Let's imagine that Summers had made the same comment but had substituted Jews, Blacks, Hispanics, or Asians for women. (And let's imagine that there was a statistical disparity for these groups as well). Would anyone be defending what he said? And if not, why is anyone defending him now?

If I read Judge Posner's post correctly, he believes that Summers acted irresponsibly by making the comment, because it was not supported by evidence and brought bad publicity to his institution. Why then was it wrong to apologize? A good leader is not just one who shows strength but also one who is willing to admit error. Summers made the right decision, politically and academically, to make clear that he did not believe there is any evidence that gender affects the ability of women to excel in science. Certainly, it is not in Harvard's interest to alienate all egalitarian-minded members of the American public, including the large portion of its students, alumni, faculty, and contributors.

I read this as yet another example of the failure of standardized tests to reflect anything
useful.

If variation on a science test between genders is unequal but performance in undergraduate science is not varied, then the obvious conclusion is, that test does not measure likely performance in undergraduate science. If the test does not even accurately predict the thing it was created to predict, then it most certainly should never be used as evidence of a biological difference between genders. It should be used to wipe the floor.

Further, Ph.D level science achievement can't be tested in a standardized way. Most if not all tests get less accurate towards the curve tails.

Discrimination has perhaps lessened in recent years but the stress should be on "recent." Since in this case, "discrimination" is likely to manifest itself in the way young girls are socialized to prefer certain activities, it would take at least a generation for any change in thinking about gender roles to translate into enough of a change in child-rearing that a change in career selection could be observed.

Seems like Harvard is never content without a scandal for all of its students to complain about in every restaurant on the square. As for Summers, all I can say is that anyone who would say that women are less willing to make time sacrifices to science careers has never tried to date in Cambridge MA.

"If variation on a science test between genders is unequal but performance in undergraduate science is not varied, then the obvious conclusion is, that test does not measure likely performance in undergraduate science. If the test does not even accurately predict the thing it was created to predict, then it most certainly should never be used as evidence of a biological difference between genders. It should be used to wipe the floor."

You're confusing the issue, as Posner noted many of Summers critics also did. You're speaking of two different populations--all women who take college admissions tests and all women undergraduate science majors. Aslong as there are fewer women than men in undegraduate science majors, there is no indication that standardized tests are producing any result which is counterintuitive.

To quote Posner from his post, "This will affect the relative number of the two populations in the tails; it may or may not affect the average quality of the members of each population who are in the tails."

What everyone is missing, or in this case ironically assuming is that men and women are equally predisposed to careers in science. My unscientific guess is this is not true. Women are not as likely pursue careers in hard science as men, not b/c of lack of intelligence, but b/c they are at some level they find it less attractive as a career than do men. See Pinker's 'Blank Slate' for a fair and balanced take on this.

I fear the first commenter didn't quite understand Judge Posner's argument. The commenter says no one would dare defend Summers if "women" had been substituted with "Jews," "Blacks," "Hispanics" or "Asians."

Posner has already replied to this claim. First, he differentiates between sexual differences and ethnic differences, saying the latter is such a hot potato that, it's true, most (all?) academics are afraid to research any biological differences, or even speak about them publicly. And if I understand Posner correctly, he thinks, nevertheless, there's no reason, if we are truly seeking knowledge at universities, not to look into it.

But more important, sexual differences, though still a tender subject, aren't quite as impossible to discuss, because there are very obvious differences between men and women, and not just cultural ones. (Posner as a judge also knows that the strict scrutiny given to racial distinctions is not quite the same when it comes to sex--note we don't have to draft women along with men, and we still have separate bathrooms by sex.) A biological explanation as to why men are better (on average) than women in basketball is completely non-controversial (I hope). In today's world of evolutionary psychology, there are solid theories as to why females and males of the same species may have different brains that view the world differently. If this is so, it's not hard to imagine that men and women may tend to have different talents, and that such biologically based differences are at least worth exploring.

One of the bigger reasons for the difference, I think, is the unusually intense drive of young males to succeed and/or distinguish themselves–to win the prize and get the girl, as the story has gone for several million years now among many or most species. This drive yields many blockbuster successes and major failures, which may be one reason why males dominate the inmate population. Is anyone saying that the domination of the convict community by males is the result of social pressures? That is one explanation for why males tend to exhibit higher variance, especially in fields that require a virtual fixation on success to excel in, such as math and science. It may not even be a significant difference between the genders in raw intelligence, whatever that might be, as it is a difference in variance regarding the drive for success-at-all-costs in a narrow, specialized, and often significantly tedious field.

I think one of the reasons people resist looking at biological determinants is because:
a- in the past this type of argument has been made to justify certain prejudices (at one point, looking at the relative width of a person's nose was considered an adequate way of determining that person's propensity to commit crimes)
b- more importantly, I think it is hard to distinguish from natural and social factors. Let's look at the olympics: a century certainly is not enough time for the gene pool to change considerably, and yet results vary greatly, not only across time, but certainly across space. Yes, certain biological characteristics matter a lot, like the fact that men have relatively more muscles than women. Still, the gap between male and female athletes is closing, and I am certain that a 19th century male athlete would not even be competitive in some events against 21st century women (whatever their genetic make up might be). That points to certain social factors being heavily influential: the fact that sports nowadays involve large sums of money (at least in the US), that it has become an accepted profession (something important specially for women), that large quantities of money are spent in trying to come up with better training, better materials, better technique... How to explain the difference in performances across countries, especially given the superiority of the US in athletic events? Certainly I think this is not a question of nature vs nurture, but that these two things interact. Unfortunatelly every major study that I've seen about the biological determinants of something end up simply using some sort of correlation between observed characteristics of a population and a socially determined outcome, without taking other factors into account (i.e., to understand why some people jump higher than others, you should have to look not only at the biological make up of a person, but also at the "nurturing" of that ability, and even at the reason why such an ability is socially valued at all...)

Solyom and Fred,

Desmond Morris (_The Naked Ape_, McGraw-Hill, 1967, p. 236) asserts that girls appear to have a more instinctive fear of spiders than boys. Supposing this to be true, the following experiment could be set up:

Economics and other sciences employ graphs, which are suggestive of spiders and their webs. Test one group of boys and girls on their learning of material that includes graphs; test the control group with material that does not include graphs.

I think you underestimate the potential benefits of Summers' comments. While Summers is not an expert on gender differences, he is in enough of a public position that any issue he highlights, especially in a contraversial manner, could become fodder for the media as it did in this case. Apology notwithstanding, perhaps Summers was hoping this would ignite a series of newspaper columns dealing with this usually-taboo subject. Saletan proved to be helpful to his cause (i.e., getting the word out that gender differences should be studied). But then he underestimated the reaction, not realizing it would be so vociferous, leading to his "weak" apology.

I disagree with Posner's evaluation that "...the benefit of Summers’s speaking out was small." The amount of attention Summers's comments generated have made many people aware of the contentiousness of the issue, and have undoubtedly inspired many to think carefully about this issue for the first time.

