Monday, April 01, 2013

Academic Orthodoxy: Official Lies

A commenter on a recent post of mine writes:
CC asked for examples of the reigning orthodoxy among academics. Here's a short list:

--There is no such thing as "race." It is not a scientific concept.

--Affirmative action is necessary because racism continues to be the primary cause of the poor performance of blacks in school.

--IQ tests do not measure anything real about human intelligence.

--IQ is not heritable.

--If government programs for the elimination of poverty have failed, it is for one of two reasons: 1) they have not been sufficiently funded; or 2) those implementing the programs have not been sincere.

--All differences between men and women are culturally determined.

If anyone doubts the extent to which these ideas dominate public discourse on college campuses, I invite that person to assert publicly a contrary view and see what happens. I say "publicly" because many people will tolerate such notions in private, but they will feel compelled to silence them if they are offered as part of the public discourse of the campus.
 I agree that everything in his list is part of current orthodoxy, with "all differences between men and women" not including obvious physical differences. Also that most or all of them are false—I'm not entirely sure that one couldn't have at least reduced poverty if a sufficiently large amount had been spent by sufficiently sincere people. 

I am curious whether anyone reading this is willing either to argue that the claims on the list are true, or at least defensible, or to deny that, in many parts of the academy, it would be imprudent for an academic without tenure to dispute them.

82 comments:

jim_gr said...

I ll bite.

As far as IQ testing/intelligence is concerned:

Do we have a sound definition of what intelligence is? I would consider human intelligence complex enough to give it a definition. Except of course in the case which we just measure it in IQ tests... which makes your arguement "tautologic" and uninteresting.

Now do you think your IQ test rating after say 1 month of "practice" in IQ tests would be the same with the rating of the 1st IQ test you took? If your rating is say 35% higher than your 1st, what your opinion would be about studies drawing conclusions on differences of about 20%?

Note:I m not trying to be political correct, I m "racist" on blacks, I believe they have stronger bodies than we whites have, they could be as well somewaht mentally "handicaped", I wouldn't mind, I m just not convinced.

Daniel said...

You asked me on my blog specifically what I thought about the last one. I have a self described "lefty" professor in my Gender Macroeconomics course (gender economics is one of my two fields). She has said on many occasions that there are non-cultural (i.e. - biological) differences between men and women. I remember in the Gender Micro course last semester we talked extensively about the impact of hormonal and brain differences that impact economic behavior.

So no, that seems completely wrong to me.

It seems silly to argue that there are no non-cultural gender differences. A blatantly obvious one is reproductive roles after all, but there are a lots of others as well.

Daniel said...

I can't speak to IQ or what "orthodoxy" is on that. I know my psychologist friend uses IQ tests... I think the concern with that is that intelligence is not considered to be as unidimensional as it was once thought. I'm not sure if this is an "orthodoxy" or not.

The one that really got me was the poverty elimination point. My mentor at AU is very insistent about lots of anti-poverty programs that don't work because they're just bad ideas. That should definitely not be on the list. There are lots of public policy and economic professors who dedicate their careers to evaluating and critiquing anti-poverty policies. I doubt they give a damn whether those implementing the programs are "sincere" or not.

Just this past weekend I presented research finding that a particular anti-poverty program did not work at a graduate research conference. The professor moderating the panel didn't seem phased in the slightest.

Daniel said...

Here's a more specifically tailored alternative:

A reigning orthodox among conservative and libertarian academics is that academia discriminates against them.

Can you think of a conservative/libertarian professor that will come out and say this is vastly overblown?

Daniel said...

I really can't emphasize enough how silly the anti-poverty one was. Who do you think does the work evaluating these programs? Academics.

Joey said...

Jim_gr:

I believe the commenter that David quotes was careful on that issue. He said

IQ tests do not measure anything real about human intelligence.

That doesn't imply that the commenter thinks that IQ perfectly captures "intelligence" (which in this context has the technical name "general intelligence"). That would be a caricature of his position.

Joey said...

I should say, I'm not accusing you of deliberately caricaturing his position. I just wanted to point it out because it's a common theme.

Anonymous said...

IQ tests do not measure anything real about human intelligence.

Which is rather absurd, if you think about it. You say things like that when the test's results are politically unacceptable, rather than when they're wrong.

The same people would have no problem saying that blacks, on average, make superior athletes to asians. Suggest that it would astonishing if similarly large divergences in intelligence didn't exist be between racial groups and those same people will deem you racist and bigoted.

How many hundreds of ways do you have to manipulate IQ tests, because you don't like what they're telling you, before you give up and say "it's not the test, it's the people taking it"?

David Friedman said...

Daniel:

I think you are misreading the anti-poverty one. The point isn't that there are no bad ideas, but that the reason none of the ideas has really worked is either lack of money or inadequate sincerity--that poverty could be solved if only enough money was spent doing it by people who really wanted to.

I agree the phrasing is a bit ambiguous, since "government programs have failed"could mean either "some government programs" or "all government programs."

David Friedman said...

Daniel:

I specified "in many parts of the academy." I think economists are a good deal less committed to these than people in some other fields--as suggested by the trouble Larry Summers got into as a result of questioning number 6. As best I can tell, it wasn't the audience of economists he was speaking to that got upset.

David Friedman said...

Daniel:

The point is the claim that these statements are both orthodox and false--in most cases, I would argue, not only false but indefensible. There is, for example, massive evidence on the heritability of IQ.

You offered some arguments on your blog against the view that conservative and libertarians are discriminated against in Academia, but what you offered was not evidence against the claim but other ways in which the evidence for the claim might be explained. So I don't think that qualifies.

Going back to the gender case ... . You have surely observed statements being made about m/f wage differences and similar matters, which simply take it for granted that a difference in outcome reflects discrimination. Most of the time, in my observation, the possibility that it is due to innate differences is simply ignored.

Try introducing it--and you observe what happened to Summers.

Tim Lambert said...

The statements are a mixture of straw men (IQ not heritable) and true statements (government programs).

Tibor said...

It is not an argument presented on the list in the blog post, but It reminded me of something.

There is one physicist from Pilsen (Luboš Motl) who also taught at Harvard for a while. I am not a physicist so I cannot judge his professional qualities well enough, but he is reputedly a very good one too. However he had to quit his position at Harvard because he publicly claimed that "women are less capable of doing physics and maths than men" and the press and academic pressure was just too much for the department, so he basically got fired.

