Tall, Dark, and Mysterious


Oh, I do believe I’ve already addressed this.

File under: Righteous Indignation, XX Marks the Spot. Posted by Moebius Stripper at 9:55 pm.

I had a long and only vaguely coherent post about the President of Harvard’s comments about the relative number of men and women in the mathematical and physical sciences, but then I realized that I’d already posted about it. Three months ago:

…no meaningful or productive discussion of [low ratio of males to females in the mathematical and physical sciences] can take place without acknowledging that there are far, far fewer women than men who are both interested in, and qualified to do, higher mathematics. Unfortunately, no discussion of that fact can take place - at least not in my presence - without someone pointing that the likely culprit is socialization, not biology - as though this is some profound insight that only those enlightened, self-proclaimed gender experts have ever considered - and one that renders any further discussion on the topic of women in mathematics completely moot. Social problem, not biological, the absence of women in math is a result of sexism, not hormones, nothing to see here.

What else did I say back then - oh, yeah, this:

‘“[W]e must increase the numbers of women in math!” has become the dominant battle cry of mathematicians and career feminists alike - the former of whom (mostly men, mostly older) typically have limited understanding of the life experiences of girls, and the latter of whom typically know next to nothing about math. The former accept the latter’s simplistic notion that the women-in-math problem would be solved if only we consciously countered our latent sexist beliefs and actively tried to recruit adult women into math programs and positions.

So I’ll leave the career feminists and male scientists to this one, and resume my role as math-chick-in-fishbowl. Honestly. Because if I don’t, I might have to explain why I can’t get too worked up over the fact that

[t]he president of Harvard University, Lawrence H. Summers… said that innate differences between men and women might be one reason fewer women succeed in science and math careers.

And I might also have to overlook Summers’ dumbass statement about his daughter’s feminine way of playing with toy trucks and how that is apparently related to women’s allegedly poor mathematical abilities, because that’s already been soundly eviscerated - by, among others, people who counter the “biology may be one of several factors” claim with fairytales like this one, cut from whole cloth:

I’ll tell you how much of a role discrimination plays in limiting female professors in so-called “elite” universities: 100%. There is no shortage of brilliant women scientists…but there is a dearth of jobs and we still have bigoted ignoramuses like Summers standing guard over the gateways.

So, there are just as many qualified women with Ph.D.’s in the physical biological sciences as there are men, or close enough? Come again? Can I please see the data on this one, because all I’ve heard on the subject in the last decade or so is that there IS a shortage of women in the above fields? Must be convenient to be able to peg the small number of tenured female scientists on the likes of Summers (who certainly isn’t the person I’d choose to head a top university) rather than doing the work to keep girls studying math and science so that they can BECOME brilliant women scientists.

I said that, too, a few months ago. Or something like it.

I might also have to say something about how it’s a crying shame that the brilliant women scientists listening to Summers’ speech (which he was invited to make by virtue of his qualfications as a leading economist - why, exactly?) walked out on him, rather than countering his claims with the data from the groundbreaking MIT study that rigourously revealed that, all other things being equal, female professors at MIT were paid and promoted less than their male colleagues. It’s a pity that they chose to walk out instead of giving data that backs up the belief they hold so strongly - that women would be men’s equals in the sciences, but for social conditioning to the contrary. It’s a shame that neither Summers nor his critics took the opportunity to outline how one could measure the impact of social conditioning versus the impact of genetics on academic performance; clearly they diverge on this issue, and it’s an interesting one to explore. I’m sure there could have been a lively discussion, had it gone that way. Instead, it became a circus, consisting of specious, deliberately provocative claims, countered not by arguments but by indignation. Welcome to Harvard, where the nation’s greatest minds are challenged, or something.


  1. The number of females offered tenured positions has declined since Summers became president. *Something* in Harvard’s environment may have changed (null hypothesis: statistical fluctuation, no relevant environmental change; hypothesis 1: the person sitting in the President’s office; hypothesis 2: something else relevant has changed).

