Then Harvard University President Larry Summers’ Controversial 2005 Speech Questioning the Continued Lack of Female Represention in the Science, Engineering and Math Faculties of America’s Best Universities
Lawrence H. Summers
January 14, 2005
I asked Richard, when he invited me to come here and speak, whether he wanted an institutional talk about Harvard’s policies toward diversity or whether he wanted some questions asked and some attempts at provocation, because I was willing to do the second and didn’t feel like doing the first. And so we have agreed that I am speaking unofficially and not using this as an occasion to lay out the many things we’re doing at Harvard to promote the crucial objective of diversity. There are many aspects of the problems you’re discussing and it seems to me they’re all very important from a national point of view. I’m going to confine myself to addressing one portion of the problem, or of the challenge we’re discussing, which is the issue of women’s representation in tenured positions in science and engineering at top universities and research institutions, not because that’s necessarily the most important problem or the most interesting problem, but because it’s the only one of these problems that I’ve made an effort to think in a very serious way about. The other prefatory comment that I would make is that I am going to, until most of the way through, attempt to adopt an entirely positive, rather than normative approach, and just try to think about and offer some hypotheses as to why we observe what we observe without seeing this through the kind of judgmental tendency that inevitably is connected with all our common goals of equality. It is after all not the case that the role of women in science is the only example of a group that is significantly underrepresented in an important activity and whose underrepresentation contributes to a shortage of role models for others who are considering being in that group. To take a set of diverse examples, the data will, I am confident, reveal that Catholics are substantially underrepresented in investment banking, which is an enormously high-paying profession in our society; that white men are very substantially underrepresented in the National Basketball Association; and that Jews are very substantially underrepresented in farming and in agriculture. These are all phenomena in which one observes underrepresentation, and I think it’s important to try to think systematically and clinically about the reasons for underrepresentation.
There are three broad hypotheses about the sources of the very substantial disparities that this conference’s papers document and have been documented before with respect to the presence of women in high-end scientific professions. One is what I would call the-I’ll explain each of these in a few moments and comment on how important I think they are-the first is what I call the high-powered job hypothesis. The second is what I would call different availability of aptitude at the high end, and the third is what I would call different socialization and patterns of discrimination in a search. And in my own view, their importance probably ranks in exactly the order that I just described.
Maybe it would be helpful to just, for a moment, broaden the problem, or the issue, beyond science and engineering. I’ve had the opportunity to discuss questions like this with chief executive officers at major corporations, the managing partners of large law firms, the directors of prominent teaching hospitals, and with the leaders of other prominent professional service organizations, as well as with colleagues in higher education. In all of those groups, the story is fundamentally the same. Twenty or twenty-five years ago, we started to see very substantial increases in the number of women who were in graduate school in this field. Now the people who went to graduate school when that started are forty, forty-five, fifty years old. If you look at the top cohort in our activity, it is not only nothing like fifty-fifty, it is nothing like what we thought it was when we started having a third of the women, a third of the law school class being female, twenty or twenty-five years ago. And the relatively few women who are in the highest ranking places are disproportionately either unmarried or without children, with the emphasis differing depending on just who you talk to. And that is a reality that is present and that one has exactly the same conversation in almost any high-powered profession. What does one make of that? I think it is hard-and again, I am speaking completely descriptively and non-normatively-to say that there are many professions and many activities, and the most prestigious activities in our society expect of people who are going to rise to leadership positions in their forties near total commitments to their work. They expect a large number of hours in the office, they expect a flexibility of schedules to respond to contingency, they expect a continuity of effort through the life cycle, and they expect-and this is harder to measure-but they expect that the mind is always working on the problems that are in the job, even when the job is not taking place. And it is a fact about our society that that is a level of commitment that a much higher fraction of married men have been historically prepared to make than of married women. That’s not a judgment about how it should be, not a judgment about what they should expect. But it seems to me that it is very hard to look at the data and escape the conclusion that that expectation is meeting with the choices that people make and is contributing substantially to the outcomes that we observe. One can put it differently. Of a class, and the work that Claudia Goldin and Larry Katz are doing will, I’m sure, over time, contribute greatly to our understanding of these issues and for all I know may prove my conjectures completely wrong. Another way to put the point is to say, what fraction of young women in their mid-twenties make a decision that they don’t want to have a job that they think about eighty hours a week. What fraction of young men make a decision that they’re unwilling to have a job that they think about eighty hours a week, and to observe what the difference is. And that has got to be a large part of what is observed. Now that begs entirely the normative questions-which I’ll get to a little later-of, is our society right to expect that level of effort from people who hold the most prominent jobs? Is our society right to have familial arrangements in which women are asked to make that choice and asked more to make that choice than men? Is our society right to ask of anybody to have a prominent job at this level of intensity, and I think those are all questions that I want to come back to. But it seems to me that it is impossible to look at this pattern and look at its pervasiveness and not conclude that something of the sort that I am describing has to be of significant importance. To buttress conviction and theory with anecdote, a young woman who worked very closely with me at the Treasury and who has subsequently gone on to work at Google highly successfully, is a 1994 graduate of Harvard Business School. She reports that of her first year section, there were twenty-two women, of whom three are working full time at this point. That may, the dean of the Business School reports to me, that that is not an implausible observation given their experience with their alumnae. So I think in terms of positive understanding, the first very important reality is just what I would call the, who wants to do high-powered intense work?