I for one do not agree with his claim, I only base it on personal experience and no research (in fact I doubt anyone would be "allowed" to do such a research today, or that any journal would publish it), but it seems to me women (on average of course) are not worse than men, they might be less interested in these subjects though which means fewer of them doing it (or doing it seriously). Also this might explain it to some degree as well (but I am not sure how singificant it really is): http://ilaba.wordpress.com/2011/03/28/why-im-not-on-mathoverflow/

But the fact that a good professional has to go because his opinions are not mainstream is troubling. One could argue that such a person will actually be biased against female students, but I don't think it is his case. Also it is an undeniable fact that there are far fewer female phycists and mathematicians than there are male and it is an interesting to ask why it is so. The official truth is basically "male opression" which however does not seem very plausible to me, or at least not at our school, and I think few mathematicians (male mathematicians) would mind more women in their departments (if anything I think they would support it). It is also true however that the article from the link above says something else - and it is from a woman's perspective. But one thing seems certain for me - allowing only one "official truth" is very unlikely going to lead to correct answers. I think both the position of women being on average less suited for maths (or interested in maths) and being pushed away from it by the school system (which in my opinion does a good job in making anyone, male or female alike, hate mathematics without ever learning it for real) or society (which seems far less plausible to me) are apriori defensible and even if not, one of them should be defeated by good arguments and not by making the issue a taboo.

sbloch said...

I think all of those statements are worded in an absolute, extreme form that few academics would agree with. I suspect that if you gave them each a bit more nuance, most academics would agree with most of them -- but then it would be harder to show that they're objectively false.

Mike said...

I think you are wrong. The items on the list, the way you phrased them, are hotly debated and would not get anyone in trouble.

What would get you into trouble is taking one of the items plus an assumption: 'Blacks score lower on IQ tests. The reason for the lower scores is innate differences in intelligence across races.'

Of course this does not mean that this proposition should not be studied or debated. It's just that the list of items is much smaller and specific than you claim.



Mike said...

What got Summers into trouble was a more specific proposition. He did not say "I believe not all differences between men and women are culturally determined"

There has been significant increase in the tail of female IQ distribution to invalidate his point. Again, I am not suggesting that this should not be debated, but it is fairly narrow and far from being obviously correct.

David Yosifon said...

I'm afraid this reads to me like a superficial caricature of what somewhat might think is orthodoxy in the academic world, but it does not reflect my own experience of academia.

Take race, for example. Sophisticated treatments of race understand it as a historically, culturally, politically, ideologically contingent concept that intersects with subjective human experience of, among other things, color. "Race" in point of fact, is not found on the top of any mountain or under any microscope. Genetic variance may be found under a microscope, but such variance becomes "race" only when it is subsumed within a discourse of power. This is basic stuff, one need not look far on our campus to hear it. To say that the orthodoxy is that is "race is not real" is, evinces perhaps a failure to listen with a sympathetic ear to what is being said.

Consider further affirmative action -- today one hears affirmative action spoken of as a device for achieving institutionally useful diversity. And again, it is only a superficial perspective which would see this call for diversity as a diversity of "color." It is a call for a diversity of perspective and experience, which in our society routinely attend differences in "race," culturally defined. We should promote affirmative action efforts to develop diverse intellectual perspectives, and there is too little of this, but the view of affirmative action described above is simply an outmoded one that one does not really hear in real classrooms or faculty discussions, in my experience.

As for social policy -- does anyone talk about government programs without discussing the deep problems of human stupidity and institutional corruption. Maybe, but I don't really hear such superficial talk (maybe I have been too well influenced by D. Friedman!)

David Friedman said...

David Yosifon writes:

"It is a call for a diversity of perspective and experience, which in our society routinely attend differences in "race," culturally defined. "

That simply isn't true, as I argued in an earlier post. A black law school applicant whose parents are California lawyers brings less diversity of perspective and experience to the school than a white applicant whose parents are farmers in Mississippi or Jewish immigrants from the ex-Soviet Union. But the former counts towards diversity, in the USNews rankings and in our school's boasting of how diverse it is, the latter do not.

A black applicants who managed to make it out of poverty, such as Thomas Sowell or Clarence Thomas, brings diversity of attitude--but that isn't either what schools are specially looking for or what they count and report.

David Friedman said...

"Race, culturally defined."

That could mean two things. One is that someone's race is defined by his culture. That pretty clearly is not true in the U.S. at present. West Indian immigrants and the descendants of northern free blacks from before the post WWI immigration from the south are classified as blacks, but they are culturally quite distinct from the inner city blacks whose ancestors came up from the south in the 20th century (and from each other, of course).

Or it could mean that the definition of race is itself cultural. There is some truth to that, but not much. Who is classified as black in the U.S. correlates pretty closely with ancestry from sub-saharan Africa, which is an objective fact. All three of the groups I described above have substantial ancestry from sub-saharan Africa, and all three are routinely classified as blacks.

People from southern India have similar skin color, having also adapted to an environment with a lot of sunlight, but aren't particularly close to sub-saharan Africans in other ways. I doubt that our school would classify them as blacks, or that you or I would, and I suspect that people who do are simply making a mistake, drawing an incorrect conclusion from the similarity of skin color.

It's true, of course, that exactly where one draws the line is culturally determined--in the U.S., someone is usually classified as black if any significant part of his ancestry is from sub-saharan Africa. But in that sense, most binary categories are culturally determined, since they involve drawing a line somewhere in a continuum. That doesn't mean they aren't real.

David Yosifon said...

Before law school I worked for a newspaper in Boston that served the black community. The conservative, African American owner of the paper used to say to me, "white people see race as color, black people see race as who you stand with."

I took that perspective to law school and thought it was pretty smart so I spoke it. But I learned in Lani Guinier's Legal Ethics course that this was too limited a view. That race is not only about how you see yourself, but about how others see you. Your experience and perspective is not only about how you see, but about you have been seen. How others treat you.

And so it may be "simply" true that a middle class black student from CA brings less diversity of perspective than a poor white student from the South, but it may not be "deeply" or "complicatedly" true.

That said, I'm all for ever deepening our conception of diversity, and I see my colleagues actively struggling for same, rather than resting complacently in the concept of "affirmative action" described in this post.

David Friedman said...

Tibor comments that he doubts women are less good at physics and math then men. I don't know what the evidence is on that question. The claim that I find convincing, mainly on theoretical grounds, is that the distribution of ability for women tends to be tighter than for men, with fewer at both the very high and very low ends.

One implication is that if average ability is the same, there will be fewer women who are extraordinarily able in those fields--hence fewer women than men who are qualified to be math professors at Harvard.

The theoretical argument is pretty simple. Living creatures are "as if designed" by their genes for reproductive success. The scarce biological resource for reproduction is womb space--a reproductively successful man can father an enormous number of children, a successful woman can bear a much smaller number. That means that being male is, reproductively speaking, a higher risk gamble than being female, which means that the optimal male design will be more willing to try for very desirable characteristics even at the risk of getting very undesirable ones. That gives you distributions with wider tails.

My understanding is that the evidence on things such as IQ supports that conclusion, but it isn't a subject I have any expertise in.

David Friedman said...

Tim:

I agree that the government program claim is the most defensible--but which others do you think are true, or at least defensible?

The one which strikes me as implausible on theoretical grounds and inconsistent with such evidence as we have is the lack of important m/f differences--which is implicit in most discussions of gender discrimination.

Do you think that one is defensible? Do you disagree that openly arguing against it will get someone in trouble in substantial parts of the academic world?

