Each year publishing agent John Brockman asks a deep question of some of the world’s leading thinkers, many of whom are his clients (via Peter Woit). It’s what you’d expect: a mixture of quite interesting ideas and rampant nonsense. This year Nobel Laureate Philip Anderson says some misleading things about cosmology, which maybe I’ll talk about later; but the prize for the worst response comes from Steven Pinker.
To make a long story short, here’s Pinker being saucily provocative:
In January, Harvard president Larry Summers caused a firestorm when he cited research showing that women and men have non-identical statistical distributions of cognitive abilities and life priorities.
That’s what’s known in studies of rhetoric as a “blatant lie.” It’s true that Summers caused a firestorm; it’s also true that he cited such research. It’s just not true that it was the citation that caused the firestorm. The firestorm was caused when Summers suggested that differences in innate aptitude were more important than systematic biases in explaining the gender gap among professional scientists. He said this despite the existence of overwhelming evidence against it, including from the very sources he was citing. How loudly do we have to shout this? The “dangerous idea” is not the possibility of innate differences; it’s using them as an excuse to ignore the obvious and pernicious effects of discrimination.
You may not agree that discrimination is important; fine, we can have a reasonable debate on the merits. This does not give you license to misrepresent why people were upset about Summers’s remarks. I can see two possibilities:
- Pinker is an idiot, and can’t understand the difference between suggesting that differences exist and claiming that they explain the gender gap.
- Pinker knows exactly what he is doing, and is lying intentionally to score some points of his own.
I honestly don’t know which it is. But I agree that spreading these lies, intentionally or not, is truly dangerous.
January 2nd, 2006 at 5:57 pm
I don’t think it’s either 1 or 2. I think Pinker’s a pretty bright guy (perhaps not exceptionally so), and I don’t think he’s intentionally lying. I just think that he’s incredibly myopic. He sees something related to his own “work” (and by “work,” I mean his writing about work that others are doing in areas that he hasn’t really researched outside of the narrow evolutionary focus), and he fixates on them. If Summers had gotten heat for comments he’d made about cheater detection, not because of the research he cited but because of Summers’ drawing of absurd social and political conclusions from that research, Pinker would have focused exclusively on the cheater detection and not on the conclusions that Summers drew from it. It’s the Evolutionary Psychologist persecution complex.
January 2nd, 2006 at 6:02 pm
Chris, I actually think you’re right, although I would count that as answer #1. Pinker is obviously not a stupid guy, but if he’s so excessively blindered in this particular area that he makes idiotic pronouncements, then it amounts to the same thing. I say this more-in-sorrow-than-in-anger (although there’s a lot of anger), as I thought The Language Instinct was a fantastic book.
January 2nd, 2006 at 7:16 pm
At least some of the blame is the idea that one has to be provocative to be interesting. I wish sometime those amazing people would be asked simply “what are you working on”, as we always ask each other, instead of trying to extract from them their least reliable ideas. I bet the answers would be much more thought provoking.
January 2nd, 2006 at 7:48 pm
To me, almost everything there tilts toward the “rampant nonsense” end of the spectrum. Tipler says something amazingly nutty about the Standard Model providing the basis for a gigantic bomb. Kurzweil spouts his usual insanity, based on mindless extrapolation of an already failing Moore’s law (how did he ever get to be famous?). Carlo Rovelli claims most physicists do not understand relativity and quantum mechanics, an insult without basis.
The rare things that aren’t nonsensical are trivial. Jared Diamond points out that tribal cultures sometimes fight and damage their environments. So they’re not all perfect “noble savages”? I’m sure Rousseau is disappointed. Philip Zimbardo points out that ordinary people have the capacity for evil and for heroism. These are things that only someone who is poorly educated or naive would doubt. I agree with Moshe — I would be much more interested in the details of Zimbardo’s current work than in a recapitulation of what everyone should know based on his classic prison experiment (if not from common sense), for instance.
Somehow this survey manages to collect a number of intelligent and interesting people (together with a few fools) and produce a lot of garbage. Even the most interesting ideas seem more like science fiction premises than thoughts about the real world. This really should be better.
January 2nd, 2006 at 8:00 pm
Unfortunately it’s not just an Evolutionary Psychologist complex. Conversations on this topic with a wide range of professionals over the past year (a lot of physicists of course, but also physicians, lawyers, business types) have been revealed a larger percentage than I would have predicted who still think (and obviously feel comfortable voicing) the idea that that the small numbers of women in physics is probably due in large part to some combination of innate differences in ability and/or interest (with some additional component due to cultural influences over which the speaker claims to have no control). Further they are completely convinced that they themselves are free of any gender biases (which none of us are) and in no way contributing to these cultural effects.
I put it down to the same reason why Bush was elected to a second term.
Part apathy, part laziness, part some human inability to think rationally about things which threaten to overturn one’s world view. Clear logical arguments, supported by data, facts, statistics, research, are blatantly ignored if they don’t agree with what one wants or somehow needs to be true. It has been depressing to recognize that scientists, who are trained in critical thinking, are not immune.
The threat here is twofold — our view of ourselves as pristine thinking machines, unsullied by such philistine behavior as interacting (as teachers, colleagues and mentors) with men and women differently; and our view of our system as essentially merit based.
January 2nd, 2006 at 8:57 pm
Well I must dissent. It seems to me that the amount of heat that you direct at someone (calling them a liar, etc) should be proportional to the degree of the misrepresentation, if any. And as directed against Pinker it seems ludicrously out of proportion. It seems to me that what Pinker is really saying is just what you deny—that not only are there intrinsic differences in aptitude but that these may well be the primary explainers of different representation. Against that you shout loudly that there is evidence against, but I have looked at your evidence and find it incredibly unconvincing. (It doesn’t, for example, distinguish between different IQs across genders and different degrees of mathematical aptitude across genders. Nor does it allow that different occupations have different IQ demands.) Pinker may well find it as unconvincing as I do. Suppose he does. Then what he has said is not wildly inaccurate. He is saying that there are differences in aptitude and ability and that these have a strong effect on the number of female physics/engineering graduates/faculty. And after all, even if there are some other factors mitigating against female involvement in graduate level physics (and who denies that there are?) these are additive: they don’t cancel out.
The flimsy evidential base that you have does not really, should not really, entitle you to lambast either Pinker or Summers. And that you have to shout it suggests that deep down you know this.
January 2nd, 2006 at 9:14 pm
Eluard, if Pinker is arguing that the differences underlie disparities in academia, then he may not be stupid or lying, just ignorant of the wealth of data that argues against his point, and the paucity of data that supports it. Of course, we already know that he’s ignorant of the fact that there is no direct evidence for innate causes of the differences in mathematics between men and women, and in fact, evidence that the differences are largely related to the types of math questions you ask.
January 2nd, 2006 at 10:01 pm
Eluard:
I’m not sure why it seems that way to you, since that is manifestly not what he said. He may think that, and it’s something interesting to discuss, but he very clearly didn’t say it. Instead, he made a completely distinct claim that misses the point. And the reason why it’s worth shouting about is because this kind of lie serves to obscure the real issue, which is one that dramatically affects the lives of a large number of people.
It’s like arguing with the Bush administration. You say “You can’t tap the phones of Americans without a warrant. That’s illegal.” And they say “You don’t want us to listen in on conversations between suspected terrorists?” And you say “The point isn’t the wiretapping, it’s the warrants. You need to go get a warrant!” And they say “Well, you might think it’s not a good idea to gather intelligence on people who are attacking our freedoms, but we do.” And you say “The warrants! Checks and balances! Rule of law!”, and then you pop a blood vessel and die. And the liars have won.
January 2nd, 2006 at 10:32 pm
I have listened to about three Pinker debates on this subject.
In each case, Pinker was extremely explicit about what he was saying, namely, as you quoted above, that all current indications are that the distribution of certain types of abilities and interests (especially at the extremes) differ between men and women. And in each case, his opponent appeared to be a complete moron who was unable to comprehend the nuances of this statement. His opponents would go on to rant about the abilities of “average” men and women, or about how children were taught in schools or various other things that have no bearing on Pinker’s point.
For what it’s worth, in all three cases, Pinker also went on to state that these were differences at the very extremes of the curves, that they had no practical bearing on the behavior of the bulk of the curve, and that the existence of such differences did not obviate our having a moral duty to treat men and women as equally valuable.
My take on this is that Pinker is damn pissed off at being surrounded by people who, wilfully and maliciously, or through stupidity, are unable to comprehend the *specifics* of what he is saying, and that this has made him perhaps a little impolitic in his words, a little more interested now in ramming across his initial point (a scientific point that I have never seen disputed) than in trying to calm things down, and has him remembering what caused what emotions in others somewhat differently than you do. You, Sean, are extremely interested in something else, namely what social issues are holding women back from being more represented in the hard sciences in the US. This is something perfectly reasonable to be interested in, but, as far as I can tell, it is of little interest to Pinker. So, for example, your “refutation” regarding bell curves, namely that fewer girls than boys in the US are interested in going into science as a career, while true, is basically irrelevant to his point of what the curve of abilities looks like. My suspicion is that if his opponents would actually concede his point, or provide convincing evidence against it, rather than attacking strawmen, he would be quite willing to stand up publicly and decry the same various social ills his opponents decry — but he’s not going to do that as long as doing so implies that he has to state that black is white with respect to what he considers to be well established scientific data.
January 2nd, 2006 at 10:40 pm
It seems that some sort of connection could be made with Judith Rich Harris’s response to Brockman’s question, The idea of zero parental influence:
January 2nd, 2006 at 10:50 pm
Pardon my intrusion, but I have a few brief questions and comments:
1. You said: “The ‘dangerous idea’ is not the possibility of innate differences; it’s using them as an excuse to ignore the obvious and pernicious effects of discrimination.” In a post about rhetorical dishonesty, this comment seems rhetorically dishonest in its own right. After all, nobody would argue that any extant discrimination is not a bad thing, and would not have “obvious and pernicious effects”. The issue at hand is the level at which this claimed discrimination is actually present.
2. I was confused by some of the “overwhelming evidence” in your old preposterous universe post. In it, you reference the plot from the AIP study, which shows that while nearly half of high school physics students are female, only a quarter of bachelors degrees in physics are awarded to women. You use this to put the kibosh on the innate ability / career choices claim. But isn’t HS vs. college an apples vs. oranges comparison, and can’t this at least partially account for the strong drop-off? It seems to me that there is limited freedom to choose your courses in HS, while there is nearly infinite freedom to do so in college. Perhaps the plot indicates that fewer females in HS physics classes want to be in those classes in the first place (much as I felt in my HS Spanish class) — for whatever reason. I certainly don’t see this HS-college drop-off as clear-cut evidence for discrimination.