The second thing that I think one has to recognize is present is what I would call the combination of, and here, I’m focusing on something that would seek to answer the question of why is the pattern different in science and engineering, and why is the representation even lower and more problematic in science and engineering than it is in other fields. And here, you can get a fair distance, it seems to me, looking at a relatively simple hypothesis. It does appear that on many, many different human attributes-height, weight, propensity for criminality, overall IQ, mathematical ability, scientific ability-there is relatively clear evidence that whatever the difference in means-which can be debated-there is a difference in the standard deviation, and variability of a male and a female population. And that is true with respect to attributes that are and are not plausibly, culturally determined. If one supposes, as I think is reasonable, that if one is talking about physicists at a top twenty-five research university, one is not talking about people who are two standard deviations above the mean. And perhaps it’s not even talking about somebody who is three standard deviations above the mean. But it’s talking about people who are three and a half, four standard deviations above the mean in the one in 5,000, one in 10,000 class. Even small differences in the standard deviation will translate into very large differences in the available pool substantially out. I did a very crude calculation, which I’m sure was wrong and certainly was unsubtle, twenty different ways. I looked at the Xie and Shauman paper-looked at the book, rather-looked at the evidence on the sex ratios in the top 5% of twelfth graders. If you look at those-they’re all over the map, depends on which test, whether it’s math, or science, and so forth-but 50% women, one woman for every two men, would be a high-end estimate from their estimates. From that, you can back out a difference in the implied standard deviations that works out to be about 20%. And from that, you can work out the difference out several standard deviations. If you do that calculation-and I have no reason to think that it couldn’t be refined in a hundred ways-you get five to one, at the high end. Now, it’s pointed out by one of the papers at this conference that these tests are not a very good measure and are not highly predictive with respect to people’s ability to do that. And that’s absolutely right. But I don’t think that resolves the issue at all. Because if my reading of the data is right-it’s something people can argue about-that there are some systematic differences in variability in different populations, then whatever the set of attributes are that are precisely defined to correlate with being an aeronautical engineer at MIT or being a chemist at Berkeley, those are probably different in their standard deviations as well. So my sense is that the unfortunate truth-I would far prefer to believe something else, because it would be easier to address what is surely a serious social problem if something else were true-is that the combination of the high-powered job hypothesis and the differing variances probably explains a fair amount of this problem.
There may also be elements, by the way, of differing, there is some, particularly in some attributes, that bear on engineering, there is reasonably strong evidence of taste differences between little girls and little boys that are not easy to attribute to socialization. I just returned from Israel, where we had the opportunity to visit a kibbutz, and to spend some time talking about the history of the kibbutz movement, and it is really very striking to hear how the movement started with an absolute commitment, of a kind one doesn’t encounter in other places, that everybody was going to do the same jobs. Sometimes the women were going to fix the tractors, and the men were going to work in the nurseries, sometimes the men were going to fix the tractors and the women were going to work in the nurseries, and just under the pressure of what everyone wanted, in a hundred different kibbutzes, each one of which evolved, it all moved in the same direction. So, I think, while I would prefer to believe otherwise, I guess my experience with my two and a half year old twin daughters who were not given dolls and who were given trucks, and found themselves saying to each other, look, daddy truck is carrying the baby truck, tells me something. And I think it’s just something that you probably have to recognize. There are two other hypotheses that are all over. One is socialization. Somehow little girls are all socialized towards nursing and little boys are socialized towards building bridges. No doubt there is some truth in that. I would be hesitant about assigning too much weight to that hypothesis for two reasons. First, most of what we’ve learned from empirical psychology in the last fifteen years has been that people naturally attribute things to socialization that are in fact not attributable to socialization. We’ve been astounded by the results of separated twins studies. The confident assertions that autism was a reflection of parental characteristics that were absolutely supported and that people knew from years of observational evidence have now been proven to be wrong. And so, the human mind has a tendency to grab to the socialization hypothesis when you can see it, and it often turns out not to be true. The second empirical problem is that girls are persisting longer and longer. When there were no girls majoring in chemistry, when there were no girls majoring in biology, it was much easier to blame parental socialization. Then, as we are increasingly finding today, the problem is what’s happening when people are twenty, or when people are twenty-five, in terms of their patterns, with which they drop out. Again, to the extent it can be addressed, it’s a terrific thing to address.
The most controversial in a way, question, and the most difficult question to judge, is what is the role of discrimination? To what extent is there overt discrimination? Surely there is some. Much more tellingly, to what extent are there pervasive patterns of passive discrimination and stereotyping in which people like to choose people like themselves, and the people in the previous group are disproportionately white male, and so they choose people who are like themselves, who are disproportionately white male. No one who’s been in a university department or who has been involved in personnel processes can deny that this kind of taste does go on, and it is something that happens, and it is something that absolutely, vigorously needs to be combated. On the other hand, I think before regarding it as pervasive, and as the dominant explanation of the patterns we observe, there are two points that should make one hesitate. The first is the fallacy of composition. No doubt it is true that if any one institution makes a major effort to focus on reducing stereotyping, on achieving diversity, on hiring more people, no doubt it can succeed in hiring more. But each person it hires will come from a different institution, and so everyone observes that when an institution works very hard at this, to some extent they are able to produce better results. If I stand up at a football game and everybody else is sitting down, I can see much better, but if everybody stands up, the views may get a little better, but they don’t get a lot better. And there’s a real question as to how plausible it is to believe that there is anything like half as many people who are qualified to be scientists at top ten schools and who are now not at top ten schools, and that’s the argument that one has to make in thinking about this as a national problem rather than an individual institutional problem. The second problem is the one that Gary Becker very powerfully pointed out in addressing racial discrimination many years ago. If it was really the case that everybody was discriminating, there would be very substantial opportunities for a limited number of people who were not prepared to discriminate to assemble remarkable departments of high quality people at relatively limited cost simply by the act of their not discriminating, because of what it would mean for the pool that was available. And there are certainly examples of institutions that have focused on increasing their diversity to their substantial benefit, but if there was really a pervasive pattern of discrimination that was leaving an extraordinary number of high-quality potential candidates behind, one suspects that in the highly competitive academic marketplace, there would be more examples of institutions that succeeded substantially by working to fill the gap. And I think one sees relatively little evidence of that. 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, because I would like nothing better than for these problems to be addressable simply by everybody understanding what they are, and working very hard to address them.