David Friedman said...

sbloch:

The initial claim is not that most academics would make these claims in the strong form, but that criticizing the claims as part of the public discourse on campus would be strongly opposed.

You will note the final sentence of what I quoted, which distinguished between what people believe in private and what they opposed publicly questioning.

David Friedman said...

Mike:

What Summers said was that one possible explanation, although not the most important, of the m/f difference in professors in elite schools in some fields, was innate m/f difference. I don't see how one can view that claim as a reason to attack him unless one is claiming that no significant difference of that sort can exist--or else that one shouldn't be allowed to mention it even if it does exist.

And the question isn't whether there has been a significant increase in the tail of the female IQ distribution, it's whether there is any good reason to believe that the shape of the distribution of intellectual abilities is the same for males and females. I've sketched in another comment in the thread why I don't think there is--why one would expect a wider spread for male distributions.

And it doesn't take a very large difference in the standard deviation of the distribution to give a several fold difference in the number far out on one of the tails, which is all it takes for that to be one of the explanations.

Anonymous said...

Of course this does not mean that this proposition should not be studied or debated.

The problem is that the academic left very much demands that proposition not be studied or debated. Get anywhere near the notion that mental abilities and traits have genetic variance among races just as physical abilities do and you'll get a shitstorm of grief for it. Unless, that is, the study shows blacks on top, then its ok. For example, a recent study scoring blacks as highest on an extroversion scale garnered no complaints.

Mike said...

"And the question isn't whether there has been a significant increase in the tail of the female IQ distribution, it's whether there is any good reason to believe that the shape of the distribution of intellectual abilities is the same for males and females."

I think you are assuming the shape of the distribution is static or pre-determined. I can imagine social factors that could cause an increase in std deviation over time. And, if I observe a large increase in the tails, and assuming a normal distribution, that would imply either the mean has increased substantially or that the std deviation has increased.


"The sex ratio of adolescents who scored 700 on the SAT-M at age 13 years, a feat achieved by only 1 in 10,000 students, has shrunk from 13:1 in 1983, to 5.7:1 in 1994, to 4:1 in 1997, to 2.8:1 by 2005 (Julian Stanley, quoted by R. Monastersky, 2005; see also Gates, 2006b)."

Simon Andersson said...

"--If government programs for the elimination of poverty have failed, it is for one of two reasons: 1) they have not been sufficiently funded; or 2) those implementing the programs have not been sincere."

My reading of this is that the lefty academic systematically overlooks systemic explanations and insists on explaining away the failure of each program with particular circumstances (not enough funding, incompetent/insincere implementation, whatever).

The taboo conclusion that we must always repress is this: Maybe expansive, paternalistic government is just a bad idea.

Examples of systemic explanations would be the ones studied by public choice economists (Buchanan &co), the possibility that problems like inequality can be intractable, and problems with centralism relating to limitations on human knowledge (as observed by Hayek, Taleb and others). I think it is characteristic of right-thinking academic progressives to have a blind spot for these possibilities.

It resembles the way, in the eighties, some people sympathetic to communism would still blame failures of individual socialist states on particular problems (Soviet harvests failed because of bad weather! The CIA sabotages progress in Nicaragua!) while dancing around the (for me) obvious conclusion that maybe communism is a bad idea.

It also made me think of this blog post.

Simon Andersson said...

BTW, I'm not sure that the main problem is the promotion of a particular set of conclusions or the discrimination against a particular group of people (say, conservatives).

Consider that Larry Summers is far from conservative or libertarian: he is a career Democrat who worked in the Obama administration. You can spend a lifetime mouthing politically correct phrases; a few heterodox comments can still derail your career. Conversely, a radical conservative or libertarian could still survive if he is slick enough.

The problem lies in creating an incentive structure resembling that of a totalitarian society: people with a knack for adapting, often unconsciously, to ortodoxy are rewarded and independent thinkers punished. In the end, most people in influential positions will be natural totalitarians.

The good news is that if and when the U.S.A decides to become a totalitarian society, very few personnel changes will be necessary :-)

suckmydictum said...

David,

The geneticist Alan Templeton at Washington U has shown that interacial genetic variation is not quantifiably more pronounced than intraracial genetic variation. Does a result like this count?

Anonymous said...

"The sex ratio of adolescents who scored 700 on the SAT-M at age 13 years, a feat achieved by only 1 in 10,000 students, has shrunk

Of course it shrunk. They keep fiddling with the scoring system to fatten up the curve under 800, making it much easier to score 700 than it was in 1985.

That's also why the modern SAT has two verbal to one math component, vs the 1:1 in 1985: girls score better on verbal, so two verbal sections offset boys advantage in the math section, evening out the scores and closing the overall scoring gap.

David Friedman said...

David Y writes:

"And so it may be "simply" true that a middle class black student from CA brings less diversity of perspective than a poor white student from the South, but it may not be "deeply" or "complicatedly" true. "

That might be true of the first middle class black student from CA we accept, but how about the twentieth? As compared to the first student from one of the non-racial categories I described?

David Friedman said...

Mike:

If educational choices and innate ability both affect how likely one is to get a high score on the SAT young, as they surely do, then one would expect a change in the former to affect the m/f ratio of how many achieve that, even if the latter remained fixed.

Some striking evidence on the other side--I don't know if it has been contradicted by later work--was a correlation between relative math and verbal abilities and a ratio of finger lengths that's a marker for hormone balance in utero. The correlation applied both across genders and within gender. Whether you count that as genetic m/f variation or not is a little complicated--one of the things affecting the hormone balance in the womb is the gender of the fetus, but not the only thing--but it's innate by the time the child is born.

David Friedman said...

"The geneticist Alan Templeton at Washington U has shown that interacial genetic variation is not quantifiably more pronounced than intraracial genetic variation. Does a result like this count?"

Quoting from a webbed story on him at his university's web page:

“I’m not saying there aren’t genetic differences among human populations,” he cautioned. “There are differences, but they don’t define historical lineages that have persisted for a long time, which is one criterion for race in a scientific sense.”

...

"The best predictor of overall genetic differences is how far apart geographically the ancestral populations are."

The final bit corresponds to the point I made about sub-saharan Africans vs people from southern India, although that's not his example. The implication is that skin color is not a reliable marker for race, not that race doesn't exist.

His central claim seems to be that the geographical separation that creates what we call races isn't long term--but in the context it looks as though his idea of long term is many thousand of years.

suckmydictum said...

David Friedman: I don't believe the academic orthodoxy; race can be discussed as a scientific idea.

Alan Templeton: historical perceptions of race disappear when variations between and within races are compared in a controlled way. Historical perceptions should yield to what we know now about our genome. It is perhaps well-posed to discuss long term lineages and geographies, but not skin color and other "racial markers".

David Friedman: but your "long term" geographic contributions to understanding of race are only long term in an evolutionary context. Really, we should consider changes in skin color to be shorter term than you propose.

