Why are women underrepresented in philosophy and should we care?

In my last post, I offered a brief commentary on a recent paper about the representation of women in philosophy journals, but I now want to discuss a bit more thoroughly the underlying issue of the underrepresentation of women in philosophy. It’s a well-known fact that women are underrepresented in philosophy. Indeed, except for engineering, computer science and physics, philosophy seems to be the field in which the proportion of women is the lowest. Now, whenever a group that has traditionally been disadvantaged is underrepresented in a prestigious occupation (which is how philosophers, though perhaps not many other people, see their field), you can be certain that people are going to conclude that it’s because the members of that group are being discriminated against. So it shouldn’t surprise anyone that, in the case of women’s underrepresentation in philosophy, a lot of people have been saying just that. What should be more surprising is that, despite the lack of evidence (more on this shortly), this explanation has become so popular among philosophers, who like to think of themselves as reasonable people who care a great deal about evidence and are less susceptible to the kind of bias that affect other people. Unfortunately, as anyone who has interacted with philosophers for long enough knows, nothing could be further from the truth. The hypothesis that women are underrepresented in philosophy because sexism is pervasive in the field is what I call the official narrative, precisely because it has become the default explanation among philosophers, at least in public discourse.

Despite how widespread that explanation is, however, there is almost no evidence to support it. As far as I can tell, when the people who promote that explanation bother to adduce some evidence in favor of it, they either rely on anecdotal evidence or draw on psychological research which is deeply flawed and/or doesn’t show what they think it shows. The anecdotal evidence they give for the most part consist in stories of women that have been mistreated because of their gender or at least who think they have been mistreated because of their gender. You can find many examples of that kind of stories on What is it like to be a woman in philosophy?, a blog that was created a few years ago to collect them. The problem is that, while anecdotal evidence may be suggestive, it’s essentially worthless as evidence, even when you have a lot of it. The right thing to do when you have anecdotal evidence which suggests that a hypothesis might be true is to collect evidence in a systematic way to test it. (As I explain below, when people have done that in the case of the underrepresentation of women in philosophy, it didn’t confirm the hypothesis that discrimination against women is pervasive in philosophy, quite the opposite.) But until you have done that, anecdotal evidence has little probative value, at least when you’re trying to use it to support a causal hypothesis about a complex phenomenon. It can’t tell you much about the prevalence of sexism and it can tell you even less about the magnitude of the effect it has. This is true even if you have a lot of anecdotal evidence, because the problem isn’t so much the amount of evidence you have as much as the kind of evidence it is and, in particular, how it was collected. It never ceases to amaze me how often I have to repeat that to philosophers, despite the fact that it’s entirely uncontroversial in science, for reasons that are familiar to anyone who has read enough of it.

When they don’t rely on anecdotal evidence, the people who promote the official narrative use psychological research which, in the best case scenario, offers some very indirect support for the hypothesis that women are underrepresented in philosophy because of bias against them. In particular, implicit bias and stereotype threat have been all the rage among philosophers in recent years, though I’m afraid it’s mostly because they haven’t read the literature. Indeed, not only is that psychological research deeply flawed, but even if it were not, it would still not offer much support to the official narrative.  It would take a whole post to explain why in details, so here I will just make a few remarks, which hopefully will be sufficient to give you a sense of the problems with that research and the way in which it’s being used by people who promote the official narrative. First, as I already noted, the research about implicit bias and, even more so, stereotype threat is often flawed in a way most people who use it don’t suspect. For instance, the research about stereotype threat suffers from both publication bias and failure of replication, especially in connection with sex-based stereotype threat. Moreover, even if we could trust the results of the studies about stereotype threat, they don’t actually show what philosophers and many other people in the public think they do. So as far as we can tell from the literature, it’s not even clear that stereotype threat exists, let alone that it can explain the underrepresentation of women in philosophy.

The literature about implicit bias also has a lot of flaws, though it’s not as bad as that on stereotype threat, which is admittedly difficult. The main problem with it, however, is that several meta-analyses have shown that, to the extent that implicit bias exists and is measured by the Implicit Association Test (IAT), it doesn’t really explain much because the effect is extremely weak. If you want to know more about this, you should read Jesse Singal’s excellent write-up on the literature in New York Magazine, which has links to the most important studies. What’s really amazing is that, even though people in mainstream publications have only started to talk about this recently, it has been very clear to people who know the literature for a long time. Indeed, it was already clear when Greenwald et al. published their meta-analysis in 2009, despite their effort to spin the results. (In general, when you read social science, you should always ignore the spin and look at the tables. I have lost count of how many studies I have read which concluded exactly the opposite of what the data actually showed.) Philosophers routinely attribute to implicit bias a power that, according to the literature, it simply doesn’t have.

Even if this were not the case, this psychological research would offer little support to the hypothesis that women are underrepresented in philosophy because discrimination against them is pervasive in the field. Indeed, there are major concerns about ecological validity with the research about both implicit bias and stereotype threat, for reasons that should be obvious if you just take a minute to think about it. (If you are interested, the fact that most studies about implicit bias lack ecological validity is illustrated by this study about police decisions to shoot, which has a very nice design.) For instance, even if the research about implicit bias really showed that, in a laboratory, unconscious bias of the sort allegedly detected by the IAT was causally related to discriminatory behavior, it would say little about whether implicit bias results in discrimination against women in e. g. a concrete hiring decision, because there are so many potential factors that might counter implicit bias in a real situation. Similarly, even if stereotype threat were a thing, it doesn’t follow that it would have any practical import.

In fact, not only is there no non-anecdotal evidence that bias against women is pervasive in philosophy, but there is actually some non-anecdotal evidence that exactly the opposite is true. In particular, in recent years, Carolyn Dicey-Jenning and her team have collected a lot of data about placement. They did a multilevel logistic regression analysis which showed that, even controlling for area of specialization and year of graduation, women were a lot more likely than men to find a permanent position as opposed to a temporary one. Moreover, as I explained at the time, this analysis probably underestimates how much more likely than men to find a permanent position women are, because it doesn’t control for the number of publications and data previously collected by Carolyn Dicey-Jennings showed that men had on average significantly more publications than women. (In another analysis of updated data that was published a few months later, the effect of gender wasn’t significant, but as I and a few other people noted at the time, this isn’t surprising since none of the models were the same as the one used previously. Indeed, not only was multilevel regression not used, but even more importantly none of the models used for that analysis compares the odds of finding a permanent position as opposed to a temporary one. Given how strong the effect was, how small the p-value and how much overlap there was in the data, I have no doubt that, if you did the same analysis with the up to date dataset, you would find a significant effect. Indeed, in a later analysis of the data, the effect was replicated as I and others had predicted it would.) Now, I know that regression analysis not the same thing as causal inference, that we need to be careful when interpreting the results, etc. But it would be profoundly dishonest to deny that it’s evidence in favor of the hypothesis that, not only are women not discriminated against, but they even get some kind of preferential treatment. It would be profoundly dishonest, but if you read the comments, you will see that it prevents many people from doing just that.