Perhaps Summers was exactly right to use his position of power to bring this issue to the public's attention (even if he didn't do it intentionally). If the president of the US's most famous university should be allowed to argue for anything, shouldn't it be for increased scientific investigation into hot societal issues?

To respond to Fred: my point is not that genetic differences should not be explored. I do not wish to stifle science. The question is whether a university president should make a remark that will be perceived by many as sexist when there is no scientific evidence justifying the statement? The answer should be a resounding no, and a president who makes such a mistake should promptly apologize. The more mea culpas the better.

No one disputes that Andy Roddick could beat Serena Williams on a tennis court. That is a real difference backed up by empirical evidence. But there is no such evidence that Roddick has a genetic advantage over Williams in chemistry or physics. Lacking any such evidence, it is politically foolish, and academically unjustified, to make such a suggestion. Especially when there is an easy and compelling explanation for the lower number of women in science, socialization, which is staring us right in the face.

This is one case where the conventional wisdom is correct. It is not an issue of free speech or academic freedom. It is a basic question of prudence and respect. Unsupported theories that play into the hands of bigots should not be uttered by a university president under the guise of "academic freedom."

I liked George Will's comment:

"Forgive Larry Summers. He did not know where he was.

...He thought he was speaking in a place that encourages uncircumscribed intellectual explorations. He was not. He was on a university campus."

I appreciate greatly the rigorous clarity of Posner's post. I will respond to his points one by one.

1. Did Summers send the wrong signal by raising the issue, given the low value of his non-expert opinion? Yes, if you find credible "stereotype threat." Summers perhaps discouraged female science students to perform worse and not to pursue a career in the hard sciences:

This theory, known as "stereotype threat," is defined by psychologist Claude M. Steele of Stanford University as "the threat of being viewed through the lens of a negative stereotype."
...
Steele and his colleagues, Joshua Aronson of New York University and Steven Spencer of the University of Waterloo, alerted black and white students that a challenging verbal quiz would measure their abilities. The black students then performed measurably worse than the whites, even though the students selected were statistically equivalent in ability level.
...
But when the researchers told the students that the quiz was a lab task that did not measure ability -- thus making the stereotype about ability irrelevant -- the blacks' performance matched the whites'.

2. What is attributable to the decline in hiring at Harvard since Summers has been President? According to Virgina Valian, a professor of psychology who studies gender bias in a Washington Post live discussion:

Prague, The Czech Republic: Hiring of women in the sciences at Harvard has declined from 37 percent since Summers has been president. Even if Summers' statements were not those later attributed to him by attendees driven to hysteria by his comments, isn't discrimination a more likely cause of the decline than differences between the structure of male and female brains? Or is that incorrect?

Virginia Valian: The decline in the hiring of women was startling, but any single year can be a fluke. You need a run of several years in order to detect a pattern. But your overall point is of course correct. If anything, the pool of talented women is larger every year, so a decline in hiring is better attributed to lack of attention to a diverse pool of candidates than to a decline in qualified women.

3. Did Summers have a fiduciary duty to shut his trap?
No. The female head of the Harvard Corporation has said Summers has done nothing wrong. He will not be fired for this. He hasn't harmed Harvard's brand, according to the people who count.

4. Is Posner permitted to comment publicly on pending or impending cases? Yes and no. He commented on the Lewinsky matter, and nothing happened after that, except major liberal gnat Ronald Dworkin gave Posner a tongue-lashing in the New York Review of Books. Dworkin may be wrong on this, but Posner is adopting Dworkin's standard which includes both IM-pending as well as pending because he doesn't want far ranging discussions like the affirmative action nonsense that raged after he wrote about profiling.

5. Does Summers' apology = weakness and does weakness = further concessions in the form of diversity hiring? Probably. But if so, the hiring won't be diverity hiring. It will be the hiring of rich, highly-educated, white women. That's who is lobbying Summers for more hiring of "women".

6. Was the apology a compounding error? Yes, so Summers should only hire conservative transgendered "women" and black conservative women scientists as a concession. That'll show those socialist feminists!

7. The substantive issue. If discrimination is conceded as major impediment to women, doesn't that imply that the average woman is batter than the average man? No! Let's say in Reality men are, on average, a 5 at science, women a 4. Our perception of male and female aptitude is respectively +1 and -1. On average then, our perception would 6 for men and 3 for women. An inaccurate gauge, but still overall correct -- men on average are better than women. So if we have a woman who is a 4 and a man who is a 5, is there a gross injustice done? No. There is an exagerrate perception, but the right hiring decision is still made (the better candidate is still chosen). When the woman is a 6 and the man is a 4 -- in other words, if the woman is exceptional and the man is sub-par -- a gross injustice will be done, because the woman will be underrated and the man will be overrated. That leads to bad decision-making. The end result is not that women are overrated, as Posner says, but that exceptional women are underrated.

8. If Summers is right and women are less likely exceptional, but also less likely idiots, that only points out that we should scrutinize more closely individuals so that exceptional women are not overlooked. Whether the basis for the dearth of exceptional women is biological is irrelevant, in part because "a decline in hiring is better attributed to lack of attention to a diverse pool of candidates than to a decline in qualified women." (Virginia Valian, Professor of Psychology, Hunter College, http://www.washingtonpost.com/wp-dyn/articles/A47055-2005Jan29.html) It may be provocative, so kudos to Summers for bringing it up! But there's no reason to underrate individuals by using an exaggerated cultural shorthand: in other words, there's no excuse for profiling women in science.

9. This ties into the "abritariness" discussion that Palooka and the other Palooka(???) were having about profiling. Generalizations may work in a good number of situations, but they aren't very good at accounting for the exceptions, and there's no reason to discourage exceptional individuals from joining academia on the basis of criteria that don't accurately evaluate what they have to offer. So, arbitariness = bad.

Two responses to the first commentator's point about this being like comments on Jews, blacks, etc. One, Posner does not say there is no evidence supporting Summers' assertion, and he mentions the issue of variance. I think a different way of making the same point is to say that the male IQ distribution has a larger standard deviation than the female IQ distribution. That is, men and women have the same average IQ, but there are more males at the extreme ends of the spectrum, more really dumb and really intelligent males than females. I am not a psychometrician, but I believe that is in fact the case (Judge Posner may be able to confirm or disconfirm this).

Two, I suspect that the comments by David and Dpin suggesting that prejudice is behind recognizing sex differences are based on prejudice - an unexamined ideological tenet. Have you two really studied this? It is my experience that few (if any) who dogmatically claim that there are no none have studied the anthropological evidence on this issue. It is just that there is a forboding environment at universities, especially law schools.

"I think a different way of making the same point is to say that the male IQ distribution has a larger standard deviation than the female IQ distribution. That is, men and women have the same average IQ, but there are more males at the extreme ends of the spectrum, more really dumb and really intelligent males than females."