3. University faculty (in sciences, humanities, everything) are overwhelmingly on the left side of the political spectrum, and the same appears to be true for high school teachers. (I am sure that you and Macho would agree that the lack of conservative professors is due to the knuckle-dragging ignorance of people like me, but that is a discussion for another day.) As we know, liberals fashion themselves as the flag-bearers for diversity, and overtly abhor discrimination of any kind. But if this is case, then how could there possibly be so much discrimination (resulting in the ultimate paucity of female faculty) in the educational institutions dominated by liberals?
January 2nd, 2006 at 11:06 pm
Charles– 1. Actually, the issue at hand in my post was why Pinker would choose to lie about the origins of the firestorm caused by Summers’s remarks. 2. In my earlier post, I said quite clearly that discrimination did not uniquely predict a drop-off between high school and college; however, “innate aptitudes” does make a prediction, and it doesn’t fit the data. 3. I see no reason to doubt that, in a group of people including a majority of self-identified liberals, there can’t be a large number of people who (gently or otherwise) discriminate against women — in fact, it happens all the time.
Maynard– Thanks for opening my eyes to this important point. I finally understand that poor Steven Pinker just wants to talk about bell curves. He didn’t want to start off his paragraph by talking about Larry Summers’s speech, which happened to be precisely about what issues are holding women back in science. It was forced on him by his irritation at all those unworthy debating opponents. Now I’m much more sympathetic.
January 2nd, 2006 at 11:11 pm
Gee thanks for that reply, Sean.
One of the posters in the Edge article, http://www.edge.org/q2006/q06_4.html#baroncohen, suggests trying a political system based on empathy. I guess we’ve just seen how well that idea would work. I try to explain to you what I think the issue is all about, as seen from Pinker’s point of view, and your response is not to concede that I/proxy Pinker may have a point, but to belittle me.
January 2nd, 2006 at 11:23 pm
Maynard, if I seem belittling it is out of an extraordinary frustration with repeated elaborate rhetorical dances that seemed designed to obscure the point, not to clarify it. If all Pinker wanted to do was to talk about scientific data bearing on the issue of people’s innate aptitudes, that would have been fine. Interesting, even. But I’m sure you can see this: he chose to frame things in terms of the Larry Summers controversy, and he did so to make his position seem more “dangerous” than it really is, and he did so by lying. I am amazed how many of his “defenders” seem to think that Pinker is incapable of expressing himself clearly and directly. He is not moving any conversation forward; he is trying to score cheap points at the expense of honest discussion.
January 3rd, 2006 at 12:01 am
At birth people with dark hair and those with light hair are separated into two groups: dark hair are brought up to be left handed; light hair trained to be right handed. The training is intensive and round the clock. They are actively discouraged from using the “wrong hand” and receive positive reinforcement for using the “correct” one. As children they are taught to lift weights with the correct hand while the wrong hand dangles uselessly at their side. Of course, it’s a more enlightened approach than that of 50 or 100 years ago, when the wrong hand might be bound or even amputated, but everyone “knows” which hand a person should be using. There is strong evidence that supports this view — starting in high school students undergo a battery of hand strength tests. The results of these tests fit very nicely onto bell curves which demonstrate that light haired people are innately right handed; those with dark hair are, on average, stronger with their left hands. And, amazingly, even in the tails of the distribution, the tests clearly show a preference for handedness that is correlated with hair color.
Pinker seems to be interested in these statistics. I find them less than useful. I can’t rule out an innate correlation between handedness and hair color, of course, but given my observations of the training and the well-documented program of reinforcement, I feel confident that to first order the training is the dominant factor.
Assume further that the scientific community is set up to accomodate lefthanders. If my goal is to train and encourage the next generation of scientists (as opposed to document the results of the hand training) I first need to recognize that
(a) the training has occurred, so that I can work to modify or eliminate it, since my experience leads me to believe that each individual does best when allowed and encouraged to find out which is their dominant hand and then use it (see, even we liberals can believe that there are innate traits — we just tend to be skeptical that they are all tied to hair color. Tendency to sunburn, ok; handedness, doubtful)
(b) that science doesn’t have to be so strongly handed either — no reason a
left-handed tool is better at constructing a dark matter detector than a right handed one, for example.
January 3rd, 2006 at 12:15 am
Sean, this dancing around is hardly the sole domain of Pinker.
Let’s ask a very simple question: do you concede the validity of Pinker’s point? Yes or no. And if no, why not?
Let’s consider the issue more generally. Presumably you will concede that there are at least *some* differences between men and women. The testosterone production curves are substantially different, the height curves somewhat different.
OK, now you will presumably concede that there are also some *mental* differences between men and women. For example the “willingness to look at naked females for long periods of time” curve is different. And you will concede that there are differences in at least *some* skills, for example the skill of taking math tests in the US.
OK, so your claim in your third article which seems most germane at this point is that this skill of taking math tests is irrelevant to the actual skills universities should value in the mathematics faculty they hire. This is plausible; for example these tests usually require one to think very fast, which punishes those who can think deeply but not fast. But one can just as easily claim that these tests, while not perfect, certainly reveal a large element of mathematical ability. So one might ask: are you claiming that a better test of mathematical ability between men and women would not reveal differences at the high end?
A second objection one might bring up is that these tests, if applied to the young, might not be useful because the brains of the young still have plenty of (biological) changes ahead of them, while if applied to teenagers, say 16-18 yrs old, they start to reflect social conditioning. A rebuttal of this might be that mathematical ability at the high end is sufficiently innate that it will out regardless of schooling if presented with the right circumstances — for example Ramanujan. This would seem to be something amenable to empirical research: is it the case in Europe, the US, and even more so India and China, that some kids with lousy schooling are discovered to have substantial innate mathematical talent when they get to college? Another direction of research would be to ask what happens when one looks at the distributions of test results when considering only girls taught under favorable conditions, eg girls from all-girl private schools in the US, or girls from those countries that have substantially higher representation in their university math/physics depts.
So we reach the following point, I guess:
1) current test results
2) how would those results change under a different social system
3) do the tests adequately correlate with what we really care about
All three of these are, at least to some extent, empirical questions. Pinker says that the data regarding 1 are pretty uncontrovertible, and that (as I recall, may be wrong here) that (3) is also covered (at the hight end, once again, the issue of the “average” student is not the issue).
You have said that you are opposed to verbal dancing around, so I’d be interested in seeing your response to these points.
January 3rd, 2006 at 12:44 am
Maynard, I’m not sure precisely what point of Pinker’s you refer to, so I can’t judge its validity. (Except his point that I actually quoted, which is a lie.) But I’m happy to say what I personally believe, which I’ve said often enough in the past.
Are there differences between how boys and girls do on standardized tests? Yes. Could there be statistically significant innate differences in the mental capacities of men and women? Sure, I don’t see why not. Do we understand what they are? No, not very well at the moment. Are we able to distinguish, with the current state of the art, between what is innate and what is determined by culture? To be honest, I’m not an expert; my impression is that we don’t really know, but there’s certainly controversy about the point. (Chris is much more of an expert than I.) Are there significant systematic biases that discourage women from becoming scientists? Yes. Is it plausible to suggest, given what we currently know, that differences in innate abilities are more important than these systematic biases in explaining the differing representation of men and women as professional scientists? Absolutely not — there are mountains of data saying exactly the opposite.
But — wearily repeating myself — none of that was the point of this post. The point was the objection to Summers’s remarks was not because he suggested that innate differences exist. Whether or not they do exist, what impact they have, etc etc etc — all quite fascinating, but not the point of the post. If you want to have an honest debate, don’t start it off with a lie. That’s all.
January 3rd, 2006 at 2:52 am
Somehow, I feel the urge to reiterate my previous comments on l’affiare Summers.
January 3rd, 2006 at 3:17 am
Alright, I just read the article in question and I really don’t see what the big deal is…
Sean, lets assume that research proves that innate diffrences really are the main cause of the gender gap in the sciences. Now if Summers then made those comments, there would still have been a firestorm.
The fact is that the mere suggestion that innate differences are part of what caused the controversy, but of course not the only part.
You have missed completely the point of Pinker’s article… congratulations.
January 3rd, 2006 at 3:22 am
He cites Charles Murray.
Ugh.
January 3rd, 2006 at 5:47 am
Sean
I’d like to suggest an alternative possible defense of Pinker, not based on the supposition that he is asserting what you deny: namely that intrinsic differences are the primary explainers. Let’s suppose he is saying no more than he seems to be saying: intrinsic differences exist. You say that this would make him a liar, because he is misrepresenting what everyone was so upset about in Summers’ speech. But was it? Certainly the complaints that I heard against Summers was merely that he had suggested that there are intrinsic differences at all. That was enough to make people mad. Now if you admit that there were people who had that view (and I don’t see how it could be denied) then Pinker is not a misrepresenting liar at all. He is just addressing a group who are asserting something that you yourself wouldn’t assert.
Second point. You say that there are mountains of evidence pointing to intrinsic differences not being the primary explainers, but rather it being mostly bias. I doubt that you are right about this. The main problem is that social research is just not that good. Certainly questionaires are almost useless. Suppose you ask someone who has just lost a job why they lost it. If they blame bias or prejudice you cannot tell from that whether they did in fact lose the job for that reason. Maybe they weren’t any good, and, rather than face that fact, they invent an irrational hostility in place of the perfectly rational one. You just don’t know. Note I’m not denying that there is bias, just that we have no way of really quantifying it. (Thus note that people don’t complain of bias when it is in their favour—but bias in favour of a group must be as common as bias against a group.) The other thing is that reasons for women not going into physics probably interact with one another in complex ways. For example people tend to pursue the thing that they do the best. Thus suppose (just suppose) that women in HS were as good as their male counterparts in physics and maths but were better in Languages. Then they may well pursue languages to the detriment of physics enrolments at College. (So equal ability would lead to *lower* enrolments.)
It is just horrendously difficult to know why people do what they do—and in a complaint culture self-reports of bias just can’t be taken at face value. We all know cases where it is self deceiving.
At any rate I’m just offering this as a reason not to think that Pinker and Summers are wrong, or at least not so vehemently think so.
January 3rd, 2006 at 6:10 am
Would it be too farfetched to hypothesize that the real reason for Sean’s recurring fits of unthinking feminist hysteria are that he’s discovered their powerful aphrodisiac effect on a segment of the female population? The answer to the question “Dangerous, stupid, or simply dishonest?” would then be “only to easily duped women; like a fox; kind of, but in love and war…”.
January 3rd, 2006 at 6:49 am
Sean: I think you are overreacting a bit. Even if what caused the firestorm for you was the suggestion that innate differences explain most or all of the gender gap, that doesn’t mean that this is what caused it for most other people. The idea of innate differences *is* controversial, and it is controversial to even suggest that it explains *part* of the gender gap. Far from everyone is open to consider the issue from your perspective, acknowledging both as possible contributors to the effect.