What is really baffling with the fact that many people find the hypothesis that women get some kind of preferential treatment difficult to believe is that, given how much people in the field talk about the importance of increasing the proportion of women in philosophy, it shouldn’t surprise anyone. Maybe it’s just me, but when everyone around me talks about how important it is to have more women in the field, it doesn’t really surprise me when I find out that most of them are doing just that… (Before you accuse me of relying on anecdotal evidence myself, which is precisely what I criticized other people for doing, I invite you to read what I just said carefully. I didn’t use anecdotal evidence to establish the claim that women get some kind of preferential treatment in hiring. I showed that we had non-anecdotal evidence that women are favored in hiring and pointed out that, given some of the anecdotal evidence we have, this should not be surprising, which isn’t the same thing.) Of course, people often say one thing and do something completely different, but they do what they say even more often. A world in which the situation for women is as dire as that described by the official narrative is unlikely to be a world in which people constantly lament the underrepresentation of women in philosophy, create task forces to figure out ways to increase the proportion of women in the field, offer various resources to women in order to help them find a job, etc. I have never been to a Ku Klux Klan rally, but somehow I doubt that if I did, I would hear a lot of people talk about the urgency of advancing the well-being of black folks.

It’s interesting that, when you talk to people in private, almost everyone agrees that women get some kind of preferential treatment (even if some think it’s justified), but most of them would never say that in public. This is hardly surprising given that, every time you say that, there is always someone to accuse you of being angry at women, as I have experienced myself. But I’m not angry at women, I’m just not statistically illiterate, which despite what some people seem to think is not the same thing. This kind of defamatory, ad hominem attack is just a way of stifling debate. As I explain below, I personally don’t think such a preferential treatment is justified, but I also think that reasonable people can disagree about that. What reasonable people can’t disagree about is that, as far as we can tell based on the evidence we currently have, women do get some kind of preferential treatment. The truth is that, although I don’t think it’s fair, I don’t care that much about it. After all, if everything I think is unfair (no matter how insignificant) kept me up at night, I would never sleep. What bothers me is not so much that women get a preferential treatment, but the amount of bad faith I see every time the issue comes up in conversation. Everyone knows that, if the situation was reversed and the evidence showed that men are more likely to get a job, we’d never hear the end of it. This is why, against my better judgment, I always seem to get myself involved in long debates about that issue.

So even if you think that women should get some kind of preferential treatment, it would be nice if you stopped pretending that we don’t have good reasons to believe that they probably do, because so far that’s what the evidence indicates. It would also be even great if people stopped claiming that sexism is rampant in philosophy, because there doesn’t seem to be any non-anecdotal evidence to support that claim. It’s kind of infuriating when, despite all the evidence to the contrary, people not only deny that women get any kind of preferential treatment, but even insist that men do since women face ubiquitous discrimination in the profession. Indeed, this is adding insult to injury, which is really annoying even when the injury is not that big of a deal. If you step on my foot because you thought it was the only way to achieve some purpose which you think is desirable, I’m not going to make a fuss about it, even if I disagree with you that it was the only way to achieve that purpose and/or that it’s desirable. But if you step on my foot and tell me that I’m the one who stepped on your foot, it’s going to piss me off, which doesn’t strike me as extravagant.

If people want to keep saying that women are routinely discriminated against in philosophy, they should give some non-anecdotal evidence to support their view. Indeed, assertions that sexism is rampant in philosophy and that women face a lot of discrimination are not evidence, no matter how often they are repeated. Similarly, appeals to cherry-picked psychological studies about stereotype threat and implicit bias that at best have indirect bearing on the underrepresentation of women in philosophy are not evidence, it’s just magical thinking masquerading as empirically informed discussion, of the sort philosophers are unfortunately quite fond. In some cases, it would actually be pretty easy to test the claims people who defend the official narrative are making, but they don’t seem very interested in doing so. For instance, not so long ago, I had a conversation on Daily Nous with someone who, among many other unsubstantiated claims, asserted that women were less often invited to give talks than men. So I did a quick and dirty test of that claim and found that not only was it not the case, but actually women were significantly more often invited to give talks. Of course, the way I did that was pretty sloppy, but it would be relatively easy to test that hypothesis rigorously. I have little doubt that, if you did, you would confirm my finding, but somehow I don’t see a lot of people trying among those who promote the official narrative…

But let’s even suppose, just for the sake of the argument, that discrimination against women is pervasive in philosophy. What people don’t seem to realize is that, even if we had good, non-anecdotal evidence that women are being routinely discriminated against in philosophy, it would not be enough to explain their underrepresentation in the field. Indeed, this would explain why women who are interested in philosophy are less likely to enter the field relative to men who are similarly interested, but it wouldn’t explain why women are underrepresented in philosophy but not in other fields. The problem for that hypothesis is that, in many other fields, not only are women not underrepresented, but they are vastly overrepresented. This suggests that, even if sexism is part of the story behind the underrepresentation of women in philosophy, there are other, far more powerful factors. If you want to insist that bias explains more than a small part of the gap, you have to show that, whatever discrimination women face in philosophy, it’s much worse than in other fields such as psychology, which are dominated by women to the same extent that philosophy is dominated by men. Unless you can do that, you may have explained part of the gap, but only a small part since the factor you are relying upon was presumably just as much a factor in fields other than philosophy where there is no gap.