I think that's exactly right; that's what Posner describes with his bell curves.

"So the benefit of Summers’s speaking out was small."

A large benefit of Summer's speaking out has been to make cyrstal clear the climate of intimidation and suppression regarding the discussion of "gender". The message below, which is circulating on the internet, is truly astonishing.

--------

For the past several days, the President of Harvard University has been apologizing profusely for what he said in offering research suggestions to a non-public conference of economists. For his statement, see http://www.president.harvard.edu/speeches/2005/womensci.html

One choice line: "I was wrong to have spoken in a way that has resulted in an unintended signal of discouragement to talented girls and women."

This is indeed Stalinist. E veryone should oppose a "signal of discouragement to talented girls and women." But the truth is that such a signal, to the extent it occurred, resulted from deliberate, intense, and misleading responses to his remark. That's classic totalitarian suppression of an unpopular view, with forced public acknowledgment of guilt and forced repudiation of the "wrong". So much for Harvard University.

The scope of the problem is well illustrated from the follow-up email distributed to all College of Engineering Graduate Students at the University of Michigan. A choice excerpt: "As a Co-PI of the UM NSF ADVANCE Program I want to communicate in no uncertain terms that the extensive research on gender in science and engineering makes clear that the 'hypothesis' about innate differences between men and women in mathematical ability has been amply tested, and is not supported by scientific evidence." This is clearly coerci ve use of organizational position to suppress legitimate scientific inquiry. Rather frightening, indeed.

This is not about general sensitivity to discussing innate differences in human ability. Consider the recent public comments of the Labor chairman of the Commons education committee: > See http://news.bbc.co.uk/1/hi/education/4079653.stm

At least some UK politicians recognize that the most significant educational problems now concern males, not females. As for leading US universities, they don't seem to be doing much leading, and don't seem much like universities. Instead, they seem to be pandering to the interests of privileged elites and peddling some gross trickle-down theory of educational progress. I suggest that socially conscious and intellectual serious persons, if they manage to stop crying or laughing, start some non-violent protests or acts of personal intellectual disobedience.

Ed

"Is Posner permitted to comment publicly on pending or impending cases? Yes and no. He commented on the Lewinsky matter, and nothing happened after that, except major liberal gnat Ronald Dworkin gave Posner a tongue-lashing in the New York Review of Books. Dworkin may be wrong on this, but Posner is adopting Dworkin's standard which includes both IM-pending as well as pending because he doesn't want far ranging discussions like the affirmative action nonsense that raged after he wrote about profiling."

Wouldn't that all depend on the definition of impending? I remember reading about this little spat with Dworkin before I read Poser's impeachment book. I don't think convention requires that Judge Posner forgo comment on any case which may potentially come before him. Or maybe prudence dictates he should. Justice Scalia commented about the Ninth Circuit's pledge of allegiance case and found he had to recuse himself when it did make it to the high court. Regardless, that is for specific cases. The Judge can comment about the constitutionality (or in Posner's case, the pragmatic implications) of broad subjects, like affirmative action, right? If Posner wanted to leave as few clues to his positions on issues as possible, I think that ship has sailed.

There are many comments here indicating that whether or not gender influences scientific ability is something that needs to be studied. Why? Certainly, we shouldn't stop people who desire to study the subject from doing so, but what will the results of the study really get us. So we can stop programs to encourage women to go into math and science? So we can stop trying to change the very real culture that discourages girls from being interested in those subjects? Worthy research, certainly, but necessary, I think not.

Erika, if we require all acedemic research to be necessary, there goes large swathes of the hard science, most of the soft sciences and as to the English Department...

"Wouldn't that all depend on the definition of impending? I remember reading about this little spat with Dworkin before I read Poser's impeachment book."

For the record, Dworkin reviewed Posner's book around the time it was published and critiqued the impropriety of its timing in his review.

Here's is a quote:

Posner acknowledges that "the ethical rules of the federal judiciary forbid public comment on pending cases," but, he says, "I do not discuss any pending cases." He has misstated the rule: Canon 3(A)6 of the Code of Conduct for United States Judges prohibits federal judges from commenting publicly on "pending or impending" cases, and though it is unclear what "impending" means, it should be understood, given the rule's evident purpose, to include not only imminent and certain prosecutions, but any possible future prosecution that has been publicly debated among politicians and officials and often mentioned in the press, particularly when the judge is prominent and his statements are likely to receive wide circulation.[9]

Footnote 9 reads:
This canon was adapted from Rule 2(B)9 of the ABA Model Code of Judicial Conduct. That code adds the qualification, however, that comment on a pending or impending case is prohibited only when the statement "might reasonably be expected to affect [the trial's] outcome or impair its fairness." The US Code pointedly omits that qualification.

In short, before Dworkin critiqued Posner, Posner was ignoring the word "impending" and giving the Rule the interpretation you are arguing he should give it NOW. So: What did I say?

"Dworkin may be wrong on this, but Posner is adopting Dworkin's standard which includes both IM-pending as well as pending because he doesn't want far ranging discussions like the affirmative action nonsense that raged after he wrote about profiling."

Posner has now adopted Dworkin's standard, which includes the word impending ("as I mentioned in our introductory blog posting, I am not permitted to comment publicly on any pending or impending court case."), but as there are no specific cases being discussed here, he must be mentioning it for another reason. Given that Posner seemed appalled by the racialist language of many of the posts ("I was however startled by the large number of comments that compare profiling to affirmative action and ask that commenters who oppose profiling as demeaning, alienating, and so forth take an equal stand against affirmative action."), I speculated that he was attempting to control the discussion and keep all responses relevant to his points. He even goes so far as to label one of his points "the substantive issue".

In short, NO. What I said has NOTHING to do with the definition of the word "impending". Posner has already conceded the point to Dworkin by including the word "impending" in his present analysis of the Rule. Prior to the "spat", Posner did not include the word impending in his analysis and he would have agreed with your interpretation. Which means including the word "impending" to reach Posner's original position (which excluded it) is plainly wrong.

"I don't think convention requires that Judge
Posner forgo comment on any case which may potentially come before him. The Judge can comment about the constitutionality (or in Posner's case, the pragmatic implications) of broad subjects, like affirmative action, right?"

Description of Begging the Question
Begging the Question is a fallacy in which the premises include the claim that the conclusion is true or (directly or indirectly) assume that the conclusion is true. This sort of "reasoning" typically has the following form.

Premises in which the truth of the conclusion is claimed or the truth of the conclusion is assumed (either directly or indirectly).
Claim C (the conclusion) is true.
This sort of "reasoning" is fallacious because simply assuming that the conclusion is true (directly or indirectly) in the premises does not constitute evidence for that conclusion. Obviously, simply assuming a claim is true does not serve as evidence for that claim. This is especially clear in particularly blatant cases: "X is true. The evidence for this claim is that X is true."

Some cases of question begging are fairly blatant, while others can be extremely subtle.