Alan Templeton: skin color and practically every physical characteristic we have historically used to dilineate race doesn't have any statistical genomic validity. I humbly posit that "race" as historically conceived before my research does not really exist.

David Friedman: ok, but just because everything we thought we knew about racial difference before modern genomics is wrong does not mean that race does not exist.

Mike said...

"If educational choices and innate ability both affect how likely one is to get a high score on the SAT young, as they surely do, then one would expect a change in the former to affect the m/f ratio of how many achieve that, even if the latter remained fixed"


That may very well be true. The fact is we just don't know the answer. Btw the differences across countries is even more stark (think Sweden).

Summers' comments about std deviations and differences exploding in the tails all sound very intuitive and appealing, but if he had mentioned this 5 fold decrease in the differences in the tails over a short time period, his argument would have been significantly less powerful. There is evidence of environmental factors that can be very significant. And one of those factors could very well be attitudes of co-workers in an academic setting, including the attitudes of university presidents.

I can imagine who are aware of such research being pissed off at someone suggesting that the differences can be largely explained by innate abilities. Of course this does not justify him being crucified the way he was. But at the same time he is not someone doing research on this topic or simply shooting the breeze with his department buddies at the water cooler. He is the president giving a deliberate speech on an important topic. He should have done his homework and mentioned other important explanations.


David Friedman said...

"I can imagine who are aware of such research being pissed off at someone suggesting that the differences can be largely explained by innate abilities."

Summers didn't suggest that.

" He should have done his homework and mentioned other important explanations. "

He did.

As best I recall, innate difference was the third in his list of possible explanations.

David Friedman said...

" ok, but just because everything we thought we knew about racial difference before modern genomics"

You seem to be assuming that, prior to modern genomics, people thought that the fact that people from Sub-saharan Africa and people from southern India and Australia all had dark skin, that meant they were the same race. Is there any evidence that any serious scientist believed that? Did Darwin?

The fact that skin color by itself is not an adequate marker for race doesn't imply that race doesn't exist, any more than the fact that height isn't.

Stella said...

Just to clear up some details, this is the gist of what Summers said:

"the first is what I call the high-powered job hypothesis. The second is what I would call different availability of aptitude at the high end, and the third is what I would call different socialization and patterns of discrimination in a search. And in my own view, their importance probably ranks in exactly the order that I just described."

So he gives three possible explanations for the relative lack of women in prestigious science and engineering academic positions. He says that in his opinion (and he's tentative about it) the availability of aptitude at the high end is probably a more important factor than discrimination but a less important factor than men and women's choices in balancing work and family.

When he explains this second possible reason in more detail, he says that there's evidence that men and women have different standard deviations in their distributions of a variety of areas that "are and are not plausibly culturally determined." Again, he's tentative, saying "Because if my reading of the data is right-it's something people can argue about-that there are some systematic differences in variability in different populations..."

He mentions socialization as theory to explain these differences in variability, and then says "there is reasonably strong evidence of taste differences between little girls and little boys that are not easy to attribute to socialization..." Notice, he's talking here about *taste differences* as opposed to innate talent or intelligence. He characterizes this view by saying "Somehow little girls are all socialized towards nursing and little boys are socialized towards building bridges. No doubt there is some truth in that. I would be hesitant about assigning too much weight to that hypothesis..." Then he lists a couple of reasons he doesn't think socialization explains the entire difference.

So that is what the man actually said, for future reference.

Joey said...

The good news is that if and when the U.S.A decides to become a totalitarian society, very few personnel changes will be necessary :-)

HAHA.

Mike said:
if he had mentioned this 5 fold decrease in the differences in the tails over a short time period, his argument would have been significantly less powerful

It's common knowledge that scores and distributions are regularly tweaked. As explained in "The Recentering of S.A.T. Scales and Its Effects on Score Distributions and Score Interpretations" (page 15), in part, this is in order to make them conform with "well known knowledge about gender differences". You can't use re-scaling of scores to correct gender differences as an argument against gender differences.

Stella said...

Also, I'll throw in my own story to add to this conversation--I'm a female engineering graduate student.

I got my undergraduate degree from a religious college where both students and teachers are generally socially conservative and traditional. Female students were more likely than your average young woman to want to be a stay-at-home mom, and male students were more concerned than your average twenty-year-old man with the need to be a good provider.

Women made up about 25% of my undergraduate engineering class, which is lower than the national average. There were also no female professors.

The department was *desperate* to increase female enrollment. While they had a hard time finding a qualified female PhD who was a member of our religion and wanted to be a full-time faculty member, they recruited female part time professors to teach intro classes and to serve as student mentors. I was offered a job as a TA for the intro class because I was a woman, because my professor wanted the freshman girls to know that, yes, women study engineering. The freshman seminar included at least one hour dedicated to persuading women to study engineering. Every year, I was asked for suggestions on how to increase female enrollment. I was sent out to the freshman chemistry classes to invite female students to special dinners where they were fed free food and listened to testimonials from professional female engineers.

Never once in my engineering education have I felt like my sex was anything but a positive. In my undergraduate classes, my professors were more likely to remember my name (and how good of a student I was) than they were to remember my male classmates. When I decided I wanted to go to graduate school, my professors were endlessly helpful. In fact, they advised me to apply to more prestigious universities than I had originally intended.

The closest thing to a sexist attitude I ever encountered was in one of my best friends. He thought that girls in general weren't as good at math, but if you were a girl *who was studying engineering* he concluded you were plenty smart. He thought I was smarter than he was. He was also resentful that every girl in our program managed to get an internship, including those girls who weren't as academically strong, and attributed this to affirmative action. I think he was partly right about that.

The point of my story is that, in light of my experience, I intend to agree with Summers that the most important factor is different expectations for work/family life balance. My department had such a hard time recruiting girls into engineering because the religious student body had girls who put more weight than average on family life. If what you really want is to work part time or stay at home, why would you suffer through what was famously the hardest major on campus? (This is why one of my friends from freshman classes switched to elementary education.) Similarly, men were more likely to major in engineering because they wanted to be good breadwinners, rather than being especially interested in the subject.

Kaj Sotala (Xuenay) said...

Both the claim that IQ measures something real and the claim that it is heritable are broadly accepted at least among psychologists. See e.g. the "Mainstream Science on Intelligence" statement, which was published both in Wall Street Journal and a leading peer-reviewed psychometrics journal:

http://en.wikipedia.org/wiki/Mainstream_Science_on_Intelligence

Similar research continues to be frequently published, see e.g. the recent review that was published in American Psychologist:

Nisbett, R. E., Aronson, J., Blair, C., Dickens, W., Flynn, J., Halpern, D. F., & Turkheimer, E. (2012, January 2). Intelligence: New Findings and Theoretical Developments. American Psychologist, 67, 130-159. http://scottbarrykaufman.com/wp-content/uploads/2012/01/Nisbett-et-al.-2012.pdf

Or pretty much any issue of the journal Intelligence.