All the same, I was wondering how you jump from what you cite as “overwhelming evidence” to that there is actually discrimination involved. You just state this as obvious, but, well, it’s obvious to precisely those people who already agree with you, not to others. So the main effect happens around college, when girls and boys make their choices as to what to study. What discrimination has happened up until then? Why are the girls victims of some bias or some conditioning? Why are the girls simply not choosing what they think will be the most interesting to study, what will be the most satisfying career for them and so on? Why can’t we expect boys and girls to make such choices differently, without resorting to explanations involving discrimination?
January 3rd, 2006 at 10:49 am
Your reaction is completely overblown, and you should apologise for even suggesting him being an idiot. First, it is not helping the debate to use emotional arguments. Second, maybe he would even agree with you that his formulation of the Summer controversy was not a very accurate one. Third, if it is an oversight, it does not in no way falsify his ideas on gender and discrimination.
You are missing out interest. The difference between the sexes is not just in the distribution of aptitudes for various tasks, BUT ALSO in the difference of inate distribution of pleasure and satisfaction experienced for various tasks.
It is dangerous to conclude discrimination just on the basis that the sexes are not equally represented. Take for example kindergarden teachers or nurses. Are we discriminating against men?? Sure, there is some teasing I guess, but the big difference comes from the fact that MEN ARE JUST NOT INTERESTED in general!!!
January 3rd, 2006 at 12:13 pm
Certainly the complaints that I heard against Summers was merely that he had suggested that there are intrinsic differences at all.
I heard a lot of rightish rags reporting that those were the “complaints” but I heard very few firsthand complaints along those lines.
January 3rd, 2006 at 1:09 pm
Noone in this debate ever suggested that such differences might not exist (other than conservatives complaining about people demanding equal rights). What they did complain about was that it is important or relevant to study such differences when very significant discrimination still exits, and this discrimination clearly has the ability to obscure any difference amongst the populations. And anyway, is it possible that inherent differences are significant enough to explain the fact that there are 15-20 times as many men in pysics as there are women?
And all it takes to realise that there is sexism in physics is to just look around with an open mind. In my personal experience (I haven’t been a member of a statistically significant number of institutions), there is abject sexism that nearly reaches the level of what I thought was relegated to Lifetime movies, AND noone cares that said sexism exists. Our department bends over backward to keep the male students in the dept, but just casually gives women their walking papers.
Noone cares about this shit, because it is the woman’s place to shut the hell up, and the one person who had called people on this was instantly labeled a bitch and discounted, and has since left the department.
January 3rd, 2006 at 1:10 pm
Interesting link somewhat related to the above…
http://www.seedmagazine.com/news/2005/12/jews_on_jews_jews_are_great.php
Per the article, and in accord with some suggestions above, I’d say the problem isn’t so much Pinker, as he’s usually pretty cautious about what he asserts and why, but rather what people take away from his statements. He’s a good funhouse mirror that way, whatever his other gifts or deficits.
Take home message: Genes certainly matter, perhaps a lot in some circumstances, when it comes to at least some aspects of intelligence. Genetically, men and women aren’t quite the same. Neither are people of differing ethinic backgrounds. Certainly bias is to blame to some significant degree for virtually any demographic trend one cares to look at, but to fixate on innate biology to the exclusion of environmental factors, or to dismiss biology out-of-hand when it seems “dangerous”, are political choices, not scientific ones. As an evolutionary psych guy, Pinker has his own biases, I’m sure, and I tend to be far less confident than he is about some of the stuff he asserts. I’m pretty agnostic about the whole thing, , really, and doubt difinitive answers to some of these sorts of questions (e.g. “Are men, on average, better at math?”) can be had, given the confounding influences of everything else not encoded in our DNA. The proper tests could probably never be ethically performed.
That said, I don’t see Pinker’s confidence in evo-psych, informed as it is by actual experimental data (even with the datas’ flaws, of which I’m sure Pinker is aware), as being less justified than the faith some other scientists put in their own ideas these days.
January 3rd, 2006 at 3:29 pm
invcit:
All the same, I was wondering how you jump from what you cite as “overwhelming evidence” to that there is actually discrimination involved. You just state this as obvious, but, well, it’s obvious to precisely those people who already agree with you, not to others.
Tom Weidig:
It is dangerous to conclude discrimination just on the basis that the sexes are not equally represented. Take for example kindergarden teachers or nurses. Are we discriminating against men?? Sure, there is some teasing I guess, but the big difference comes from the fact that MEN ARE JUST NOT INTERESTED in general!!!
Guys, please go back and read over the details of the arguments and evidence Sean cited in previous posts on this subject instead of just jumping to conclusions based on the summaries of them in this post and comments section. Sean was not arguing that just because the sexes are not equally represented, that automatically implies discrimination; for example, in an earlier post he pointed to the extreme differences in female representation in physics and math between different countries, which I think you’d agree is extremely unlikely to have a biological explanation (and to head off anyone bringing up racial differences, note that these huge disparities in female representation can be found between different european countries).
January 3rd, 2006 at 3:36 pm
Overlooking the primary point in Sean’s post that he repeated in comment #18, there are a couple of points about the role of gender differences in career selection that don’t seem to have been emphasized enough yet in the interesting comments above.
First, regarding empirical studies of the role of discrimination, I wonder about regional differences. I have little doubt that there are important regional differences in the degree to which traditional role models and career choices are pounded into children. Taking the US as an example, there is little doubt (in my mind, at least) that socially conservative, less educated population centers are less likely to be supportive of women entering the hard sciences or engineering than a technologically oriented area like Silicon Valley (where I live). Certainly among the people I know, parents seem quite supportive of their daughters taking career paths into the sciences, especially since those careers tend to pay significantly higher than most other areas of employment. Assuming there have been studies that consider these regional differences, do they deomonstrate a trend toward nearly equal representation in the sciences?
Second, a key point mentioned in comments #21, #23 and #24 needs to be reemphasized — even if there are no innate differences in aptitude, there can be innate differences in interest. Taking a strictly evolutionary point of view, if one of the sexes didn’t have a nurturing instinct and the other have a more adventurous instinct (so as to explore for food and shelter), there would probably have been a lot of ignored or inadequately fed children. And those children would definitely have been at an evolutionary disadvantage… So it doesn’t seem like much of a stretch to expect differences in level of interest in certain “exploratory arts” (science/math/engineering), distinct from aptitude. As a personal example, my daughter is in high school and has an aptitude for math and science, but has little interest in pursuing math or hard science once she enters university. She will be a contributor to the APS statistics that show a significant fraction of girls who took physics in high school don’t do so in university (hey, taking physics in HS looks good when applying). My son, however, has aptitude but also an interest in math and science, and so his choices are likely to be different. Both of them have have been encouraged to think independently, both parents have careers in science/engineering, and neither have been significantly exposed to the kinds of discrimination that Sean deems to be the primary culprit.
January 3rd, 2006 at 4:14 pm
M:
This is exactly the problem with taking such “evolutionary” viewpoints. Why does the distribution of these “instincts” have to be determined by gender? Why can’t both sexes have both instincts? Why can’t some men have a “nurturing” instinct, and some women have an “adventurous” instinct? A priori, this type of reasoning doesn’t give you any evidence as to why not.
And anyway, these types of arguments would be much more convincing if there were any effort toward eliminating the gender discrimination that exists.
January 3rd, 2006 at 4:16 pm
“It is just horrendously difficult to know why people do what they do—and in a complaint culture self-reports of bias just can’t be taken at face value.”
This pretty well sums it up. I could tell you some horror stories…but I would just be a whining, untalented female that should know better than to pursue a ‘career’ in physics.
January 3rd, 2006 at 4:18 pm
I would go with “dishonest.” I wikipediaed (wikipeded? wikipediaëd?) Larry, and noted that Pinker has been an active participant in the Summers debate ref, and is surely aware of the details of the fight, and the quote:
“So my best guess, to provoke you, of what’s behind all of this is that the largest phenomenon, by far, is the general clash between people’s legitimate family desires and employers’ current desire for high power and high intensity, that in the special case of science and engineering, there are issues of intrinsic aptitude, and particularly of the variability of aptitude, and that those considerations are reinforced by what are in fact lesser factors involving socialization and continuing discrimination. I would like nothing better than to be proved wrong.”
That said, it seems helpful for Pinker to claim otherwise in other places as well.
(PS: you know, re-reading that Summer’s quote, I was actually kind of shocked. I had forgotten that Summers was claiming a whole bunch of crap, including the apparent implief “fact” that women are “low power and low intensity”.)
January 3rd, 2006 at 5:02 pm
S. DeDeo,
Summers implied that men and women are “equal power and equal intensity” in the mean. The differences were suggested to exist in the variance, with men having wider tails in science and engineering (therefore lower lows and higher highs). His speech is linked in the wiki article that you reference.
January 3rd, 2006 at 5:24 pm
Hi Ken — I would reply substantively, but doing so would violate a number of New Year’s resolutions I’ve just made.
January 3rd, 2006 at 5:28 pm
Why do people assume that the tail of the bell curve is all that is relevant for people doing science? In my experience this is far from true. The people in the tail walk many different paths. It might be true that scientists tend to sit at the upper end of the curve, but this is very different to discussing extremes.
January 3rd, 2006 at 6:03 pm
Kea – “Why do people assume that the tail of the bell curve is all that is relevant for people doing science?”
It’s only relevant if you assume that tenured faculty in physics and math rank two or more standard deviations above the general population in mathematical ability. In that case, it’s absolutely central.
Otherwise, it would be as silly as assuming that location on the bell curves of height or athletic ability was relevant to success in the NBA.
January 3rd, 2006 at 6:21 pm
The sarcasm in the previous post was inappropriate, and I apologize for it, but the point is that just because some tall or athletically talented people don’t play in the NBA is irrelevant to the fact that every NBA player is athletically talented and most are three or more standard deviations taller than the general populace.
I pretty sure that most tenured physics and math faculty are also well out on the upper tail of that bell curve.
January 3rd, 2006 at 7:05 pm
CIP
I wouldn’t know about tenured physicists, since I cannot imagine ever being one myself. But I have many anecdotes to illustrate my point. For example, in primary school (under 12s) I was sent to a special school for gifted children (this was quite a long time ago). As far as I am aware, I am the only member of that class that eventually took up physics.
January 3rd, 2006 at 7:07 pm
OK. You will make the same objection again, I guess. We are simply defining the ‘end of the Bell curve’ differently.
January 3rd, 2006 at 7:38 pm
If you want to see the relative importance of the width of the distribution vs. the central value, read
http://www.lagriffedulion.f2s.com/math.htm
La Griffe du Lion verifies that the representation of the sexes in target groups of various degrees of selection fully agrees with the Gaussian curve. Finally, a female Fields medal is predicted once per 103 years.
January 3rd, 2006 at 7:52 pm
Buried somewhere on this blog or Sean’s previous blog is a citation that claims that the “leaky pipe” isn’t quite true for US women; more end up with bachelor’s degrees in science than start out in a bachelor’s science program. The authors of that citation use that kind of evidence to suggest that one has to study the life-histories of women to understand why and where they end up specializing rather than these statistics of average and 95-percentile standardized scores for men and women and so on.
