Today's post, Of Gender and Rationality was originally published on 16 April 2009. A summary (taken from the LW wiki):

 

Analysis of the gender imbalance that appears in "rationalist" communities, suggesting nine possible causes of the effect, and possible corresponding solutions.


Discuss the post here (rather than in the comments to the original post).

This post is part of the Rerunning the Sequences series, where we'll be going through Eliezer Yudkowsky's old posts in order so that people who are interested can (re-)read and discuss them. The previous post was Bayesians vs. Barbarians, and you can use the sequence_reruns tag or rss feed to follow the rest of the series.

Sequence reruns are a community-driven effort. You can participate by re-reading the sequence post, discussing it here, posting the next day's sequence reruns post, or summarizing forthcoming articles on the wiki. Go here for more details, or to have meta discussions about the Rerunning the Sequences series.

New to LessWrong?

Mentioned in
New Comment
26 comments, sorted by Click to highlight new comments since: Today at 12:15 AM
[-][anonymous]11y60

I can only speak for myself, but the example used in this post decreased the affinity I feel for LW:

http://lesswrong.com/lw/ro/2place_and_1place_words/

The details of the function chosen as an example are pretty gratuitous. Couldn't we just have made it about food, or something?

(And not just the example -- some of the comments! This is what you call rational? What goes on here? They're pretty entertaining though.)

I didn't cherry pick this example after combing through many. It's one of the first and few things I've read here so far.

And I have noticed some people here think being offended is, without exception, A Bad Thing. But I also noticed that there isn't consensus on this point. So the question of whether or not it is irrational remains open. I would concede that it's useful to avoid being offended as much as possible, but there are obviously limits to this.

Anyway, the relevant question here isn't, "Is it wrong to be offended?" but, "Will reasonable and intelligent people who are new visitors to LW be potentially discouraged by examples like this?" I think the answer is yes. It's perhaps worth finding the time to use more gender-neutral examples in your posts.

Footnote: The reason this is discouraging is because it sends the message that the author and his community view men as intellectual peers and women as things for looking at and fucking. I'd rather be places where I am an intellectual peer though. I thought this was a really nice, relevant discussion: http://lesswrong.com/lw/4vj/a_rationalists_account_of_objectification/

Will reasonable and intelligent people who are new visitors to LW be potentially discouraged by examples like this?

Would they be discouraged more or less if instead they found a website with completely abstract and unspecific examples (to avoid offending anyone, because some people are sensitive about food)? How about a website with no articles at all? Perfect is the enemy of good.

Or, from a different angle: Would some other reasonable and intelligent people be discouraged by a website with strict self-censorship norms?

You can't make everyone happy. There is always a trade-off. Whatever choice you make, someone will always criticize you for making it. (Then you might optimize for people who are most loud about their dislikes; but they are probably the people who would dislike you anyway, so that would probably be a bad choice.)

I'd rather be places where I am an intellectual peer though.

Uh, this will be offensive, but I don't know how to express it otherwise: If you want to be treated as an adult person, then behave like an adult person; and not whining is a good way to do that!

If someone has a problem to read an example using sexiness as a 2-place word, then I have a problem to consider them my intellectual peer. I could probably make the person happy by treating them like a child and walking on my toe tips around them... but that certainly is not like I behave towards my peers.

If someone has a problem to read an example using sexiness as a 2-place word, then I have a problem to consider them my intellectual peer

How confident are you that none of the people you unproblematically consider your intellectual peers also find it offputting to choose agents as illustrative objects of desire in contexts where their agency is irrelevant, and simply don't articulate that judgment in public?

Any evidence can be countered by other evidence. If the only information I have about someone is that they have this taboo against perceiving agents as something else than disembodied minds, my estimates would be rather low. But of course, add other information to the picture, and the results may change.

If the only information I have about someone is that they [react in this way] my estimates would be rather low. But of course, add other information to the picture, and the results may change.

(nods) Fair enough.

