Comment by rayiner | Hacker News Reader

Comment by rayiner | Hacker News Reader

Parent Full thread

It's not just the history of computer science. Damore's essay was based on time-tested pseudo-scientific reasoning, the sort that has been invoked to exclude women in many different fields. When law firms offered Sandra Day O'Connor a secretarial job after she graduated third in her class at Stanford, do you think they said "we just don't like women?" Of course not. They couched their bias in appeals to the differences between men and women, and how those differences supposedly were relevant to the practice of law.

Damore is more of the same. Take, for example, his invocation of studies showing that women are more prone to be "neurotic." Aside from the fact that "neurotic" covers an immense range of personality traits--Damore leaves it as an exercise for the reader why that would be a bad thing. Among other things, "neurotic" encompasses "obsessive compulsive" or "anxious." Why isn't that a good trait in a programmer--someone who needs to be constantly on the lookout for the errant buffer overflow? He also invokes the idea that women are less able to handle "stressful" situations. Which explains why nurses--who are literally dealing with life and death every day--are mostly women? It's not that there isn't evidence of differences in personality traits between men and women, it's that all the inferences after "infant boys prefer trucks" is complete handwaving.

Back in the day, people would, with the pretense of science, invoke differences in cranial measurements to explain the economic plight of African Americans. Damore's is just the latest in a long line of pseudo-scientific gibberish. And it blows my mind that so many in the tech community are falling for it. You actually have to be bad at logical reasoning to find Damore's points compelling.

> When law firms offered Sandra Day O'Connor a secretarial job after she graduated third in her class at Stanford, do you think they said "we just don't like women?" Of course not. They couched their bias in appeals to the differences between men and women, and how those differences supposedly were relevant to the practice of law.

Note that he isn't saying that women do have these traits; they are saying that on average, they are more likely to. In this example, O'Connor would just be one of the ones to not have them.

There are more male programmers than female. Please explain how invoking genetic differences explanations is pseudo scientific, whereas invoking systematic discrimination explanations is not pseudo-scientific.

Consider: There are more male rapists than female rapists. Is this because female rapists are systematically discriminated against? Does our society shamefully teach children that raping is only for boys? Or are males simply genetically more predisposed to rape relative to females?

If males are genetically more likely to rape, should we lock them all up at birth? I guess not?

Tangent, but men and women are heterosexually assaulted at similar rates, it's just that men on average are better able to resist due to strength advantages,and have more of a stigma against reporting. And not as much in the workplace since on average men have more positions of power in workplaces

I'm not sure about that, for instance the below link claims 91% of victims of sexual assault are female.

https://www.nsvrc.org/sites/default/files/publications_nsvrc...

The difference is that we have a theory of causation and evidence that the causal factors exist. We can point to studies showing that identical resumes with female names are half as likely to get called back, and we have a workable theory of how things like that affect gender ratio. These theories are backed by a real-world social experiment: when we eliminated many explicit forms of discrimination that existed in past decades, we saw the proportion of women rise rapidly in many formerly male dominated professions. That gives us a rich source of empirical evidence to draw from. (E.g. we saw in law and medicine that affirmative action led to self-perpetuating long term changes in gender ratio; once going into a law career wasn’t signing up for a life of being an extreme minority, womens’ representation shot way up).

With the biological explanation, we have evidence of certain gender differences, and we have evidence of disparities in representation, but we have no theory of how the two are causally linked. That’s the hallmark of pseudo-science. Say we know that women are more cooperative than men. What does that mean for programming? Is programming inherently less cooperative than fields like accounting where there are more women than men? What is the sensitivity of programming choice to differences in “cooperativeness?” The biggest confounder is that while we have scientfic studies associating genders with traits, we have nothing linking traits with professions. That’s where Damore’s memo goes off the rails into handwaving and begging the question: he repeatedly assumes that because men are over represented in programming, that programming is necessarily characterized by male traits. (E.g. thing oriented versus people oriented: why does Goldman Sachs have more women creating abstract financial instruments than many tech companies have creating social apps?)

Yes, when affirmative action becomes status quo, you get a stable population of whoever you are affirming into the profession. However, that is not removing an explicit form of discrimination but rather introducing one.

Certainly, it is quite difficult to predict what psychological profile will succeed in a particular profession. But observing that males tend to succeed in programming is not begging the question whether males have relative genetic predisposition to programming (especially at the top 0.1% google level) but rather serves as evidence for the proposition.

