historical baggage.”

      • Andrew says:

        That’s not an informative answer. What is it that “people mean when they say ‘historical baggage'”? I genuinely don’t see what’s (supposedly) lacking in the case of height.

        • Harald K says:

          Racism has been overt, political, institutional and justified by scientists. Discrimination on height has been almost entirely unconscious and accidental.

          Part of the reason for this, I think, is that people form identity groups around “races” – they do not form them around height.

          • Andrew says:

            I don’t think your facts are correct.

          • Nick T says:

            Andrew: Why?

            (I think height discrimination is sometimes conscious and I don’t know what “accidental” means, but the rest of what Harald says is AFAIK right and probably what sjw means by “historical baggage”. I agree that inequalities that are both innately-driven and non-politicized are greatly underrated.)

  57. Andrew says:

    I think you’re right about myopia. It’s caused by spending many hours of the day with the eyes in short focus — usually, that means reading.

    Perhaps the solution for readers is to read from giant digital displays, from a distance?

  58. Andrew says:

    I think you’re right about myopia. It’s caused by spending many hours of the day with the eyes in short focus — usually, that means reading.

    Perhaps the solution for readers is to read from giant digital displays, from a distance?

  59. Morendil says:

    Cross-posted from Gwern’s G+ feed comments section:

    In statistics you often have to control for factors for which random assignment is unfeasible or unethical.

    “Gender” is typically thought as one of those factors.

    And then there are clever study designs where it turns out that the situation screens out just the right things and you can in fact randomly assign subjects to one group or the other.

    And then it turns out that what gender you are, _all other things equal_, does in fact strongly affect your career outcomes.

    http://www.pnas.org/content/109/41/16474.full

    This isn’t “just that women prefer working shorter hours to have time with their families”, agreed?

  60. One way that it being racial makes it worse is that the burden of unfairness doesn’t just affect some individuals, it also affects a high proportion of the people they know, so the cost is concentrated in non-obvious ways. If one of your close friends or family members is in prison, this affects you.

    Also, if a high proportion of people in a community are being economically debilitated, the whole community is worse off.

    If the unfairness is concentrated by some characteristic, then it’s easier for people who don’t share to characteristic to think that the unfairness is deserved.

    All this being said, justice system atrocities happen to white people too, and I want procedural and emotional changes to prevent atrocities in general.

  61. Woman’s housework tends to be more constant and time-consuming than men’s housework.

  62. If I had my druthers, “racism” would include a phrase indicating which race is being favored and which is being disfavored.

  63. It just occurred to me that short people tend to be related to other short people– would there be familial effects that add up because of prejudice against short people?

  64. What do you think of Cracked?

  65. Eric Garner’s death was a result of being choked by a policeman. The choking was caught on video. The policeman was not indicted, but the Feds may take some action.

    Video only matters if the people who can take action care about what’s seen.

    • Ilya Shpitser says:

      I think on the margin police cameras will help, but I think we need to reform also how prosecution incentives work. In particular it would be nice if we could stop the obvious collusion/symbiosis between the police and people that prosecute.

  66. I meant random relative to actual criminal behavior.

  67. victoria says:

    Yes, precisely. Two things:

    1.) In a Norwegian-style system this doesn’t matter so much. Norway’s system has a ten-week maternal block (part of which must be used for physical recovery from childbirth) a ten-week paternal block, and another forty-six weeks that can be split between the parents as they choose. But in the U.S. that’s so far outside the Overton Window that the physical issues of childbirth and pregnancy matter a lot more. A lot of people are relying on unpaid FMLA leave, banked sick leave, short-term disability, and/or the good graces of their employers to get any leave at all, which means in a lot of families only one person can take leave and it’s pretty much got to be the birth mother.

    2.) The percentage of women for whom pregnancy and postpartum is complicated enough to mean they’d need more than minimal leave is difficult-to-impossible to find, but I would guess it’s pretty substantial — over 10%, just based on how often various conditions occur and how severe they tend to be. Things like hyperemesis can cause repeated hospitalizations during pregnancy; pre-eclampsia or preterm labor can result in bedrest or activity restrictions in the third trimester; complicated deliveries, clotting disorders, postpartum psychiatric disorders that are difficult to treat, etc., can make it difficult to go back to work on schedule postpartum.

  68. James Miller says:

    I discussed inequality today in my microeconomics class at Smith College and used “Professors who identify as feminist give twice as many As to female students as they do to male students.” Thanks for the material!

