Bryan Caplan  

What Are You Saying, Scott Aaronson?

PRINT
The Two Davids Converge... Noah Smith on Brad DeLong...
Scott Aaronson's effort to mediate the virtual war between feminists and nerds has gotten a lot of attention.  But only a handful of people have remarked on a rather strange belief Aaronson has avowed:
I believe there still exist men who think women are inferior, that they have no business in science, that they're good only for sandwich-making and sex.  Though I don't consider it legally practicable, as a moral matter I'd be fine if every such man were thrown in prison for life.
A few possibilities:

1. This is a joke.  Maybe, but it sure doesn't seem like it.  Nothing else in Scott's list of beliefs sounds jokey.

2. It's not meant to be taken literally.  Perhaps, but highly unlikely.  Scott seems like a careful, literal-minded guy.

3. He spoke in haste, and doesn't really believe it.  Happens to the best of us.

Which brings us to the scarier stories:

4. He is signaling loyalty to intolerant feminists, even though he doesn't really agree with them.

5. He said what he meant, and meant what he said.

My question for Scott: Where does the truth lie?  Thanks in advance.

P.S. If you're tempted to reply, "He admitted that jailing men for their retrograde views is legally impracticable, so who cares?," you should read Mike Huemer on the power of hypothetical reasoning.


Comments and Sharing






COMMENTS (31 to date)
James Miller writes:

I asked him about this in the comments saying "my guess is that you don’t really believe this and, for example, you would oppose a constitutional amendment legalizing imprisoning people for having sexist beliefs."

Scott Responded with

Yes, you’re right, I’d strenuously oppose such an amendment. I was writing not from a legal standpoint but from a Godlike one, where I get to judge everyone’s ultimate moral worth.

Lawrence D'Anna writes:

Strage? It seems like a typical feminist thing to say to me. If Scott is 97% on board with feminism like he says, why be surprised that he would say it?

I honestly don't understand what's unusual about this statement. I don't agree with it but it seems like the sort of thing one hears all the time.

Miguel Madeira writes:

I think this polemic feminists x nerds a bit silly:

1) nerds are probably of the more pro-feminist of men (if you go to GSS, the men in the ISCO88 codes from 2121 to 2139 - mathematicians and computing profissionals - seems to have strong pro-feminist opinions)

2) I bet that most feminist consider nerds "less bad" than "alpha male-type" mens

I suspect that the only reason for this "fight" is that the only men who have the mental disposition to engage in an internet polemic with feminists are... nerds (by definition?)

Carl writes:

Scott is trying to score victim class Pokemon points with his remark but unfortunately he'll never level up enough to take on a Laurie Penny Pokemon A-class creature with special Female Pokemon power points. The battle for Earth's Ultimate Victim continues. Stay tuned, folks!

Sieben writes:
I believe there still exist men who think women are inferior, that they have no business in science, that they're good only for sandwich-making and sex.

The majority of women are exasperating to the Serious Male. Miss Netflix + Mac&Cheese complaining about liberal arts midterms really doesn't deserve to be treated like a full human being.

But humans in general have devolved into lazy consumerists. The counterpart to the whiny couch-potato female is the "Shy Male Nerd", which both SAs self-describe as. Can you talk about uninspiring?

This is why feminists don't see why men should be in charge. Men aren't men anymore. I've never heard a feminist complain about her powerlifting boyfriend for being too jacked and wanting to have sex all the time.

How mad can you be about making that guy a sammich?

I'd be mad if I had to make SA a sammich.

Andrew_FL writes:

His conflation of believing the representation of women in science accurately and rightly represents the relative IQ distributions with a belief that women have no business in science, and conflation of both these beliefs with beliefs about sandwiches and sex, reflects rather poorly on his ability to understand nuance.

Do women have any business in science? If they're cut out for it, sure. There are certainly some very smart women: Lisa Randall, for example, who may be one of the smartest women (if not the smartest woman) on Earth. She not only has business doing high level science, science would be impoverished without her.

But this doesn't mean it's desirable to buy into conspiracy theories about why women are "under represented" in various scientific field, or justify attempts to try and up female representation by hook or by crook. The number of Ed Wittens is just greater than the number of Lisa Randalls. Various affirmative action programs based on denial of this fact will only hurt science, and won't really help women.

