5 ms·

Well, sexist in the case in question. And I don't believe a reasonable person can conclude that someone believes things like "women score higher on the psychia

Well, sexist in the case in question. And I don't believe a reasonable person can conclude that someone believes things like "women score higher on the psychiatric test for neuroticism" and then sets that belief aside when evaluating the behavior or performance of women on their team.

zahlman 11mo ago
I know I said I was done, but I do review my own comment history and what you've said (to a third party) here is so fallacious I feel compelled to interject. Zeroth, statistical results like this are not at all like the "scientific racism" of the past. They are well accepted across a broad field of study and have shown themselves to be useful in understanding human behaviour. (Admittedly, a psychiatric evaluation is not a precise measurement. But it is not fundamentally based on a flawed assumption, and it is not being conducted by people with an axe to grind.) First, OCEAN traits are basic material in psychology. By your argument, nobody with a relevant degree should be allowed to be employed anywhere; but they learn (not "believe") these things exactly because it makes them more qualified for their intended career path. It's the same as how doctors nowadays are expected to understand how culture mediates responses to pain, or how men's and women's health care facilities have proven themselves necessary (as has the need to disentangle what's caused by physiology or hormones from what's caused by psychology). Second, Damore obviously understood the material he presented reasonably deeply. Nobody who looks into the psychiatric tests on that level would misunderstand what they mean. And, I mean, really, this is such an incredibly tangential thing to actual work. If I thought you belonged to a group with more propensity to self-doubt (I'm still not exactly being scientific here, but that's closer to what I've understood from my own reading), are you really saying that this would prevent me from fairly "evaluating your behaviour or performance"? The quality of your code doesn't change because of how you felt about it when you committed it. Third, this is a claim that Damore fully understands as a statistical claim about the general population. He knows very well, and he makes it very clear that he knows very well, that this does not transfer to "women on one's team"; and furthermore he didn't argue that trait neuroticism makes people unfit for purpose (as above). The entire point of the argument is to explain why women disproportionately don't apply for the job, and thus relieve Google of the moral burden of correcting for that. (But he even offered suggestions for what they could do instead!) Throughout all of that, Damore makes it abundantly clear what he's trying to debunk — the idea presented in the tl;dr: "Extreme: all disparities in representation are due to oppression". If disparities in representation aren't caused by something purely external to the group (what "oppression" means), then they must be caused by something at least partially internal: i.e., a statistical group difference. And Damore is really making the mildest possible arguments about such differences. If you consider this objectionable, then logically you require everyone to believe that every group is statistically identical to every other group in every aspect except the one that defines the group (or at least: in every aspect relevant to performance in any job, etc.); that this is something that must be true a priori and cannot in principle be falsified even by actually doing the statistical work; and that disagreement is thoughtcrime. This is not a strawman; your argument requires this level of absolutism, because otherwise it is not fairly engaging with what Damore actually said.
- shadowgovt 11mo ago
  > If I thought you belonged to a group with more propensity to self-doubt (I'm still not exactly being scientific here, but that's closer to what I've understood from my own reading), are you really saying that this would prevent me from fairly "evaluating your behaviour or performance"? The quality of your code doesn't change because of how you felt about it when you committed it. Quality of code is by no means the only, or even dominant, evaluation metric for Google performance reviews. There is significant opportunity for bias in how those evaluations work, and that is by design; an engineer that puts out pristine code but shirks oncall duties, responds poorly in performance reviews, is hostile to their coworkers, or pushes blame while pulling credit, is not ready for higher ladder rungs. It's one of the reasons performance reiviews are done by peers, so that being stuffed in a team with a bad manager is less likely to hamstring one's career. But, that does mean that if you're on a team with someone perceived to be sexist (here, I mean "believes different sexes have different inherent natures of behavior"), it sows a lot of doubt in their ability to objectively evaluate you. I don't think Damore intended to create an uncomfortable working environment, but he did, sadly. There's an interesting book, "The Myth of the Rational Voter," that touches upon how higher education (in economics, in the case of that book) influences one's political views. Damore may not have been scientifically wrong, but if he's expected to keep working with people who haven't been read-in on the science and don't agree with it even when they are... He made a career-limiting mistake suggesting that the company's approach should pivot to that avenue of research's conclusions. I think he made the mistake honestly, but it was incompatible with his continued employment and he didn't back down when that was made clear. To answer your question more directly: "Yes. If I were working with a coworker that I know has publicly and thoroughly quoted science that indicates men are categorically more impulsive than women and he gives me, a man, a bad performance review, I would wonder if he did because he doesn't think men have the self-control to make good leaders." That's the hostile work environment, sadly. > He knows very well, and he makes it very clear that he knows very well, that this does not transfer to "women on one's team" Two issues with this in practice: 1) If it doesn't transfer to women on one's team, then why was it relevant? 2) If it applies specifically to the general population, than why was it relevant? Google already isn't hiring from the general population. If that's actually the argument he put on the table, he committed a category error that undermined the whole exercise, which makes it more suspicious when he doesn't back down. > If you consider this objectionable, then logically you require everyone to believe that every group is statistically identical to every other group in every aspect except the one that defines the group (or at least: in every aspect relevant to performance in any job, etc.); that this is something that must be true a priori and cannot in principle be falsified even by actually doing the statistical work; and that disagreement is thoughtcrime. You have the meat of the situation yes. Title VII does make that assumption, and does not tolerate working protected-category differences into how employees are treated. I believe Damore was trying to make an argument merely regarding Google's diversification initiatives, but the argument he made had immediate implications for the current workforce (it's not like individuals from the categories with the personality traits Damore's manifesto suggests magically stop being of that category when they are hired, and I think reasonable people made that inference whether it was intended). There is a non-specious argument to make that the Civil Rights Act is behind the science (although, as I've noted previously, any claim it is should be treated with maximum scrutiny given where "the science" has led society in the past in this problem domain). ... but it is the law of the land, and as consequence, one cannot be "just asking questions" about protected categories in an employment environment without career risk.
gitaarik 11mo ago
What do you mean? There is a statistic that's been supposedly well studied, and it concludes some differences between woman and man. Now you say as soon as you accept that statistic, you cannot set that aside when you're evaluating individuals? I mean, yeah of course, in a way statistics give us bias about everything. Just like a woke kind of bias will also affect someones perception. That's just always going on. But that doesn't mean we can't realise that every individual is unique, and can be a unique asset in a company / project / whatever?