  69. Daniel H says:

    I like this idea of trying to confine causes to their immediate effects. I wouldn’t say that my computer isn’t working because of power failures if this is because the power is failing at my Internet provider.

    Completely coincidentally, I was looking for an old post on this blog today and in the process ran across somewhere you wrote about something similar with reasoning instead of with actions: the first part of this article, which I am still reading and wish I’d read before a recent conversation I’ve had. I hadn’t thought of that in the context of abortion before, but I’ve always been annoyed when Christians failed to try to save me from an eternal torment for being atheist, or ethical vegetarians didn’t try to stop my mass-murdering habits (I later decided the ethical vegetarians were right, and simultaneously realized that trying to stop people eating meat would usually accomplish nothing except bad feelings all around).

    • Lizardbreath says:

      That old article hit one of my pet peeves with the line: “In fact, isn’t the person who says “This is murder, so let’s not do it” a whole lot more honest than the person who says “This is murder, but I think we should continue to permit it anyway”?”

      This line conflates “it’s wrong, so let’s not do it” with “it’s wrong, so let’s ban it.” There are some things I think are wrong and would like to encourage people to *choose* not to do, *without* banning them. So I don’t like seeing that kind of conflation. “It’s wrong, so let’s ban it” is not the only way there is to say “it’s wrong, so let’s not do it.”

      That kind of conflation seems to be very common these days…characteristic of the times perhaps. Too often, when someone argues, “XYZ is wrong,” others reply as if they had argued, “Ban XYZ.” This helps no one.

      There needs to be room in society for private citizens to make moral arguments–or even non-moral arguments and advice (e.g., “Drinking 48 ounces of soda at one meal is bad for you, don’t do it”)–that aren’t about what the *government* should do. That are just about what *private individuals* should *choose* to do. There used to be more such room than there is now.

      And no. I did not just argue that the government should ban this conflation. 😉 😉

  70. Gilbert says:

    I agree with everything you say here. Now let me use it as a starting point to talk about something different.

    Suppose you’re in a failed state where everyone cheats on their taxes and the tax collectors regularly pose as charities to catch the rich, so nobody will be telling you their income. In other words, you won’t be measuring the actually causal factors. In that society there would be lots of good science establishing skin tone spectrography as the best predictor of discrimination. And now perhaps the first guy doing such a study had a fancy statistical method to establish this as the actual causal structure (he thought the correlations between the risks for various kinds of discrimination vanished after controlling for the skin tone factor), but that turned out to be a fluke and we now know that argument doesn’t work. Perhaps there were attempts to control for income by various statistical indices, but it turns out those indices are based on lots of indefensible mathematical assumptions. Also, there is good science showing that giving $1000 to black parents of four-year olds doesn’t measurably change their risk of police discrimination twenty years later. So now lots of people go for the kind of solutions you decry, thinking they have science on their side. After all, race is the best predictor of discrimination and early monetary interventions are ineffective, so we know it’s not a poor people’s issue. And there’s nothing wrong about the actual science part, just a lot about the way they draw conclusions from it. We know the conclusion is wrong because we have research basically impossible in that world, but they only have the science outlined. From our outside it still looks like the should try an anti-corruption program, no?

    Back in the real world, IQ is the best predictor of various kinds of success. There used to be some evidence that different kinds of problem solving success don’t correlate after controlling for IQ, and that is the known historical cause of it being considered a causal thing in the first place, but that’s now known to be a fluke. Also there are some attempts to control for environmental effects with twin studies, but it turns out that depends on a lot of mathematical assumptions we positively know to be false. And early childhood interventions don’t help much decades later. Based on that, a lot of people conclude the achievement gap is biological.

    • roystgnr says:

      there are some attempts to control for environmental effects with twin studies, but it turns out that depends on a lot of mathematical assumptions we positively know to be false.

      Specifically, please?

      • Gilbert says:

        Two big ones:

        First the basic modeling assumption that genetic and environmental causes are strictly non-interacting and additive. I.e. you get X IQ points from your genes and Y IQ points from your environment and then your total IQ is X+Y. This rules out effects like some foods being bad only for people missing a given enzyme, people genetically differing in the time they need to learn skills but not in being able to learn them, etc. We actually know that model to be bunk for IQ purposes, because fitting it with identical and fraternal twins yields totally different results, which doesn’t happen with well specified models.

        Second, the assumption that two separated twins face social environments about as different as two random people. This is not how the adoption system actually works. It can be a reasonable assumption for things the system doesn’t care about but certainly not for anything class-related.