What’s to be done? And what further questions should one know the answers to? Let me take a second, first to just remark on a few questions that it seems to me are ripe for research, and for all I know, some of them have been researched. First, it would be very useful to know, with hard data, what the quality of marginal hires are when major diversity efforts are mounted. When major diversity efforts are mounted, and consciousness is raised, and special efforts are made, and you look five years later at the quality of the people who have been hired during that period, how many are there who have turned out to be much better than the institutional norm who wouldn’t have been found without a greater search. And how many of them are plausible compromises that aren’t unreasonable, and how many of them are what the right-wing critics of all of this suppose represent clear abandonments of quality standards. I don’t know the answer, but I think if people want to move the world on this question, they have to be willing to ask the question in ways that could face any possible answer that came out. Second, and by the way, I think a more systematic effort to look at citation records of male and female scholars in disciplines where citations are relatively well-correlated with academic rank and with people’s judgments of quality would be very valuable. Of course, most of the critiques of citations go to reasons why they should not be useful in judging an individual scholar. Most of them are not reasons why they would not be useful in comparing two large groups of scholars and so there is significant potential, it seems to me, for citation analysis in this regard. Second, what about objective versus subjective factors in hiring? I’ve been exposed, by those who want to see the university hiring practices changed to favor women more and to assure more diversity, to two very different views. One group has urged that we make the processes consistently more clear-cut and objective, based on papers, numbers of papers published, numbers of articles cited, objectivity, measurement of performance, no judgments of potential, no reference to other things, because if it’s made more objective, the subjectivity that is associated with discrimination and which invariably works to the disadvantage of minority groups will not be present. I’ve also been exposed to exactly the opposite view, that those criteria and those objective criteria systematically bias the comparisons away from many attributes that those who contribute to the diversity have: a greater sense of collegiality, a greater sense of institutional responsibility. Somebody ought to be able to figure out the answer to the question of, if you did it more objectively versus less objectively, what would happen. Then you can debate whether you should or whether you shouldn’t, if objective or subjective is better. But that question ought to be a question that has an answer, that people can find. Third, the third kind of question is, what do we know about search procedures in universities? Is it the case that more systematic comprehensive search processes lead to minority group members who otherwise would have not been noticed being noticed? Or does fetishizing the search procedure make it very difficult to pursue the targets of opportunity that are often available arising out of particular family situations or particular moments, and does fetishizing and formalizing search procedures further actually work to the disadvantage of minority group members. Again, everybody’s got an opinion; I don’t think anybody actually has a clue as to what the answer is. Fourth, what do we actually know about the incidence of financial incentives and other support for child care in terms of what happens to people’s career patterns. I’ve been struck at Harvard that there’s something unfortunate and ironic about the fact that if you’re a faculty member and you have a kid who’s 18 who goes to college, we in effect, through an interest-free loan, give you about $9,000. If you have a six-year-old, we give you nothing. And I don’t think we’re very different from most other universities in this regard, but there is something odd about that strategic choice, if the goal is to recruit people to come to the university. But I don’t think we know much about the child care issue. The fifth question-which it seems to me would be useful to study and to actually learn the answer to-is what do we know, or what can we learn, about the costs of career interruptions. There is something we would like to believe. We would like to believe that you can take a year off, or two years off, or three years off, or be half-time for five years, and it affects your productivity during the time, but that it really doesn’t have any fundamental effect on the career path. And a whole set of conclusions would follow from that in terms of flexible work arrangements and so forth. And the question is, in what areas of academic life and in what ways is it actually true. Somebody reported to me on a study that they found, I don’t remember who had told me about this-maybe it was you, Richard-that there was a very clear correlation between the average length of time, from the time a paper was cited. That is, in fields where the average papers cited had been written nine months ago, women had a much harder time than in fields where the average thing cited had been written ten years ago. And that is suggestive in this regard. On the discouraging side of it, someone remarked once that no economist who had gone to work at the President’s Council of Economic Advisors for two years had done highly important academic work after they returned. Now, I’m sure there are counterexamples to that, and I’m sure people are kind of processing that Tobin’s Q is the best-known counterexample to that proposition, and there are obviously different kinds of effects that happen from working in Washington for two years. But it would be useful to explore a variety of kinds of natural interruption experiments, to see what actual difference it makes, and to see whether it’s actually true, and to see in what ways interruptions can be managed, and in what fields it makes a difference. I think it’s an area in which there’s conviction but where it doesn’t seem to me there’s an enormous amount of evidence. What should we all do? I think the case is overwhelming for employers trying to be the [unintelligible] employer who responds to everybody else’s discrimination by competing effectively to locate people who others are discriminating against, or to provide different compensation packages that will attract the people who would otherwise have enormous difficulty with child care. I think a lot of discussion of issues around child care, issues around extending tenure clocks, issues around providing family benefits, are enormously important. I think there’s a strong case for monitoring and making sure that searches are done very carefully and that there are enough people looking and watching that that pattern of choosing people like yourself is not allowed to take insidious effect. But I think it’s something that has to be done with very great care because it slides easily into pressure to achieve given fractions in given years, which runs the enormous risk of people who were hired because they were terrific being made to feel, or even if not made to feel, being seen by others as having been hired for some other reason. And I think that’s something we all need to be enormously careful of as we approach these issues, and it’s something we need to do, but I think it’s something that we need to do with great care.