I also think that if average human abilities and temperaments differ among groups for genetic reasons, then we should find the genetic causes for treating people as group stereotypes and not as individuals and attempt to remedy those.
January 3rd, 2006 at 7:55 pm
Kea
I said that such reports can’t be taken at face value, not that they can all be discounted. But then, you went to a gifted school so you know perfectly well how just plain irrelevant your response was.
My point was that getting any reliable data on bias is going to be incredibly hard. I don’t think people have really faced up to the methodological difficulties. I wasn’t saying—how tedious to have to repeat it—that people don’t face bias and prejudice in their lives.
January 3rd, 2006 at 7:58 pm
Here is a story from India about women in science there.
http://www.flonnet.com/fl2226/stories/20051230003203900.htm
January 3rd, 2006 at 8:06 pm
“I said that such reports can’t be taken at face value, not that they can all be discounted. But then, you went to a gifted school so you know perfectly well how just plain irrelevant your response was.”
I do not recall having told you about the rest of my 38 years.
January 3rd, 2006 at 8:07 pm
“My point was that getting any reliable data on bias is going to be incredibly hard.”
There is reliable data. HOW MANY TIMES DOES SEAN HAVE TO TELL YOU.
January 3rd, 2006 at 8:19 pm
M #30:
I love this kind of “reasoning”. But I’m confused — can you outline the reasons for the innate lack of interest in physics among African Americans?
(they’re even more underrepresented, which is the data being used to hypothesize that females may be innately less interested — and I certainly know some very talented African Americans who are not interested in science, so there’s your anecdotal evidence as well).
and
So they have been completely unexposed to other children and adults, no tv movie, radio, magazines, popular culture of any kind? And you and your spouse were also raised in completely gender blind environment?
Flaming liberal that I am, I know that there is no way that I am gender neutral in every interaction with my children. Good intentions are extremely helpful and awareness of the issues involved are esssential, but we all live in the real world.
I can put my WIMP detector far underground in order to minimize the backgrounds, but I still need to monitor and account for cosmic rays, radioactivity in the rock walls, etc. It may be the best WIMP detector around, and I’m very proud of it, but pretending that cosmic rays don’t exist because I don’t want them to or because they mess up my data just doesn’t work.
I don’t mean to pick on you, because you sound like someone who truly wants to do the right thing, but I’m frustrated by your response, which is all too typical of too many of my colleagues. The people who are making a difference in recruiting and mentoring outstanding women scientists (and their efforts are working, which also argues against any innate differences as the dominant source of the disparity in numbers) are those who have first carefully assessed the biases and then worked to equalize the playing field.
I have to run now but will post some references later.
January 3rd, 2006 at 8:59 pm
Goodness, Lubos! You refer to Bell curves for 15 year olds from the 1960s! I don’t quite remember the 1960s, but the 1970s were bad enough.
P.S. Thank you for banning me from your blog. I am wasting too much time doing this sort of thing.
January 3rd, 2006 at 9:44 pm
Kea: “Thank you for …”
On the contrary, thank you – pleasure is on my side. The Bell curves don’t change significantly in a couple of decades because they are of biological origin as you obviously can’t understand, and 15 years is the critical age in which the gender gap fully develops.
January 3rd, 2006 at 9:48 pm
macho
your response seems to be indicative of many of the things that are wrong in this debate. No matter how positive a story you are told you insist that there must be some residual uncorrected bias that is responsible for why a talented high school girl would not go into physics. You then list a whole lot of environmental factors that you have no idea the effect of and proclaim that this good news story is not really good news at all. Yours is not the attitude of science it is that of a religous believer—of that fundamentalist paranoid stripe that can always find yet tinier evidence of the devil’s handiwork.
And your own account of equalising the playing field sounds much more like supplying an affirmative action non-even playing field.
This debate alienates people because it does tend to take on the form of a religous dispute. It should be a dispute about the weight of evidence. Maybe as someone (Aaron?) has pointed out, Summers was oafish, even boorish. But Pinker I think just wants this to be a scienfic discussion. That should be good thing.
January 3rd, 2006 at 10:04 pm
Hehe, if the raw scores from 2000 on the IQ tests were scored based on the 1960s distribution, then we’d see the apparent fact that IQ has increased by more than 1 standard deviation! This is known as the Flynn Effect. For where there is data for the same test available over a century (Raven’s Progressive Matrices) it turns out someone who scored in the top 10% a century ago would have a score in the bottom 5% today! So, in a sense, and to the extent IQ measures something objective, the average person today is smarter than Einstein. But really all that IQ can be measuring is ranking within a population. IQ does not give one any way of relating two populations separated sufficiently in space or time without some ad hoc assumptions.
January 3rd, 2006 at 10:30 pm
Just to reiterate the obvious, for the benefit of those who may join the middle of the conversation: no one here is denying that men and women are different. OK?
January 3rd, 2006 at 10:50 pm
With all due respect, I think the real “lie” here — though I would prefer the word “misrepresentation” or perhaps “ignorant misrepresentation” — is not in what Pinker said about Summers speech, but rather in what Sean says about Summers’s speech.
I defy anyone to quote three consecutive sentences in Summers’s remarks [without elision . . .] that are even remotely guilty of the sins that Sean charges. The full text is here: http://www.president.harvard.edu/speeches/2005/nber.html
January 3rd, 2006 at 11:00 pm
Me:
Summers:
What do I win?
January 3rd, 2006 at 11:31 pm
Below is a sampling of references on this topic. It is obviously incomplete — I am not a social scientist, but someone who is very interested in this issue I have found these useful. They represent a range of studies and viewpoints (from particle physicists to sociologists to the Harvard Business review).
Ivie et al present the latest numbers/statistics; Xie and Shauman’s book is excellent, and Kimberlee is a very interesting speaker. Valian includes a lot of the references detailing gender biases in our society; Rosser is a Dean at Georgia Tech and head of their NSF Advance program — her book is specific to women in science (she is a zoologist). I assume everyone has already read the MIT report, which documents very specific biases, and the more recent update on the effects of the institutional changes in response to this report.
Wenneras and Wold is a famous paper that documents an unsuspected gender bias is worth reading in the original; Rouse and Goldin determine the effects of blind auditions (where the selection committee does not know if the auditioner is male or female) for top 5 orchestras in the US — a different field than science, but similar in that it is extremely competitive, has a tenure like system and the selection process has some striking similarities. There has been a significant increase in the number of women who have been hired, and the authors conclude that “the switch to blind auditions can explain about one-third of the increase in the proportion female among new hires (whereas 34 percent is due to the increased pool of female candidates).” Well worth reading — and the parallels with physics should be obvious to any physicist who’s been part of the hiring process.
Paludi and Bauer is interesting, although too often misquoted in talks — a lot of second and third hand versions going around. It’s short and easy to read, but I recommend being a bit cautious about over-generalizing its conclusions. References in #8 are studies looking at responses to women who are successful in what are considered traditionally male ways.
Fels article is excellent (and from that bastion of conservative thinking, the HBS review) — from the business world but very relevant to physics.
For those who can’t bear to listen to any who are not in the physical sciences, check out the article by Howard Georgi (particle physicist at Harvard and former chair of the department there). Howard also stands out as someone who has made definite difference — his list of advisees is almost a who’s who of women physicists field of a certain generation in his field. Lisa Randall, Anne Nelson, Liz Simmons are outstanding physicists (of either gender). This would be part of what I was referring to as “equalizing the playing field”…
Finally, for a personal essay from a successful woman scientist, Meg Urry’s article is excellent.
References
1.Ivie, Rachel and Ray, Kim Nies, Women in Physics and Astronomy, 2005. College Park Maryland: American Institute of Physics (2005).
2. Xie, Yu and Shauman, Kimberlee, Women in Science:
Career Processes and Outcomes Cambridge: Harvard University Press (2003).
3. Valian, V., Why So Slow? The Advancement of Women, Cambridge: The MIT Press (1998) and the many references within.
4.Wenneras, C. and Wold, A. Nature 387, 341 (1997).
5. Rouse, C. and Goldin, C. “Orchestrating Impartiality: The Effect of ‘Blind’ Auditions on Female Musicians”, American Economic Review (September 2000).
6. Paludi, M.A. and Bauer W.D. Sex Roles: A Journal of Research, 9, 387 (1983).
7. Steinpreis, R.E., Anders, K.A. & Ritzke, D. (1999). The impact of gender on the review of the curricula vitae of job applicants and tenure candidates: A national
empirical study. Sex Roles, 41, 7/8, 509-528.
7. Heilman, M.E., Wallen, A.S., Fuchs, D. & Tamkins, M.M., Journal of Applied Psychology 89, 416 (2004) ; Butler, D. & Gies, F.L., J. of Personality and Soc. Psych. 58, 48 (1990)
8. The Pasadena Recommendations for Gender Equality in Astronomy, http://www.aas.org/~cswa/ (2005).
9. Fels, A., Harvard Business Review, April 2004, 50 (2004).
10. Rosser, S. The Science Glass Ceiling: Academic Women Scientists and the Struggle to Succeed New York: Routledge (2004).
11. Etzkowitz, H., Kemelgo, C., and Uzzi, B. Athena Unbound: The Advancement of Women in Science and Technology Cambridge University Press (2000)
12. A Study on the Status of Women Faculty in Science at MIT. (1999). The MIT
Faculty Newsletter, Vol. XI, No. 4. http://web.mit.edu/fnl/women/women.html
13. Georgi, Howard. (2000). “Is There an Unconscious Discrimination Against Women
in Science?” APS News Online. College Park, Maryland: American Physical
Society. (This can be found at http://schwinger.harvard.edu/~georgi/women/cfd.html along with other interesting references).
14. Whitten, B. Foster, S. and Duncombe, M. What Works for Women in Undergraduate Physics? Physics Today (September 2003).
15. Urry, M., Diminished by Discrimination We Scarcely See Washington Post February 6, 2005.
See also some of the NSF Advance program sites, for example the one at U. Michigan:
http://sitemaker.umich.edu/advance/STRIDE
which contain more references, as well as descriptions of programs they have instituted.
January 3rd, 2006 at 11:44 pm
Eluard: see above references.
I would agree that the most frustrating aspect of this issue is the many comments that are thrown around without any research or data to support them.
As a scientist I try to avoid that.
I am sure everyone can find some article in the list that they object on some grounds, but the data and the research are actually very interesting to read.
For those who are antagonistic toward these ideas I recommend starting with Georgi. His talk at Fermilab last year (available on the listed website) that was very well received by an audience that was almost all male physicists.
As for Pinker, I haven’t weighed in too much because I agree completely with Sean, who is much clearer and more eloquent than I.
Sean: how about a nice bottle of single malt?