Examples of Begging the Question

Bill: "God must exist."
Jill: "How do you know."
Bill: "Because the Bible says so."
Jill: "Why should I believe the Bible?"
Bill: "Because the Bible was written by God."

"If such actions were not illegal, then they would not be prohibited by the law."

"The belief in God is universal. After all, everyone believes in God."

Interviewer: "Your resume looks impressive but I need another reference."
Bill: "Jill can give me a good reference."
Interviewer: "Good. But how do I know that Jill is trustworthy?"
Bill: "Certainly. I can vouch for her."

Of course the research doesn't need to be necessary for people to do it. Heck, my research isn't necessary except to a very specialized group of people. I am just commenting on the fact that many people seem to be saying that Summer's comment shows that such research is necessary.

"I am just commenting on the fact that many people seem to be saying that Summer's comment shows that such research is necessary."

Erika, research presented at that very conference discussed innate sex differences. His comment doesn't prove that such research is necessary; but his comment was probably contextually proper.

"Posner acknowledges that "the ethical rules of the federal judiciary forbid public comment on pending cases," but, he says, "I do not discuss any pending cases." He has misstated the rule: Canon 3(A)6 of the Code of Conduct for United States Judges prohibits federal judges from commenting publicly on "pending or impending" cases, and though it is unclear what "impending" means, it should be understood, given the rule's evident purpose, to include not only imminent and certain prosecutions, but any possible future prosecution that has been publicly debated among politicians and officials and often mentioned in the press, particularly when the judge is prominent and his statements are likely to receive wide circulation."

I'm not sure exactly what set you off, but you seem to have misunderstood me. I was not defending Posner's actions, just merely pointing out that "impending" can have a very narrow or a very broad interpretation. Whether or not a judge can commment on a case if it's theoretically possible it may come before him I think is a matter of debate, as yourself has conceded.

As far as "begging the question." I didn't commit a logical fallacy, I actually WAS asking a question! I wouldn't think, though I am not a a lawyer or a law student, that the rule against commenting on "pending or impending" cases would apply to broad questions of social policy--i.e. affirmative action. It's my impression that you think Posner's apparent reluctance to speak about affirmative action is related to this point. If that is the case, I believe you are mistaken.

I still don't think, for the reason that "impending" is subject to varying interpretations, that the Judge is required to remain silent on all cases which may potentially come before him or any federal court. Maybe I am wrong, but you concede the meaning of impending is uncertain, and therefore concede there are some cases where (I believe Judge Posner's book is one of them) reasonable people can disagree.

I think I read Dworkin's sentiments as your own. Sorry. Most of the post is still relevant though.

No.

You quoted Dworkin and assumed that his article was my words. I rarely agree with Ronald Dworkin, which is why I called him a liberal gnat. I did not concede that "impending" has variable meaning, Dworkin did. The meaning of the word "impending" is irrelevant to my point.

Here:

Posner's original interpretation(P1): pending.
Dworkin's interpretation: pending or impending.
Posner's new interpretation(P2): pending or impending.

Your interpretation is (P1). You cannot get to (P1) with the word impending added in, because it excludes the word impending, unless, of course, the word impending means nothing. That isn't interpretation: it's absurdity.

My point is that there are no specific cases being discussed here. Therefore, Posner must have another reason for mentioning he cannot reply to comments outside of the scope of what he has actually written. My guess: he doesn't care to read a dozen posts expressing racial animus like last time. Are you arguing that he wasn't "startled"?

"You quoted Dworkin and assumed that his article was my words."

Yes, I noticed the error and apologized. I understand what you are saying about Posner's new inclusion of the term "impending." My point, which I guess you have no qualms with, is that there does seem to be question about what impending means.

I think Posner was discussing his limits on free speech to illustrate that not all authority figures (CEOs and judges) are free to speak their minds.

"My guess: he doesn't care to read a dozen posts expressing racial animus like last time. Are you arguing that he wasn't "startled"?"

Though the discussion was heated, I would not characterize either side as expressing any racial animus. But I do agree Judge Posner may wish to contrain comments to the topic he has chosen. On that note, the "impending" conversation is closed, for me.

This thread has taken on a decidedly self-referential bent!

"But I do agree Judge Posner may wish to contrain comments to the topic he has chosen."

Yeah but guess what, this is the internet. One of its chief values lies in not being subject to the normal social and regulatory norms. I do not
see the utility of restraining our commentary to arbitrary interpretations of Posner's will when an interesting analogy or tangent presents itself. Occasionally, unexpected truths may be discovered. For instance, one could conclude from the last topic that people are more concerned about and emotionally invested in Affirmative Action vs. airport profiling.

Posner gets enough chance to control the terms and norms of debate in his role as one of the most respected Judges in the country. Perhaps the advantage here is access to a more free-form and public forum. The Federal Judiciary is constantly being criticized for being isolated and variously out of touch. I am no fan of Law & Economics but
I welcome the idea of any judge seeking input from "The People" regarding public policy. Indeed one of the most predominate critiques of the L&E movement is its failure to ground assumptions in the day to day realities of human narrative.

Of course we should all try to be civil, and remember that we have Posner and Becker to thank for creating this forum and attracting smart people to it in the first place. I should stress that I am expounding merely my own reasons for posting here.

Incidentally, I am going to have to agree with Erika and further suggest that Harvard should devote some energy to studying ways to bring about greater representation of women in the sciences rather than dredging up biological determinism/eugenics inquiries that should have died in WWII. And if we are going to investigate biological gender differences in intelligence, lets agree to let women run the study eh?

“men and women have the same average IQ“

In fact adult men have an IQ score of something like 4 points higher than women. AND there are more men at the top of the distribution. Since we know that IQ is in fact highly correlated with scientific ability there is ex ante reason to suspect the sizable gender differences may have something to do with (the even more sizable) gender differences in the sciences.

Just look at GRE numbers from 2001-2002. Men have average quantitative scores of 650, women of 556. Those are no small differences; we are talking something like the 60th percentile and the 35th percentile. I realize there are selection issues. But men also beat women in the verbal test (by only 9 points) and the analytical section (by 23 points). Since we know by axiom that “there are no biological differences” I guess we now have to throw out the GRE along with the IQ and the SAT. (In the SAT men beat women by 9 points in verbal and 34 in math, thus proving gender bias. As Jerry Seinfeld said, “maybe the test was gender bias, you know a lot of questions about hunting and testicles”)

“If discrimination is conceded as major impediment to women, doesn't that imply that the average woman is batter than the average man? No!”

Actually, according to your own argumentation the answer would be “yes!”. If your example the marginal (least competent) woman hired would have a score of 6, while the marginal man would have a score of 4.

I mistrust any argument based on discrimination. Why don’t the people who believe women make 20% less than men for the same job or that more qualified female scientists are not hired simply put their money where their moths are? Start a company and only hire women, and compete the others out. Same for academic departments. I am sure with all the brilliant and well publicized women Harvard and U-Chicago turn down there must be hundreds of potential stars to recruit. What are you waiting for?