Let me just conclude by saying that I’ve given you my best guesses after a fair amount of reading the literature and a lot of talking to people. They may be all wrong. I will have served my purpose if I have provoked thought on this question and provoked the marshalling of evidence to contradict what I have said. But I think we all need to be thinking very hard about how to do better on these issues and that they are too important to sentimentalize rather than to think about in as rigorous and careful ways as we can. That’s why I think conferences like this are very, very valuable. Thank you.
Questions and Answers
Q: Well, I don’t want to take up much time because I know other people have questions, so, first of all I’d like to say thank you for your input. It’s very interesting-I noticed it’s being recorded so I hope that we’ll be able to have a copy of it. That would be nice.
LHS: We’ll see. (LAUGHTER)
Q: Secondly, you make a point, which I very much agree with, that this is a wonderful opportunity for other universities to hire women and minorities, and you said you didn’t have an example of an instance in which that is being done. The chemistry department at Rutgers is doing that, and they are bragging about it and they are saying, “Any woman who is having problems in her home department, send me your resume.” They are now at twenty-five percent women, which is double the national average-among the top fifty universities-so I agree with you on that. I think it is a wonderful opportunity and I hope others follow that example. One thing that I do sort of disagree with is the use of identical twins that have been separated and their environment followed. I think that the environments that a lot of women and minorities experience would not be something that would be-that a twin would be subjected to if the person knows that their environment is being watched. Because a lot of the things that are done to women and minorities are simply illegal, and so they’ll never experience that.
LHS: I don’t think that. I don’t actually think that’s the point at all. My point was a very different one. My point was simply that the field of behavioral genetics had a revolution in the last fifteen years, and the principal thrust of that revolution was the discovery that a large number of things that people thought were due to socialization weren’t, and were in fact due to more intrinsic human nature, and that set of discoveries, it seemed to me, ought to influence the way one thought about other areas where there was a perception of the importance of socialization. I wasn’t at all trying to connect those studies to the particular experiences of women and minorities who were thinking about academic careers.
Q: Raising that particular issue, as a biologist, I neither believe in all genetic or all environment, that in fact behavior in any other country actually develops [unintelligible] interaction of those aspects. And I agree with you, in fact, that it is wrong-headed to just dismiss the biology. But to put too much weight to it is also incredibly wrong-headed, given the fact that had people actually had different kinds of opportunities, and different opportunities for socialization, there is good evidence to indicate in fact that it would have had different outcomes. I cite by way of research the [unintelligible] project in North Carolina, which essentially shows that, where every indicator with regard to mother’s education, socioeconomic status, et cetera, would have left a kid in a particular place educationally, that, essentially, they are seeing totally different outcomes with regard to performance, being referred to special education, et cetera, so I think that there is some evidence on that particular side. The other issue is this whole question about objective versus subjective. I think that it is very difficult to have anything that is basically objective, and the work of [unintelligible] I think point out that in a case where you are actually trying to-this case from the Swedish Medical Council, where they were trying to identify very high-powered research opportunities for, I guess it was post-docs by that point, that indicated that essentially that it ended up with larger numbers of men than women. Two of the women who were basically in the affected group were able to utilize the transparency rules that were in place in Sweden, get access to the data, get access to the issues, and in fact, discovered that it was not as objective as everyone claimed, and that in fact, different standards were actually being used for the women as well as for the men, including the men’s presence in sort of a central network, the kinds of journals that they had to publish in to be considered at the same level, so I think that there are pieces of research that begin to actually relate to this-yes, there is the need to look more carefully at a lot of these areas. I would-in addition looking at this whole question of the quality of marginal hires-I would also like to look at the quality of class one hires, in terms of seeing who disappoints, and what it was that they happened to be looking at and making judgments on, and then what the people could not deliver. So I think that there is a real great need on both sides to begin to talk about whether or not we can predict. I hate to use a sports metaphor, but I will. This is drawn basically from an example from Claude Steele, where he says, he starts by using free throws as a way of actually determining, who should-you’ve got to field a basketball team, and you clearly want the people who make ten out of ten, and you say, “Well, I may not want the people who make zero out of ten,” but what about the people who make four out of ten. If you use that as the measure, Shaq will be left on the sidelines.
LHS: I understand. I think you’re obviously right that there’s no absolute objectivity, and you’re-there’s no question about that. My own instincts actually are that you could go wrong in a number of respects fetishizing objectivity for exactly the reasons that you suggest. There is a very simple and straightforward methodology that was used many years ago in the case of baseball. Somebody wrote a very powerful article about baseball, probably in the seventies, in which they basically said, “Look, it is true that if you look at people’s salaries, and you control for their batting averages and their fielding averages and whatnot, whites and blacks are in the same salary once you control. It is also true that there are no black .240 hitters in the major leagues, that the only blacks who are in the major leagues are people who bat over .300-I’m exaggerating-and that is exactly what you’d predict on a model of discrimination, that because there’s a natural bias against. And there’s an absolute and clear prediction. The prediction is that if there’s a discriminated-against group, that if you measure subsequent performance, their subsequent performance will be stronger than that of the non-discriminated-against group. And that’s a simple prediction of a theory of discrimination. And it’s a testable prediction of a theory of discrimination, and it would be a revolution, and it would be an enormously powerful finding in this field, to demonstrate, and I suspect there are contexts in which that can be demonstrated, but there’s a straightforward methodology, it seems to me, for testing exactly that idea. I’m going to run out of time. But, let me take-if people ask very short questions, I will give very short answers.
Q: What about the rest of the world. Are we keeping up? Physics, France, very high powered women in science in top positions. Same nature, same hormones, same ambitions we have to assume. Different cultural, given.