And I have noticed some people here think being offended is, without exception, A Bad Thing. But I also noticed that there isn't consensus on this point. So the question of whether or not it is irrational remains open. I would concede that it's useful to avoid being offended as much as possible, but there are obviously limits to this.

A comrade agrees. Generally, I'm not quite certain where I stand myself, but quite certain that the mean (pardon the pun) LW reaction to people signaling "offense" is arrogant, privileged and sanctimonous - not to mention treating "offense" as a one-place word, etc.

I've never liked the word "privilege" to describe this problem, but there is definitely something to this. Compare the modal reaction on Lesswrong between someone saying something plausibly offensive about women or blacks to the modal reaction to, Omega forfend, someone suggesting that autistic people might actually be inferior in some ways to neurotypical individuals.

It appears to matter quite a bit whose ox is getting gored.

[-][anonymous]11y60

Men are the majority at the high end of the IQ / social success spectrum, and also the low end. At the low end are the homeless, the mentally ill, those in prison, those who drop out of school, and in each case the majority are men. If there is anything to be solved, it will be solved at both ends of the IQ / social success spectrum. I'd like to hear about efforts for gender equality at the low end as well as the high end. If I knew of even one, I'd mention it here. I don't.

There are more homeless shelters where men can stay than where women can stay. The jail where I work is 90% male and thus a male inmate has access to classes, exercise equipment, outdoor space, and jobs that a female inmate can't access.

It's not clear that mental illness is gender-skewed overall. Particular illnesses tend to be more male or more female. Substance abuse is probably slanted male, but so is the research on the topic, and I don't think men are being shortchanged compared to women in substance abuse services. I don't know about services for high-school dropouts.

Men are the majority at the high end of the IQ / social success spectrum, and also the low end.

I'd be interested in reading citations on this, if you have any handy.

[-][anonymous]11y150

Citations suggesting men are over-represented at the low end of the IQ / social success spectrum.

Prison Population (2010) http://bjs.gov/content/pub/pdf/p10.pdf Male 1,445,628 Female 104,629

Homeless (2011) http://homeless.samhsa.gov/ResourceFiles/hrc_factsheet.pdf 62% were male 38% were female

Mental retardation higher among males than females (2006) http://www.cdc.gov/mmwr/preview/mmwrhtml/ss5501a1.htm

IQ test scores of males consistently have larger variance than of females (1995) http://www.ncbi.nlm.nih.gov/pubmed/7604277

Boys are over-represented at the low and high extremes of cognitive ability http://web.archive.org/web/20120211085904/http://www.psych.umn.edu/courses/spring06/mcguem/psy8935/readings/deary2003.pdf

Counter-claims and only-partially confirming claims can be found here: http://en.wikipedia.org/wiki/Sex_differences_in_human_psychology

Awesome, thanks. I've got a lot of reading to do.

[-][anonymous]11y80

You should also give this blog post a read:

http://scienceblogs.com/notrocketscience/2009/06/01/gender-gap-in-maths-driven-by-social-factors-not-biological/

It summarizes recent research suggesting that the gap between boys and girls in math performance is largely cultural. This isn't about IQ, but is closely related.

You'll want to search for things like 'male variance' or look at Baumeister's Is There Anything Good About Men?.

Thanks, that seems like a good place to start looking.

Incidentally, if you question the claims about numbers of males successfully reproducing vs female, we've already tracked down the citations and jailbroken them in http://lesswrong.com/lw/h4e/differential_reproduction_for_men_and_women/

Not so much that I question the claims, more that I'd like to know more in-depth about the subject.

I think that a downside of the karma system is that it is likely to put off people who have good ideas but are nervous about sharing them from commenting. I think these people are disproportionately women.

However, the karma system works really really well and massively contributes to the high quality of discussion here. So I don't have a solution!