January 3rd, 2006 at 11:57 pm
bittergradstudent: You are going to have to forgive me, but I remain mired in confusion. From everything I have read on this website (and others like it), I’ve deduced that liberals are our intellectual and moral betters, while conservatives are racists, sexists, bigots and homophobes. However, I just can’t get this to square with the claim that pervasive gender discrimination exists in departments and institutions dominated by people who are firmly ensconced on the political left. Is there something I’m missing here? Perhaps you can expand upon Sean’s rather terse response.
January 4th, 2006 at 12:08 am
macho, a bottle of the 16-year-old Lagavulin would do nicely. Perhaps we could set up a Paypal account so that Luke could send his contribution. Even better, we could set up a system where it costs $1 to leave a comment. If the bloggers agree that the comment makes sense and is on topic, you get the dollar back; if it’s actually a constructive contribution, you get two dollars! I’d be getting rich off this thread.
January 4th, 2006 at 12:14 am
Kea – I don’t understand the concept “end of the bell curve.” I prefer to think terms of so many standard deviation from the mean. Can you translate.
Sean – I find it a bit ironic that you rage against Steven Pinker, when there are people out there like lagriffedulion putting up facts and figures with far more racist and sexist implications – and I can’t find anybody refuting him. Pinker’s rather banal comments are essentially without specific content, and hence much easier to rail against than refute. lagriffedulion is not just a random nut – he has statistics and appears to know how to use them – you can’t just casually dismiss him – you need to dispute his numbers.
If lagriffedulion can be refuted, somebody ought to do it. If he can’t, never mind, please resume the academic food fight.
If you feminists get an attack of the vapors everytime Steven Pinker says something that most of the non-academics in the country already believe, I sure don’t like your prospects against the real racists.
January 4th, 2006 at 12:29 am
From everything I have read on this website (and others like it), I’ve deduced that liberals are our intellectual and moral betters, while conservatives are racists, sexists, bigots and homophobes.
Funny. From your posts, I’ve deduced that conservatives have a persecution complex.
January 4th, 2006 at 12:34 am
Sean said ”a bottle of the 16-year-old Lagavulin would do nicely”. Actually that was my favourite tipple—when I had a tipple. But I gather from your remarks that I’ll be among those paying for it.
cheers (Ah those Islay malts)
January 4th, 2006 at 1:29 am
Charles Martel (cute name, are you a French patriot?)–
You’re the one who brought up the left/right divide. I’ve found physicists to be generally quite apolitical, really. I think that might even be part of the problem, as apolitical = unaware that some behavior is inappropriate. Even ceding your point that physics departments are dominated by “liberals,” it hardly changes the point that said people might be somewhat hypocritical about their beliefs.
In the end, however, if there were a single day of my career in grad school where I wasn’t around during some sort of rape/domestic assault/”women are whores for my amusement”-themed joke that I was around for, not to mention the numerous more serious offenses that I’ve witnessed, I’d be more willing to seriously listen to the claim that there is no problem. I actually wasn’t very outspoken at all on women’s issues before coming to grad school. Coming from a 65% female campus, I thought that gender isssues were mostly resolved, and mostly just related to helping out the worlds’ GLBT. My experiences at grad school have shown me otherwise (though I should add that there are several faculty at my institution that are sensitve to gender issues).
January 4th, 2006 at 5:39 am
Jesse M.:
“for example, in an earlier post he pointed to the extreme differences in female representation in physics and math between different countries, which I think you’d agree is extremely unlikely to have a biological explanation (and to head off anyone bringing up racial differences, note that these huge disparities in female representation can be found between different european countries).”
Fair enough, but if that is what he really thinks, then why “sum it up,” as you say, by the following:
“The “dangerous idea” is not the possibility of innate differences; it’s using them as an excuse to ignore the obvious and pernicious effects of discrimination.”
He really is pitching discrimination against innate abilities, if not in his mind, then at least in his rhetoric (which was what upset him so about Pinker, in any case). The sources he cites talk about bias, and that we have a long way to go. Nowhere is the idea that women are capable of making rational choices on their own – it’s always some bias or discrimination making it for them, someone to blame, and something wrong with the situation. I just want to know where the evidence is that there is actually discrimination going on, and the evidence that it contributes so greatly to the whole state of affairs.
January 4th, 2006 at 6:06 am
It would also be interesting to hear some good arguments against the evidence cited by Lubos. The curves produced at the link he provided perfectly reproduce the shape of the curve for how many men/women study physics at various levels.
January 4th, 2006 at 9:04 am
The very first mistake that La Griffe Du Lon makes is assuming that the pool of women and men is of equal size.
E.g.,
That is wrong -
http://math.scu.edu/putnam/rulescJan.html
The basis of computing the numbers of faculty, members of National Academy of Sciences members, etc., is wrong, because it does not take into account considerable foreign input, at the graduate school level, and the IQ distributions of men relative women are not universal, and the pool of foreign applicants will be skewed.
Then, to get to a faculty or NAS position, one has typically gone through several filters: school – undergraduate – graduate – post doc – faculty – NAS. If each step is purely IQ based, there should be a prediction for each step of the way, e.g., given roughly equal number of women and men in high school, how many women and men sign up for undergraduate mathematics. If it turns out that a step is not IQ based (e.g., if school to undergrad. is not an IQ filter), then to compute the next step, undergrad to grad, one needs the women and men undergrad IQ curves and their relative numbers.
That the filtering is non-IQ based is quite clear:
http://whyfiles.org/220women_sci/index.php?g=3.txt
January 4th, 2006 at 9:40 am
CIP: Of course that La Griffe du Lion cannot be refuted. La Griffe du Lion offers a set of perfectionist calculations. They make sense. The make sense nearly to everyone. They work. The predictions match. Even if there were some errors, there is simply no one in the left-wing blogosphere who could compete with people like La Griffe du Lion.
Sean’s article is a fair illustration of the left-wing “counterpart” of Lion’s analyses. There is no one significantly better on the left wing. Look at this blog, look at your blog, and make your conclusions. Everything is just a pile of emotions, ad hominem attacks, misinterpretations, and stupidities addressed to below-the-average left-wing voters who are not terribly demanding. They only want to hear the dogmas. They don’t want to hear any quantitative justification even if there were any. You don’t have any La Griffe du Lion on the Left. And in fact, you are the only one who seems to care about the intellectual deterioration of this part of the political spectrum.
Your challenge to Sean that he should try to refute Lion is hopeless. It is impossible. Sean will never write an article about these issues that would dramatically exceed the quality of this one. It’s just impossible, and you know it as well as I do. This article was real, and it was no fluctuation.
Arun: you are confused about so incredibly elementary things that there is no chance that you will ever contribute anything to this debate. Anyone who can either read (Lion’s calculation) OR think – you don’t need both of them – can easily determine that the statistical pool for the Putnam competition are all people of the same age, and it has the same percentage of women and men. It is a completely flawed calculation to use the distribution of the general population and assume that the undergraduate students follow the same distribution. Undergraduate students have a completely different distribution. Lion used the curve for the whole population, and with this input, it is absolutely clear that the pool contains everyone – it is just that those who did not make it into the college have virtually no chance to win the contest. Whether or not the Putnam competition is officially open to non-student youth is absolutely irrelevant for the calculation because the non-students would not win it anyway.
January 4th, 2006 at 10:02 am
It may be a good challenge.
Whoever believes that the differences have mostly social origin, try to write down your refutation – or analogy with the opposite conclusions – of
http://www.lagriffedulion.f2s.com/math.htm
with a comparable amount of numbers and formulae. My prediction is that no one will come even remotely close to La Griffe du Lion. Why? Because, first of all, your beliefs are not right – and second of all, you are not used to talk about these issues rationally and quantitatively.
January 4th, 2006 at 10:15 am
Yeah Lubos, that’s what I like about reading you and other wingers – never any resort to that ad hominen stuff.
Your buddy,
The Chimp
January 4th, 2006 at 10:20 am
Dear buddy,
thanks. You are definitely not the first one who likes to read my texts because of the very same reason, namely the emphasis on the merit rather than ad hominem issues and emotions. But still, I am pleased to hear it from you, too.
All the best
Luboš
January 4th, 2006 at 10:27 am
Kea, and Bitter – You each have a number of “men are pigs” type complaints, and who am I to deny it. Let me just mention two things, though. Doctors, lawyers, and construction workers are no better, and women have been very successful at penetrating those fields. Second, those guys in grad school with you are your competitors, and they all understand that. You are all competing for grades, post docs, faculty positions, and tenure. It’s a harsh Darwinian competition, and those who show weakness are bound to have a tough time.
Physics and science in general are not unique in those respects, but physics especially may be disproportionately testoserone drenched, because physicists are at the top of the intellectual heap (though astronomers could argue that point).
January 4th, 2006 at 11:23 am
First, only students taking undergraduate courses are eligible to take the Putnam. Therefore using the entire population as a basis for computing the women/men ratio in the top 200 slots in the Putnam either assumes that college selection is also IQ-based, or else is meaningless. The pool is simply not the whole population. If college selection is IQ based, then here are the 2003 college enrollment figures:
http://www.census.gov/population/www/socdemo/school/cps2003.html
Explain why there are 9.319 million women and 7.318 million men in college based on IQ.
January 4th, 2006 at 12:04 pm
More on the Putnam competition :
http://www.maa.org/news/050505putnamwomen.html
Notice that the women’s winners include those who did their high school in Romania, Canada and Korea. The pool is not America but America, Canada and an elite crust from the rest of the world as well, with cultural filters applied to enrollees from abroad (e.g., they need to be well off-enough and overcome cultural barriers to women going overseas to study).
January 4th, 2006 at 12:08 pm
Of course that the college selection is IQ-based in each well-defined group of the people. Women’s percentage in these overall numbers is a matter of affirmative action and fashionable social trends and systematic bias, and it is not too important.
What is more important is that the youth outside the college has a completely negligible chance to succeed in the Putnam contest.
You can get clear inequalities between the students’ IQ and the non-students IQ, but you simply can’t assume that the average IQ of male students equals the average IQ of the female students because this selection is heavily affected by various quotas and affirmative action and other social bias.
It is very important to do the calculations properly – which means to consider the whole population as competing for the Putnam award, and make the reasonable assumption that the non-students do not have a chance anyway. It would be impossible to do similar calculations using the ensemble of the students only – because the distribution of IQ among the students is different and it is not even Gaussian because the students are essentially selected by a rather sharp cutoff.
January 4th, 2006 at 12:19 pm
“Between 1980 and 2000, the percentage of Ph.D. scientists and engineers employed in the United States who were born abroad has increased from 24% to 37%. The current percentage of Ph.D. physicists is about 45%; for engineers, the figure is over 50%. One fourth of the engineering faculty members at U.S. universities were born abroad. Between 1990 and 2004, over one third of Nobel Prizes in the United States were awarded to foreign-born scientists. One third of all U.S. Ph.D.s in science and engineering are now awarded to foreign born graduate students. We have been skimming the best and brightest minds from across the globe, and prospering because of it; we need these new Americans even more now as other countries become more technologically capable.”
http://www.aau.edu/homeland/Wulf_Testimony91505.pdf
The pool of eligible people for the Putnam prize, faculty positions, NAS membership is rather different than postulated.