When we observe for instance that Asians are overrepresented as programmers, we don't use that to infer that non-Asians are discriminated against, but rather that for whatever reason Asians seem to possess some trait profile, possibly genetic, possibly not, advantageous for programming work.

I'm not claiming that women were never discriminated against, just that it does not appear to be the best explanation for the observed large trends in gender ratios. For instance it doesn't explain male/female behavior differences in other arenas like criminality while genetic differences does.

>Yes, when affirmative action becomes status quo, you get a stable population of whoever you are affirming into the profession. However, that is not removing an explicit form of discrimination but rather introducing one.

This is super clearly not what GP was talking about. Removing implicit and explicit barriers to participation is not affirmative action. You are asserting the opposite of history here.

>Certainly, it is quite difficult to predict what psychological profile will succeed in a particular profession. But observing that males tend to succeed in programming is not begging the question whether males have relative genetic predisposition to programming (especially at the top 0.1% google level) but rather serves as evidence for the proposition.

It in no way serves as evidence for that proposition. By logic, quite the opposite is happening: you are proposing that as an explanation for the observation. There is no inherent connection between them.

>When we observe for instance that Asians are overrepresented as programmers, we don't use that to infer that non-Asians are discriminated against, but rather that for whatever reason Asians seem to possess some trait profile, possibly genetic, possibly not, advantageous for programming work.

Such a "trait profile" could include things like contemporary trends in national cultures, economic development policies, attitudes toward education, random vagaries of history and much much more. Anecdotally, I don't think I've ever heard someone, with a straight face, include genetics in such a discussion. But just as broad a spectrum of influences could be at play with the current gender balance of a given profession. And that's exactly the point: we don't have the clean data to make ANY strong claims about how much of the status quo was shaped by genetics.

>I'm not claiming that women were never discriminated against, just that it does not appear to be the best explanation for the observed large trends in gender ratios. For instance it doesn't explain male/female behavior differences in other arenas like criminality while genetic differences does.

It doesn't have to be the "best" or only explanation. But if there's good reason to believe that it's a factor, it should be taken into consideration and attempts should be made to solve it just like any other problem.

> yes, when affirmative action becomes status quo, you get a stable population of whoever you are affirming into the profession

You ignored the “self perpetuating” part. The even gender ratio stayed stable long after those measures were removed.

> But observing that males tend to succeed in programming is not begging the question whether males have relative genetic predisposition to programming (especially at the top 0.1% google level) but rather serves as evidence for the proposition.

What? No. That’s textbook begging the question. You’re trying to explain the overrepresetnstion of men in programming, by pointing to that overrepresentation as evidence of predisposition. Textbook begging the question.

The memo itself debunks most of what you're saying here - and the insane thing here is that people like you are opposing other's rights to even state an opposing theory. You're not arguing against the memo so much as you are arguing against this technical employee's right to make what he deemed a technical argument.

There's nothing wrong with stating pseudo-science. It's just that people won't take you seriously.

More very disingenuous arguments. People took him so seriously he got fired!

I wont answer the Consider* part of what you said because it's a straw man argument and should be treated as such.

There are more male programmers than female NOW* but that wasn't the case 50 years ago.

Systematic discrimination has been proven as fact, with palpable examples (I shouldnt have to google this for you).

The 'genetic differences' that this memo quoted have been proven wrong by science... making it psuedo-science.

Many jurisdictions define rape in a manner that requires that the person assaulted be penetrated. Under this definition, it is literally impossible for a woman to rape a man, except in cases where that definition would include anal penetration with an object other than a penis. In most jurisdictions, this would be defined as sodomy, not rape.

I don't have stats on sexual assault handy. But I can tell you that, yes, our society does, in fact, teach children (and adults) that rape is something only boys do. This is a fact.

I think you'll find that many feminists and activsts are not happy with most juristictions rape laws.

I have zero idea why you feel the need to point that out to me.

Yep they did the same to me... If they don't like your argument they just flag it as offensive. Admins are obviously on the side of oppression since I was prevented from posting for the last few hours and they threatened to ban me. It's like... Employee discusses theory and gets fired to silence the unwanted opinion. Then we discuss whether it should have happened or not - and we're at the point now of suppressing any opinions that strongly disagree with the firing. So not only aren't you allowed to support an idea - but you're not allowed to defend anyone who has supported it. This is why this discussion always comes back to free speech! Why can't we defend each other's rights to free speech? After participating in this thread it is clearer than ever.

Reply on news.ycombinator.com