In fact, even if you could show that women face a lot more of discrimination in philosophy than in other fields where they are not underrepresented, it still wouldn’t be enough. After all, since women are overrepresented in psychology, it wouldn’t be surprising if they faced less discrimination than in philosophy, where they are underrepresented. But this wouldn’t explain why, after the restrictions that prevented women from entering academia were lifted, some fields quickly reach gender parity and some, such as psychology, even came to be dominated by women, whereas philosophy remained dominated by men. However, there is no reason to believe that, 50 years ago or so, sexism was more widespread in philosophy than in psychology, law, medicine, literature, etc. For instance, I have no doubt that women are far more likely to be affected by sexual harassment in philosophy, because men are presumably more likely to engage in that kind of behavior than women and they are overrepresented in philosophy. Thus, when it comes to sexual harassment, there is no doubt that men are privileged in the field. But there is also no reason to think that, when the restrictions that prevented the entry of women in academia were lifted (at which point most fields had the same, extremely low proportion of women), there was more sexual harassment in psychology, law, medicine, literature, etc. than in philosophy.

Thus, the official narrative can’t really explain why women are underrepresented in philosophy but not in many other fields, because we’d have to assume that philosophers have historically been a lot more sexist than e. g. physicians. But given the long history of sexism in the medical profession, this seems completely implausible to me. This is why I think that a much better explanation of the fact that women are underrepresented in philosophy is that, for whatever reason, men tend to be more interested in philosophy while the opposite is true for other disciplines such as psychology. Unlike the official narrative, the hypothesis that women have different preferences than men explains not only why they are underrepresented in philosophy, but also why they are not underrepresented and are even overrepresented in other fields. Indeed, it’s interesting that, in the case of psychology, nobody has any problem accepting the idea that men are underrepresented because they have different preferences than women. It’s only when women are underrepresented that people start protesting against this explanation. I get that, since women have historically been disadvantaged, the situation isn’t exactly symmetric, but this doesn’t justify that double standard.

Moreover, unlike the hypothesis that women face pervasive sexism in philosophy, the hypothesis that women tend to be less interested in philosophy than men has a lot of direct empirical support. For instance, a study from 2015 found some evidence that, even before their first philosophy course, women at the University of Sidney were already less interested in majoring than men. I also calculated that, according to a survey taken by more than 2.4 million students in the US between 2000 and 2005, among those who entered a four-year college with the intention of majoring in philosophy during that period, only 33.2% were female despite the fact that women made up more than 56% of the respondents. It’s true that, in 2012, another study had found that the proportion of women declined significantly between the population of philosophy introductory courses and the population of philosophy majors. Many people took that to be evidence that women are discouraged from majoring in philosophy after they experience a philosophy classroom, but since many if not most students take introductory philosophy courses to satisfy a requirement, this was clearly a mistake. Indeed, not only did that same study also found that men still outnumbered women in introductory philosophy courses, but it found no statistically significant decline in the proportion of women between the population of philosophy majors and the population of philosophy graduate students or between the population of philosophy graduate students and the population of philosophy faculty.

Yet another study based on a survey of 1,540 undergraduates at Georgia State University, which they filled at the end of their introduction to philosophy, showed that women found the course less interesting, were less interested in taking more philosophy courses or majoring in philosophy, thought philosophy was less relevant to their lives, etc. They said that at a time when they had only taken only one philosophy class, whose instructors were clearly interested in drawing women to the field, since they were conducting a study to understand why so few women were majoring in philosophy. Indeed, according to the survey, the women in the class felt the instructors, but also the men in the class, treated everyone with respect no matter their gender. In order to test whether the small number of women may have something to do with the fact that women showed less interest than men in philosophy, the instructors intervened by increasing the proportion of women on the syllabus after the first year, but it had no effect whatsoever. There is also nothing surprising about it, since it’s just a specific instance of the more general phenomenon that women have different occupational preferences than men, which is largely uncontroversial. Again, even among those who peddle the official narrative, few would dispute that women tend to be more interested in psychology than men. It’s only when you suggest that women might be less interested in philosophy than men that people start being uncomfortable with that kind of explanation.

Not only does the evidence I just discussed support the hypothesis that women are less interested in philosophy, but it’s also hard to reconcile with the official narrative, since it suggests that women are already less interested in philosophy before they have even taken any philosophy class. I understand that, for the people who promote that narrative, bias is a powerful factor, but they will have to forgive me if I continue to think that backward causation is not a thing… On the other hand, unlike the official narrative, the hypothesis that women are underrepresented because they tend to be less interested in philosophy explains the data extremely well. Indeed, as we have seen above, approximately 33.2% of the students about to enter college who declared the intention to major in philosophy between 2000 and 2005 were female. Now, according to the National Science Foundation (NSF), 27.9% of the people who received a PhD in philosophy between 2010 and 2015 were female. However, during that period, women only made up 46.3% of the people who received a PhD. When you account for that fact, based on the proportion of women among the students who declare the intention to major in philosophy when they enter college, you would only expect approximately 28.6% of the people who received a PhD in philosophy to be women, which is barely more than the actual proportion. Of course, by applying this correction, I may be in part accounting for bias against women that prevents them from getting a PhD, but if so this has nothing to do with philosophy in particular.

In fact, this is not just true of philosophy, it seems to be true across the various academic fields. I used the data from the CIRP Freshman Survey, which asks more than 400,000 students every year what they plan to study as they enter college, as well as the data from the NSF about PhD recipients, to see what correlation there was between the women/men ratio of incoming freshmen who declared their intention to major in a field between 2000 and 2005 and the proportion of women who received a PhD in that field between 2010 and 2015. There was data about dozens of fields/majors, but it wasn’t always possible to match the data about majors from the survey to the data about PhD recipients from the NSF and the preparation of the data was pretty time-consuming, so I only compiled them for 21 fields. They include philosophy, mathematics, physics, history, psychology, biology, etc. The criteria I used to choose which fields were going to be included in the analysis were how easy it was to match the data from the survey to the data from the NSF, what proportion of the incoming freshmen declare their intention to major in them and how prestigious they are. I may have made a mistake, so if you want to check my analysis, I have uploaded the data here. (It was created with Calc, so you need OpenOffice to open the spreadsheet.) If you want to include more fields/majors, you also have everything you need. 

As it turns out, the proportion of women among freshmen who declared the intention to major in a field upon entering college between 2000 and 2005 predicts the proportion of women who received a PhD in that field between 2010 and 2015 almost perfectly, as you can see on this graph.