Anyway, ’nuff of this, the chimps are calling.
January 4th, 2006 at 12:31 pm
Sorry, the chimps let me back to tell you that maternity apparently produces changes in female brains, increasing their cognitive abilities; therefore, even though teenage pregnancy may seem to be of epidemic proportions, a IQ curve of 15 year olds as a predictor for life may not be appropriate.
January 4th, 2006 at 12:41 pm
Lubos:
snip…
Yeah, there are absolutely no ad hominem attacks present there…
Regardless, anyone who is saying that there is gender discrimination, no matter how widespread, also has to admit that it is unevenly distributed, and difficult to place a number on, as all types of discrimination are not equal. This model does nothing to explain why there are different proportions of women faculty in different countries, when, by this IQ model, all countries should have roughly the same proportion of men to women. Instead, the proportions range over a huge breadth.
And, I also think that the main argument that I’m making is that it is impossible to measure whether or not the end result we get is due to discrimination or due to inherent differences, because discrimination is present and makes any serious measurement difficult.
Furthermore, dsicrimination is an inherent bad thing, so why not get rid of it? If it affects one or two people, it is still bad, no? Why should it be tolerated? If it has made one person drop out of grad school or choose not to pursue a faculty position, is it still not something that should be gotten rid of?
January 4th, 2006 at 12:46 pm
CIP:
fair enough, but that is no reason for faculty to discount female students out of hand, or for them to try and grope them. Intense competition is also no reason to abondon professionalism. Also, a Darwinian system that produces anything other than the best physicists should probably be rethought in favor of a system that does produce the best physicists.
January 4th, 2006 at 1:03 pm
All anyone has to do is click the links in the post. “La Griffe du Lion” and his friends push a straightforward hypothesis: that the observed distribution of genders in scientific professions can be explained as a refelection of an underlying distribution of innate abilities, disregarding any purported systematic biases. This hypothesis makes several falsifiable predictions: the distribution of women in science should be constant over time, the same in different countries, and there should be a dramatic falloff at the higher levels even taking into account the time lag required for people to move through the system. All of these predictions is spectacularly, embarassingly wrong. In principle all you should have to do is notice that the fraction of women has been growing linearly in the past couple of decades — but the enthusiasts are, predictably, unmoved.
But they do change their hypothesis a little bit. With a straight face, and without any apparent appreciation of the self-undermining irony, the innate-aptitude folks will explain to you that these discrepancies are accounted for by — wait for it — systematic biases. Only that the biases actually favor women. Then you will point them to the book by Xie and Shauman, and all the other mountains of data in the references mentioned by macho in #54 above. (Complete with equations!) And they will declare themselves “unimpressed.” This is the advantage of religious conviction: it is not disturbed by confrontation with reality.
Of course this is precisely why the innate-aptitude folks would like to pretend that the controversy was about whether it’s politically correct to contemplate the possibility of differences between men and women, rather than about whether any such differences are more important than systematic biases in accounting for the observed gender gap. A well-fed persecution complex is more comforting than an honest discussion.
January 4th, 2006 at 1:29 pm
“This hypothesis makes several falsifiable predictions: the distribution of women in science should be constant over time, the same in different countries, and there should be a dramatic falloff at the higher levels even taking into account the time lag required for people to move through the system.”
The percentage of women is only predicted to be approximately constant with time if you remove social bias, affirmative action, and fashionable trends – which is what La Griffe does in the best way we can but you oppose it. Of course that if you impose a 50:50 quota, then the percentage will be 50:50.
The Gaussian curve that measures the actual innate aptitudes is of course predicted to be more or less constant, and it agrees with the observations.
It is not true that the distribution of men and women is identical in different countries. There are also dependences of the distribution curves on the ethnic origin, and (smaller) correlated dependences on the ethnic origin combined with sex. (Look at the distribution of black men vs. women using the internet, to get a good example of the last category.) All these things are subject to a scientific evaluation, and indeed, this evaluation is done by those who ask these questions seriously. You seem to be a victim of your own flawed assumptions. No one who studies these things seriously would say that various different groups of people are predicted to have the same outcomes.
You are spectacularly wrong that the percentage of women continues to grow. In computer science, for example
http://motls.blogspot.com/2005/12/internet-gender-gap.html
the percentage of women among the CS students dropped from 37 percent in the 1980s back below 20 percent right now (the article about Tufts university) – which is just one thing among many that show that the increased percentage in the 1980s was a result of artificial social engineering.
Also, among the Nobel prizes, women got 9 or so before 1964, but since 1964, they got none. The rate of the female science Nobel prizes has dropped from one per five years to zero exactly in the last 40 years in which feminism was influencing our societies. Of course, feminism is not the primary reason to be blamed. The real reason is that physics and chemistry simply got even more difficult, picking an even more selective piece of the tail of the Gaussians.
This dramatic falloff at the highest level is a very well-known thing, and La Griffe du Lion demonstates that the observed falloff exactly matches the prediction from the two Gaussians.
When one isolates these observations properly, of course that it is found that both biological as well as social influences affect the final outcomes. But they affect it exactly in the opposite direction than you say, Sean. The biological differences are responsible for the natural curves and expected ratios demonstrated by La Griffe du Lion; the social influences mostly generate the noise of the affirmative action that artificially raises the percentage above the natural average, towards the constructivist ratio around 50:50. It is very obvious in the case of the genders and less obvious, but still convincing, in some other cases.
You don’t believe that your confused, vague, non-quantitative, and negativist statement can be viewed as a serious answer to La Griffe du Lion’s calculations, do you?
January 4th, 2006 at 1:32 pm
Sean – I don’t think you have constructed a refutation. You made the falsifiable predictions based on an innatist theory you impute to him – he didn’t. For one thing lgdl doesn’t deny the effects of past discrimination and doesn’t make any explicit predictions. What he does do is make post-dictions from the 1960 age 15 math talent tests, using essentially only the differing means and standard deviations for the males and females, and shows that these postdictions are highly consistent with the relative distributions of men and women in the National Academy, and other measures of elite mathematical performance.
Without trying to defend any innatist ideas, it’s pretty clear that the changing distribution of women in math and science in time can be explained by the former existence of massive insitutional and social obstacles to women. Ditto the different distributions in different nations.
By the same token, it’s pretty clear that you could artificially impose gender parity in elite college faculties by instituting adequate barricades to men.
If you really want to refute lgdl you need to either challenge his data or his analysis. I’m still trying to find someone who does either.
January 4th, 2006 at 3:14 pm
CIP,
1. How does lgdf constitute an explanation?
2. There are really 2 worked out examples with significant statistics – the NAS and the Putnam Prize (the 200 finishers in 2004). (The Fields Medals and the Putnam Fellows don’t have significant statistics – this decade’s women Putnam Fellows beat the odds, as per lgdf). The NAS is in the millions of competitors per slot level, and factors of two in the competitors per slot are not very relevant. There is really just one prediction for this range, that women will occupy around 5% of the slots. The Putnam Prize is more interesting – it is in the hundred thousand competitors per slot level, where the results are much more sensitive to getting the competitors per slot correct. The same would hold good for faculty appointments and so on.
-Arun
January 4th, 2006 at 3:47 pm
CIP,
One can find current members of the NAS in Mathematics and Applied Mathematical Sciences, using the search page here:
http://www.nasonline.org/site/Dir/1115426753?pg=srch&view=basic
Notice that the 52 applied math and the 116 math members have election dates that go back to 1962.
If we look at the elections in these two sections for 2000-2005, we see the following names, I’ve asterisked those that are women: 3 out of 29, or more than 10%.
*Grace Wahba
*Margaret Wright
*Marsha Berger
Arthur Jaffe
Benedict Gross
Charles Newman
David McLaughlin
David Siegmund
Efim Zelmanov
George Andrews
George Papanicolaou
Gregory Margulis
Haim Brezis
Iain Johnstone
Jacob Palis
James Berger
Janos Kollar
Joseph Bernstein
Kenneth Ribet
Michael Powell
Nicholas Katz
Peter Sarnak
Phillip Colella
Robion Kirby
Sergiu Klainerman
Simon Donaldson
Stanley Osher
Yakov Eliashberg
Yum-Tong Siu
If we go by the competitors/slot using N=190 million for 29 slots, there are 6.6 million competitors/slot. Remember, the higher the competition, the more skewed the ratio in favor of males, and the expected value for women should be somewhat smaller than the 5% of the lgdl 143-slot NAS example. Or, if we scale the ratio of election of the last half decade to 143, there would be 14.79 women (the lgdl number is 7.1, 95% confidence interval of [2,12]). Therefore, the last six years has been highly unlikely as per lgdl-type calculations.
The question for you is what happens this election ratio holds up at the 10% rate for a while, would you still buy lgdl?
-Arun
January 4th, 2006 at 3:55 pm
1995-2005 inclusive, 6 out of 62 elections to NAS for Mathematics and Applied Mathematical Sciences were women. Just under 10% and way unlikely as per lgdl, according to which it should be less than 5%.
January 4th, 2006 at 5:08 pm
Arun – This is exactly the kind critique I would like to see, but I can predict (falsifiably) that the response would be that recent results contaminated by “affirmative action.”
January 4th, 2006 at 5:52 pm
That’s what’s known in studies of rhetoric as a “blatant lie.”
Cosmic Variance
January 4th, 2006 at 6:09 pm
Dear Arun,
your calculations are completely nonsensical. When you’re naming 29 NEW members, then the pool is not 190 million people. The pool is just a small fraction of the population that are expected to join the society right now.
190 million is only the total pool for all the members of the group.
Without loss of generality, you can assume that everyone has a chance to join the elite bunch once per life – for example at the 35th birthday. (The precise moment plays absolutely no role.) If you assume that people live for 75 years, the pool of candidates for these 29 annual slots is 190 million divided by 75 which is less than three million. So the ratio is less than 100,000 thousand per slot.
If you look at La Griffe’s graph, you will see that the prediction for 100,000 per slot is 10% of women which, once again, perfectly matches the observations. You must just avoid trivial errors that screw your results by two orders of magnitude, understood?
Best
Lubos
January 4th, 2006 at 6:19 pm
For differences in cultures see:
http://edu.technion.ac.il/TechnionWomen/Women_iupap.pdf
The data are for 21 countries in 1997-98
The range in percent of women granted PhD’s in physics:
8% (Japan and Korea) to 27% (France)
“First degrees” (bachelors):
5% (Netherlands) to 37% (Turkey)
What is also very interesting is that while some countries rank similarly in both
degrees (Poland is strong at 23% (PhD) and 36% (first degree))
some show a much sharper dropoff. In Korea, for example, 30% of first degrees are earned by women (near the high end of the range), but only 8% of PhDs
(near the bottom).