Kaj Sotala (Xuenay) said...

Also the original "Intelligence: Knowns and Unknowns" report, of which the 2012 Nisbett et al. report is an update:

http://en.wikipedia.org/wiki/Intelligence:_Knowns_and_Unknowns

Anonymous said...

Race is not scientific. there is no such thing. a white person such as me would have more in common with Mongols than with a white person who is from scandaniavia, even though we are both white. just like a black person is not at all some how "close" to another black person. for example a black person from southern sudan will have very little in common with a black person from zambia.

skin colour is a VERY bad indicator of anything apart from skin colour.

Tim Lambert said...

David, the statement "All differences between men and women are culturally determined" is a strawman, not "orthodoxy". Orthodoxy would be some thing like "gender discrimination is real". This is true -- see, for example, Goldin and Rouse who found:

"Discrimination against women has been alleged in hiring practices for many occupations, but it is extremely difficult to demonstrate sex-biased hiring. A change in the way symphony orchestras recruit musicians provides an unusual way to test for sex-biased hiring. To overcome possible biases in hiring, most orchestras revised their audition policies in the 1970s and 1980s. A major change involved the use of blind' auditions with a screen' to conceal the identity of the candidate from the jury. Female musicians in the top five symphony orchestras in the United States were less than 5% of all players in 1970 but are 25% today. We ask whether women were more likely to be advanced and/or hired with the use of blind' auditions. Using data from actual auditions in an individual fixed-effects framework, we find that the screen increases by 50% the probability a woman will be advanced out of certain preliminary rounds. The screen also enhances, by severalfold, the likelihood a female contestant will be the winner in the final round. Using data on orchestra personnel, the switch to blind' auditions can explain between 30% and 55% of the increase in the proportion female among new hires and between 25% and 46% of the increase in the percentage female in the orchestras since 1970."

suckmydictum said...

"You seem to be assuming that, prior to modern genomics, people thought that the fact that people from Sub-saharan Africa and people from southern India and Australia all had dark skin, that meant they were the same race. Is there any evidence that any serious scientist believed that? Did Darwin?"

For Darwin, in The Descent Of Man, the evidence is not clear. He had no problem sorting groups into "savage" and "civilized" races, but believed that a gradation existed between established races.

As for other "serious" scientists, of course. George Culvier categorized men into white, yellow, and black races (making no South Indian destinction). Christoph Meiners explicitly broke humanity into two (white and black) races. Benjamin Rush thought that dark skin didn't occur naturally but was a skin disease transmitted to children through their parents. It doesn't really matter and I'd have to check to be sure, but I think all these men lived within Darwin's lifetime.

Even if none of this were true, and these men had somehow managed to figure out that skin color was not a good predictor of what is called race, your point lacks force. Democritus believed the world was made of atoms. That didn't make him scientifically correct about atomic theory.

Ben Nader said...

My views on the orthodoxies:

1. There is no such thing as race.

2. Affirmative action is a very poor way of achieving social justice, potentially even exacerbating problems it intends to ameliorate, but it would be justified if both (a) really improved things for discriminated-against and deprived social groups and (b) was the only means we could use to deal with this sort of injustice.

3&4. I have no view on IQ really, since I have very little knowledge about it.

5. I don't think this is really an orthodoxy; I think most mainstream economists would agree that many government welfare programmes are badly designed, and in cases where they've failed this is a key contributing factor. And I agree with DDF that enough sincerity & enough money might work.

6. Gender undeniably is a social construction like race, but obviously male- and female-bodied people have different roles in the purely physical elements of reproduction. Also, inasmuch as hormones affect or control behaviour, male-and female-bodied people will be hit with different hormones and so be different due to that. But those hormones are effectively optional now. And bear in mind that gender has nothing to do with what body one lives in.

But the huge leaps and bounds we've made towards equality in, say, the last 50 years, together with the obvious oppression of women in society now, imply to me that with a good bit more effort we could almost completely eliminate gender and sex inequality.

Tibor said...

David Friedman: That is an interesting hypothesis I have not heard of before with explaining potential, so thanks for that. However it sure does not explain everything. It would explain fewer female Newtons and Gausses but it would not explane fewer female mathematicians in general, I think. What we are talking about here is talents. Something that comes with your genes and that's it. But I think that although talent is what makes a difference between very good and brilliant, it is less important on lower levels. Definitely not all math professors (maybe at Harvard they are) at even good universities are brilliant. I believe talent is not much more than 5-10% of the success, while other things like being organized (which it seems to me - just based on personal experience - that women tend to be better at then men), focused, having good teachers and motivation and a very many hours spent improving the skill. The lower variance in the distribution of talent in women would explain fewer women in maths to some degree, but the difference is far sharper than that. Of course another thing is that women -including those with talent- could be on average interested in an abstract pursuit that mathematics is. I still like the hypothesis of yours, but right now I think (not that you said otherwise) it can only explain it very partially. Of course, I might change my opinion based on more arguments or evidence.

One way or another, it is an interesting question that seems to be pretty much a taboo today. I think the problem might be the way people look at other people - while saying "women tend to be more average" does not mean that a particular woman is average, or that a particular man is either great or terrible, people (I think) often see it that way, they take statistics personally and that is why anything that suggests any statistical differences in people based on some basic easy to observe trait (like sex) is offending to a lot of them.

Also, I am not really sure whether women being more "centralized" is a bad thing from a woman's perspective (because it also means there are many more complete losers among men than there are among women), I suspect it would be considered offending by quite a lot of women, again because of the "taking-statistics-personally approach" - noone wants to feel just average. Then again, maybe I am wrong. Nobody will know if things like this stay a taboo.

Tibor said...

Also as far as IQ goes, I have doubts about the methodology. If you take a lot of IQ tests, you will improve your score pretty fast - you will learn how to do IQ tests, but that does not mean you're suddenly (much) smarter than you were.

I see intelligence in general as something that can be learned, even though different people are able to learn it at a different pace and therefore to a different degree. But it seems to me IQ tests only give out information about the current level of your abstract thinking skills...and only a very partial information anyway.

I think it would be a very interesting to make an experiment with high school graduates - measure their IQ and then measure it again after they finnish college. I suspect people with math degrees would improve much more than people who studied e.g. journalism.

I believe abstract thinking is a skill like any other and it can be developed. So a lot people who are labeled "stupid" by an IQ test might be just unskilled (some might say lazy) - they have never practiced the skill and therefore have never improved it. Similarly - a lot of people with high IQ may not be brilliant (defined as having exceptional talents for abstract logic), they just invested a lot of energy into improving the skill IQ tests (imperfectly) measure, that is logical thinking...or into the skill IQ tests measure perfectly - the ability to do IQ tests. I suspect quite a few people in mensa have spend a considerable amount of time improving their IQ test taking skill before taking the mensa IQ test, but maybe that is just my bias against that organization :)

Now, don't take me wrong - I don't think people come to the world as blank shells, some have talents for this and some have for that. But it is really extremely hard to measure the talents and not the skills, which depend considerably on many things other than the talents. And I don't think IQ tests do the job very well.