LHS: Good question. Good question. I don’t know much about it. My guess is that you’ll find that in most of those places, the pressure to be high powered, to work eighty hours a week, is not the same as it is in the United States. And therefore it is easier to balance on both sides. But I thought about that, and I think that you’ll find that’s probably at least part of the explanation.
Q: [unintelligible] because his book was referred to.
Q: I would like to make an on observation and then make a suggestion. The observation is that of the three. There is a contradiction in your three major observations that is the high-powered intensive need of scientific work-that’s the first-and then the ability, and then the socialization, the social process. Would it be possible the first two result from the last one and that math ability could be a result of education, parenting, a lot of things. We only observe what happens, we don’t know the reason for why there’s a variance. I’ll give you another thing, a suggestion. The suggestion is that one way to read your remarks is to say maybe those are not the things we can solve immediately. Especially as leaders of higher education because they are just so wide, so deep, and involves all aspects of society, institution, education, a lot of things, parenting, marriages are institutions, for example. We could have changed the institution of those things a lot of things we cannot change. Rather, it’s not nature and nurture, it is really pre-college versus post-college. From your college point of view maybe those are things too late and too little you can do but a lot of things which are determined by sources outside the college you’re in. Is that…
LHS: I think…
Q: That’s a different read on your set of remarks.
LHS: I think your observation goes much more to my second point about the abilities and the variances than it does to the first point about what married woman….
LHS: Yeah, look anything could be social, ultimately in all of that. I think that if you look at the literature on behavioral genetics and you look at the impact, the changed view as to what difference parenting makes, the evidence is really quite striking and amazing. I mean, just read Judith Rich Harris’s book. It is just very striking that people’s-and her book is probably wrong and its probably more than she says it is, and I know there are thirteen critiques and you can argue about it and I am not certainly a leading expert on that-but there is a lot there. And I think what it surely establishes is that human intuition tends to substantially overestimate the role-just like teachers overestimate their impact on their students relative to fellow students on other students-I think we all have a tendency with our intuitions to do it. So, you may be right, but my guess is that there are some very deep forces here that are going to be with us for a long time.
Q: You know, in the spirit of speaking truth to power, I’m not an expert in this area but a lot of people in the room are, and they’ve written a lot of papers in here that address ….
LHS: I’ve read a lot of them.
Q: And, you know, a lot of us would disagree with your hypotheses and your premises…
LHS: Fair enough.
Q: So it’s not so clear.
LHS: It’s not clear at all. I think I said it wasn’t clear. I was giving you my best guess but I hope we could argue on the basis of as much evidence as we can marshal.
Q: It’s here.
LHS: No, no, no. Let me say. I have actually read that and I’m not saying there aren’t rooms to debate this in, but if somebody, but with the greatest respect-I think there’s an enormous amount one can learn from the papers in this conference and from those two books-but if somebody thinks that there is proof in these two books, that these phenomenon are caused by something else, I guess I would very respectfully have to disagree very very strongly with that. I don’t presume to have proved any view that I expressed here, but if you think there is proof for an alternative theory, I’d want you to be hesitant about that.
Q: Just one quick question in terms of the data. We saw this morning lots of data showing the drop in white males entering science and engineering, and I’m having trouble squaring that with your model of who wants to work eighty hours a week. It’s mostly people coming from other countries that have filled that gap in terms of men versus women.
LHS: I think there are two different things, frankly, actually, is my guess-I’m not an expert. Somebody reported to me that-someone who is knowledgeable-said that it is surprisingly hard to get Americans rather than immigrants or the children of immigrants to be cardiac surgeons. Cardiac surgeon is about prestigious, certain kind of prestige as you can be, fact is that people want control of their lifestyles, people want flexibility, they don’t want to do it, and it’s disproportionately immigrants that want to do some of the careers that are most demanding in terms of time and most interfering with your lifestyle. So I think that’s exactly right and I think it’s precisely the package of number of hours’ work what it is, that’s leading more Americans to choose to have careers of one kind or another in business that are less demanding of passionate thought all the time and that includes white males as well.
Q: That’s my point, that social-psychological in nature [unintelligible].
LHS: I would actually much rather stay-yes, and then I’m on my way out.
Q: I have no idea how you would evaluate the productivity of the marginal hire if this person is coming into an environment where [unintelligible] is marginal and there’s [unintelligible].
LHS: You’re absolutely right. You’re absolutely right. I used the term-I realized I had not spoken carefully-I used the term marginal in the economic sense to mean, only additional, to only mean…
LHS: No, to mean only the additional [unintelligible]. Yeah, obviously [unintelligible] going to identify X is the additional hire, is the marginal hire, the question you can ask is, you know, here is a time when, as a consequence of an effort, there was a very substantial increase in the number of people who were hired in a given group, what was the observed ex post quality? And what was the observed ex post performance? It’s hard to believe that that’s not a useful thing to try to know. It may well be that one will produce powerful evidence that the people are much better than the people who were there and that the institutions went up in quality and that made things much better. All I’m saying is one needs to ask the question. And as for the groping in the kitchen, and whatnot, look, it’s absolutely important that in every university in America there be norms of civility and proper treatment of colleagues that be absolutely established and that that be true universally, and that’s a hugely important part of this, and that’s why at Harvard we’re doing a whole set of things that are making junior faculty positions much more real faculty positions with real mentoring, real feedback, serious searches before the people are hired, and much greater prospects for tenure than there ever have been before because exactly that kind of collegiality is absolutely central to the academic enterprise.
Lawrence Henry Summers (born November 30, 1954) is an American economist and the Director of the White House’s National Economic Council for President Barack Obama. Summers is the Charles W. Eliot University Professor at Harvard University’s Kennedy School of Government. He is the 1993 recipient of the John Bates Clark Medal for his work in several fields of economics and was Secretary of the Treasury for the last year and a half of the Clinton Administration.