January 4th, 2006 at 7:02 pm
Lubos,
First, let’s take your computation as entirely correct. You claim that 10% intake of women annually into the NAS Math/App.Math is consistent with ldgl. Then, cumulatively over time, if male and female longevities are the same (in fact, women live longer), won’t the NAS have 10% instead of the current 5% membership? Remember, the 5% is what the ldgl paper predicts. How is a 10% annual intake and a 5% cumulative average consistent?
Second, the number of slots per year is less than 6 (62 in 11 years) and not 29. If one buys the rest of your computation, that is still 420,000 competitors per slot. Take a look at the lgdl graph for that ratio, it is well into the male-saturated range. And in reality, people have multiple chances to get into the NAS, and the pool is larger than your estimate. In principle, in any given year, everyone who hasn’t been elected yet has a chance to be elected.
-Arun
January 4th, 2006 at 7:52 pm
I know that Claude Steele’s work on stereotype threat is anathema to conservatives on this issue, and suspect (hope) that many of the readers of this blog are already familiar with his research, but for those who aren’t his papers are definitely worth reading:
http://steele.socialpsychology.org/
If you’d rather listen to a brief National Academy of Sciences interview:
http://www.nasonline.org/site/PageServer?pagename=INTERVIEWS_Claude_M_Steele
esp. track 6.
In short he proposes (from his website) “a theory of how group stereotypes — by posing an extra self-evaluative and belongingness threat to such groups as African Americans in all academic domains and women in quantitative domains — can influence intellectual performance and academic identities” and has studies supporting this view. Read his 1995 paper for more info.
January 4th, 2006 at 7:55 pm
Some time ago, Steve Hsu compared the income of scientists with other academic professions. He concluded that scientists (and I assume this is especially true for physicists and mathematicians) are significantly underpaid.
The fact that less women than men become scientists could then just indicate that they are on average smarter
January 4th, 2006 at 8:09 pm
Dear Arun,
La Griffe calculates the people-per-slot ratio for the current 143 members of NAS which result from the pool of 190 million people. This is indeed 1.4 million people per slot, which gives 5% women.
You calculated a different thing which is less selective and it is about 100,000 people per slot, giving 10% of women.
If they will accept 30 new people every year, and assuming that they don’t expel members
, then the total number of members will be much higher than 143 pretty soon, and therefore it will be a less selective club with the number of women that will be predicted to approach those 10 percent from the current 5 percent. In other words, the currently new accepted members are less selective than they were in the past.
Makes sense?
Best
LM
January 4th, 2006 at 8:15 pm
If you reduce the rate of new members below 30 per year, I hope that the calculation can be done again. At any rate, it is obvious that the more member the NAS has, the less selective club it becomes, and the higher percentage of women is predicted.
Moreover, even if you predicted 5% of women, which is 1.5 among 30 people, it is still pretty likely that the actual number will be 3. It’s just too small a statistics. Seven among 143 is a much better statistical ensemble.
January 4th, 2006 at 9:01 pm
Wolfgang,
It is not 30 people an year, it is less than 6 ( 29 in the 6 years 2000-2005 inclusive or 4.8 per year; 62 in the 11 years 1995-2005 inclusive or 5.6 per year). If you follow Lubos’s calculation the pool is 190 million/75; if you follow the ldgl recipe, then it is 190 million/61 (for the 25-85 years; the US population in that age group is 190 million, but almost 300 million total). This is between 420K per slot and 520 K per slot and still is in the 94%-95% men range of the ldgl.
As to whether the NAS is becoming significantly less selective, here
are the number of living members in Math/App. Math per decade of selection (remember mortality for those from the 60s). Also the census counts are given.
1960-69 11 – 1960 US population = 203.3 million
1970-79 32 1970 US population = 226.5 million
1980-89 38 1980 US population = 248.7 million
1990-99 58 1990 US population = 267.7 million
2000-05 39 (on pace for 65 for 2000-2009)
2000 US population = 281 million
The elections over the past 6 years have been
2000 – 6
2001 – 5
2002 – 4
2003 – 4
2004 – 5
2005 – 5
January 4th, 2006 at 9:33 pm
Sorry, I goofed, it is 29 elections in 2000-2005, on pace for 48 for 2000-2009, compared to 58 for 1990-99. In the 2000s so ar, the NAS has been **more** selective than in the 1990s.
January 4th, 2006 at 11:19 pm
[...] Steven Pinker seems to be the first one to start this round, with his contribution to the question posed by Edge this year: “What’s your dangerous idea?“. Sean Carroll catches him in the act of willfully misrepresenting the ‘dangerous idea’ in the Summers episode. [...]
January 5th, 2006 at 5:20 am
bittergs – faculty who engage in groping are in vilolation not only of professional ethics but of the law. They need to be exposed and punished. Schools that tolerate such behavior are also subject to harsh penalties. Exposing such people and getting them fired and/or imprisoned is the only way to rid the profession of them – another plus – it opens up tenure slots.
January 5th, 2006 at 5:43 am
Hi macho (#46) and bittergradstudent (#30),
This thread certainly moves fast! Sorry I didn’t responsed sooner — haven’t looked at the thread for a couple of days. Will primarily respond to macho:
I love this kind of “reasoning”.
Just to help you further cast a negative light on my post, you could also note that my speculation (you called it “reasoning”) was entirely unsupported by empirical data, except a single anecdote (very uncontrolled variables!), and that generalization from a single example (especially an anecdotal one) is, well, poor argumentation.
However, my goal wasn’t to give a strongly-reasoned argument, but simply to emphasize a couple of points that I thought deserved added discussion. My speculation about possible evolutionary factors was also not a statement of a strongly held point of view, but more of an informal plausibility “argument,” i.e., that there are other explanations for the relative paucity of women in science and engineering besides/beyond aptitude or discrimination. I was not implying (nor do I believe) that environmental factors are unimportant. But just because a cause is important does not imply that it is the only important cause, nor that a single cause alone will account for the relative paucity.
(Any serious discussion of the role of genes should include analyses of the closest relatives of humans among the primates. For example, do male and female chimpanzees take different roles when it comes to nurturing vs. “exploration?” To the extent that females play the dominant role in raising their young, would anyone here seriously argue that it is due to social discrimination and stereotypes held by chimpanzee groups?)
Just to be clear, I am not aware of unambiguous evidence that women are less genetically endowned than men in the mental capacity to succeed in science or engineering, and personally I doubt that mental capacity is the real cause. For any person to have a successful career as a scientist, she or he must choose to enter that career in the first place! And there I think that it is plausible that genetics COULD play a role, although this seems like something to be sorted out by science rather than anecdotes or personal belief about how things should be in an ideal world. That is, a person who believes that scientific inquiry is the best way to separate belief from fact should be willing to carry out the necessary studies as best as possible and be prepared to change personally dear beliefs if scientific studies repeatedly show they are unmerited. That viewpoint seems uncontroversial among educated people.
can you outline the reasons for the innate lack of interest in physics among African Americans?
Don’t know whether you think this represents my viewpoint, but in case what I wrote above isn’t clear enough, I don’t hold that view at all (and doubt that you do either). I agree that African-Americans are much less represented than women in the sciences and engineering, certainly in my experience. I can also think of some environmental reasons that seem fully capable of explaining the poor participation without any need at all for invoking genetic arguments.
So they have been completely unexposed to other children and adults, no tv movie, radio, magazines, popular culture of any kind? And you and your spouse were also raised in completely gender blind environment?
Probably a lot more so than you might imagine. We have never had broadcast television in our house since before our children were born, have been selective about what movies they watched, avoided most video games, continue to discuss the commercial motives behind the mass-media imagery promoted to the public (including stereotypes), etc. They have seen both parents working in science and engineering (not just the male parent). I grew up in a one parent home (my father died when I was young) and my mother was a business woman (several health care related businesses at once!) who evidenced no prejudice and who by her example dispelled any belief I might have otherwise had about the competence of women. My wife grew up on a farm where the kids all worked hard (!); they had no television, her father was a peace activist, both parents were liberal and definitely didn’t raise daughters to believe stereotypes about women. So I feel confident in saying that our upbringing was probably as gender blind as one might reasonably expect.
Certainly our daughter is not insulated from popular culture, nor should she be. She sees it all the time at school, people who are immersed in popular culture. We don’t tell her who she can have or not have as friends; nonetheless, her friends are academically oriented, some of whom plan to go into science or technology. But regardless of “holes” through which popular culture can seep into the lives of children, surely you wouldn’t make the argument that having some popular influences will overcome the dominant message children receive as they grow up, or that they are so malleable by advertising that they will cave in to stereotypes at the slightest exposure?
Not that this example really matters to the debate — it is but an anecdote, and certainly no substitute for scientific study. But it gives an example of how personal interest can be the critical factor in whether a female enters science.
Before making my last point, a prefatory remark: I believe that environmental factors are very important in whether someone becomes a scientist or engineer, probably dominant in most cases. Hopefully it is obvious that I think discrimination on the basis of stereotypes or personal beliefs is disgusting. (It shouldn’t be necessary to explicitly say that, but given the quickness of some commenters here to jump on anyone without the “right” social viewpoints and thereafter ignore their valid points, it sadly seems appropriate.) So…
There is a certain aspect of the controversy that I find puzzling. If studies ultimately show that females are less likely to pursue science or engineering because genetic factors predispose them to prefer something with more of a “social” component (and who would consider physics to be a primarily people-oriented field?), why is that harmful? To argue it is somehow “bad” (and therefore to fear any study that shows genetic factors) is to make an unjustified value judgement that a career in science is somehow “better” than a career in a people-oriented field. This seems like an arrogant and self-serving viewpoint to me, and hopefully it isn’t implicit in the world view of anyone participating here. (This is coming from someone who is currently a full time graduate student in a physics doctoral program.)
January 5th, 2006 at 7:49 am
M wrote: “(Any serious discussion of the role of genes should include analyses of the closest relatives of humans among the primates. For example, do male and female chimpanzees take different roles when it comes to nurturing vs. “exploration?” To the extent that females play the dominant role in raising their young, would anyone here seriously argue that it is due to social discrimination and stereotypes held by chimpanzee groups?)”.
Fortunately, we count both chimpanzees and bonobos as our closest primate relatives, and chimps and bonobos are different in their behaviors in many significant ways, so that to the extent we do not know the evolutionary history of man (which we do not) we cannot trace the “innateness” of human behavior back to the “innateness” of chimpanzee or bonobo behavior. Anyway, it would be interesting to see how plastic these species are by raising one with the other.
January 5th, 2006 at 8:13 am
Coming back to lgdl, one of the arguments is that the Putnam exam. manages to attract most of the mathematically gifted, one way or another, and that is the basis for arguing it is a selection from the entire 17 million or so US population from 18-21.