Tom West writes:

But this doesn't mean it's desirable to buy into conspiracy theories about why women are "under represented" in various scientific field, or justify attempts to try and up female representation by hook or by crook.

It doesn't require a conspiracy, just knowledge of the human psychology. We overgeneralize and actively try to force reality to conform to our expectations. (Expensive wine doesn't *seem* to taste better, it *does* taste better!)

Between the two tendencies, I fully expect *any* inequality that naturally occurs to be reinforced by human behaviour to a point well beyond what would occur "naturally".

The fact that blind orchestra auditions significantly increased promotion of women in orchestras didn't require "evil men", it just required men (and women) whose brains work the way human brains work, and who made no attempt to compensate for our brain's idiosyncrasies.

Andrew_FL writes:

@Tom West-can you give evidence that there are fewer women in the hard sciences than there should be?

Because the fact that there are fewer women than men is not evidence of this, since there are good reasons for the numbers not to be equal. As I said, there are not as many Lisa Randalls as there are Ed Wittens.

Jeff writes:

Is there any evidence that women are particularly good at making sandwiches?

Matt Skene writes:

If the cost of entering certain male dominated fields is having to deal with the blatant misogyny casually displayed by some of the posts here, it's hardly a surprise that intelligent, capable women might choose to use their talents in other fields. Usually, smart people have choices, and will choose not to pointlessly subject themselves to baseless hostility and unjustified condescension like this.

Tracy W writes:

Sieben:

The majority of women are exasperating to the Serious Male.

I suspect that this opinion is generally reciprocated.

This is why feminists don't see why men should be in charge.

I don't think it's just feminists who don't see why a Serious Male should be in charge. After all, you're commenting on a blog dedicated to the advantages of markets and decentralising generally.

I'd be mad if I had to make SA a sammich.

I think most commentators here would be mad if they had to make anyone a sammich. Except under a freely-agreed contract of course.

Tracy W writes:

Andrew_FL:

Because the fact that there are fewer women than men is not evidence of this, since there are good reasons for the numbers not to be equal.

But there are also bad reasons for these numbers not to be equal. And at the moment we don't know. But the example Tom West gave, of orchestra composition changing sharply when blind auditions were introduced is reason to think that it's worth investigating.

Sieben writes:
I don't think it's just feminists who don't see why a Serious Male should be in charge. After all, you're commenting on a blog dedicated to the advantages of markets and decentralising generally.
The extent to which markets decentralize depends on many things, including competitive advantage.

My point is that when feminists think of "men" being in charge, they think of petty alcoholic douchebags getting upper-level management positions from their pledge brothers. It's a mistake to label this phenomenon "patriarchy" because those aren't really men.

I think most commentators here would be mad if they had to make anyone a sammich. Except under a freely-agreed contract of course.

I think feminists are complaining about the *expectation* (rather than the obligation) to be servile. Again, my speculation is that this issue is related to the pathetic state of modern men. I can't see the same complaints about a cultural expectation for women to make sammiches for devoted athletic millionaire husbands.

Men are supposed to be wired to convert protein into excellence. The fact that there's a social conspiracy for men to be as lazy as possible and take all the resources shouldn't discredit all forms of cultural patriarchy.

Andrew_FL writes:

@Tracy W- I have no objection to trying to figure out whether the gender ratios of the hard sciences correspond to the right ratios given the different abilities of men and women. I object to assuming the ratios are unjustified already and from there enacting elaborate social engineering projects aimed at making those ratios conform to some feminist's preconceived notion of the appropriate ratio.

I would also add that the hard sciences are for the most part meritocratic: people succeed to the extent they produce good science. This doesn't seem to me to leave much room for orchestra scenarios. But I could be wrong. My request for evidence was completely sincere. This requires determining what the appropriate ratios are (and not assuming they are 1:1 a priori), and then citing the actual ratios as significantly different from the appropriate ratios.

Sean Mack writes:

I believe there still exist people who use the word "inferior" without explaining "with respect to what", and though I don't consider it legally practicable, I believe such people should be forced to fight someone two weight classes above them, on the grounds that every decent person knows bigger people are not "superior" to smaller ones.

This will drive home the lesson that terms like "inferior" and "superior" really should be attached to something before they go out in public.

I further submit that anyone who fails to understand this is unfit to broker a peace on behalf of nerds.

Tom West writes:

Andrew_FL:
can you give evidence that there are fewer women in the hard sciences than there should be?