zahlman 11mo ago

I know I said I was done, but I do review my own comment history and what you've said (to a third party) here is so fallacious I feel compelled to interject. Zeroth, statistical results like this are not at all like the "scientific racism" of the past. They are well accepted across a broad field of study and have shown themselves to be useful in understanding human behaviour. (Admittedly, a psychiatric evaluation is not a precise measurement. But it is not fundamentally based on a flawed assumption, and it is not being conducted by people with an axe to grind.) First, OCEAN traits are basic material in psychology. By your argument, nobody with a relevant degree should be allowed to be employed anywhere; but they learn (not "believe") these things exactly because it makes them more qualified for their intended career path. It's the same as how doctors nowadays are expected to understand how culture mediates responses to pain, or how men's and women's health care facilities have proven themselves necessary (as has the need to disentangle what's caused by physiology or hormones from what's caused by psychology). Second, Damore obviously understood the material he presented reasonably deeply. Nobody who looks into the psychiatric tests on that level would misunderstand what they mean. And, I mean, really, this is such an incredibly tangential thing to actual work. If I thought you belonged to a group with more propensity to self-doubt (I'm still not exactly being scientific here, but that's closer to what I've understood from my own reading), are you really saying that this would prevent me from fairly "evaluating your behaviour or performance"? The quality of your code doesn't change because of how you felt about it when you committed it. Third, this is a claim that Damore fully understands as a statistical claim about the general population. He knows very well, and he makes it very clear that he knows very well, that this does not transfer to "women on one's team"; and furthermore he didn't argue that trait neuroticism makes people unfit for purpose (as above). The entire point of the argument is to explain why women disproportionately don't apply for the job, and thus relieve Google of the moral burden of correcting for that. (But he even offered suggestions for what they could do instead!) Throughout all of that, Damore makes it abundantly clear what he's trying to debunk — the idea presented in the tl;dr: "Extreme: all disparities in representation are due to oppression". If disparities in representation aren't caused by something purely external to the group (what "oppression" means), then they must be caused by something at least partially internal: i.e., a statistical group difference. And Damore is really making the mildest possible arguments about such differences. If you consider this objectionable, then logically you require everyone to believe that every group is statistically identical to every other group in every aspect except the one that defines the group (or at least: in every aspect relevant to performance in any job, etc.); that this is something that must be true a priori and cannot in principle be falsified even by actually doing the statistical work; and that disagreement is thoughtcrime. This is not a strawman; your argument requires this level of absolutism, because otherwise it is not fairly engaging with what Damore actually said.