        (And yes, I’m riffing all of this from the standard Shalizi essay, as usual. Because it is obviously right and large parts of the Internet just stubbornly refuse to get it.)

  71. Mary says:

    “Woman’s housework tends to be more constant and time-consuming than men’s housework.”

    Constant, yet, but since the surveys have a history of asking what you do on a typical day they are less reliable about how much time is consumed.

  72. 27chaos says:

    Oftentimes the description of controls as the mechanisms through which a larger problem like racism manifests is accurate and useful. The key question, as you note, lies in what description of the problem is likely to lead to correct solutions. In order to find that out, we need to look at counterfactuals. Suppose racist police are using clever excuses to discriminate against black people. Correcting the individual excuses will not fix the problem, as the police will simply create new excuses in response to new efforts. So both you and Klein have a point here.

    You might recognize this already, but it’s not explicitly acknowledged in this essay so I’m not sure.

  73. Captainbooshi says:

    I had heard the about the problem with deciding what exactly to control in studies like this before, but the way Klein described it made me realize something I had missed before. Feminists are really missing out on a great tool for deciding what they should focus their efforts on! By seeing how controls affect the various gender and racial problems, we can get an effective, numerical measure of exactly how bad these problems are, and which ones are having the worst effect, so that we can focus on those. It might even give a way to more efficiently measure whether or not we are having an effect. I wonder if anyone is doing anything along these lines, and I simply haven’t heard about it?

  74. Captainbooshi says:

    I think this post perfectly crystallizes how conflicted I am about your posts on feminism and other social justice issues. It’s insightful, intelligent, and then unnecessarily unfair to feminists and liberals. You take a problem that could easily be explained by feminists being as bad as statistics as everybody else on the planet, and decide instead that they “wanted to throw a match on a gunpowder keg for s**ts and giggles.” Instead of the majority of feminists simply not understanding statistics very well, an extremely common problem, it must be that they decided to ‘actively misrepresent’ the real situation.

    I would also note that this also assumes that they do not care at all about fixing the problem, and just want to take advantage of it, instead of the far more charitable explanation that they are just bad at it. This is the sort of thing people are talking about when they say you treat feminists differently. In most other things you deal with, you don’t start by assuming that the majority of people involved are straight-up lying about their motivations.

    I can’t stop reading your posts, because are genuinely intelligent, and have important observations that often make me think about issues from a different perspective, but I also always cringe whenever I see one of your social justice posts, because you cannot seem to help but take these little digs. You’ve actually gotten a whole lot better in the past few months, ever since your post “I Can Tolerate Anything Except the Outgroup,” but I can still see the remnants, and like I said, it still makes me cringe.

    • Ilya Shpitser says:

      I think if you don’t understand statistics enough that you can’t evaluate evidence for claims sensibly, does that not mean you should stop talking, and go do some reading? Ignorant activism is pretty dangerous.

    • Matthew says:

      I would link to Scott’s hilarious evisceration of Jim, to show that he does actually attack to his right as well, but sadly that post was deleted because Scott decided the other half of it had been unnecessarily provocative toward David G./rationalwiki to to his left….

      • Anonymous says:

        Wow! It’s been deleted from the internet archive. I’ve never seen that happen before. I’ve seen whole sites explicitly blocked, but never a single page removed with no explanation. It’s on other archives, though.

      • Susebron says:

        Actually, he posted the bit about RationalWiki as a reply to a David G. post about him on LessWrong.

  75. matt says:

    You might be interested in this essay, by veteran black socialist Adolph Reed, Jr. It makes many of the same points.

  76. Pingback: The problem of doing rigorous social science research | Beyond Labels

  77. The Sprat says:

    Scott, I only recently discovered your site, and am coming back regularly both for new updates and also to slowly crawl through your back catalog. As someone who is in ideological opposition to you on a whole host of things (I’m pretty solidly”red tribe”), I am in complete admiration of your almost unparalleled thoughtfulness, integrity, and intellect.

    I say all that in order to minimize just how much of a backhanded compliment this next sentiment might otherwise feel like: I *eagerly* await the day when you realize you shouldn’t be taking Ezra Klein and his ilk seriously. Watching a guy like you get his news from Vox is like Superman lining his own cape with Kryptonite.

  78. Pingback: Lightning Round – 2014/12/10 | Free Northerner

  79. Pingback: How it Feels to be a Young American in 2014 (Hint: Shitty) | Refine The Mind

  80. Pingback: The Practice of Protest | Andrew Glidden

  81. Pingback: Quotes & Links #20 | Seeing Beyond The Absurd