No. After all, how do we define "should be"?

I would also add that the hard sciences are for the most part meritocratic: people succeed to the extent they produce good science.

I had about 10 years of experience in the Physics faculty of my local large university. In it, I saw very little out-and-out discrimination. Yet, I saw quite a variety of women driven out of physics and computer science by the requirement to continually "prove you're a physicist" or "prove you're a computer scientist".

Time and time again in conversation with my male colleagues (both students and staff), I saw confirmation bias at work. "We know there are few women in science, therefor the women I see aren't really scientists." (It was slightly subtler, but real.)

The women I saw leave the faculty didn't have lower grades. The discrimination was in the fact that their peers remembered their every failure and forgot their successes because you remember what matches your expectation and forget the deviations. They weren't forced out. They left because they were bright and had lots of other options where their colleagues could subconsciously accept their competence.

So, no, I don't know that the "real" ratio should be. I simply know that I saw a large number of competent women leave the faculty because of pressure not faced by their male colleagues.

And no, none of those men were consciously discriminating. Many would have been happy to have a women peer. Except to be considered a peer, such a woman would have had to have been almost super-humanly good to counteract our expectations.

As I said, humans are really bad at handling correlations that are positive, but not 1.

Sean Mack writes:

Tom,

I hate to say I don't believe you, but I don't believe you. You speak of seeing "large numbers" of women driven out of the faculty in a physics department.

Physics departments can't drive out large numbers of women because they don't get large numbers of women to begin with.

Isn't it just possible that the person working from confirmation bias here is you?

Tracy W writes:
The extent to which markets decentralize depends on many things, including competitive advantage.

The key point about markets is that they don't have anyone in charge, be that a Serious Man or a Funny Woman.
And of course, centralisation and decentralisation is a difficult concept to apply to markets, after all, how do you define the limit of a market?

My point is that when feminists think of "men" being in charge, they think of petty alcoholic douchebags getting upper-level management positions from their pledge brothers. It's a mistake to label this phenomenon "patriarchy" because those aren't really men.

Interesting approach to argument, if I can summarise it, it consists of "let's define the problem away!". Let's try your approach in some other cases.

'My point is that when libertarians think of totalitarian dictarships, they think of petty alcoholic douchebags inheriting leadership positions from their fathers. It's a mistake to label this phenomenon "totalitarian" because those aren't really dictators.'

'My point is that when South African blacks thought of apartheid, they think of alcoholic Boer farmers with tanned skin being elected by their fellow countrymen to try to aggregate privileges and wealth to them. It's a mistake to label this phenomenon "apartheid" because those men aren't really white.'

Nope. Doesn't work.

I think feminists are complaining about the *expectation* (rather than the obligation) to be servile.

I strongly suspect that commentators here would complain about an expectation to be servile as well as an obligation.

Again, my speculation is that this issue is related to the pathetic state of modern men.

I think it's related to objecting the idea that some people should be servile to others.

I can't see the same complaints about a cultural expectation for women to make sammiches for devoted athletic millionaire husbands.

Out of curiousity, do you have a devoted athletic husband, or personally know (as opposed to knowing of some celebrity) someone who does? Indeed, do you have a millionaire husband at all?

Because my first speculation on this point is that the number of people with millionaire husbands is fairly small, and presumably some millionaire husbands are not athletic or not faithful.

My second speculation is that people expect millionaires to buy their own sammiches, or even employ a chef.

If you do indeed have a devoted athletic millionaire husband, and you do find that many people expect you to make sammiches for him, that would be interesting data.

The fact that there's a social conspiracy for men to be as lazy as possible and take all the resources shouldn't discredit all forms of cultural patriarchy.

Indeed, what discredits cultural patriarchy is the superior outcomes of market societies in terms of wealth and happiness.

Richard writes:

@Tom West

I've also spent time in academia, and although the phenomena you describe is real, you give no weight to the conscious desire of the vast majority of educated people to give women and minorities advantages in the name of equality.

There's affirmative action, which is conscious discrimination promoted by law and explicit norms. Such a phenomenon more than balances out unconscious biases.

Richard writes:

@ Tracy W

Indeed, what discredits cultural patriarchy is the superior outcomes of market societies in terms of wealth and happiness.