Well, the list of Putnam Fellows is here:
http://www.maa.org/awards/putnam.html
The list of NAS members can be found by search (by section) here:
http://www.nasonline.org/site/Dir/38278516?pg=srch&view=basic
The interesting question is – how many of NAS Math/App Math members were Putnam Fellows or wrote the Putnam exam. ?
The thought is as follows: as far as I can see, Edward Witten was not a Putnam Fellow. Then one stares at the NAS members list and the Putnam list some more, and notices things, looking at names that sound familiar e.g., NAS members Nicholas Katz, Robion Kirby, Simon Donaldson, Andrew Wiles, John Nash, Robert Langlands, Eugene Dynkin, William Thurston, were not Putnam Fellows; the question would be did they write the Putnam? Unfortunately, that is a hard thing to find out; but it would enable us to test whether or not the Putnam does lure most of the mathematically gifted.
We may have to settle for comparing the NAS list against the top 100 finishers, “The lists of the top ten teams and the top hundred individual contestants, in the categories described above, will be published in the American Mathematical Monthly, together with the problems and solutions.”
(http://math.scu.edu/putnam/announcecJan.html). The expectation would be that virtually all NAS Math/App Math members who did undergrad in the US should be found on the top 100 list; or else the claim that Putnam attracts most math talent is wrong.
Unfortunately, I cannot do this exercise; having no easy access to a univ. library.
January 5th, 2006 at 9:10 am
Some of the entries you wrote are really entertaining, Arun. For example, Eugene Dynkin is a Russian who emmigrated to the US when he was 52 years old. So don’t be surprised if he did not compete in the Putnam competition.
Then you start with Katz. The Katz I know best is Sheldon Katz who was a finalist (or fellow) in 1974, 1975, and so forth.
John F. Nash (Nobel prize for economy, game theorist) was among the best in the examination in 1947 and received a prize of twenty dollars. Congratulations. He also scored high in 1948.
I don’t have access to all the lists, they are just published year by year and I only see some years. But let me remind you that even in the hypothetical case that not all of them were finalists, most of them were competing – and even if they were not and the pool was a bit lower than the whole population, it won’t change the predicted percentage of women significantly.
When I look at the 1947 winners: Murray Gell-Mann was a member of the silver team in 1947.
January 5th, 2006 at 9:21 am
Perhaps a little of topic. I remeber seeing a BBC documentary some years ago (it could have been Horizon, I don’t remember) about differences in cognitive abilities between men and women and in particular the role of hormones such as testosterone. If women were given testosterone they did better in some tests in which men are usually better in and worse in other tests in which women are usually better in.
Women scientists/engineers turned out out to have a ”brain profile” (I don’t remember exactly what tests they performed here) similar to males, while males who work in occupations in which usually mostly females work, such as nursing etc. had brain profiles more similar to what you see in most women.
January 5th, 2006 at 9:36 am
[...] The Edge questions are up. I wasn’t aware that they are from an agent to his clients. I’m not sure this weakens the exercise so long as anyone can contribute. [...]
January 5th, 2006 at 2:56 pm
M:
Don’t know whether you think this represents my viewpoint, but in case what I wrote above isn’t clear enough, I don’t hold that view at all (and doubt that you do either). I agree that African-Americans are much less represented than women in the sciences and engineering, certainly in my experience. I can also think of some environmental reasons that seem fully capable of explaining the poor participation without any need at all for invoking genetic arguments.
Of course I don’t hold this view and I assumed that neither did you. What I was trying to do (in an admittedly more aggressive way than was probably useful) was to point out that people who are willing to consider genetics as a possible explanation for male/female representation in science (and can then outline a “plausible” reason why — your nurturing example) are much more reluctant to do the same for other underrepresented groups. (When it is equally illogical to do it for either)
Your last sentence:
I can also think of some environmental reasons that seem fully capable of explaining the poor participation without any need at all for invoking genetic arguments.
sums it up very well for both cases. (The evidence for this can be found in the many studies that I listed in an earlier post above. ) This is exactly the point I and others have been trying to make (with the support for the environmental reasons found in the references and studies).
My frustration here, which I really do not mean to direct at you, is that the data point clearly to strong environmental factors that are still very much in play for both women and minorities. When you outline why you think that women might be genetically less interested, you are reinforcing strongly held cultural beliefs that only serve to underscore them. My point is that you recognize the dangers (and offensiveness) of doing so for another underrepresented group. Yet are willing to do so for women in science without any more data to support this than for racial differences.
Yes men and women are different in some ways — they respond differently to some medical treatment, for example. There are also various race specific genetic disorders and response to medical treatment. Different upper body strength, different skin color, different facial structures. No correlation between any of these and scientific ability/interest has ever been shown. Some differences in brain activity have been claimed– again no correlation between these and scientific ability/interest. History is replete with attempts to correlate brain size, physiognomy, skin color, presence of a uterus, with various limitations on mental ability/interest. In most cases it was a thinly veiled attempt to somehow “scientifically” justify continued discrimination against a particular group, and so far none of the “results” are still standing.
As recently as 100 years ago members of groups currently underrepresented in physics were directly forbid access to the ranks of physics, and were openly discouraged from entering the field well into the 1960s. To continue to use their underrepresentation as evidence that they are innately less interested or qualified is not sensible. To use studies that find small differences that have no demonstrated correlation with the prerequisites for the field is also not sensible.
Am I denying the possibility that there may be some innate differences between various groups in science ability/interest? No. But neither do I deny the possibilty that one of the events in a WIMP detectors was due to a WIMP, not a cosmic ray (if wimps exist, certainly a low probability possibility). However, I do deny that we yet have any evidence of such an event. (Let’s leave DAMA out of this for the sake of the argument). The cosmic ray background is very real, well- documented, and at this point clearly dominates over any wimp signal. We can and should argue over the details of this background — maybe a new Monte Carlo analysis will show that it’s somewhat stronger or weaker than I’m currently assuming. But by this time it’s well enough documented that I can’t claim that there are no CRs, or that they don’t really affect my data.
The bottom line is that any such innate differences, if they exist, are not useful in trying to understand the numbers of women in physics at the moment.
Here is where my beliefs come in — I belive that physics is an incredibly rewarding endeavor, as the writers of this blog have made clear, and I believe in distributing opportunities to engage in such a profession as fairly as possible. (Nor is it exclusively for non-nurturers — many outstanding physicists I know are also excellent and very nurturing fathers.) Toward that goal, focusing on any possible innate differences before accounting for the documented environmental factors is illogical and, worse, counterproductive. Further, institutions and individuals that have made progress in increasing the numbers have done so by addressing the environmental factors. (read the Blind Auditions ref for an interesting take on this or Georgi’s talk for practical info).
On a different note, our children are obviously older than yours (we have some of each gender) and we also tried hard to minimize stereotypes in favor of encouraging each child to play to their individual strengths and work on their own weaknesses. But it isn’t easy. And it’s both genders that feel pressured to conform to certain behaviours — try reading Valian’s book for more insight on this.
In any case, best of luck with both the kids and the PhD.
January 5th, 2006 at 5:49 pm
Lubos,
Well, I do not find Nash on http://www.maa.org/awards/putnam.html
The 1947 Putnam Fellows were
Clarence Wilson Hewlett, Jr, Harvard University
Maxwell Rosenlicht, Columbia University
W. Forrest Stinespring, Harvard University
William Turanski, University of Pennsylvania
Eoin L. Whitney, University of Alberta
As I wrote (but you probably didn’t bother to read through, a look through the top 100 each year is what is needed).
-Arun
January 6th, 2006 at 9:31 am
Dear Arun, I did not write that Nash was a Putnam fellow. I wrote (quoted the journal) that he was among (5) best during the “examination” part of the competition, whatever it means, and he won 20 dollars for it. I don’t have the lists of 100 from each year. Even this list contained many less names than 100 from one year separately.
I don’t think that your exercise is so terribly important because it is clear that most of the talented ones did the contest, some of them may have failed, but you will certainly not prove that most of the NAS members did not even try to participate. This is what you would need to reduce the pool by a factor of two, and even such a heroic act would only change the prediction for the representation of women by a very small amount.
January 8th, 2006 at 8:57 am
Thanks for the criticism of Pinker, who deserves it very much. I guess only a contemporary psychologist can come up with something so superfluous and even tautologous, but sounding so much like a Harry Potter soundbite. As Cicero wrote, the bigger effort some people invest in public speaking, bigger nonsense will come out of their mouths… Look only at his sources — more bizarre (and less persuasive) bunch I couldn’t find if I searched with the lamp in daylight (like Diogenes): a university bureaucrat, an op-ed in NYT, poor old Murray who in web reviews always oscillates between being “political scientist”, “social scientist”, “sociologist”, “anthropologist”, “social psychologist”, whatnot, reiterating Bell’s Curve (what else? as an old Russian saying goes, who talks about what? nannies about pancakes; whores about honesty; cowards about heroism; etc.), another newspaper gossip (why hasn’t he cited primary research source? perhaps since it doesn’t exist). By using sources like that, I guess one could prove that George Bush is one of the finest intellectuals of our age.
(All this apart from the obvious fact that his “dangerous statement” is a truism; just consider the logically opposite statement that “Groups of people never differ genetically in their average talents and temperament”, and you’ll see a true Woody Allen rendering of a “great truth”, and how senseless Pinker’s posturing really is.)
But Roger Schank’s and Haim Harari’s contributions to Edge’s question are excellent (and truly disturbing).
January 16th, 2006 at 3:30 pm
[...] Here at CV we occasionally pat ourselves on the back at the high quality of some of our comment threads. So it’s only fair that we acknowledge our dismay at the depressingly consistent character of the discussions about women in science; posts by Clifford and me being just the most recent examples. What a depressing exercise to poke a finger into the turgid world of pseudo-scientific rationalizations for inequality that people will believe so that they can feel better about themselves. Among other things, it makes it nearly impossible to have a fruitful discussion about what we could realistically do about the problem; it’s as if Columbus were trying to equip his ships to voyage to the Indies and a hundred voices kept interrupting to point out that the world was flat. [...]
March 26th, 2006 at 12:04 pm
[...] P.P.S On a more serious side-note, and in view of our discussions of women in physics (see also e.g., here, and here) not so long ago on this blog, notice how more balanced the demographic is in this subfield. Refreshing, frankly. Here’s to the future. [...]
June 27th, 2006 at 11:31 pm
[...] I hope, as opposed to this, or that, no one considers Dr. Virdee’s statement a particularly dangerous idea. [...]
July 18th, 2006 at 8:36 am
[...] So who are these unnamed people who think that Barres “should take scientific hypotheses less personally?” That sounds suspiciously like a straw man — most careful scientists would be reluctant to stoop so directly to an ad hominem attack, rather than dealing with the aforementioned mountains of data. Sadly, it’s a direct quote from our old friend Steven Pinker, himself a master of the straw-man technique. [...]