Mike said...

"the first is what I call the high-powered job hypothesis. The second is what I would call different availability of aptitude at the high end, and the third is what I would call different socialization and patterns of discrimination in a search. And in my own view, their importance probably ranks in exactly the order that I just described."


I agree about the last part not being as important. But, just like std deviations I mentioned previously, these preferences for high powered jobs can be driven by social factors. Preferences for nurture, child care, etc. are not created in a vacuum. Just like David was implicitly assuming that the std deviation differences are innate, Summers is assuming these preferences are largely innate. Again we see large differences m/f ratios of high powered jobs over time and especially across countries. Presenting such evidence would have made Summers' argument significantly less powerful.


Btw, I think these preferences have a larger impact on men. Let me give you an example. I was offered a job in a new country, which meant the end of my wife's legal career. She became a housewife. We could have stayed and I could have been the one to stay at home. I would not have been happy with that decision. The way I was brought up the man is the bread winner. If I had become a stay at home dad, my family and friends would have judged me as being a failure, and I would have felt self worth. Women don't have this problem.

David Friedman said...

Tim:

Of course gender discrimination is real--probably in both directions depending on context. But what one routinely sees is the claim that differential outcomes demonstrate gender discrimination and measure the size of its effect.

That depends on the unstated assumption that other explanations can be eliminated, part of which is the assumption that there are no important differences between the distribution of intellectual abilities m/f.

And pointing that out would be risky in some of the sort of contexts being discussed.

David Friedman said...

"And bear in mind that gender has nothing to do with what body one lives in."

I assume you are distinguishing psychological gender from biological gender. I agree that it's a useful discrimination--but your "nothing to do with" is obviously false, since the two correlate highly. Most people with XX think of themselves as female, most people with XY think of themselves as male.

Simon Andersson said...

I think Steven Pinker's The Blank Slate is a great read.

The book specifically addresses the notion that the "blank slate" view is a straw man. Pinker notes that people who invoke nature and nurture as opposed to nature alone have been "picketed, shouted down, subjected to searing invective in the press, even denounced in Congress" as well as "censored, assaulted, or threatened with criminal prosecution."

He concludes that

"When it comes to explaining human thought and behavior, the possibility that heredity plays any role at all still has the power to shock. To acknowledge human nature, many think, is to endorse racism, sexism, war, greed, genocide, nihilism, reactionary politics, and neglect of children and the disadvantaged. [It] strikes people not as a hypothesis that might be incorrect but as a thought it is immoral to think."

Simon Andersson said...

Should be "as opposed to nurture alone"

David Yosifon said...

David,

Certainly I agree that at some point there is a declining marginal contribution to diversity when the next admitted student is similarly situated to prior admitted students. It is not a simple calculus, given what we understand about the importance of critical mass to achieving the kind of expressive confidence that would deliver on the promise of diversity. But my point is that these considerations are a live, active part of our institutional conversation about diversity, and that this conversation does not therefore represent the caricature about "affirmative action" that you described in your list of "official lies."

Eric Rasmusen said...

One of the oddities of the discussion of women in academia is the notion that in our culture boys are encouraged to be scientists and mathematicians. Academics may think that, but are those really the jobs a real he-man would pursue? It's more plausible to make the case that those jobs are considered effeminate and that that's one reason fewer black men have college degrees than black women.

Even if we put aside being a lumberjack or a football player, law or business is more the kind of career men are steered towards. To be a mathematician requires fanatical devotion, a love for the subject and a willingness to endure people wondering why someone as smart as you earns so little and doesn't yet have a secure job at age 26. It isn't something you get sucked into because of cultural expectations.

Rebecca Friedman said...

Ben Nader,

"But the huge leaps and bounds we've made towards equality in, say, the last 50 years, together with the obvious oppression of women in society now,"

That's odd. I don't feel very oppressed.

At least, not as a -woman-.

(Unless you mean societies very different from the modern American one, in which case I'm willing to concede you the point. But I was brought up here; here, I -can- speak to.)

Mike,

As a woman, I'm pretty sure there are social groups (or sub-societies) where my choosing to be a stay-at-home housewife would expose me to serious criticism. (This is the basis of many of my issues with modern feminism - wasn't the idea to -expand- the range of choices available?) Fortunately for me, the groups that would have problems with me making that choice aren't the groups that contain the people on whose opinion I rely for my sense of self-worth. Unfortunately for others, I believe a good number of women find themselves in the opposite position. I agree that the pressure is higher for men, but sadly, I believe it applies to both.

Ben Nader said...

"I assume you are distinguishing psychological gender from biological gender. I agree that it's a useful discrimination--but your "nothing to do with" is obviously false, since the two correlate highly. Most people with XX think of themselves as female, most people with XY think of themselves as male."

Fair enough point. Most people are cis. I should have said something like "there is no necessary link". Gender is just a set of social attitudes we have evolved, imposed, created, found, and society has generally said that sex = gender (i.e. everyone is necessarily cis).

"That's odd. I don't feel very oppressed."

I'm glad to hear that, but you'd agree that not all women (or even necessarily any of them) would have to feel oppressed for it to nevertheless be the case that they are oppressed.

I think it's pretty undeniable that – though much reduced – there is a level of oppression of women in the US (I don't mean through law). E.g. http://www.newstatesman.com/laurie-penny/2013/03/steubenville-rape-cultures-abu-ghraib-moment

Joey said...

Ben:
In light of the Newtown Tragedy, are you going to claim that society oppresses first graders as well?

Tim Lambert said...

David, since we agree that gender discrimination exists, then if we observe differential outcomes then it is likely that discrimination is partly to blame. There is no need to assume that women and men are the same.

David Friedman said...

Tim:

Gender discrimination exists, in both directions. But the routine claim is that the size of outcome differences measures the effect of discrimination and that the existence of outcome differences in some particular context demonstrates the existence of discrimination in that context.

Both of those require assuming away alternative explanations.

James said...

Take race, for example. Sophisticated treatments of race understand it as a historically, culturally, politically, ideologically contingent concept that intersects with subjective human experience of, among other things, color. "Race" in point of fact, is not found on the top of any mountain or under any microscope. Genetic variance may be found under a microscope, but such variance becomes "race" only when it is subsumed within a discourse of power. This is basic stuff, one need not look far on our campus to hear it. To say that the orthodoxy is that is "race is not real" is, evinces perhaps a failure to listen with a sympathetic ear to what is being said.

This is all true of most categories. "Car" is also an historically and culturally contingent concept that intersects with subjective human experience. Where does "car" stop, and "truck" begin? Does "car" have a necessary and sufficient definition? Given a list of 100 car-like objects, would everyone parse them into exactly the same groups of "car" and "not-car"?

Yet few would deny that "car" is a useful distinction.