Average happiness or total happiness? From a utilitarian perspective, the low birthrates arguably due to women's liberation have to be taken into account. If everyone is 50% happier but there are half as many people after one generation, that's a net loss.

Matt Skene writes:

There have been a ton of recent studies on implicit bias that seem to all have the same results. Among the more relevant ones are those showing that psychology professors rank an identical CV higher if they think it is a man's, that professors respond less often and less quickly to emails from women, and that students will rank an online instructor higher on evaluations if they think it's a guy.

Since getting feedback and interaction with their professors usually leads to better work, and since student evaluations are important in making hiring decisions, women already fail to have CVs that accurately reflect their abilities. Add to that the fact that women who overcome these obstacles to have a CV otherwise identical to a man's are still less likely to be thought of as a good candidate, and it's clear that bias is going to lead a number of qualified women to be passed over during the hiring process.

While things like affirmative action might cause those making hiring decisions to choose a minority who is similarly qualified in their final decisions, if their bias makes them think that those candidates are less qualified than they are, then those sorts of decisions will occur far less often than they should. In fact, given this information, a rationally self-interested person ought to voluntarily engage in affirmative action as much as possible to offset the probability that their decision is biased against certain candidates.

As for why it's important that women are under-represented, it should be pretty obvious to those interested in economics. One thing we should learn from those studies showing open borders would double global GDP is that one of the largest social problems is the inability for people to use their talents properly due to social obstacles. Right now, we are missing out on lots of valuable contributions by hindering a number of qualified people from being an a position to make use of their skills.

Richard writes:
There have been a ton of recent studies on implicit bias that seem to all have the same results.

Well, thanks to affirmative action, you should rationally assume a white male is smarter than a similarly credentialed beneficiary of affirmative action. A white male simple needs much better test scores to get into Harvard Medical School or Yale Law School than a black woman. That's a simple fact.

http://www.aei.org/publication/acceptance-rates-us-medical-schools-2014-reveal-ongoing-racial-profiling-affirmative-discrimination-blacks-hispanics/

Here's a simple way to prove your point. If there's discrimination against women and minorities and it's a major factor in markets, simply give IQ tests to women and minorities in certain jobs and see how their IQs compare to white men with the same jobs. Women and minorities should at the very least have equal to higher IQs relative to similarly qualified males.

Knowing what I know about college and graduate admissions, I'm guessing such a study would show discrimination against whites, males, and Asians.

Matt Skene writes:

Given the results I mentioned, white males should be able to get higher test scores on exams even if they aren't smarter. They are more likely to have the sort of help and encouragement they need to succeed than others. So the fact that they have higher test scores isn't great evidence that they're smarter.

IQ tests measure certain things well. I don't have any idea how those results would go if we did such a test, but speculation about the results certainly isn't much in the way of evidence. In addition, it's not clear that for a large number of jobs high IQ is going to be a decisively important job skill. They are likely to be relevant for academic jobs. However, one thing to notice about academic jobs is that they aren't subject to the same market forces as other jobs, and so bias is likely to stick around more easily in those jobs than in ones with significant competition.

As for evidence that markets are influenced by this bias, there was a fairly good natural test of this with the last economic crisis. It seems likely that firing decisions would be better informed than hiring decisions. If pay inequity is fully grounded in merit, then one would expect that in a time where there was less leeway to overpay anyone, there would be no difference in the rates at which men and women are fired. If there is a bias in favor of men, the first ones fired would be men. When the financial crisis happened, there were a lot of people fired, and they were disproportionately men, (hence the term "man-cession"). This at least suggests that they were less valuable to their employers than similarly positioned women. It's hardly decisive, since other factors could have played a part, but it's certainly better evidence that hiring and pay-level decisions are irrationally influenced by bias than speculation about IQ test results we don't have would be that they aren't.

Richard writes:
If there is a bias in favor of men, the first ones fired would be men. When the financial crisis happened, there were a lot of people fired, and they were disproportionately men, (hence the term "man-cession").

An absolutely meaningless statistic when looking at the workforce as a whole. Women disproportionately work in government employment and government supported fields like education and healthcare. Their employment prospects therefore depend much less on economic conditions.

You're assuming employers are rational when it helps your argument. But I'm guessing you believe women are payed less than men at least partly because of discrimination.

And you're wrong about IQ tests. They are the best predictor of productivity we have, have predictive power in every occupation, and do not underestimate the productivity of certain demographics.