The correlation between the proportion of women among freshmen who declared their intention of majoring in a field and the proportion of women among the people who received a PhD in that field was a whopping 0.963, which means that the former explained almost 93% of the variance in the latter. This is as strong a correlation as you are ever going to find in social scientific data. In fact, although I expected the correlation to be strong, even I didn’t think it would be that strong.

Thus, for the most part, the proportion of women in a field seems to be already determined by the time students enter college. Of course, this doesn’t explain why women are underrepresented among PhD recipients across the board, which could have something to do with bias or some other form of injustice against women. (My guess is that it has a lot to do with the fact that, in order to get a PhD, women often have to delay motherhood, which many are not willing to do. Universities could probably do more to accommodate women who want to have a child while in graduate school, but it’s not that easy and, in any case, this is more complicated than people taking the work of women less seriously or something like that.) But even if bias or some other kind of injustice explains this fact, it seems to operate more or less equally across fields. In particular, you can see on the graph that philosophy is not an outlier, on the contrary. This should put to rest the narrative that philosophy is uniquely bad for women and that sexism discourage them to pursue a PhD in philosophy in a way it does not in other fields. According to the data I analyzed, insofar as sexism discourages some women from pursuing a PhD in philosophy, it does not seem to do so any more than in other fields. Moreover, given how much variation in the proportion of women there is between fields, even if bias is a factor in the underrepresentation of women in academia as a whole (which is not clear), it’s presumably not very powerful compared to the factors that explain variation in the proportion of women between fields.

Another possible explanation of the underrepresentation of women in philosophy, which is not incompatible with the hypothesis that women are less interested in philosophy, is that men are more likely to have the kind of cognitive abilities that are required to succeed in philosophy. Although it’s often summarily dismissed for what I think are ideological reasons, I don’t think we can rule out this hypothesis, but I also think that we are not currently in a position to know whether sex differences in cognitive abilities play a role in the underrepresentation of women in philosophy and, if so, how much of the gap it explains.  Sex differences in cognitive abilities are a complicated issue that would require a lengthy discussion, but it’s better left to another post specifically on that question, so I will just make a few brief remarks here. (If you want to read more about that question, you can read this great, balanced review of the literature by Halpern et al. from 2007.) There is a lot of evidence that there are sex differences in cognitive abilities, although it’s not always conclusive. For instance, it seems that men have higher visuospatial abilities, but lower verbal abilities. Moreover, for some cognitive abilities, there seems to be more variability among men. It’s less clear whether these differences are genetic or result from environmental factors, although the evidence suggests that genetic and environmental factors interact in complicated ways to produce them.

But it’s unclear how much, if at all, sex differences in cognitive abilities bear on the underrepresentation of women in philosophy. First, we don’t know what specific cognitive abilities are correlated with philosophical aptitude, but as we have seen differences between the sexes vary a lot across different cognitive abilities. Moreover, even when there are differences between the sexes in cognitive abilities, they are often relatively small. Now, even small differences can result in large men/women disparities, but only at the tails of the distribution of cognitive ability. However, not only is it not clear what cognitive abilities are important to be successful in philosophy, but it’s also not clear which region of the distribution of these cognitive abilities philosophers typically occupy. To be clear, I do not doubt that philosophers are smarter than people in most other fields, nor do I deny that many philosophers are superiorly intelligent. For instance, I read John Hawthorne’s paper on Benardete’s paradox a few months ago and it left me no doubt about the fact that he is extremely smart, but I don’t really know how high philosophers typically score on the cognitive abilities that are associated with success in philosophy. (Indeed, as I already noted, we don’t even know exactly what cognitive abilities are associated with success in philosophy.) Thus, even if men are overrepresented among people who have very high cognitive abilities of the kind required to succeed in philosophy, it may not matter that much, although this could also be part of the story behind the underrepresentation of women in philosophy. Finally, as we have seen above, the fact that women are already less interested in philosophy as they enter college can already explain pretty much the whole gap. This suggests that, even if differences in ability contribute to explain the gap (which is hard to know at this point), their effect is probably mediated by differences in preferences. Indeed, as Halpern et al. explain, even among individuals who are academically gifted across the board, people tend the prefer the fields in which they are strongest. So I think it’s reasonable to assume that, even if men are more likely than women to have the kind of cognitive abilities that are required to succeed in philosophy (which again is pure speculation at this point), the fact that women have different preferences than men explains why they are underrepresented in the field.

Now, even if you accept this explanation of the underrepresentation of women in philosophy, you may worry that the fact that women have different preferences from men is not innocent. It’s certainly true that, in general, this fact is probably not innocent. It seems likely that, if women tend to prefer low status occupations, it’s at least in part because of the way they are socialized, which probably reflects some illegitimate gender-based discrimination. But it doesn’t follow that every difference in occupational preferences between men and women reflects some kind of injustice, even if it’s partly a product of cultural factors. In particular, it doesn’t mean that, if men are more likely to be interested in philosophy than women, we should worry about it. Indeed, while I can see why someone might worry that women are being socialized to prefer low status occupations, I really don’t see why we should worry if women are being socialized to prefer psychology to philosophy. Of course, this could be a byproduct of the factors that make women prefer low status occupations, but this is hardly obvious. Moreover, even if this were the case, philosophy departments would probably not be the place where change needs to happen.

Leslie et al. published a study in 2015 which found that women were underrepresented in fields that were perceived as requiring innate talent. Since philosophy was among the fields in question, this result was taken to show that women may be discouraged from entering the field by the perception that it requires some kind of innate talent. If this were the case, it might be a reason to worry about the fact that women are less interested in philosophy than men, for it could be that women are socialized to believe that they don’t have the necessary innate talent. The problem is that, as Scott Alexander explained on Slate Star Codex, the study failed to account for the possibility that the effect of perceived required ability was mediated by actual required ability. As it happens, when you perform the same kind of analysis as Leslie et al. but control for the quantitative score on the GRE, the effect of perceived required ability disappears. Now, you may not like the idea of using the quantitative score on the GRE as a proxy of innate ability (whatever this means exactly), but the fact is that it predicts the share of women in fields of study extremely well and, in any case, it’s clear that Leslie et al. should have accounted for the possibility that people’s beliefs about required ability may be accurate. This isn’t specific to Leslie et al., but is actually a widespread problem in the literature about bias and self-fulfilling prophecies, because studies in that literature rarely try to measure the accuracy of people’s beliefs and stereotypes. As Lee Jussim argued in a book published in 2012, which I have recently started to read, this omission has resulted in a completely distorted picture of human cognition in the psychological literature.