Summers also served as the 27th President of Harvard University from 2001 to 2006. Summers resigned as Harvard’s president in the wake of a no-confidence vote by Harvard faculty that resulted in large part from Summers’ conflict with Cornel West, conflict of interest questions regarding his relationship with Andrei Shleifer, and a 2005 speech in which he suggested that the under-representation of women in the top levels of academia could be due to a “different availability of aptitude at the high end.” Summers has also been criticized by some liberals for the economic policies he advocated as Treasury Secretary and in later writings. Since returning to government in the Obama administration, he has come under fire for his numerous financial ties to Wall Street.
The Girls of Summers
What Harvard’s president and his critics got wrong
By William Saletan
Updated Tuesday, Feb. 22, 2005
For more than a month, critics have accused Harvard President Larry Summers of using genetics to explain away sexism in society and academia. They’ve demanded that he release transcripts of the remarks in question, delivered at an academic conference on Jan. 14. On Thursday, facing calls for his resignation, Summers released the transcript. It shows his critics misconstrued or misrepresented him on numerous points. It also shows what he got wrong and why.
Let’s start with his caveats, which eyewitness accounts omitted.
1. He reaffirmed the need to address discrimination. The transcript shows him affirming Harvard’s commitment to “the crucial objective of diversity” and urging his audience to address factors that cause women to drop out of academic career paths. Women are among the groups “significantly underrepresented” in an advanced field, he said, and their absence “contributes to a shortage of role models for others.”
2. He questioned the rationality of work expectations that discriminate against women. Earlier accounts suggested that when Summers cited very long work hours as a standard women were less likely to accept, he was justifying that standard and its discriminatory result. The transcript shows him making the opposite point: “Is our society right to expect that level of effort from people who hold the most prominent jobs? Is our society right to have familial arrangements in which women are asked to make that choice and asked more to make that choice than men?” He worried about employers’ defiance of “legitimate family desires” and suggested that they offer “different compensation packages that will attract the people who would otherwise have enormous difficulty with child care,” as well as “extending tenure clocks” and considering other “family benefits.”
3. When he said discrimination was the least of three factors in women’s underrepresentation, he was talking about discrimination in academic hiring, not discrimination earlier in life. The transcript shows him describing the third factor as “different socialization and patterns of discrimination in a search”—i.e., the search for a new faculty hire. Earlier accounts suggested he blew off discrimination as a factor on the grounds that there weren’t enough qualified women to hire in the first place. But the transcript shows him drawing a different conclusion from the inadequate pool of female candidates: He and his audience should be “thinking about this as a national problem rather than an individual institutional problem.”
4. When he spoke of differences between male and female test scores, he was confining his analysis to a tiny subset. “If one is talking about physicists at a top 25 research university,” he argued, the population in question was “in the one-in-5,000, one-in-10,000 class. Even small differences in the standard deviation will translate into very large differences in the available pool.” Summers explicitly said he wasn’t talking about a difference in average scores.
5. He rejected socialization as the sole factor—not as one factor—in test score differences. Summers said there was “reasonably strong evidence” of differences “that are not easy to attribute to socialization.” Afterward, when a critic suggested that the evidence supported an alternative explanation based on socialization, Summers replied, “I don’t presume to have proved any view that I expressed here. But if you think there is proof for an alternative theory, I’d want you to be hesitant about that.”
6. His story about his daughters was grossly misrepresented. Numerous reports of Summers’ remarks noted damningly that he had mentioned his daughters as evidence of innate gender differences. And indeed he did cite “my experience with my two-and-a-half-year-old twin daughters who were not given dolls and who were given trucks, and found themselves saying to each other, ‘Look, Daddy Truck is carrying the baby truck.’” But not one report mentioned that this was a minor anecdote appended to a more serious case study: the Israeli kibbutz movement, which, according to Summers, “started with an absolute commitment … that everybody was going to do the same jobs: Sometimes the women were going to fix the tractors, and the men were going to work in the nurseries.” Despite this sex-neutral commitment, he said, individual choices “in a hundred different kibbutzes … all moved in the same direction”—toward traditional gender roles. Summers’ point wasn’t that nature accounted for everything, but that attempts to erase it as a factor had failed. The kibbutzim were the evidence; his daughters were an afterthought.
In short, Summers got a bum rap. So, was his analysis of biological and cultural factors sound? The transcript answers that question, too. The answer is no. Summers grossly overreached the evidence, and he made a couple of glaring logical blunders.
Summers proposed “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.” In other words, biology outweighs environment. No evidence he presented justifies this hypothesis. So how did he reach it?
First, he rashly extrapolated from the limits of socialization in one area to the limits of socialization in another. “Most of what we’ve learned from empirical psychology in the last 15 years has been that people naturally attribute things to socialization that are in fact not attributable to socialization,” he said. “We’ve been astounded by the results of separated twins studies. The confident assertions that autism was a reflection of parental characteristics … have now been proven to be wrong.” For this reason, he was “hesitant about assigning too much weight” to the idea that girls and boys are socialized differently.
In the Q&A, a questioner pointed out that the environmental differences affecting identical twins (which are always of the same sex) are nothing like the environmental differences affecting boys and girls. Summers replied,
The field of behavioral genetics had a revolution in the last 15 years, and the principal thrust of that revolution was the discovery that a large number of things that people thought were due to socialization weren’t, and were in fact due to more intrinsic human nature. And that set of discoveries, it seemed to me, ought to influence the way one thought about other areas where there was a perception of the importance of socialization. I wasn’t at all trying to connect those studies to the particular experiences of women and minorities who were thinking about academic careers.
