I'll eat the pages of the law if anyone can find a clause that calls for quotas | Hacker News Reader

I'll eat the pages of the law if anyone can find a clause that calls for quotas

latimes.com

16 pointsnhchris25 comments

The key observation that drives modern affirmative action policy is that you can have systemic bias even in a system where there is no hiring bias at all, if past bias has resulted in under-representation for minorities. If young people in the minority population rarely encounter role models to follow in a specific career, they will be less likely to pursue that career, resulting in under-representation of that minority in the career population, which perpetuates the dearth of minority role models, leading to a vicious cycle. Adding in the economic consequences of historic bias makes the situation even worse. To ignore this systemic bias would be to endorse it, which is in effect the same as endorsing institutionalized bias against the minorities affected.

It also relies on blank slate theory - ie that all groups should perform equally, and if they don't, that's evidence enough that there is a conspiracy against them.

Perhaps a way to test the "different groups have different abilities" vs "past bias" theory, would be to find people with high IQ / historical record of accomplishment who suffered great historical discrimination and deprivation (eg Ashkenazi Jews from the holocaust, Chinese families purged in the great leap forward) and then compare their descendants test scores and job outcomes in the USA.

You could look at outcomes a generation or two afterwards and tell whether their performance was high/low ie was "reversion to high group accomplishment" or "underperformed due to present or previous oppression"

There is really no basis for any sort of theory that says underperformance of Black Americans is purely due to genetics.

Primarily because genetic diversity among Black people is too broad. In other words, if you randomly compare the genetics between two randomly selected Black people, the difference, on average, will be far greater than comparing any two randomly selected white people.

This is simply because all of humanity stems from Africa, so that is where the most diversity lies.

There are hundreds of genetic markers involved in something as comparatively simple as skin color. The genetic markers involved in something as complex as intelligence are likely orders of magnitude more complex.

You might be able to extrapolate something from certain sub-groups that branch from very narrow branches of the human diaspora, but you certainly could never, ever come up with something that made any sense to apply to the extremely broad strokes that we use to "group" people here in this melting pot we call the USA.

Nobody has ever claimed a 100% genetic basis of difference. The individual heritability of IQ is .5 to .8 for example

Sub-Saharan diversity is a real thing, eg Bantu, Nilotic, Pygmy and Bushmen are extremely diverse (African Americans are not Pygny or Bushmen so not as diverse)

But if you use cluster analysis on human genomes and ask it to divide humanity into 2 clusters, it divides humans into Sub Saharan Africans and everyone else.

The key is selection not just randomness (most variation does little or nothing) you can't say eg "Africa is the most diverse therefore it will have the best adaptations for altitude" (that would be Tibetans, courtesy of interbreeding with Denisovians)

The number of variants responsible for skin color is actually extremely small. It's why you can have such large variation between siblings.

In contrast yes intelligence is highly polygenic. However, given sufficient sample sizes you can calculate genome wide association scores and work this all out.

I think the more telling test for bias (genetic, social, or otherwise) is to compare groups with long standing US ancestry with recent immigrants of the same background and education.

My understanding is that while we attribute much of inequality to bias in treatment, recent immigrants and their decedents vastly outperform comparable individuals with the same skin color.

To me, this indicates that much of the differences we observe are due to biases of past treatment, opposed to discriminatory treatment in the present.

Comparing modern African immigrants to modern day Black Americans descended from enslaved people is not going to get you a clean comparison by any means. Modern African immigrants are one the most highly educated groups of people in the US. And many of them come having already attained that education due to family money, intelligence, whatever. The poor or even average African immigrants has little chance of making it here. In other words, you are comparing an all star team to the general population.

And yes, biases of past treatment is one of the main issues that many are trying to correct with affirmative action-like programs. There doesn't have to be any modern, active racism at all to try and correct the terrible damage done in the past on a systemic basis. In many cases, this past injustice is exactly what systemic racism is referring to.

To be fair, I did say comparable background, which can include education. There easily is enough immigrants, even poor uneducated ones lacrosse decades to make this comparison.

From what I read, poor and average immigrants do great, even African ones. Even more so, the first generation born in the US.

My greater point doesn't negate the impact of past treatment. If anything, it's strongly supports it. What I think it adds to the conversation is the idea that the challenge is very different than overly simplistic model of skin color discrimination which most people around usually try to reduce everything to.

Miss attribution of the root cause leads to ineffective Solutions.

I agree that historic impacts can be scoped into the definition of systemic racism, but that doesn't mean that other tenants of systemic racism are not overstated or incorrect.

I doubt matching for education matches for potential. An American who drops out of highschool is not the same as a malnourished African who never had the chance

I think that is exactly what you would want such a study to detect, not something that you would be trying to control for.

