Eugine_Nier comments on Don't Get Offended - LessWrong
You are viewing a comment permalink. View the original post to see all comments and the full post content.
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (588)
I'm not sure whether we should or not. However, given that we currently have race-based policies and this is likely to continue for quite some time, they might as well be based on accurate beliefs about race.
IOW, you're assuming that changing the US's race-based policies so that they be based on accurate beliefs would be less hard than letting go of them altogether?
Let's just be clear on where the status quo is. Eugine_Nier has mentioned disparate impact analysis several times. In addition to that is the far more important disparate treatment prohibition.
In the US, a worker can get fired even if it is not for cause. If you boss thinks you are a terrible worker, you can be fired even if you could prove your boss is wrong and you actually are a great worker. But you boss would be liable for wrongful termination if the boss said "I think [blacks / whites / Germans / Russians] are likely to be bad at your job, and you are [black / white / German / Russian], so you're fired." Likewise, an employer can't refuse to hire on that basis.
Proving that is a separate issue, but having no public policy based on race implies repeal of both disparate impact prohibitions and disparate treatment prohibitions (and lots of other stuff, but it's complicated).
In practice firing a black worker even if he is a terrible worker leaves employers open to wrongful termination suits. Whereas it's harder to prove that with not hiring a black worker, so frequently the safest route for employers (especially small employers) is to find excuses to avoid hiring black workers rather then risk getting stuck with a bad employee that you can't fire.
In practice, complying with laws has costs, some of which fall on innocent and semi-innocent third parties. As a lawyer, this is not news to me. The question is whether the benefits of implementing those social policies outweighs the costs. Clarence Thomas, a black Justice of the Supreme Court of the United States thinks the answer is no.
But that is a very different question from asking, as a matter of first principles, whether certain kinds of discrimination are allowed even if the facts don't support the discrimination. In the United States, most discrimination of this kind is allowed, and restrictions on the factors employers and others may consider are fairly narrow (race, gender, religion, national origin - not ok. youth, poverty, moral character, basically anything else - ok).
Accurate beliefs about what? If a group (however defined) has been subject to negative discrimination, however arbitrary, then there is an argument for treating them to a period of positive discrimination to compensate. That has nothing to do with how jusitfied the original negative discrimination was.
But “black people in 1960” (for example) isn't the same group as “black people today”, as many of the former are dead now and many of the latter hadn't been born in 1960, and it's not obvious to me that it makes sense to treat people according on who their grandparents were.
If someone was wrongfully executed, killed in a medical bliunder, etc, it is typically their families who are compensated.
It is morally right to do so. But society is deeply conflicted about doing so (for reasons good and bad), so I'm not sure that "typical" is an accurate description of how often it happens.
Regardless of the frequency of compensation, you really should address head on why you think society should do so. The fact that society occasionally does provide such compensation is barely the beginning of the discuss of whether it should and says almost nothing about how much compensation should be provided, or who should pay.
To put it slightly differently, Eugine_Nier is not wrong when he asserts anti-discrimination laws impose significant cost on society as a whole. I think the benefits are worth the costs, but that is a fact-bound inquiry, not a statement of first principles.
I am not arguing that Affirmative Action/Positice Discrimination is necessarily right. Just that it doesn't necessarily have anything at all to do with any facts about DNA.
If actually significant differences in competence have a genetic component, then public policy should reflect that difference. Particularly if the differences are easy / cheap to identify anyway. (I don't think this is true about race / ethnicity, but that's a different issue).
Otherwise, our preferred policies won't work in the Least Convenient Possible World.
Are you assuming all other things are equal? They never are.
If history and practice led to blacks being treated as if the mean IQ was 20 points lower, and the actual difference is 5 points, then the proper public policy is to act as if the difference is 5 points, not zero points to remedy the history and practice.
I suspect that g is not interestingly different between race / ethnicity, and that the IQ test, which seeks to measure g, is culturally biased. But if there is a difference in g that cannot be attributed to environment, then we should consider it in making policy.
In the real world, I think all the important observed difference is culturally driven, so this nod towards facts doesn't change my policy preferences. I think the facts are in my favor. I just think that we should be explicit about how policy should change if the facts turn out to be different.
And are they?
And why do I need to be told "that there exist genetic differences between races that give rise to behavioral difference". I have said nothing about affirmative action/positive discrimination one way or the other. You raised that issue. But you didn't say how they two relate.
I'm not sure whether the cause is genetic or cultural but there are most definitely behavioral differences between the races. Furthermore, the fact that it's politically impossible to talk about this is causing a lot of problems. Consider the state of US cities with large black populations as discussed in this blog post by Walter Mead. The behavior in question is probably purely cultural since historic "white ethnic" political machines lead to the same problems, on the other hand the fact that this political machine is black means most people would rather pretend the problem doesn't exist than talk about it and risk getting called "racist".
For another example, consider the campaign to force Rhodesia to accept majority rule. Given the subsequent history of Zimbabwe this campaign almost certainly resulted in a worse situation for everyone involved.
No, that's the problem.
By the way affirmative action is by no means the only race-based policy, just the one simplest to describe.
In what what are they not?