Speaking of WWII, don't you think Lamarckism should have died along with Stalin and Lysenko? Or is the fanatical left still in the business of denouncing biology as a invention by the ruling classes? At least these days you are not starving people with your blind ideological view on science, only stifling dissent in academia. Guess we have come a long way.

I agree that the above post raises valid issues. The feminist response to the Summers controversy has consistently been that we must remove all environmental impediments before we really even begin to discuss the innate differences. This is a clever argument, because as a practical matter it means that genetic questions are never properly on the table, because we will never change the different ways in which we treat men and women, given that there are so many other differences aside from the sciences. It also is a covert argument for persistent affirmative action, as today’s editorial in the NYT, yesterday’s NPR “All Things Considered” editorial, and several others I have seen include a subtle plug for those types of policies.
One thing bothers me about that line of reasoning–the idea that only women are discouraged from succeeding in the sciences. Well, boys who are interested in math and the sciences are constantly ridiculed as geeks and nerds among peers as well as popular culture, go through years of not being popular with friends of both sexes, and the like. It is particularly directed towards boys often, because it’s seen as seriously not macho to be a school geek. But, you don’t hear men complaining about it. It’s just something to suck up and take, instead of get flustered about. On average, the social pressures against success among both sexes are significant. In my view, it's time to abate the victim mentality problem. It is a significant impediment to just getting the work done and creating effective departments.

I can’t get past this nonsense of Harvard being the “nations most famous university.” I say it is the most over rated and over hyped university. What has Harvard done that dozens to hundreds of other universities haven’t done? What great new breakthroughs in technology or medicine or engineering? Every program in every other university is second place to Harvard?? Ha ha chortle guffaw!

Your assumption that only people at the very top of scientific aptitude tests become great scientists is quite a leap. Einstein the "Genius" relied on Marcel Grossman to do the heavy mathematical work because he didn't like math much. Most of the successful scientists in my field are not geniuses; instead, they are creative people with the self-confidence to follow their own path. I've heard countless stories from them of how difficult it was to pursue their ideas, to get funding, to have their early papers rejected. Those scientists, men and women, who prefer to do safe, incremental research will rarely make a splash in their field. Only the bold, who are usually men, with a "boom or bust" attitude can succeed. If women are so fragile that they are scared off by senile, misogynist professors and loose-lipped university presidents, then those women aren't likely to do well in science anyway.

I echo Corey in thanking Posner and Becker for creating this forum to allow a free-flowing public discussion on these issues. This website is intellectually stimulating and is an example of how the internet can be a very useful tool for learning and debate.

Just a thought: we seem to have only one female contributor here so far (Erika), and I think she asks a good question that no one here has really addressed: WHY would we think it valuable to study innate differences in science ability between men and women? We know from history and experience that both women and men can become great scientists. So obviously both genders have the ability. Why does it really matter whether one gender has a slightly increased tendency toward scientific greatness?

The answer, I think, is all about politics. On one side of the debate are those who claim that Harvard and other schools have fewer women science faculty largely because of entrenched discrimination (perhaps the subtle kind of "old boys" network that we don't really notice on the surface but that affects numbers nonetheless). On the other side are those in the academy who believe, ideologically, that gender discrimination has been largely conquered -- at least among the educated elite -- and disparities must therefore be the result of something else. So this is fundamentally an ideological debate among two warring camps, and Summers set off a lightning rod by giving some credence to the views of one of those camps. Maybe the problem here is that certain academics are too invested in the bigger ideological debate to treat this issue with the prudence and rigor that it deserves.

No one disputes that Andy Roddick could beat Serena Williams on a tennis court. That is a real difference backed up by empirical evidence. But there is no such evidence that Roddick has a genetic advantage over Williams in chemistry or physics. Lacking any such evidence, it is politically foolish, and academically unjustified, to make such a suggestion.

Fortunately, nobody, including Lawrence Summers and Judge Posner, did any such thing. You're still confusing individual and group distinctions. The statement that the population of men has more outliers in science/math than the population of women says nothing about any individual man or woman.

Especially when there is an easy and compelling explanation for the lower number of women in science, socialization, which is staring us right in the face.

"Easy and compelling" is not valid evidence. You cannot pick answer B over answer A merely because the former makes you more comfortable. Nor is there any clear reason, other than discomfort, why "socialization" should be more "compelling" than biology when discussing sex differences.

I'd like someone to explain for me why any Tom, Dick or Harry can't be President of Harvard University, provided he has a knack for fundraising?

Here we have a former Secretary of the Treasury, who has been mentioned (perhaps implausibly, given the current Republican administration) as a possible President of the World Bank, embroiled in a public controversy that could not have done him or his institution any good, over remarks that his critics either refuse or are unable to understand and for which he has found it expedient to apologize repeatedly. What can Larry Summers accomplish as Harvard's President that would be worth putting up with this?

[[[Actually, according to your own argumentation the answer would be “yes!”. If your example the marginal (least competent) woman hired would have a score of 6, while the marginal man would have a score of 4.]]]

No, I never said that at all. You're assuming that raw numbers have anything to do with actually being hired. My point was only that the differential in one case is unjustified IF it affects hiring, not that it necessarily must. I never posited that any men definitely would be hired at 4, or that a woman needed at least a 6 to be hired. For all I know, women might need a 94 to be hired. The point is that women applicants as a class are underrated, not overrated. Now, the women who are actually hired (a subset of the women applicants) are not overrated, because they actually are exceptional. Their reputation is justified.

Sorry not to spell it out in my previous post, but: good decision-making is justifiable; bad decision-making is not.

[[[Actually, according to your own argumentation the answer would be “yes!”. If your example the marginal (least competent) woman hired would have a score of 6, while the marginal man would have a score of 4.]]]

I scrolled up, and it clearly says above "IF the woman is exceptional and the man is sub-par" -- it doesn't say that that those numbers necessarily must represent exceptionalness or subparness, nor does it say that a hiring decision must be made at that point. I don't see how you get from contingency to necessity. I never said that any men with a 4 were being hired, just that if they were, given the implicit evaluations that hirers tend to make, it would be unjustifiable if he were chosen over an exceptional woman.

You're also assuming that we KNOW what people's perceptions are and how hiring is done. We don't. I never said that the numbers in my example would be freely available to everyone. (This is actually a major error, because you are assuming everyone has perfect information.) The only way that women would acquire a reputation as exceptional if they are hired is if the actual averages and actual bias were widely known. I never assumed that. Hiring is not that transparent. You have no way of knowing what is in the hirer's head (the +1, -1 part) -- and that is what creates the bad decision-making.

So my example in no way says what you insist it does unless you distort it out of context. I'd appreciate if you not do that!

