Hacker News Reader

Comment by vkazanov | Hacker News Reader

This is a well-known fact, actually.

My math teacher in Lithuania mentioned that back in his days (50-s) it was practically impossible for a non-Jewish student to get into any math- or physics-related faculty: Jewish families had a very strong multi-generational tradition of both in-school and additional education. He also said he was the only non-jewish in his group in Moscow State University that year.

So at some point it was decided to, ehm, discriminate jewish - they were supposed to be a minority in all the main universities/institutes, under 10%, I think, although this is not a precise number.

Note that I don't advocate anything or anyone, this is just the way it was.

View on HN

Wouldn't that be called "affirmative action" in the US?

Yep, sounds like it to me. The fact that Jews are involved sometimes clouds the issue.

The differences:

In the US, they'd ask everyone the same questions, but have different standards for the results. The practice would not be secret, and so there would not be subterfuge of this scope (the plausible-deniability angle is gone). Finally, the practice would be used to boost the effective scores of the select less-qualified candidates, instead of damaging the scores of select more-qualified candidates. (This is not strictly a difference of terminology: in a competitive assessment affirmative action on a less-successful candidate would consistently crowd out candidates from the bottom end of the 'successful' range, whereas the Jewish questions would sabotage everyone who got them, including those who would otherwise be at the top of the field).

True, it's important, if ever a mechanism like this is applied, that it is made official and mandated by law. I don't agree that the system described works in a substantially different way from affirmative action, though, as you can consider the current admission tests to already include the "Jewish problems", for a number of points equivalent to the "score boosting" applied to the other candidates.

Most importantly, if things were like described by the GP, the intended effect is not to discriminate against an ethnic group, but balance its overwhelming predominance in a particular field. Which is precisely the objective of affirmative action.

If the intended effect was merely to balance its overwhelming predominance, sabotage in the exams is the most problematic way to do so, because of the way that it can efface the best talent in the group; diversity-quotas of some sort would damage the field less.

More broadly, though, if you think that this was "merely" about balance and that there was no anti-Semitic sentiment in academia in Soviet Lithuania in the 1950s, shortly after the "Rootless Cosmopolitan" campaign and squarely in the era of the "Doctors' Plot", then I may have a bridge to sell you.

This is relevant because employment discrimination law in the USA prohibits the use of any tests where certain protected racial categories score worse than the national average by a specific amount. Usually that amount is specifically that each protected race must achieve four fifths of the passing scores of any other group. (Yes, that is a vague and easily gamed standard. It was made up by lawyers in the Department of Justice, not by game theorists in the DoJ.)

The result in, for example, fire and police departments has been that all the tests, especially written tests, have been dumbed down so that 80% of everyone passes. That is automatically considered to fulfil the four fifths standard. Then candidates are chosen by political patronage or random selection.

The result is much worse average qualifications. If there were just a racial quota like US universities use, you could pick the top X% of people from each protected underperforming racial category but also the top people from each of the high performing categories. In the Jewish Problem and four-fifths scenario, you don't get the top people from the high performing category except by chance and with four-fifths not from the protected categories either.

The new, state of the art federal civil service exam written under the Carter administration was abandoned because protected racial groups did badly. the obvious solution of race norming was considered out of bounds because Republicans could demagogue against the practice. The entire civil service exam system was dismantled instead and race based hiring is promoted with no objective standards. The result is the worst of all possible worlds. It's just lucky that federal jobs are still plush enough that good people work hard to game any system to get them.

> The result is the worst of all possible worlds.

With respect to the rest of your points: given the number of Jews who emigrated from Russia for fear of being shipped off to Siberia to die, what you describe as the "worst of all possible worlds" is remarkably civilized by comparison.

Those tests were not everything you say they are. 'Objective' tests have historically been used to exclude minorities from many things, even from voting. Those lionized tests didn't select for quality of applicant, from what I know.

> The result is the worst of all possible worlds

I think it's probably a much better world, where the second-rate candidates, who only got the job because their competition was excluded, have been replaced by the best candidates.

As a simple example, think of professional sports, which for a long time wouldn't hire black or latino athletes. To imagine the impact on quality, just imagine the reverse, if today Major League Baseball announced, 'we're firing all the black and latino players and replacing them with white people - but don't worry the quality will be the same'. It would be absurd.

It's also very sad to say it's the 'worst of all possible worlds'. The worst was when all these people were excluded from jobs and every other part of our society. That was criminal and tragic.

How is "discriminating against a group" different from "balancing its overwhelming predominance in a particular field"? The dictionary definition of "discrimination" says there's no difference. Is the definition that says there is a difference some sort of a moral standard? What are the thresholds of that standard - at what point group members outdoing others warrants affirmative action and by how much the numbers of those members should be dwindled by affirmative action? Which groups should be punished by affirmative action - ethnic and gender, or is religion a legitimate target, what about height, family history, a preference to wear clothes of a certain color - whom is it OK to count and then hold to a higher standard to balance their overwhelming predominance?

Also - what cost should society pay in lost output due to the average ability of people allowed to enter certain occupations having been lowered by affirmative action (as it has to be if less capable candidates are admitted instead of more capable)?