    It does not matter which hypothesis is correct. Summers is the President and bears responsibility when things go badly (he also get to take credit if they go well). *This* is the context within which Summers made his rather stupid remarks, and they were, predictably, seen as crass attempt to shift blame away from the President’s office. Someone who is this politically tone-deaf probably should not hold an inherently political office of such importance as the presidency of Harvard, eh?

    - Alex Merz — 1/17/2005 @ 11:11 pm

  2. Actually, when people went back and compared the female scientists’ work to the male scientists at MIT, they found the guys had published more and had been referred to more in other papers. So by academic lights, they were producing more. (Though we know the worth of people churning out papers.)

    Lets back up to a =really=huge= disparity. How many blacks are tenured profs in the mathematical sciences? Hispanics? In most departments you’ve got a big, fat zero… though I do remember at N.C. State we had one in the math department - a guy from Africa (who had grown up there, and did his basic education there.) I think he was from Kenya. Oh, and we had Hispanics… from Brazil. No American-grown blacks or hispanics… wonder why…

    So… anyone want to claim innate differences there?

    That said, I’m tired of the assumption that the truly smart people will all end up in academia. Hmmm, I think the truly smart people might look at the pay and the hours and decide to go into business (I’m prejudiced, of course.) The truly smart people might say “To hell with other people’s expectations, I’d rather be a car mechanic.” or a caterer. or a financial planner. or a plumber.

    It doesn’t seem terribly smart to me to make your life miserable just because other people think a certain profession is above all others.

    - meep — 1/18/2005 @ 1:59 am

  3. Meep: you got a citation for the claim that male scientists @ MIT are more highly cited? And are we talking mean or median? I strongly suspect that citation rates have a skewed distribution, even at MIT, so mean citation rates could be misleading even if they are different. As for your speculation about where the smart people go: I’d advise my kids to learn a trade (plumbing never goes out of demand) before considering college, that’s for sure. I wish that had been pushed on me, and I’m a college professor who really, really likes his job.

    - Alex Merz — 1/18/2005 @ 10:30 am

  4. While I certainly respect TD&M’s oppinion and credentials, I believe attempting to place blame on either a director who must base his staffing decisions almost soley on accolades, or on the set of female mathematicians who are currently teaching mathematics at the university level may be yet another form of the blame game which you so despise in that same set of mathematicians.

    I would think that if anything is truely to change with the system then one must instead focus on the early high school classroom. While an ability at algebra in no way constitutes an ability in higher math, it is almost assuredly the ones who truely excel at and understand high school math (algebra, geometry, precalc) who are willing to take college math above and beyond the required calc I-II. To attempt to encourage these students to go into higher math from a calculus class they were woefully underprepared for, and thus struggling in is ludicrous! Therefor it is hardly apt to place blame on college professors for not encouraging their female students to enroll in higher math classes.

    As for why girls don’t seem to get the same level of education in the maths at the high school level? I have no sound basis for this, only speculation and stereotype… I firmly believe it is because the majority of men who truely enjoy and excel at math are “nerds” and proud to be considered so. Few are the female students who are willing to be ostracised so.

    - Arima — 1/18/2005 @ 12:21 pm

  5. Good point, Alex. I prefer seeing the entire distribution as opposed to a statistic like an average (whether mean or median). You’re right in that one person, if popular enough, could skew the entire distribution… especially if they’ve been a prof longer.

    Back to a short point I made at my livejournal — I want to give Summers credit in that he also noted that far more males than females were on the “dumb side” when it came to math, too. Basically, he noted that the standard deviation for math scores/math achievement/what have you over the entire male population is larger than that of the female population. He was saying we don’t know why this is, and it’s possible there could be biological reasons. We won’t know unless we ask. So the population means could be exactly the same, but the differing std. devs will mean that, if all else is equal (inclination, opportunities, etc.), one would still end up with more men at the upper echelons than women.