Just to clarify:

There are two groups of women:
A. applicants
B. those actually hired

There are three stages of evaluation:
C. as applicants, when hired or not
D. as academics, when earning a reputation

"if the woman is exceptional and the man is sub-par -- a gross injustice will be done, because the woman will be underrated and the man will be overrated. That leads to bad decision-making. The end result is not that women are overrated, as Posner says, but that exceptional women are underrated."

The exceptional woman here and sub-par man here are underrated and overrated when they are C, not when they are D. The gross injustice is that an exceptional woman was not hired, or that an A did not become a B.

The "end result" is the result at the end, or D. Women are not overrated at D if they made it to B, because in that case they actually are exceptional. An exceptional woman being considered exceptional is not an "over-rating." So their reputation would be justified.

In any event, as I said before, exceptional women are underrated. They are: at C. They are not overrated at D, because they actually are exceptional. Posner's point was that women academics are overrated at D. My point is that they are underrated at C, and if they made it to D, then they actually are exceptional.

The argument is not based on "discrmination." It's based on implicit evaluations. They aren't the same. Both women and men rate equally credentialed male candidates higher than female ones. You aren't ignoring an argument based on discrimination, but one based on empirical data.

I think I supplied the link to Virginia Valian's chat forum. She provides links to empirical data on implicit evaluations (the +1,-1 part) there.

Sorry I meant two stages of evaluation. I do this crap at work.

"You're also assuming that we KNOW what people's perceptions are and how hiring is done. We don't. I never said that the numbers in my example would be freely available to everyone. (This is actually a major error, because you are assuming everyone has perfect information.)"

Just to anticipate: I don't mean that scientists cannot acquire empirical data about what the perceptions of employers are; I mean that such information is not so distributed throughout the public as to be such a part of common knowledge that it forms a part of stereotypes held about women academics. In other words, neither the people submitting their resumes, nor everyone else "watching" in society actually knows that the implicit evaluations are being made. Most people think men are better at science on average, much like the hirers do, and they don't realize this belief is exaggerated and overlooks exceptional women. Why? Because they never encounter any. They're exceptional. That doesn't mean they don't exist.

"Women who want to have children, as most do, must expect to devote more time to child care that men do. That is a brute fact and has nothing to do with scientific careers as such."

Why? What if a woman has a busy and successful career and her husband, who has a less demanding and lucrative job decides to take an extended leave to care for their infant child? That situation may be unusual, but it's not impossible.

"The point is that women applicants as a class are underrated, not overrated."

Note that this is not because each and every woman is exceptional, but because some women are exceptional, and their exceptionalness is dscounted. In other words, there are ten women, with ratings of 7,7,7,7,7,7,7,7,10, and 10. That averages out to 7.6. But if you assume that all of the women are 7s, then you're off by .6. That's an under-rating.

What if a woman has a busy and successful career and her husband, who has a less demanding and lucrative job decides to take an extended leave to care for their infant child? That situation may be unusual, but it's not impossible.

HAHAHAHAHAHAHAHAHAHAHHAHA! Only gay men would do that! Unusual indeed!

"What can Larry Summers accomplish as Harvard's President that would be worth putting up with this?"

Zathras, have you seen the Harvard chicks? I'd give anything to be near them! Smokin'!!!!!

I'm glad that Imposter Palooka is trying to be funny instead of just disruptive.

I, like Corey and others, express my gratitude to Judge Posner and Mr Becker for providing this forum. I share with Corey the belief that this is an intellectually engaging blog, and thank all those who contribute.

I do think, however, if Posner and Becker ever decide to terminate comments (as many other blogs have), that it will be precisely because of juvenile antics like Imposter Palooka's (and his many, many handles). I see no reason in engaging you any further, as you have (as recently as your last post of yours on a prior thread) proven you cannot evolve to a higher standard of debate which respect for this forum demands.

"Especially when there is an easy and compelling explanation for the lower number of women in science, socialization, which is staring us right in the face." [Emphasis added]

This argument fails to separate the dependent variable (socialization) from the independent variable (natural inclination). It is a basic rule of human behavior that where individuals have a greater propensity toward a trait or desire, society always encourages greater investment in its development among such persons because society, and the individuals, get a greater return on their investment. Example: Musically talented people invest more time in music than the less talented. Example: Persons with more talent for mathematics are encouraged to invest more in this activity than people who have a knack for history. Example: A person who makes $30/hour will work more overtime than the person who makes only $10/hour, because of the greater return on the marginal effort (up to a point at least).

Thus, assuming without concluding that men have some advantage over women in math, men will naturally invest more in it because their returns are greater on two levels: (a) they will be even more succesful in the field itself, and (b) the status and financial rewards of success in math are not only valuable intrinsically, they make it easier for them to attract mates, which is not usually true for women.

Note that this analysis does not ignore socialization, but assumes that socialization is important as a dependent variable. The underlying trait is the independent variable. (Obviously the first half of the equation would not apply where the trait is negative, such as a tendency to violence, since society doesn't want to encourage that, but the second half might apply anyway, as long as the individual benefits.)

Erika and David both addressed this interesting question of "what good is it to look for differences?" That is, if we find a difference in ability that's meaningful between men and women, what possible good can it do?

What if the differences exist, but aren't just raw ability differences? Maybe there are differences in best teaching strategy, or in how to express some result so that women and men can understand it quickly. That kind of result could lead to a lot more capable women ending up in physics and math and computer science and all the other heavily male-dominated fields, which would benefit everyone.

If we're afraid to look at differences for fear of giving fodder to bigots, we will never find out if we could have addressed any of those differences in a meaningful way.

--John

I am afraid people are still making a basic mistake here:
- social outcomes cannot be used to describe biological differences. Different mean IQ scores do not necessarily mean that there is a natural difference between men and women (although it would be interesting to have a study analyzing how much of it is because of biological factors and how much of it are social). On top of this, people seem to assume that biology is always the independent variable. People must keep in mind that social factors help shape who we marry, and what abilities we try to develop. On top of this, current social conditions mean that people with certain disabilities (which are biological) can survive and reproduce. This, to me, is the difficulty of the topic: where do social factors begin and where do biological factors end?
I remember reading this article in bio-sociology a few years ago where the author claimed that displays of femininity are highly correlated with high levels of strogen in the mother during pregnancy. What this article didnt explain, though, was why certain things were considered to be feminine and others weren't, especially given the large variations accross time and space.
So yeah, biology matters, but it certainly matter allied to, and not despite of, social characteristics.
Saying that men and women have different IQ scores or different GRE scores is irrelevant because:
- these also take into effect many social factors (education, desired career - e.g. nursing versus engineering, and so on)
- these have varied widely over time.

Reading Posner's and Becker's posts on this issue has reminded me that you should never send a lawyer to do an economist's job.

What is an economist's job anyway? I have yet to
figure that one out.

Evidence from this Blog suggests that females are overwhelmingly under-represented in the field of Blog commentating.

Do you think this result is the result of a biological difference between men and women?