Perhaps most interestingly - how much cost should a group supposedly "helped" by affirmative action pay for the dubious favor in (A) people being unable to successfully function at institutions who admitted them, not because of their abilities, but due to affirmative action and (B) perfectly capable people of the "helped" group being stigmatized because "everyone knows they only got to where they were due to affirmative action, and not due to their ability?"

(I guess you might notice that I'm not a huge fan of affirmative action, but if we could at least agree that there's no reasonable way to distinguish between "affirmative action" and "discrimination", that would be in itself awesome, even if we disagree about the merit of, well, that one thing with two names...)

> How is "discriminating against a group" different from "balancing its overwhelming predominance in a particular field"?

Since you ask, I'll repeat myself. Suppose you have lots of people applying for 100 slots in the university. Affirmative action at a US university designed to give you a quota of, say, 10% minorities could, in the worst case, crowd out qualified persons #91-100. "Jewish questions" deployed against highly qualified Jews in a Jewish-dominated field in Lithuania could have easily excluded most qualified persons in the range #1-100.

I really don't like affirmative action either, but it differs substantially in intent, technique, and impact. Considering them morally equivalent slights against the ideals of fairness and merit is a very narrow, black-and-white world view - and without abandoning the ideal of justice, I think it's important to see that there are many shades grays in this world, and some are much much darker than others.

I'm sure your argument makes #95 feel much better.

How would they know they were #95?

It's not clear why failed applicants would automatically leap to the theory that they didn't get in because of affirmative action, rather than just assuming they weren't good enough. Competition for good academic institutions is generally stiff enough that this is the most likely reason anyway.

"It's not clear why failed applicants would automatically leap to the theory that they didn't get in because of affirmative action, rather than just assuming they weren't good enough."

Why wouldn't they? People generally look for an explanation of bad events that doesn't put the blame on themselves. Even better, if it defines a specific culprit and paints them as a villain. "I didn't get into X because they discriminated against me, those racist bastards" is therefore a very appealing narrative.

It's not designed to make #95 feel better. It's designed to make you feel bad about equating #95 to a guy who probably would have gotten shipped off to Siberia with all his family and many of his friends if Stalin had lived a few more years than he ended up doing.

What happens to minorities if they exceeded their "quota"? Do you put them back to their place?

> a quota of, say, 10% minorities could, in the worst case, crowd out qualified persons #91-100

You're assuming #91-100 are better candidates than the minority candidates. I think the resasoning behind affirmative action is that #91-100 are worse students who just had access to far better resources and thus produced better quantitive results. I'd rather hire the minority candidates.

I wasn't arguing with your analysis of the differences between asking different people different questions and asking everyone the same questions but then holding different people to different standards. I was only arguing that "discriminating against a group", taken literally, is equivalent to "balancing its overwhelming predominance in a particular field", taken literally. That is, I was arguing with that sentence, without assuming that these two names for what I claim are the same general thing are also names for two different discrimination methods (which are two different specific things as you pointed out.)

Regarding "intent, technique and impact" - I agree on technique and impact, intent is a thornier issue; certainly nobody ever instituted discrimination for reasons they proclaimed evil, it's always done in the name of "righting wrongs," and then real intent as opposed to proclaimed intent is very hard to establish. From my own point of view, at the gut level, some discriminators seem to act much more maliciously than others, but I'm not sure there's always a reasonable argument to support my gut feeling.

I'm not a great fan of affirmative action either. Some of the things you say make perfect sense to me. I was only remarking that a similar mechanism exists today in the US and it is considered ethically valid.

As for the difference between AA and simple discrimination, the latter is usually intended to increase the power of the current dominant group, while affirmative action should act in the interest of a minority or disadvantaged group. The mechanisms are the same, it's the objective that's different. Much as the same substance or tool can be used to cure or to kill, it's the intention that matters.

> the intended effect is not to discriminate against an ethnic group, but balance its overwhelming predominance in a particular field. Which is precisely the objective of affirmative action.

This is clearly false. The intent of affirmative action is to give certain oppressed groups get a little assistance; it is not to reduce the predominence of any other one. The latter might be a consequence, but it's certainly not the intent.

The differences are mostly aesthetic then.

I wouldn't overlook the distinctions between them. This kind of policy against Jews was a "quota", specifically "numerus clausus". Participation from a specific minority is prohibited once it reaches a specific number.

Not exactly, but not that different from Harvard capping Asians to around 17% for the past 25 years.

Third-party academic research showing that Asian-Americans have to score significantly higher on the SATs than other students to gain admission to Harvard.

The reason given was that Harvard uses uses participation in non-academic events as a mechanism to exclude Asian students.

You can read about it, and the lawsuit filed, here:

http://www.wsj.com/articles/complaint-alleging-discriminatio...

Hmm ... your citation's headline is: "Harvard Asian-American Bias Complaint Dismissed". The rest is paywalled.

No. Affirmative action is giving extra support to a minority that suffers widely from discrimination and who lack social and political power. It recognizes that admissions are not a meritocracy for many (and in many ways, for anyone - people with the right social network have a big advantage) who because of discrimination don't have access to the same playing field or equipment as everyone else.

Jews in the USSR were the ones suffering widely from discrimination.

Reply on news.ycombinator.com