Another thing people often say is that, even if women are underrepresented in the field because they are less interested in philosophy than men, we should worry about it and try to change this because it means that we are losing out on a lot of potential talent that would improve the quality of the field. This argument has obvious problems, but it’s almost never challenged, because people are afraid to say what they think. Indeed, this line of reasoning implicitly assumes that we could change the preferences of women, without changing the field in a way that would harm its quality. But it seems rather implausible that we could change the preferences of women without changing the field in a way most philosophers would find undesirable. To the extent that women are less interested in philosophy than men, it’s presumably the result of several factors: natural differences between the sexes, social forces operating in society at large, the nature of the subject-matter and the culture of the field. Since philosophy departments have presumably little to no influence over the natural differences between the sexes and the social forces operating in society at large,  they can only increase the interest for philosophy among women by redefining the subject-matter and/or changing the culture of the field. It’s probably not easy to change the culture of the field without also redefining the subject-matter, because to a certain extent, the difference between the culture of a field and the nature of its subject-matter is artificial or, at least, they are causally related.

I know that many people think we could also increase the proportion of women in the field without radically changing its boundaries, but the notion that departments of philosophy can adopt policies that change the culture of the field in a way that would affect the preferences of women toward philosophy without also redefining the subject-matter in a significant way strikes me as extravagant. As we have seen above, when people tried to attract more women by increasing the number of women on the syllabus, it had absolutely no effect. Merely increasing the proportion of women in departments of philosophy might attract more women to the field, but it’s unlikely to have a large effect, since otherwise psychology, which started with the same proportion of women as philosophy, would probably not have become female-dominated while philosophy remained male-dominated. Similarly, given how enormous the difference between the gender ratio of philosophy and that of psychology is, the notion that what is sometimes described as the adversarial culture of philosophy could explain a large part of it, which is often floated, strikes me as magical thinking. (Beside, it neglects the fact that, as Joseph Heath argued, there is a lot to be said in defense of the adversarial culture of philosophy.) If philosophers really want to attract significantly more women, they have no choice but to change the nature of the subject-matter, i. e. what counts as philosophy.

In practice, if they refuse to do that but nevertheless insist on increasing the proportion of women in the field, departments of philosophy will just engage in some kind of affirmative action. Indeed, as we have seen above, there is quite a lot of evidence that it’s already the case. But if the reason why women are underrepresented in philosophy is not that they are discriminated against, but that they have different preferences than men, this will probably lower the average quality of the field, which is precisely the opposite of what this policy was intended to do. Indeed, the idea was that we should try to increase the proportion of women in philosophy because, at the moment, we are not tapping into a large pool of talent. But if women are underrepresented in philosophy because they have different preferences than men, then insofar as affirmative action policies don’t affect women’s preferences, we’ll just tap in the same pool but go deeper in it and recruit less talented people than we would have in the absence of such policies. This is true even if, as is no doubt the case, hiring decisions don’t perfectly track philosophical talent even when there is no affirmative action.

On the other hand, if we really try to increase the proportion of women by making them more interested in philosophy, then for the reasons I gave above, I worry that we’ll just end up changing philosophy rather than women. In particular, I worry that, in order to make philosophy more “women-friendly”, we’re going to make it less like philosophy and more like the rest of the humanities or fields such as gender studies, critical race theory, etc. But I don’t want that to happen, because I quite like the way philosophy is currently done and, to be perfectly honest, I have a rather dim view of the value of these other fields. It’s not that I think everything is perfect about the way philosophy is done and that we couldn’t improve some things. But I doubt the changes that people who promote the official narrative have in mind would improve the field and, on the contrary, I’m quite sure they would have the opposite effect. If I liked meaningless jargon and poorly argued, empirically uninformed nonsense, I would have pursued a PhD in another field and perhaps I would have written a dissertation about why the fact that white people make ethnic food is a symptom of white supremacy. But instead I chose a field in which people have traditionally avoided that kind of bullshit and I would like it to stay that way.

EDIT: I have made substantial changes to the original version of this post, especially in the last part, which I had kind of botched and which I think is now much better. Among other things, I added a brief discussion of Leslie et al.’s paper, though you should read Alexander’s post for a more thorough treatment.

ANOTHER EDIT: After someone raised that issue upon reading my post, I added a few words about sexual harassment, which I should probably have done in the original version of this post. I also added something about why I don’t think anecdotal evidence is very useful.

YET ANOTHER EDIT: I have added a regression analysis which shows that, when you use data from a large-scale survey to account for the fact that women are already less interested in philosophy than men by the time they enter college, you can explain pretty much the whole gap.

ONE MORE EDIT: Someone pointed out in the comments that the independent variable I had originally picked for the regression did not make sense, so I updated the analysis and the post to fix that issue. I guess that will teach me not to accuse other people of being statistically illiterate… Anyway, when I use a more sensible independent variable for the regression, the correlation actually goes up and some fields which used to look like outliers no longer do. Other than that, it doesn’t change anything, for reasons that should be obvious.

ONE LAST EDIT: I was able to get the data from the CIRP Freshman Survey, so I didn’t have to rely on what the Higher Education Research Institute publishes anymore, which is a good thing because they round everything to one decimal places in their publications and this means that my calculations sometimes lacked precision. (I would like to thank Ellen Stolzenberg, Assistant Director of the CIRP, who helped me get access to the data. Naturally, she can’t be held responsible for any of the views I express in this post, let alone for any mistake it may contain.) It didn’t really change anything to the result, but now we know it’s reliable.

33 thoughts

  1. A traditional history of philosophy, say, Russell’s History of Western Philosophy, only includes males, so it just may be that women have less interest in studying what males have to say about the problems philosophers talk about.

    It would be interesting to see if in another field, say, literature, women specialize in the 16th and 17th century, when all known authors were male or tend to specialize in the 19th and/or 20th century, for example, from Jane Austen on (in English literature) when there are certain number of recognized female authors.

    1. S. Wallerstein,

      I don’t think you want to push that line of argument. You will end up having to defend the claim that women and men have substantially different ways of thinking, and that they each much prefer their own gender-specific way of thinking.