Any Harvard student who gave this answer on an exam would be flunked. If you aren’t claiming that a highly abstract resemblance to another subject has any bearing on this one—and you present no evidence to justify the cross-application—you have no business bringing it up.
Second, Summers confused two different causal conflicts. In the course of arguing that socialization was a less persuasive explanation for differential outcomes than biology was, he observed, “When there were no girls majoring in chemistry, when there were no girls majoring in biology, it was much easier to blame parental socialization. Then, as we are increasingly finding today, the problem is what’s happening when people are 20, or when people are 25, in terms of their patterns with which they drop out.” In other words, even after we’ve substantially canceled out differences in socialization by getting women to major successfully in sciences, they still drop out of the academic race. Well, yes. But that doesn’t show that the alternative factor is biology. It just shows that there’s an alternative factor—and Summers had already mentioned two other alternative factors that would more plausibly affect 25-year-old women: bias against women and bias against people who bear and raise children. The limits of egalitarian socialization in controlling a woman’s career prove nothing about the limits of sexist socialization in shaping a girl.
At one point, Summers acknowledged, “It’s pointed out by one of the papers at this conference that these tests are not a very good measure and are not highly predictive” of academic success. “And that’s absolutely right,” said Summers. “But I don’t think that resolves the issue at all. Because if … there are some systematic differences in variability in different populations, then whatever the set of attributes are that are precisely defined to correlate with being an aeronautical engineer at MIT or being a chemist at Berkeley, those are probably different in their standard deviations as well.”
What? This is pure abstract inference at an absurd level. It’s also incoherent. You can’t presume that men and women differ in the second respect while inferring this presumption from a likeness to their difference in the first. Either you presume similarities, or you presume differences.
Why did Summers make these mistakes? The transcript suggests two conflicting reasons. One is that he’s stubborn and argumentative. He repeatedly deflected cultural explanations by saying things like, “No doubt there is some truth in that,” “This kind of taste does go on,” and “Yeah, look, anything could be social”—and then minimizing these explanations. The consistent tone of his remarks was “Yeah, but …” There are two possible explanations for that tone in this context. One is that he’s a sexist. The other is that once he offers a hypothesis, he’d rather defend and extend it than listen objectively to the alternatives. He’s got an open mind but not an open heart.
I suspect this, rather than sexism, is the root of Summers’ errors, because a sexist wouldn’t have said what he said while displaying a second intellectual flaw evident in the transcript. Again and again, Summers warned his listeners to be skeptical of what they’d prefer to believe. We all want to believe socialization explains differences in male and female outcomes, he observed. Therefore, he reasoned, we should distrust that hypothesis and look for evidence to the contrary. He was so busy being skeptical of the popular explanation that he forgot to be skeptical of the unpopular one. He overstated the case for innate sex differences not because he wanted to believe it, but because he didn’t.
If you think this explanation is too kind to Summers, ask yourself why he told the story about his daughters. An old-fashioned sexist wouldn’t have told that story, because he wouldn’t have been surprised at his daughter’s maternal behavior—never mind that he wouldn’t have given her a truck in the first place. Summers brought up the incident not because it would rock the academic world—it didn’t—but because it rocked him. As he put it, the incident “tells me something.” He wasn’t speaking as the president of Harvard or even as a scholar. He was speaking as a modern dad who thought he could overcome nature and discovered he couldn’t.
When we talk about gender or any other controversial topic, we “have to be willing to ask the question in ways that could face any possible answer that came out,” Summers implored his audience. What brave and wise counsel. Now he just needs to follow it.
William Saletan is Slate‘s national correspondent and author of Bearing Right: How Conservatives Won the Abortion War.
Don’t Let Larry Summers Off the Hook Yet
Why the Harvard president’s tactless social science was a bad idea.
By Meghan O’Rourke
Posted Friday, Jan. 28, 2005
In the weeks since debate erupted over Harvard President Lawrence Summers’ suggestion that “innate differences” between men and women help explain the lack of top-level female professionals in science and engineering, a remarkably consistent narrative has emerged in the mainstream media: Summers is a martyr to political correctness. He’s the inquisitive freethinker asking the hard questions that need to be asked, while his small-minded critics are thwarting serious debate. These women, a Washington Post columnist argued, epitomize the “unwillingness of the modern academy to tolerate … freewheeling inquiry.”
But this is a facile narrative. You need not be animated by 1990s-style political correctness, or guilty of suppressing academic freedom, to suggest that both the manner and the substance of Summers’ comments at the conference convened by the National Bureau for Economic Research were counterproductive rather than usefully controversial. While Summers’ trademark bluntness is sometimes useful in prodding an institution entrenched in complacency to change its ways, in this case he was singularly ill-positioned to play galvanizing provocateur.
The issue on the table at the National Bureau for Economic Research conference was the underrepresentation of women at the upper levels of some of the physical sciences and in engineering. No one is talking about achieving 50-50 representation; women constitute approximately 20 percent of science and engineering departments nationwide and hold few senior positions. The possible explanations are either sociocultural or genetic, or both. Summers allegedly offered these three reasons as explanation: 1) Women want to have children, and as a result they don’t put in the 80-hour work week that would make them competitive with their male peers; 2) the innate differences between men and women lead men to outperform women at the top end; 3) discrimination discourages women from pursuing science and engineering past their undergraduate education. (According to Nancy Hopkins of MIT, who walked out of his presentation, he ranked these reasons in order of descending importance. Summers was traveling and couldn’t be reached for comment.)