Richard writes:

On men losing jobs because women are disproportionately working in health, education, and government.

http://www.theatlantic.com/business/archive/2009/07/its-not-just-a-recession-its-a-mancession/20991/

The fact that females have the best representation in fields with the least market pressures should tell us something too.

Matt Skene writes:

To know the value of the statistics we would need to control for a number of other factors. Suggesting that the only relevant ones are explained by this one consideration is certainly jumping the gun. Another factor that appears relevant, for example, is that men are far more likely to enter manufacturing jobs. These aren't magically disappearing, they're disappearing in part because the employees are getting paid more than they're currently worth. (Perhaps I should say that the fact that men are best represented in fields where they expect to be paid more than they're worth tells us something.) It's not at all obvious that the factors affecting the results in one direction or another don't cancel out when we look at the economy as a whole.

As for the rationality of employers, I assume they are more rational when they have more information to go on and more time to evaluate the smaller group of candidates they are deciding about. Bias usually creates rationality in quick decisions and decisions with little available information, like cutting out people from consideration in a hiring process as opposed to deciding which employee you have to let go.

If IQ is really the determining factor in productivity, then we should see women starting to get paid more than men in the coming years since they now have higher average IQs than men do. If discrimination is part of the reason for the pay gap, though, then it will continue to exist for the next generation of workers as well.

Richard writes:
To know the value of the statistics we would need to control for a number of other factors. Suggesting that the only relevant ones are explained by this one consideration is certainly jumping the gun.
But you're the one who brought up the fact that men were more likely to lose their jobs during the recession and used it to make your point. Yet, you're comparing vastly different workforces, and unless you can show that within any given sector men were more likely to lose their jobs, it doesn't help your argument.

Blacks, by the way, have seen their unemployment rate increase more than whites during the great recession. Is this evidence that black employees are less productive? Although blacks and whites have different jobs, I would bet the gap between the races in what jobs they work is small relative to differences in employment by gender.

Tom West writes:

Sean Mack, I misspoke. By large numbers, I mean large percentages. I'd say about 12 of the approximately 20 I knew personally (over several years) left the graduate program.

And yes, perhaps there's confirmation bias at work.

But I certainly know what several of the women complained about, and boy, did a lot of the male students and staff happily, if inadvertently, confirm their attitudes in conversation.

Sorry, but in my opinion this is a no-brainer. After all, do you suppose that the vanishingly few men as kindergarten teachers also has nothing to do with the pervasive suspicion that such men could be predators?

You suffer from a pervasive suspicion that you're not good enough, or are a danger, and sure, all but the most devoted of qualified applicants will look elsewhere.

Also, in Canada, we have no Affirmative Action.

Tracy W writes:

Richard: The data is indeed average happiness, otherwise Australia would be ranking considerably lower.

From a utilitarian perspective, the low birthrates arguably due to women's liberation have to be taken into account. If everyone is 50% happier but there are half as many people after one generation, that's a net loss.

Then, based on observed behaviour, either most people aren't libertarians and/or most people ultimately find such arguments unconvincing. (Note: *most* people. If Sieben wants to churn out babies like pancakes while being servile to their devoted athletic millionaire husband and making him sammiches then that's Sieben's business).

Personally I have no desire to maximise the number of children I have. If that means that I cannot be a utilitarian, well, then, I will not be a utilitarian.

Hazel Meade writes:

Personally, I do not think underrepresentation of women in STEM fields has anything to do with mean intelligence or mathematical ability.

I speak as STEM female, myself.

I think it has more to do with cultural norms about sex roles and expected gender behavior. Lots of girls grow up learning how to do their hair and their makeup instead of learning how to program a computer. This is really strongly reinforced by the social milieu in teen and pre-teen years. If girls don't spend as much time focusing on being good at math and science in those years, the distribution going into college is going to skew downwards, in terms of preparation, and that flows through into success in those fields later in life.

Tom West writes:

Obviously cultural norms do have a lot of influence, but it's the number of young women that show themselves capable in university, but switch out, that I find dismaying, and find it likely that its an indication that they find the field somewhat unwelcoming.

And sure, the creme-de-la-creme will probably be okay. It's the 40th-95th percentile students who are good, but can get discouraged, and whose contribution to the field are lost to its detriment.

Comments for this entry have been closed
Return to top