    2. Funny how men never consider that male writers are just mockingbirds for women of their time, and that might be the reason why women find male philosophy uninteresting, because it is old hat method. Oral history or gossip is a tradition that has kept women alive. Women relate to the emotional side of philosophy, not because we are more emotional, but because we are more evolved in creating a language around it. In my experience, unless a man is deeply practiced in identifying emotions, men want to explore what they already know and which cannot challenge them, or steal ideas from women that they feel they access enough mentally and manipulate the situation as theirs. Hence this blog.

  2. Thank you, we needed that. Philosophers should get back to our roots of asking uncomfortable questions (and considering unconventional answers) or get out of the profession and do some honest work for a living. We are swimming in a sea of motivated reasoning in moral and political philosophy, and if things don’t change soon we can only hope someone will come along and hasten our death by ripping off our water wings and holding our heads under water.

  3. Even if women had less natural aptitude for philosophy, it seems that explanation would be swamped by their lack of interest. After all, as you demonstrate, by the time they are first year undergraduates women already have less interest in philosophy than men, and this can account for most of their under representation. So it is not as though women are trying philosophy and getting forced out because they are not as good at it. Perhaps women are less interested because they have less aptitude – but that seems a different thing to some degree

    1. Yes, I completely agree with you and was actually making that point already, but I realized after reading your comment that it wasn’t clear so I made a change in the relevant paragraph. (The original version of that paragraph, where I made the point but not very clearly, had been written before I added the regression analysis, so at the time I didn’t know that preferences explained the gender ratio in philosophy so well.)

  4. Are there any studies on the effects of including philosophy in primary/secondary curricula for representation of women in philosophy at the third level? I believe that philosophy is part of the secondary school curriculum in France. Do we find a similar under presentation at the third level there? This is not to say that better representation would be explained by the inclusion of philosophy in the secondary school curriculum. But if we want to decide why women are less interested in pursuing philosophy at the age of 17-18, we need to identify what exposure they have had to philosophy by that age. Any? The same as their male peers? I don’t know, for instance, whether all boy schools are more or less likely to include philosophy in their curriculum than all girl schools. It wasn’t that long ago when girls were expected to study home economics in British schools while boys were expected to study different subject. I don’t know whether any schools in the US do include philosophy in their curriculum, and, if any do, which types. A comparison to France, if possible, would be helpful here given that philosophy is included in the secondary school curriculum there.
    Any info? (might be too big a request while you are working on your dissertation)

    1. Actually, I already looked this up a while ago, so it’s no work at all! According to this article, which gives the figures for 2012, women made up only 20% of professors of philosophy at university in France, which apparently is the lowest in any non-scientific field. So either the inclusion of philosophy in the high school curriculum has no effect on the representation of women or there are other factors which cancel that effect.

      1. Thanks! I found some relevant data on Ireland where philosophy was actively excluded from the secondary school curriculum for years. For 2012, women received a little over half of all higher level degrees, diplomas, and certificates (all disciplines). They received 47% of third level philosophy awards that year. Most study philosophy in Ireland by pursuing a mixed humanities degrees. Women received about 68% of these, but that means little given that I don’t know what subjects they pursued as part of their mixed degree. Also interesting is that a higher percentage of women received awards in philosophy than were enrolled in philosophy programs. More years are needed to draw any conclusions from that, but women in Ireland who start a program in philosophy might be more likely to complete their studies than their male peers.

        The Irish data also shows that 35% of postgraduate awards in philosophy in 2012 went to women. I’d need to look at all the data for more years, but if this held up, then the percentage of women earning undergraduate degrees in philosophy is significantly lower than those who earn postgraduate degrees in philosophy.

        What’s interesting is that the percentage of women earning any type of post grad award in any field in 2012 was 58%. The percentage of women earning an undergrad award in the same year was 52%. Focusing just on humanities and arts, the percentage of women earning undergrad and postgrad is about the same (58%). So it’s possible that Ireland has a very specific problem about representation of women in philosophy that is not part of a larger problem of representation of women in the academy.

        I didn’t dig in further, but I wonder whether this kind of detailed data would allow us test some proposed hypotheses. For instance, if women who are enrolling in philosophy programs are more likely to graduate, that would suggest their aptitude is equal to, if not greater than, their male peers for philosophy. Other explanations are possible of course.

        The French and Irish data also raise the question whether exposure to philosophy in secondary education might be less rather than more likely to improve the representation of women pursuing philosophy at third level.

        The data for 2012 is here: http://www.hea.ie/en/statistics/2012-13

        They government posts detailed data for other years too. Given that the data is so complete–it includes all fields, levels (degree, diploma, etc), full time, part time, enrollments, awards, degree level earned (honor, ordinary, etc)– I think someone with the right statistical skills, which I don’t have, might be able to learn something interesting from it.

        1. Thanks, I really don’t have time to look through this right now, but hopefully someone else will. I was actually a bit surprised by the low proportion of women among French professors of philosophy, because departments of philosophy in France are heavily skewed toward history of philosophy, which I think has more women in the US. They are also predominantly continental, which may suggest that whatever makes women less interested in philosophy, it transcends the analytic/continental divide. But I would like to see more data from other countries before I draw this conclusion, because I’m a bit skeptical.

      2. Have you also checked how many women enter the undergraduate programs in philosophy in France? And how many pursue a PhD? The small ratio of women professors does not say much about how many women wanted to study philosophy at the beginning (in your graph above you compare freshmen intentions to how many pursue a PhD — and not to how many enter the profession as professors). (Another country to look at is Italy that has 3 years of philosophy taught in high school).

  5. Regarding your regression analysis and your plot: Why did you plot w/(w+m) for awarded PhDs against w/m for declared interest (purely formally: probability versus odds)? I haven’t checked your spreadsheet in detail, but it seems to me that if you plot w/(w+m) for awarded PhDs against w/(w+m) for declared interest, the correlation goes up to 0.959.

    1. I did that for no good reason at all, I just wasn’t thinking, but you’re right that it doesn’t really make sense. (In my defense, when I ran that regression, it was 3 in the morning ^_^) I updated the spreadsheet and the graph to correct that oversight. I also find that, when you do that, the correlation goes up to 0.959. Otherwise things are essentially the same, which is unsurprising, except that some fields which used to look like outliers no longer do. Thanks for catching this.

  6. “Not only does the evidence I just discussed support the hypothesis that women are less interested in philosophy, but it’s also hard to reconcile with the official narrative, since it suggests that women are already less interested in philosophy before they have even taken any philosophy class. I understand that, for the people who promote that narrative, bias is a powerful factor, but they will have to forgive me if I continue to think that backward causation is not a thing.”