I did a quick google search and found that women are more risk adverse than men (at least in financial decisions - that's what most of the studies were about). If that generalizes to social situations, then women would have more fear of disapproval of their comments than men, independent of their intelligence. (I've talked to people who say they are scared to post because they think they're not as smart as other LessWrong users, but I don't know how common that is.)

I don't think it's just because of the karma system. People here also aren't shy about telling others when they're wrong, so even without a karma system, people will worry about the response more than they do on most other sites. I think it's really important that we tell each other when we disagree, so it wouldn't be worth removing that. Maybe we could change how we tell each other, but I'm not sure what form that would take.

I like this sort of attitude a lot more than other attitudes I see on gender here. If there are intelligence differences between different groups of people, I want to believe that so that I can help fix that and make everyone as intelligent as is right (once more important issues are solved, of course). But it seems to be more common for people to stop once they notice a difference, and just conclude that that group is inferior.

It is interesting to me that possibility 1 in this post (hypothesis: men and women are annoyed by different mistakes) manages to avoid giving the most obvious (to me) difference in what mistakes men and women are annoyed by, namely: in comparison to women, men tend to be much less sensitive to, and much less annoyed by, sexism in otherwise rational discourse. In fact, there's even a "mirror image" counterpart: men tend to be more annoyed than women by mention of sexism in situations where it is not clear that sexism is relevant, or where it is clear that no one was explicitly attempting to be sexist.

I suppose that this may have been a conscious omission, as part of the attempt to avoid making things worse merely by bringing up the topic for discussion. But for me it was very confusing, and somewhat alienating. Many male-dominated fields and communities stay that way (sometimes despite the explicit desires of the majority of the community) in part because members of the community engage in casual sexism, sometimes not noticing that they do so, and the other people present, who are mostly men, either don't notice or don't think it is important to point this out. Women who enter the community tend to notice the sexism, but may not address it directly (because they are insufficiently self-confident or disagreeable, or because addressing it directly has been a failure in the past) and instead allow it to prompt them to leave the community sooner than they otherwise would have. I don't know why there should be an assumption that "rationalists" would not share this problem to some degree, and not seeing it listed among possible causes of the gender imbalance here suggests to me that there is a taboo on discussing sexism here, even when doing so is necessary to understanding the phenomena we are discussing. That taboo, if it exists, certainly would make it harder for me to engage with this community. Since I don't know if there in fact is one, or what general rationalist principles it would follow from, though, I'm currently mainly confused.

One of the original commenters wrote:

You ask evo-psych why we have a problem, and evo-psych provides the answer. The gender that has a biological reason to pursue low risk strategies - shockingly! - tends to not show much interest in weird, high-risk, high-payoff looking things like saving the world.

Ask evo-psych how to solve the problem then. We already know that women tend to like doing highly visible charitable activities (for signaling reasons). Maybe we should provide a way for people to make little sacrifices of their time and then make it visible over the web. I am thinking of a rationalist social network that allowed people to very prominently (perhaps even with a publicly visible part here on LW) show off how many hours they had volunteered next to a picture of themselves. I once attended an amnesty international letter writing group that was 90% female, for example.

This brought up an anecdote that I experienced recently.

I'm part of the swing/blues dancing community in DC. At a dance a few weeks ago I noticed that all of the volunteers for helping out at the door taking cash were women. Considering that this is DC - a heavily politicized area - and the vast majority of the dancers are liberal/feminist, I wondered why that would be. I made a comment about it to someone and they noticed that it was weird as well, and then I mentioned (as a pretty obvious counterpoint to our liberal community) that the same phenomenon happens in churches; women volunteer a lot more than men do.

This may not even be evo-psych, but an overall Western/socialization thing, where women - liberal or conservative - are subliminally primed into being more helpful. If that's the case, then using the tools of Western socialization to encourage more women to join LW (as well as other male-dominated fields) might be a solution.

Regarding points 1) and 2) of the original article, I'm wondering how one would measure the 'perfectly-rational amount of disagreement'. Would you even have to in order to consider how likely those possibilities are?