Or perhaps it represents a revulsion to certain types of ideology or ideological expression, but if true, is THAT biologically determined?

No, and No

As a lawyer and the son of an economist, I think the two professions have one thing in common: both believe that they are experts on everything! :)

I just found some very interesting data: the results of the "Trends in International Mathematics and Science Study" just came out in december. It measures achievement in Math and Sciences in 4th and 8th graders across the globe. They analyzed differences in scores in several countries, and while they found that in the United States boys did score higher than girls, in many other countries there was no difference, and in some women actually did better (Women did better in science in Jordan and the Phillipines, and a number of countries had differences that were nono significant). On top of that, they've that from 95 to 2003 the gap between boys and girls reduced considerably. So, repeating my previous posts, while I encourage looking at biological determinants, I am afraid that what Summers did (drawing conclusions from tests) is poor science, and certainly not directly related to biology. Here's a direct quote: "In terms of gender differences between girls and boys, there were significant differences in some countries in 1999, as in 1995. An important finding, however, is that fewer countries had these differences (3 compared to 6) in 1999. In 1995, the Czech Republic, Iran, Japan, Korea, the Netherlands and Israel all had significant differences in achievement that favoured the boys. In 1999, only three countries still have these significant differences favouring boys; the gap has decreased in others. There were no significant differences found in 1995 or in 1999 for South Africa pupils. Although in 1995, there was a 29-point difference between girls and boys, this gap has narrowed to a 16 point difference in 1999."

It seems like the consensus in the comments is that Summers remarks were legitimate, and that work on discovering any innate differences between the sexes should continue.

To critique this position (which I, by and large, hold), I'd like to consider something Posner alluded to. Posner pointed out that "no reputable" academic pursues or comments on innate racial differences or even the possibility they exist. Most here would, I would guess, hold the position that work of this sort is a far more difficult question. As purely an intellectual question, a pursuit of knowledge, they are indoubtedly similar (though they of course might have different answers). What, then, is causing this difference? The answer emerges from political and moral considerations--not scientific. I think scientists and many intellectuals fear how any apparent differences would appear to the public at large and how the information would be used. Is society mature enough to digest that kind of information? The net utility gained from new knowledge of this sort, then, seems to a definite negative in many minds.

The same concerns present themselves with the question currently under consideration. Can society digest and use this information in a mature and rational fashion? Does the information itself have a pyschological impact (self fulfilling prophecy)? How likely is it the conclusions will be useful or ultimately correct?

Part of me is deeply uncomfortable with the idea a scientist should limit her inquiry in accordance with political and moral beliefs. But if a scientist really believes the knowledge will hurt society or that society is not mature or rational enough to digest it, are they wrong from abstaining from that research?

Dpin,

Thanks for that new information. Certainly the possibility that the issue is just too complex to ever reduce to the variable we want to measure is reason enough to leave the question alone. Not to mention any potential costs the information may produce (see my last post).

I don't think it is a question that should be left alone (especially since I think in the end the results will deny sexist and racist remarks), just that is one that is incredibly complex and that should be considered carefully. But I do think that what Summers (and many here) did by simply comparing certain test results is not only incredibly naive, but it is also very bad science (it doesn't matter if he only used those results to illustrate something), and certainly has fueled sexist remarks.
I think it is particularly ironic given how upset economists (myself included) become when we see people with no knowledge of economics talking nonsensical things about economic policy and so on (which is true, especially when we look at the relationship of economists and sociologists)

Dpin said "social outcomes cannot be used to describe biological differences." I'm trying to figure out how to parse this. Surely, there can be a lot of correlation between social outcomes and biological factors--this is why the best womens' tennis players aren't as good as the best mens' tennis players, and why this is really unlikely to change in the future.

...but that was Summers's only point--that all things should be on the table. Reading his remarks makes that pretty clear. He was not making any conclusions, only positing hypotheses. That's neither naive nor bad science to do so. In addition, the basic idea that men exhibit greater variance is hardly a bogus scientific theory, having been backed up by numerous reputable studies. If anything, I think Summers was a lot more hesitant to accept the conclusions of those studies than he would be were this not such a political scientific question.

I am scandalised Mr Posner that you don't think Summers has the requisite intellectual background to make such a comment. Any economist worth his salt ought to have come to the same tentative first hypothesis..

Its neither here nor there but he did his first degree in science. Any thinking person could have made the same remarks. He messed up when he came to the apology though. You are right that he shouldn't have made a condescending apology. He shouldn't have made any apology at all.

You get this sort of idiocy dealing with people trained in the physical sciences. Lots of them have very high IQs but they are not taught to think. And they are arrogant about their smarts. And it is this arrogance and contempt that halves their EFFECTIVE IQ.

Plus what we are really talking about here is left-wing idiocy. Most economists these days have shorn themselves of left-wing idiocy before they are very old.

Economics is a study that teaches you mental flexibility. Because there are so many concepts everywhere but you can't fool yourself that any of them are precisely like the real World. Instead of one rigid big model you run about with many small models.

What Summers said would be the very first thing to occur to an economist as an hypothesis. Ask your colleague he'll tell you.

John Kelsey,
I did not say that biology and social outcomes cannot be correlated. What I meant was that simple correlations that do not control for social factors cannot be used to describe biological elements. IQ scores are not purely (if at all) biologically determined, so to use them as something that accurately describe biological differences is nonsense. As the TIMSS (the study I quoted) shows, the differences between boys and girls vary a lot across countries and accross time. To take these test scores (or IQ scores, or GRE scores) as pure reflections of biological factors is absurd, especially since it doesnt explain the narrowing of the gap in most countries, or even the superior performance of women in others. Even in sports the gaps in performance between males and females are decreasing over time, so to treat differences in performances as pure reflections of biology is not, in my view, the way to go.
Ps: keep in mind that measuring biological differences in men and women with regards to sports is a lot different that with regards to math and sciences. Height, muscle mass, and all these factors are easily observable, whereas there is still no consensus of how more "intelligent" brains differ from less "intelligent" ones.

GMB,
first of all, I would appreciate it if you were a little less condecending with all the "left wing idiocy" talk. You simply group together a bunch of people and then ascribe them motives without ever thinking much. If you have ever read Elster, you should know that economics and biology are different sciences that should be studied using different methods. The functionalist/observational character of biology is very different from the deductive/intentional approach of the economist, and while the deductive approach can be used quite succesfully in a number of different social factors, it is really not appropriate for the biological sciences (you test hypothesis in economics differently than in biology.) Just remember Friedman's seminal article on economic methodology and the "leaves on the trees" example that he gives: that might be great for economics, but not so much for biology. To prove biological differences, you have to use biology, and actually observe such differences. My problem with Summers remarks is not so much that he should not have mentioned biology, but that he ignored that the differences between genders that exist in the US are certainly not universal.

Dpin:

1. The biological differences between men and women are already well established. Even the biological differences between men and women’s brains are already established. What is not established is how much these differences matter in explaining social and economical behavior and outcome, i.e.is exactly what Summers was talking about.