    This is disingenuous. Presumably in first year women make their course choices based in part on prior personal interests and the “reputation” of certain courses of study (based on reported the experience of friends, relatives, info online, etc). If first year women avoid phil in part because it has a reputation for being unwelcoming to women, then the lack of enrolment would be partially explained by the sort of factor’s identified by the narrative you are criticising. No commitment to backwards causation is necessary.

    1. I was being a bit cheeky in this passage, but you are right that it’s a possibility. That being said, I doubt that it’s the case or at least that it has a large effect, because I doubt that 18-year-old students have any view on how welcoming or unwelcoming for women philosophy is. But suppose that I’m wrong. In that case, it would be all the more urgent to push back against the narrative that philosophy is uniquely bad for women, because it’s not warranted by the evidence and it might prevent women from entering the field.

  7. This is a very interesting post – thank you for taking the time to write it. I’m very sympathetic to the worry that narratives can relatively easily become established with relatively little evidence, and can then be very difficult to displace: this seems like a straightforwardly epistemically bad way for things to be.

    A few thoughts

    (i) You mention Scott Alexander’s response to Leslie. I would be more wary than you of repeating his conclusion that ‘the effect of perceived required ability was mediated by actual required ability’, for various reasons. But the main one is this: it seems like he is happy to treat GRE *Quantitative* scores as measuring innate ability. First, it is genuinely unclear to me – as it seems to be to you too – what sorts of cognitive ability make for good philosophy. In part this is because different bits of philosophy seem to require very *different* cognitive skills: to be good at logic, you need to be good at different kinds of stuff than you need to be good at ethics. And in part I think it is because people who are *very* good philosophers tend to have an unusual combination of cognitive talents: they are good at both quantitive and verbal reasoning. So it simply is not clear to me that it is legitimate to take GRE Quantitive scores as an indication of ability within philosophy, let alone as an indication of academic ability per se. Even if this is right though, we might think Alexander has shown something very interesting: that quantitive ability predicts the number of women in a field. But unless you are antecedently committed to the idea that good qualitative scores are a better indication of philosophical ability than good overall GRE scores – and I don’t think any of us *should* be antecedently committed to that – it Alexander seriously undermotivates his sloganeering conclusion. (I say this having learnt a lot from his piece when I first read it.)

    (ii) I worry about this inference:

    ‘But if the reason why women are underrepresented in philosophy is not that they are discriminated against, but that they have different preferences than men, this will probably lower the average quality of the field, which is precisely the opposite of what this policy was intended to do.’

    Part of my worry is that it just is not clear to me that someone’s being less interested in X makes them less good at X. People can be really good at things that don’t interest them enormously – I’m sure most accountants think accounting is basically pretty boring, and I have lots of students who are very enthusiastic but absolutely hopeless at philosophy. I mean, it *could* be true that being interested in X is a good predictor for being good at X, but I’d want to see some data. A second worry I have is this. You note that women are underrepresented as PhD recipients across the board, and note:

    My guess is that it has a lot to do with the fact that, in order to get a PhD, women often have to delay motherhood, which many are not willing to do. Universities could probably do more to accommodate women who want to have a child while in graduate school, but it’s not that easy and, in any case, this is more complicated than people taking the work of women less seriously or something like that.

    I too think this is probably a big part of it (though I also think aversion to financial risk plays a large role – there’s plenty of evidence that women are more risk averse than me). The trouble is, if the preferences women have which stop them from pursuing graduate work are (i) financial risk aversion or (ii) desire to have children, it seems totally unclear to me that having more women in the field would mean more people in the field with the sorts of preferences one would expect to lower philosophical quality.

    Again – many thanks for an interesting and informative post.

    1. Regarding (ii), you seem to be misreading the inference. The point is not that the women added by affirmative action in philosophy are less talented because *they* are less interested than men who might be hired instead. One can assume that everyone seeking employment as an academic philosopher is equally interested and that the distribution of talent among interested female philosophers is the same as the distribution of talent among interested male philosophers. The point is just that if women are less interested, then the pool of interested female philosophers is smaller, so if one draws from that pool disproportionately, then the field overall will be less talented than if one hired purely on the basis of talent.

    2. Thanks for this thoughtful comment. I briefly reply to the points you make.

      (i) I think it’s a mistake to focus on the case of philosophy when discussing Leslie et al.’s paper. There seems to be a widespread belief that philosophy requires some kind of innate genius. Leslie et al.’s hypothesis is that this belief explains in part why there are relatively few women in philosophy. But the fact that philosophy is both a field where there are relatively few women and a discipline that is believed to require some kind of innate genius, on its own, offers essentially no support to that hypothesis, because many other things could explain why there are so few women in philosophy. This is why Leslie et al.’s regression analysis is important, i. e. because it’s supposed to show that the degree to which a field is believed to require some kind of innate genius explains the proportion of women in that field across the board. If this were true, it would make it plausible that, in the case of philosophy in particular, the fact that people believe it requires some kind of innate genius explains the underrepresentation of women. But Alexander’s results undermine Leslie et al.’s interpretation of their regression analysis, namely that the extent to which a field is believed to require some kind of innate ability is causally related to the proportion of women in that field, which in turn undermines their explanation as it’s supposed to apply to philosophy in particular. You may doubt that what is measured by the quantitative section of the GRE is correlated to whatever people have in mind when they think about innate genius, but as I point out in my post, the fact is that it explains the data very well, much better than perceived required ability. It could still be that, if women are underrepresented in philosophy, it’s in part because people believe that it requires some kind of innate genius and this keeps women away from the field, but Leslie et al.’s data don’t really offer much support for that claim.

      (ii) I think you are right that differences between men and women in risk aversion, which are well documented, may also play a role. As for the rest, Greg already replied on my behalf. The inference I was making in this passage is exactly what he explains in his comment.

    3. “I’m sure most accountants think accounting is basically pretty boring”–I don’t know why you think that. I’ve never been an accountant, but I used to be a tax lawyer, which is close, and most tax lawyers don’t think tax law is boring. (I eventually decided that being a real estate lawyer would be more fun, but that may have been fairly idiosyncratic.)