To start with, when it comes to talking about innate biological differences between the genders, some tact and intellectual rigor is required—if only because of the long history of genetic explanations being used to justify discrimination. So how and what Summers said matters. So, too, does the position from which Summers is speaking. As a university president, Summers has a different role than a professor or researcher and a different set of public obligations. On a local level, he oversees tenure appointments and questions of policy and hiring. Any generalizations he makes about the genetic inferiority of women might easily lead individuals at his institution to question his faith in their ability and, in the best of situations, make it hard to attract talent to Harvard. And the situation at Harvard is not the best of situations. Summers has not done a good job of reassuring women he is battling discrimination at Harvard. Under his leadership, tenure appointments of women have declined every year; only four of the last 32 appointments were of women. Harvard does not have a senior female math professor; in a department of 18 chemistry professors, only one senior position is held by a woman. The chairwoman of the sociology department has told the New York Times that after a meeting with Summers about this issue many female faculty had left feeling he did not understand their concerns.
On a national level, Summers, as president of Harvard, has a stature and cachet few professors have. If he suggests in even the most nuanced way that women are innately inferior to men at top-level science and math, his words will inevitably be twisted, in a game of nationwide Telephone, into something far cruder by those whose latent sexism is in search of intellectual validation. (Where Harvard leads, others follow.) This might be a necessary price to pay if Summers’ comments actually shed light on debate over representation of women. It’s hard to see that they do. In the first place, the suggestion that genetic differences play some role in the discrepancy between gender representation in top-level science departments is hardly a revelation. Quite the contrary; as the many media responses to his comments have made clear, the terrain of innate differences is well-studied (if still poorly understood). Even Nancy Hopkins of MIT told me over the phone that she herself is thinking of switching to this “very intriguing” field. In the second place, Summers casually downplayed, and actually ended up reinforcing, the social consensus that girls are innately less good at math than boys—as well as the consensus, less widely embraced in other societies, that math and science are fields particularly shaped by innate ability rather than hard work.
This matters because, whatever the influence of genetics may turn out to be, there is no doubt that the enduring social consensus that women are on average worse than men in math and science plays a major role in shaping women’s careers and their career choices. It does so in two ways: through discrimination and through socialization. Contrary to the pie-in-the-sky assumptions of many of Summers’ media defenders, studies show that discrimination against women in the academy is alarmingly widespread, if often unconscious. M.A. Paludi and W.D. Bauer conducted a study in which 180 men and 180 women were asked to grade a paper on a five-point scale. When the author was “John T. McKay” rather than “Joan T. McKay,” the men on average graded the paper a point higher—and the women scoring the test weren’t much more egalitarian. And studies have shown that men writing mathematics papers are less likely to cite women than women are (1.2 percent of the time, compared to 4.8 percent)*. Scientists and engineers may say they aren’t biased. But consider the case of classical musicians: Until blind auditions were held for national orchestras, women were radically underrepresented in field of classical music. Many argued that women had less wind power and were biologically incapable of performance at highest levels on many instruments. Since blind auditions have been held, though, the participation of women has risen precipitously—evidence that it was almost entirely discrimination that was keeping women out.
Perhaps even more important than discrimination are the socialization biases—the impact of our collective belief that men are better than women at science and math. Whatever may or may not be the case about genetic differences, there’s clearly something going on that keeps even the larger percentage of women who now major in math/science from continuing on in those fields—something that a university, eager for a bigger pool of Ph.Ds from which to pick to augment its female faculty, should care about a lot. Claude Steele’s work on gender differences in learning gives solid—rather than impressionistic—grounding to the concern that comments like Summers’ are exactly what work against the continuing advancement of women.
Steele studies the way stereotypes affect people’s performance. And he has found that when women are told that a test is going to measure cognitive differences between genders they tend to do much worse than men. But when they’re told a test is gender-blind, they tend to perform as well. The pressure of the “stereotype threat,” as Steel terms it, actually leads women to do worse, in other words. The amazing thing is, as Steele convincingly argues, stereotype threat most affects those at the high end of the spectrum in math and science, because they’re the ones who are the most identified with the field and have the most to lose as they move upward and are increasingly identified as, say, a “female engineer.” This doesn’t mean that men aren’t outperforming women at the very high end of the bell curve, as my colleague Will Saletan points out; but it makes it look as though socialization is a weighty factor in gender disparities at top levels.
This is why discussion of genetic superiority of men in math and science needs to be especially rigorous—which is not to say Summers’ critics think it oughtn’t be discussed. But talking about genetics, in the age of genome-mapping, makes it very hard to take other factors seriously; people hear “genetics” and then draw broad, ill-considered conclusions—even educated columnists like Robert Samuelson, who concluded earlier this week that “many women probably reject science and engineering for another reason: They simply don’t find the work appealing, just as they generally don’t like football.” (In fact, 43 percent of the NFL fan base is women—nevermind, though; surely scientists are about to find the male football gene.) Summers has to be aware of this problem. He also ought to know that as real as genetic differences may be, the percentage of engineering majors has risen six times since 1971, to 18 percent in 2004—which, unless you think the human genome has changed since 1971, shows that factors other than genetics play a major role in women’s career choices.
Yet Summers used poorly digested social science to propose the innate superiority of one gender—”I hope to be proven wrong,” he allegedly said—while invoking the protection of the role of the “provocative” intellectual. (There is a tape of the event, but its release has not been authorized by Summers—a fact that has not been made much of by the reporters who have named him a defendant of free inquiry.) For this he has been praised for his scientific curiosity, while those who criticize him are indicted as obstructionists. It’s a curiously unscientific conclusion.
Meghan O’Rourke is Slate‘s culture critic and the author of Halflife, a collection of poetry. She is at work on The Long Goodbye, an account of grief, forthcoming in 2011.
Copyright 2010 Washingtonpost.Newsweek Interactive Co. LLC