shadowgovt 11mo ago
> If I thought you belonged to a group with more propensity to self-doubt (I'm still not exactly being scientific here, but that's closer to what I've understood from my own reading), are you really saying that this would prevent me from fairly "evaluating your behaviour or performance"? The quality of your code doesn't change because of how you felt about it when you committed it. Quality of code is by no means the only, or even dominant, evaluation metric for Google performance reviews. There is significant opportunity for bias in how those evaluations work, and that is by design; an engineer that puts out pristine code but shirks oncall duties, responds poorly in performance reviews, is hostile to their coworkers, or pushes blame while pulling credit, is not ready for higher ladder rungs. It's one of the reasons performance reiviews are done by peers, so that being stuffed in a team with a bad manager is less likely to hamstring one's career. But, that does mean that if you're on a team with someone perceived to be sexist (here, I mean "believes different sexes have different inherent natures of behavior"), it sows a lot of doubt in their ability to objectively evaluate you. I don't think Damore intended to create an uncomfortable working environment, but he did, sadly. There's an interesting book, "The Myth of the Rational Voter," that touches upon how higher education (in economics, in the case of that book) influences one's political views. Damore may not have been scientifically wrong, but if he's expected to keep working with people who haven't been read-in on the science and don't agree with it even when they are... He made a career-limiting mistake suggesting that the company's approach should pivot to that avenue of research's conclusions. I think he made the mistake honestly, but it was incompatible with his continued employment and he didn't back down when that was made clear. To answer your question more directly: "Yes. If I were working with a coworker that I know has publicly and thoroughly quoted science that indicates men are categorically more impulsive than women and he gives me, a man, a bad performance review, I would wonder if he did because he doesn't think men have the self-control to make good leaders." That's the hostile work environment, sadly. > He knows very well, and he makes it very clear that he knows very well, that this does not transfer to "women on one's team" Two issues with this in practice: 1) If it doesn't transfer to women on one's team, then why was it relevant? 2) If it applies specifically to the general population, than why was it relevant? Google already isn't hiring from the general population. If that's actually the argument he put on the table, he committed a category error that undermined the whole exercise, which makes it more suspicious when he doesn't back down. > If you consider this objectionable, then logically you require everyone to believe that every group is statistically identical to every other group in every aspect except the one that defines the group (or at least: in every aspect relevant to performance in any job, etc.); that this is something that must be true a priori and cannot in principle be falsified even by actually doing the statistical work; and that disagreement is thoughtcrime. You have the meat of the situation yes. Title VII does make that assumption, and does not tolerate working protected-category differences into how employees are treated. I believe Damore was trying to make an argument merely regarding Google's diversification initiatives, but the argument he made had immediate implications for the current workforce (it's not like individuals from the categories with the personality traits Damore's manifesto suggests magically stop being of that category when they are hired, and I think reasonable people made that inference whether it was intended). There is a non-specious argument to make that the Civil Rights Act is behind the science (although, as I've noted previously, any claim it is should be treated with maximum scrutiny given where "the science" has led society in the past in this problem domain). ... but it is the law of the land, and as consequence, one cannot be "just asking questions" about protected categories in an employment environment without career risk.

gitaarik 11mo ago

What do you mean? There is a statistic that's been supposedly well studied, and it concludes some differences between woman and man. Now you say as soon as you accept that statistic, you cannot set that aside when you're evaluating individuals? I mean, yeah of course, in a way statistics give us bias about everything. Just like a woke kind of bias will also affect someones perception. That's just always going on. But that doesn't mean we can't realise that every individual is unique, and can be a unique asset in a company / project / whatever?