2. Your examples with international studies, while interesting, do not refute Summers point. You seem to be talking about averages; he is specifically talking about the tails of the distributions.

3. Economics is no longer the study of economic behavior. It is the study of human behavior. Economics is the only true social *science*, i.e using scientific methodology in order to understand the social world, as the natural sciences do for the natural world. I honestly do not want to be condescending. The other fields (like sociology, anthropology, political science, history etc.) study subjects that are often as interesting as economics. But it is sadly difficult not to be condescending, when looking at the dismal state of most those fields, with no stringency, method and often not even the goal to drive to be as objective as possible.

That is why the by far most influential sociologist since Weber was a economics (wink wink), why political science has been swallowed by economics, why economic history is dead (except in a descriptive role) etc.

Of course, I do not want to overgeneralize, economics is not perfect and there are still many competent researchers in humanities and social sciences.

But This debate is a perfect example of the extent of the problem. Most those in gender studies are so blinded by ideology that they refuse to even engage in an open scientific debate. How can we expect them to come up with good science when that is their attitude?

4. Lastly ignoring the issues argument and instead questioning formal credentials is unscientific. Even if Summers was a Entomologist the only thing that should matter is the validity of his arguments.

Dear Dpin,

Here's a caution about inferring differences, or their absence, between male and female scientists from studies of boys and girls. I haven't read TIMSS, but accepting your characterization of the subjects as fourth and eighth graders, the study is dealing with a life stage in which females are still more mature than males. Post puberty lots of cognitive relationships change. See Nancy Cole's ETS study of gender fairness in testing.

For an example of an international comparison that might be mistaken for a cultural difference when it's primarily chronoligically conditioned, consider the country I'm in, where the university matriculation examination is taken by sixteen year olds after ten years of education. Since the girls perform overwhelmingly better than the boys and those at the top of the list go to medical school or engineering, a local form of affirmative action kicks in and girls must meet a higher standard to enter those fields. A wiser local chauvinist has advocated raising the age for career selection.

Best regards,
Jungle Gymn

Tino,
At no point did I say that differences between men's and women's brains are not well documented. What I did say was that "there is still no consensus of how more "intelligent" brains differ from less "intelligent" ones." And if you go back and re-read my post, you will see that at no point I am talking about sociology and economics, but about economics and biology (although I would suggest you ask Becker about people like Jim Coleman before throwing all work done in sociology in the trash - keeping in mind that Kahneman is also not an economist). The methods of economics are not well suited to biological topics (again, Jon Elster, who is certainly well respected among economists, wrote on this. And I doubt that when Friedman wrote about the methods of economics he thought they could be extended to biology - if have read this article, just remember his example about leaves in a tree.) My problem with Summers was certainly not his lack of credentials, as I was referring once again to how different sciences have different methods (hence why being an economist and not a biologist matters).

Gymn,
The reason that I mentioned the TIMSS study was not to "prove" anything, but simply to mention that results from the U.S. are certainly not universal. And regardless of the impact of the chronological differences, that still doesnt explain variation over time.

Anyways, I think we are getting too much into a topic that is certainly not going to be solved here. So, to conclude, my basic response to Posner's post is:
- Yes, biological differences should be studied
- No, test results do not accurately reflect "biological" differences
- Biology is not an independent variable
- the methods of economics are poorly suited to understand how biology shapes individuals. It works great when discussing things that can be solved using deduction about intentions (maximizing behavior) but hypothesizing a difference between genders and calling it "biological" does not work (the model would look exactly the same if you called it "socialization" or "the effects of using skirts"). Biology requires a great deal of "induction," of actually looking why A is causing B (again, I refer to the example Friedman used in his seminal article)

PS: in my last post where it reads "biology is not an independend variable", it should read: Biology is not ALWAYS an independent variable

I would want to point out that evolutionary biology and economics are practically parallel sciences. The relevant work for this Larry Summers subject was done by an economist/sociologist named Thomas Sowell.

Whats upsetting is the knee-jerk assumption that if their isn't homogenised representation its got to be evidence of discrimination. Thats just silly. But in any case Sowell is the best reference for this question.

An Economist

Not a biologist.

GMB,
The issue of whether economics should resemble evolutionary biology is not a new one, but I would certainly argue that, unless you take the kind of institutionalist/path dependence work that North, Hodgson and others to be the main work of economists, it is not a majority opinion. Most still follow Friedman's article on methodology (construct deductive models and see if the data fits: if it is a good predictor, then it doesn't matter how real the assumptions are. This works great for GDPs and even crime as Becker has shown us, but when it is about disentangling effects I dont think that is the way). Just look at AER and other publications and you will see the type of "evolutionary" economics (chaos theory math, path dependence)is not the norm there. I guess it is always possible (and it should be encouraged) to challenge the opinion of the majority, but I must, once more, say that the current state of economics is not apt to properly distinguish between biological factors and social ones. Before you group me with all left wingers in the world, let me say it again: I think that biological effects should be studied. I just think that the current mainstream economic way of doing things is not good enough for that. As Friedman says is his article, good economic theories are measured by their capacity to predict future outcomes, even if they are not so great at explaining current ones. To analyze the biological differences in genders, I think that we should use the average biologist's tools (not evolutionary biologists either): controlled experiments where you can actually observe certain things. I know this is difficult to do with humans, but you can always start by analyzing how the brain of otehr creatures work, and then see how the current state of experimental economics might help (perhaps this is already being done, I dont know). Doing the standard "create-an-abstract-model-and-relax-assumptions according-to-the-data" would not work here, simply because it is too complex an issue. My issue with Summers was more of a methodological one than an ideological one (especially after seeing so many people here showing some test statistics that alledgedly prove one thing or another) and I really do think that he should NOT have apologized. I think that a statement in the lines of "I never said that there were biological differences, only that it was a possibility and that it should be studied. I was simply doing my job of stimulating academic debate" would be more than enough.
As for the politics of it all, the "left" does not have the monopoly over external influences on scientists. Just look at this issue: Summers simply said that differences should be studied. He never said that they exist for sure. Left wing feminists interpreted him as saying that there actually are differences, and complained about it, and some chauvinist conservatives also thought that he was actually saying that there differences, and thought that Summers was on "their" side. (just look at some of the posts here). I just heard one of these right wing talk show hosts on radio who really thought that Summers was affirming that a biological difference existed (instead of just suggesting that people should look into it) and complained about the "left's" intereference in science. The irony is that this same person supports Bush on stem cell, and he has actually campaigned against some scholars (and sometimes entire departments) who he disagreed with. So, while I don't think any one here would question the fact that the left really does dominate academia, I do think that they are not the only ones exerting outside pressure on what topics should and shouldn't be studied or taught (evolution, anyone?)

It was far from condescending to describe this stoush as being about dumb-leftwingers. Thats the be-all and end-all to this situation.