The orthodoxy is that "race" is an especially flawed concept, when I say, for definite reasons, it is unexceptionable. Mainstream discourse on race adopts far stricter than normal criteria for classification, which if used consistently would render communication impossible; and also perpetuates outright fallacies.

Will McLean said...

“The classification into races has proved to be a futile exercise for reasons that were already clear to Darwin. Human races are still extremely unstable entities in the hands of modern taxonomists, who define from 3 to 60 more races. To some extent, this latitude depends on the personal preference of taxonomists, who may choose to be 'lumpers' or 'splitters'. Although there is no doubt that there is only one human species, there are clearly no objective reasons for stopping at any particular level of taxonomic splitting. In fact, the analysis we carry out..for the purposes of evolutionary study shows that the level at which we stop our classification is completely arbitrary." (Cavalli-Sforza)

Natural Language Processor said...

Race doesn't exist! Lulz.

http://www.amazon.com/The-History-Geography-Human-Genes/dp/0691087504/ref=cm_cr_pr_product_top

http://genomebiology.com/2002/3/7/comment/2007

http://evoandproud.blogspot.com/2010/10/evolution-of-cavalli-sforza-part-vi.html

http://discovermagazine.com/2009/mar/09-they-dont-make-homo-sapiens-like-they-used-to#.UV3s3nd4Ido

http://seattletimes.com/html/health/2009268868_apmixedracedonors.html?syndication=rss

http://www.nytimes.com/2013/02/15/science/studying-recent-human-evolution-at-the-genetic-level.html?hpw&_r=1&

IQ isn't heritable! Lulz.

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3182557/

This entire thread is a game of semantics. How about instead of "race" we just use "phenotypically varied human populations." And instead of "IQ" we use "cognitive aptitude for problem solving and pattern detection." And instead of "gender" we use . . . well, I don't know about that one. Anyone who thinks that gender differences don't run biologically deep either failed Bio 101 or is addicted to ideology.

Mr. Friedman, the patience you have with your commentariat is truly to your credit. I would just post more links directing them to the latest scientific research.




James said...

The classification into races has proved to be a futile exercise for reasons that were already clear to Darwin. Human races are still extremely unstable entities in the hands of modern taxonomists, who define from 3 to 60 more races. To some extent, this latitude depends on the personal preference of taxonomists, who may choose to be 'lumpers' or 'splitters'. Although there is no doubt that there is only one human species, there are clearly no objective reasons for stopping at any particular level of taxonomic splitting. In fact, the analysis we carry out..for the purposes of evolutionary study shows that the level at which we stop our classification is completely arbitrary."

A perfect example.

The classification into types of vehicle has proved to be a futile exercise for reasons that were already clear to Henry Ford. Vehicles are still extremely unstable entities in the hands of modern engineers, who define from 3 to 60 or more kinds of vehicle. Road vehicles, planes and ships. Or, lorries, vans, motorbikes and cars. Or sedan, station wagon, hatchback? Totally arbitrary! To some extent, this latitude depends on the personal preference of the engineers, who may choose to be 'lumpers' or 'splitters'. Although there is no doubt that some things are vehicles and some are not, there are clearly no objective reasons for stopping at any particular level of taxonomic splitting. In fact, the analysis we carry out...for the purposes of Top Gear shows that the level at which we stop our classification is completely arbitrary.

A good reason to renounce prejudiced distinction between "different" vehicles?

Natural Language Processor said...

@ Will McClean

Cavalli-Sforza throws out nuggets like that to keep the PC police off his back. Look at the cover of his book, for God's sake, especially as regards Africa and everywhere else.

http://www.amazon.com/The-History-Geography-Human-Genes/dp/0691087504/ref=sr_1_3?ie=UTF8&qid=1365110885&sr=8-3&keywords=Cavalli-Sforza

Anonymous said...

Take colors on the spectrum, for example. Sophisticated treatments of colors understand it as a historically, culturally, politically, ideologically contingent concept that intersects with subjective human experience of, among other things, COLOR. "Color" in point of fact, is not found on the top of any mountain or under any microscope. Wavelength variance may be found under a microscope, but such variance becomes "color" only when it is subsumed within a discourse of power.

Or: "Clearly the existence of red and yellow proves there's no such thing as orange!"

Anonymous said...

"The classification into colors has proved to be a futile exercise for reasons that were already clear to Darwin. Colors are still extremely unstable entities in the hands of modern taxonomists, who define from 3 to 60 more colors. To some extent, this latitude depends on the personal preference of taxonomists, who may choose to be 'lumpers' or 'splitters'. Although there is no doubt that there is only one Visible Spectrum, there are clearly no objective reasons for stopping at any particular level of taxonomic color splitting."

Natural Language Processor said...

Here's a non-PC nugget buried deeper in Cavalli-Sforza's work. Here's him commenting on his genetic map:

"The color map of the world shows very distinctly the differences that we know exist among the continents: Africans (yellow), Caucasoids (green), Mongoloids, including American Indians (purple), and Australian Aborigines (red). The map does not show well the strong Caucasoid component in northern Africa, but it does show the unity of the other Caucasoids from Europe, and in West, South, and much of Central Asia" (pg 136).

Natural Language Processor said...

@ Ben Nader, re: your New Statesman link.

I think a better place to look for an explanation of why males rape females (it happens all over the world you know; it's--shockingly!--not just a Middle America football thing), anyway, I think a better explanation might be found in this excellent little book, "Demonic Males: Apes and the Origins of Human Violence":

http://www.amazon.com/Demonic-Males-Origins-Human-Violence/dp/0395877431/ref=pd_sim_b_7

Biology, Ben. Not culture.

Rejoicing in the lamentations of women goes way back. It's old school, Ben!

I think people like you need to a) admit that humans are animals, and b) watch more wildlife documentaries. Do these two things, and you'll stop feeling the need to search your soul and write nonsense about "rape culture" or whatever else you feel explains something that doesn't really need explaining, i.e., that humans often act like animals.

Will McLean said...

Natural:

Using phenotypes only reproduces the error of classic 19th c. racial theory: they assumed there was a black race, identifiable by superficial indications like dark skin and other features.

We now know that the "blacks" of South India, the Andaman islands, Melanesia and Australia were all closer genetically to Asia than Africa.

Will McLean said...

Anonymous:

The race/color analogy would be more effective if colors had a history of having sex with other colors and producing new hues that had never existed before.

David Friedman said...

Will:

As best I can tell by a little googling, Darwin, at least, distinguished between Australian aborigines and sub-saharan Africans. I don't know whether he thought other dark skinned groups were the same race as sub-saharan Africans or not.

Your point about colors and sex implies that one would expect intermediate cases, not that there is nothing for them to be intermediate between--and, of course, it is a point that Darwin discussed, explicitly disagreeing with contemporaries who thought human races were not cross-fertile.

Rebecca Friedman said...

Ben,

"I'm glad to hear that, but you'd agree that not all women (or even necessarily any of them) would have to feel oppressed for it to nevertheless be the case that they are oppressed."