    I think it’s interesting that std. devs for men tend to be larger than women’s over a wide range of attributes, and I wonder why that is. Declaring that one possible cause is ideologically unacceptable does not get one closer to the truth.

    - meep — 1/18/2005 @ 2:58 pm

  6. And I don’t remember where I read the MIT study followup, and I’m too lazy to look it up right now. I followed this latest brou-ha-ha through several blogs… and as you say, the reported results that I saw were in terms of averages. But then, the original study complaining about unfair treatment was done in anecdotes and averages - and not controlling for certain factors.

    - meep — 1/18/2005 @ 3:00 pm

  7. Meep - right, lack of black mathematicians. I think I’ve met two, ever. Both very talented. At the risk of drawing conclusions from such a small, anecdotal sample, if they’re at all representative of what blacks could do if they were given better opportunities in math, then there’s where we should be putting our energies.

    Arima - Where was I placing blame? I agree with this statement of yours - I would think that if anything is truely to change with the system then one must instead focus on the early high school classroom. - and in fact I said something quite similar in the older post I linked, though I went further - we need to focus on kids even earlier. By the time it comes time for people to apply for tenure-track positions, there just aren’t as many women in the pool. If biology is a large contributing factor, there’s nothing we can do. If socialization and discrimination are the main reason, then they need to be addressed with GIRLS, not with women, because it takes years of study to even get to the point where you can think of applying for a tenure-track job.

    Here are a number of interesting, realted questions that I haven’t seen addressed (perhaps I should put these in a separate post:

    1. In the last 30-40 years, the proportion of women in law school, med school, and various graduate programs has increased dramatically. I recall reading that women now comprise over half of all law students. Yet the percentage of women studying math and physics seems to have hit a plateau (20% or so in certain areas). If discrimination is a large factor, why do these fields disproportionately suffer?

    2. Further to #1 - it seems that there would be less potential for discrimination to take place in fields as objective as mathematics and physics; humanities are so much more culture-dependent, and hence, I’d imagine, more prone to harbour research done from the standard white, male, middle-class viewpoint. Why, then, have women and minority academics found more success there than in less political fields? (I should mention that a lot of bullshit has been published in math journals over the last 2 decades, so I’m not saying that math is immune to political crap. It just seems that it should be more robust than the humanities and social sciences.)

    3. What percentage of women in math and physics is enough? Do we work until we have a 50% share of those fields, or could we stop before if discrimination has been adequately dealt with? If the latter, how do we measure when that has happened? if the former - it’s a common and well-documented phenomenon that as we diverge from an equilibrium state, more and more work is required to achieve a difference. It may take X amount of effort to increase the proportion of women from 20-25%, but 10X effort to increase the proportion from 45-50%. At what point does the extra effort stop being “combatting discrimination” and become an excessive expenditure that could perhaps be diverted to assisting the actual women who are studying math and physics, as opposed to the ones who might if only they were given uber-ideal conditions?

    I’ve seen very little discussion on THOSE questions; what I’ve seen has been eclipsed by the indignant claims that biology can’t possibly be playing a role in the gender imbalance, and that until we have 50-50 representation, sexism and only sexism are afoot. And I ask those three questions sincerely and with an open mind; I’m not trying to prove anything here, I’m honestly curious.

    - Moebius Stripper — 1/18/2005 @ 5:24 pm

  8. Very interesting (and complicated) questions.

    In computer science, I’ve seen the percentage of female undergrads vary between 5 and 20 percent, with a slow but steady increase up until the .com craze started, and then a fast dropoff. Currently, it seems that a larger fraction of graduate students in CS are women than undergrads. I don’t know if my impressions are backed up by real data, though.

    Rob Kirby has some interesting thoughts about sexism in math. But take them with a grain of salt; Kirby was the target of several accusations of sexism while my wife and I were PhD students there, me in CS, she in math. At the time, the department was still reeling from the aftermath of Jenny Harrison’s tenure case.