It is a matter of what the hypothesis being tested is. If your hypothesis is that racial inequality is due to ongoing discrimination on the basis of skin color, finding that the cause is a difference in potential undermines that hypothesis.

If the difference in outcome is due to different potentials, not skin color, you can ask why are the potentials different.

I think this is the interesting and most relevant question.

Why are outcomes so different for a black person born dirt poor in the US with immigrant parents compared to a black person born dirt poor with slave ancestors?

This Cuts directly to the heart of why racial inequality is so persistent in the United States.

It's going to almost impossible to match backgrounds between the two groups and selection effects are huge.

Another question is what are people owed if someone's ancestors were under artificial selection by slavers for many generations?

I think a claim there may be legit. However I'd rather keep objective admissions and hiring and just give cash transfers. I can understand though that people want status as well as cash

I don't think matching backgrounds is particularly difficult depending on how specific you want to be. Would be pretty simple to do a controlled comparison black people born dirt poor in the US slave ancestry versus immigrant parents.

There are also cultural differences, I would bet they have different mindsets of 'grievance' vs 'opportunity' towards the USA

It would also be hard to not be affected by selection effects, eg compare Obama's dad who came to America for an economics PhD at Harvard vs someone whose ancestors were enslaved for hundreds of years

I think that I strongly agree. Any model of racial inequality that ignores such factors is flawed at best.

I think a candid discussion of the persistent cultural impacts of slavery are crucial too making meaningful progress towards equality.

I'm not saying that modern discrimination doesn't take place, just that there is a huge elephant in the room when many people discuss proposals to address inequality

Yes, sub-saharan diversity is even more diverse. But diversity among African Americans is still much higher than that of the white population. In fact, most African Americans descended from enslaved have 20-40% European/white DNA (and for unspeakable reasons, most of that DNA is of the supposedly wealthy, elite, most intelligent white progenitors of this country). So I just see no way to make any useful extrapolations based on genetics. Especially considering the externalities involved in being Black in America.

The white admixture could allow you to do admixture studies. Ie examine the inherited white regions see how white polygenic educational attainment scores vs pigmentation scores affect outcomes

This is explicitly forbidden in terms of use of the best databases however

https://stuartritchie.substack.com/p/nih-genetics

From a policy perspective, it makes more sense to assume that past and present bias---which we know exists---is a more likely explanation than genetic inferiority (for which there is no evidence). The burden of proof is on the racial supremicists to provide support for their position, not on the government.

What's your explanation for the over-performance of the Ashkenazi and Chinese vs US whites?

(The comment you replied to)

Cultural emphasis on educational achievement and hard work, for one.

Why does everyone find much stronger genetic effects than parental in adoption studies?

“Little intergenerational correlation in education was observed in the absence of genetic similarity between parent and child—that is, among adoptees.”

https://gwern.net/docs/genetics/heritable/adoption/2021-lude...

“By examining parent-offspring resemblance in a sample of offspring that are among the oldest of any adoption study of IQ to date, we have effectively tested for the presence of parenting effects that would have persisted for more than a decade after the conclusion of the typical rearing period. No such persistence is found to occur in our unique sample.”

https://gwern.net/docs/iq/2021-willoughby.pdf

In an adoptive sample of Korean Americans parental income was unrelated to offspring income.

https://gwern.net/docs/genetics/heritable/adoption/2007-sace...

I'm not sure, that's not my area of expertise! I think it's pretty well-replicated that IQ is more predicted by genetics than any other factor; however, my understanding is that variation in IQ within an ethnic population is much greater than variation between populations. If a population mean is significantly different from others (which appears to be the case with Ashkenazi Jews and Asians), I'd expect the effect to be primarily environmental, not genetic.

I don't know enough about the inheritance of intelligence to be sure of this at all, though.

Did you actually read your own references? From the abstract of the study of adopted Korean Americans:

"I find large effects on adoptees' education, income, and health from assignment to parents with more education and from assignment to smaller families. Parental education and family size are significantly more correlated with adoptee outcomes than are parental income or neighborhood characteristics."

>To ignore this systemic bias would be to endorse it, which is in effect the same as endorsing institutionalized bias against the minorities affected.

Which is an attempt to imply that you must take race into account when hiring or you're a racist, without actually saying that, because it would be roundly rejected by most people if made plain.

Sure, no one wants to admit they're endorsing racism. But it doesn't change the fact that they are. There is no easy solution to this problem, it means facing some hard truths.

Title is "Battle Against Bias Waged on Shifting Legal Ground : Civil rights: Discrimination was banned in a 1964 law. But the prevailing view of the measure changed greatly."

Article is from 1995.

Thanks for pointing out the date, I was shocked at how un-woke it was for a mainstream news article.

Reply on news.ycombinator.com