Actually, I would prefer you do not assume what I would agree with. Especially when I do not, in fact, do so.

Your statement assumes a limit case wherein women are oppressed, but not a single woman notices. This assumes that the oppressors are sufficiently more capable – brighter, more subtle, etc. - than the oppressed to be able to oppress them without even a single one of them noticing. I find this extremely improbable, both on factual evidence (we have had the opportunity to observe that, in societies that are genuinely oppressive, at least some individuals do notice; how do you think our society got the way it is?), and on grounds of logic; I am unwilling to accept the proposition that women are, frankly, that incapable, and I do not logically see how your statement can be true without it. So no; if no woman feels herself oppressed, I will take that as plentiful evidence that there is not, in fact, oppression going on.

As a general rule, I find the viewpoint of “Oh no, you don't understand yourself/your life, let us experts tell you how it really works,” inherently problematic. I recognise that in a few exceptional cases – cases of mental illness, say, or a few situations where you're so involved you don't notice things becoming problematic until someone points it out – people may in fact not be aware of all the facts, and may need outside guidance. I still feel these cases are the exception, not the rule. As a general rule, the people who are actually living their lives will have a certain amount of experience and perspective on those lives that an unrelated person – simply doesn't.

As per the article you link – I am sorry, but while it is certainly describes an extremely regrettable state of affairs, I am not sure I would refer to it as evidence of “oppression of women” - though the article chooses to interpret the evidence that way. To throw something unrelated into the mix – the idea that a woman can rape a man has been treated as a joke, just in and of itself. Does that reflect at all a healthier culture? Is what we are dealing with here a situation of "Women are oppressed" – or one of rape (or, put more generally, people's rights to make choices about what they do and do not do with their own bodies) not being taken seriously enough?

Natural Language Processor said...

if no woman feels herself oppressed, I will take that as plentiful evidence that there is not, in fact, oppression going on.

@ Rebecca, that's far too logical and simple for the likes of Dan, who enjoy slicing their wrists on Occam's razor.

@ Willy

Using phenotypes only reproduces the error of classic 19th c. racial theory

Oh dear! I'll have to inform every biologist on the planet! This whole time they've been using phenotypes as part of their classification schemes; I'll let them know a publisher says it's just wrong, wrong, wrong. Goodness, Darwin was really on the wrong track with that whole finch thing.

Phenotypes are much more than physical characteristics. Genotype plus Environment equals Phenotype. So, phenotypes are a pretty good way to figure out the current state of genetic variance and how environment is affecting it.

We now know that the "blacks" of South India, the Andaman islands, Melanesia and Australia were all closer genetically to Asia than Africa.

A gross oversimplification of what anyone in the 19th or early 20th centuries thought, as someone has already pointed out. Hell, I've got a collection of missionary training guides on my shelves, part of my old books collection. Each one that discusses South India and India generally says that Indians--I quote from a Lutheran book--"are part of the white race, though swarthier in color than the Nordics."

You must be one of those fellows who also thinks that blacks in the American South were still being lynched en masse in 1964?

Will McLean said...

David:

He also distinguished between the Khoisan, other black Africans, and Australian Aborigines, so the ability to draw a distinction does not indicate that he thought they were unrelated. In any case, many 19th c. writers on race grouped the black ethnic groups together, although some did not.

My point on color is that race is not, like color, a fixed property, but fluid and mutable.

David Friedman said...

Will writes:

"My point on color is that race is not, like color, a fixed property, but fluid and mutable."

I'm not sure I understand what you mean. Colors can be and routinely are blended together to give intermediate colors. The definition of color is at least potentially mutable--it isn't clear, for instance, where one draws the line between green and blue, just as there isn't a sharp line between black and white racial classifications.

The fact that some people got racial classifications wrong doesn't make race meaningless, nor does the fact that there aren't sharp lines distinguishing races--unless you are willing to claim that color is meaningless as well.

Will McLean said...

We can and do define colors objectively as either parts of the visible spectrum (in the case of saturated colors)or mixes of frequency for unsaturated colors. The pure spectral colors are all common English words and hardly anyone complains that we need to insert a new color between yellow and orange. For more precise distinctions we can use more detailed taxonomies like the Pantone numbering system.

Ethnic groups, on the other hand, are like trying to nail Jello to the wall. You design a nice rational taxonomy and those crazy Europeans spoil it by sailing about, shipping Africans to the New World, interbreeding promiscuously and also interbreeding. Did I mention the interbreeding?

David Friedman said...

1. Actually, colors can't be fully defined in the way you describe. As Edwin Land demonstrated quite a long time ago, the Newtonian analysis of colors does not accurately describe how our color vision works. He demonstrated how it was possible to get what was perceived as a color picture by superimposing, if memory serves me correctly--it was a very old Scientific American article--two images, both pink.

2. The effect of interbreeding is the same as the effect of mixing pigments, so that doesn't make race a different sort of thing from color. And people mix pigments all the time--paint stores have special machines for the purpose.

Jim Rose said...

On race as a social construct, there are frequent public health advisories on diabetes advising people of certain ancestries that they are more prone to diabetes and should be tested more often and watch their weight.

Most of these ancestries were late to the industrial revolution.

A genetic ability to put on weight fast was an advantage in subsistence society where food supply was insecure.

Moving quickly into a rich society with plenty of food makes these genetic dispositions not such an advantage.

Natural Language Processor said...

Ethnic groups, on the other hand, are like trying to nail Jello to the wall. You design a nice rational taxonomy and those crazy Europeans spoil it by sailing about, shipping Africans to the New World, interbreeding promiscuously and also interbreeding. Did I mention the interbreeding?

This is like Henry Louis Gates' program on PBS, on which he tests celebrities for genetic ancestry. For example, some African American studies professor discovered she was 60% Euro, 5% Asian, and 35% African. So, therefore, race is a social construct and there's no such thing as a Euro population, an Asian population, or an African population!

Despite your protestations, this line of reasoning is EXACTLY like saying, "Orange is really just a mixture of red and yellow, so clearly colors are a social construct!"

Will McLean said...

The effect of interbreeding is the same as the effect of mixing pigments, except that mixing yellow and red produces a hue that has long existed, and the first interbreeding of, for example, Europeans and American Indians produced genotypes that had never existed before. And two brown parents can produce children who are white, brown or black.

I'm not convinced that traditional broad racial categories don't mislead more than they reveal. It was commonly believed that sickle cell disease was something that affected blacks very disproportionately. But in the old world it is quite rare in South Africa, but prevalent in Southern Italy, where Malaria was once common.

Milhouse said...

Colour perception is very much goverened by culture, specifically language; indeed our language wires our brains to perceive colours differently. For instance, are pink and red different colours? What about dark blue and sky blue? To English speakers the answers are "obviously" yes and no; but to Russian speakers both distinctions are equally obvious, and in fact native Russian speakers are better able than English speakers to distinguish shades of blue, because they think of dark blue and sky blue as separate colours.