    The percentages are weird. Almost half of mathematics bachelors degrees go to women, but only 20-25% of the PhDs. Or maybe it’s closer to 30% and rising. The difference between ugrad and grad fractions may be explained by the predominance of women in math education programs, which are almost always lumped together with regular math majors when the question of gender balance comes up. Roughly 15-20% of assistant professors in math are women, and about 5-10% of the tenure-track faculty overall. (I found the tables here.) The fraction of women math faculty is still growing, albeit slowly. On the other hand, I’ve heard that almost half of adjuncts/lecturers/instructors in math are women! (I’m still looking for supporting data for that.)

    Math and physics may be more objective, but mathematicians and physicists are still very human! Perhaps because of the objectivity of their research, they’re less willing to admit that their personal biases are not Objective Truth. (This certainly applied to disagreements over which subfields are Real Math as opposed to Applied Junk or Useless Nonsense. What does “geometry” mean in your department?) On the other hand, perhaps math and science people are less willing to buy into the 50-50 gender equality myth.

    “When should we stop?” When the pipeline stops leaking? When the same fraction of men and women complain about gender discrimination? When ethnic imbalance finally becomes too embarassing to ignore?

    To Arima: There’s plenty of blame to go around. Yes, socialization at the high school level (and earlier!) is a much bigger problem, but professors who steer female students away from higher math make things worse.

    - Jeff Erickson — 1/19/2005 @ 7:58 am

  9. Here is an objective fact that I find illustrates that there is still a problem. (It comes from the annual AMS studies on the subject; I am still trying to track down exact numbers).

    The proportion of math PhD graduates which are female (which is at an all time high) is significantly higher than the proportion of new (tenure track) hires which are female. The proportion granted tenure is even a smaller fraction, particularly at elite american universities (like Harvard) where the number of females with tenure is almost 0%. (The AMS has different categories for the different universities).

    This year for the first time the proportion of women in graduate school (across *all* subjects) is greater than that of men, but university hiring does not reflect this.

    I think a hiring rate comparable to the phd graduate rate is a reasonable (first) target. Or, it begs one to ask why are the women so adequately unprepared for the market? There are lots of possibilities like the way hiring is actually done, specifically the importance of a mentors’ influence.

    Finally, the AMS in a recent meeting had a huge session on, not just minorities in mathematics, but how to better accomodate less traditional families, for example, families with severly disabled children. These topics are definitely being discussed in the framework of how to make academia more accessible to those who do not fit the profile of (upper) middle class white man with a wife/mother at home who can dedicate his every waking hour to mathematics.

    - Marni — 1/19/2005 @ 10:19 am

  10. No, I don’t think hiring rates should match graduation rates. If 60% of all Ph.D.s are female, I don’t think it follows we should have 60% of all profs being female. It is a bit simplistic.

    It could be that female Ph.D.s favor other lifestyles. It could be they don’t to go live in another state, want to remain close to family. It could be they are truly more likely to be interested in other types of jobs, like teaching at the high school level or working in industry.

    To be frank, getting a professorship is not necessarily a sane thing to do.

    Now, I’d be a lot more worried if getting tenure was more difficult for women… if that’s true, then we have to give it a really hard look…

    - Daniel Lemire — 1/19/2005 @ 8:45 pm

  11. DL: I agree that a hard rule about percentages would be excessive, but I fear your argument sounds a little like the classic “but women just don’t like to be management”, (see current walmart case, for example).

    Here is why I am happy about my university job: lots of freedom and power over what I do and how I spend my day; excellent pay, benefits and security; intellectually stimulating; my research is associated with me, and not some large corporation … sure there are bad parts, but in all honesty as jobs go, it is not such a bad lifestyle, you must admit.

    My fear that the “other lifestyle” that keeps them out is the one in which they get pregnant, and choose a family in the crucial years of phd/ postdoc = file with maybe one less publication = …

    - Marni — 1/22/2005 @ 7:34 am

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.