  8. This is an excellent essay. You have balls too, publishing it under your real name. Thank you. However, I don’t like this part:

    “If you step on my foot because you thought it was the only way to achieve some purpose which you think is desirable, I’m not going to make a fuss about it, even if I disagree with you that it was the only way to achieve that purpose and/or that it’s desirable. But if you step on my foot and tell me that I’m the one who stepped on your foot, it’s going to piss me off, which doesn’t strike me as extravagant.”

    The analogy is bad. It’s not like someone stepping on your foot. More like someone stealing half your life savings after you worked very hard for that money. Men who are highly qualified and dedicated and accomplished, and who were never properly informed of the extreme systemic bias working against them when they started their educations, are being deprived of work and opportunities as a result of these policies. I’ve been told by people in the profession, more than once, that some TT position is really open only to women, regardless of who may apply. This is never explicitly stated, of course, but that’s how it seems to work at least sometimes. And this is a very serious injustice, on many levels. Quite apart from the fact that people are struggling to find a job and putting time and effort into applications that, unbeknownst to them, will never even be given serious consideration, it’s a serious injustice that an entire class of people are missing out on jobs and promotions simply because of their sex. Whole families–including women, of course, when those are wives or daughters dependent on male philosophers–have lives that are far more impoverished, stressful and disappointing than they could have been if universities didn’t have these policies. And students are being short-changed also, though this is probably not such a deal. Moreover, it’s going to be a lot harder for men who are excluded to find mates and start families given that, rightly or wrongly, women just aren’t attracted to men who are lower in status or less independent than themselves. (And that makes things worse for women too, since the pool of attractive men is much smaller than it should be.)

    And it’s made worse, ironically, by “sexist” social norms. Men aren’t supposed to show weakness or complain. So the vast majority of victims won’t admit how crushing and demoralizing it is for them to be passed over in favor of less capable or accomplished women, and they won’t want to talk about how these policies have blighted their lives and the lives of their wives and children. They’re not going to get together and start some self-pitying thing called “What it’s like to be a (non-elite) man in philosophy”.

    Anyway, great post but I wish you hadn’t been so cavalier about misery and poverty and dysfunction that these policies generate. They’re evil policies, and I hope the people responsible for them will someday be held accountable for all that needless suffering and wasted potential.

    1. As I explain in my post, I agree that it’s unfair, but I also think that you overestimate how bad it is. People with a PhD, even if they can’t find employment in academia, rarely end up in poverty, because they typically have other options even if they don’t realize it. I may not find a job in philosophy, especially after writing this post, but if I don’t, I’ll just do something else. It’s unfair but I don’t think it’s the end of the world and, in any cases, there are certainly many far worse injustices in the world.

      1. There are worse things than most things. But what we’re talking about here is a lot worse than someone stepping on your foot. The injustice is not (or not only) that people can’t find jobs. It’s that they won’t be considered for jobs when they do at least deserve to be considered, when they have invested years of work in the hope of being considered, and when the profession is actively deceiving them about their very poor prospects.

        Add to this that many of them _will_ find jobs, as adjuncts, so they’ll be doing the real work for half the pay (or whatever) and with no job security in order that less capable women can enjoy one of the most privileged positions in our society.

        They’ll have been deceived, and wasted prime years when others are saving money and building a life, and then the system that deceived them will exploit their cheap labor. (I wonder whether you’ll be as cavalier about it if you find yourself in that situation for a few years.) And _then_ they’ll have to listen to those same women whining about how they’ve been treated.

        It’s not like someone stepping on your foot. Not even close. But thanks again for writing this.

        1. My brother-in-law did several years of graduate study in philosophy at a very elite U.S. university. When my sister became pregnant, he realized that it was going to be hard to support a family as a philosopher, and using the same skills that made him a promising philosopher, got a job as a computer programmer. He makes a good living, probably more than he’d make as a philosopher and studies philosophy in his leisure time (he’s now on Kant).

          No harm done really. I think that my brother-in-law and many others study philosophy because of the joy of doing it and maybe even because they seek wisdom, and it is possible to continue doing philosophy and seeking wisdom even if one spends one’s days programming computers (which is far from the worst job in the world).

        2. My use of that analogy doesn’t imply that I think the fact that women get preferential treatment is no worse than someone stepping on my foot. I think it’s worse, but I clearly don’t think it’s as bad as you do. I think people in academia need to understand that they can flourish outside of it. I don’t mean to sound callous, but if people who have a PhD are exploited, it’s because they accept it. Nobody is forcing them to work as adjuncts under very poor work conditions. They could easily find another, more rewarding career. Moreover, as s.wallerstein pointed out, they could also continue to do philosophy in their spare time. I know that, if I leave academia, that’s what I will do. I’m not denying that a lot of people who adjunct are very unhappy, and I really sympathize with them, but I think it’s largely self-inflicted. If more people who can’t find a permanent position in academia realized that they could do something else, there would be less miserable people who are exploited by working as adjuncts. Not only would more of them just find another career, but as a result, universities would have to offer better work conditions to adjuncts in order to attract or retain them. Again, I still think it’s unjust that women get preferential treatment in philosophy, but I also think that, once you realize what I have just said, it doesn’t look as bad.

  9. Yes, very unfair indeed – women pretending to do jobs their are not predisposed for and taking them away from us white men who have worked so hard for them.

  10. Fascinating graph.

    Quick question though…I could be reading the Freshmen Survey incorrectly, but don’t think their % numbers by gender account for the higher number of women enrolling in college. It looks like their sample for the years you used was about 55% women and 45% men, so the percentages would have to be weighted accordingly to get the % of women among interested freshmen.

    Running a quick and dirty version with a blanket 55/45 weighting, it doesn’t look like it changes your correlation value (I got .957) it just shifts the % freshmen values up by 2-5% depending on the field.

    Thanks for putting together the number though, interesting comparison!

    1. Yes, that’s a very good point, I completely forgot about this when I created the spreadsheet. I’m working on getting the full dataset from the Freshman Survey, so I have something more precise for my analysis than what is published by the Higher Education Research Institute, which rounds everything to one decimal place. This will take care of the problem and will make the results more robust, so as soon as I have the data, I will update this post.

      EDIT: I have now updated the post with the data from the Freshman Survey and, as you can see, it doesn’t really change anything.

  11. My mother graduated in Philosophy but then went to study further and become a prof in Aesthetics. It makes sense to me

Leave a Reply