UPDATE: A new piece at the Heterodox Academy by Sean Stevens and Jon Haidt, ‘The Google Memo: what does it say about gender differences?“, examines Damore’s claims about sex differences in ability and preferences by reviewing a great deal of the relevant literature. Some of it supports his claims; other bits don’t. Their general conclusions are below, but you should read their piece:
3) OUR CONCLUSIONS
The research findings are complicated, as you can see from the many abstracts containing both red and green text, and from the presence on both sides of the debate of some of the top researchers in psychology. Nonetheless, we think that the situation can be greatly clarified by distinguishing abilities from interests. We think the following three statements are supported by the research reviewed above:
1. Gender differences in math/science ability, achievement, and performance are small or nil.* (See especially the studies by Hyde; see also this review paper by Spelke, 2005). The one exception to this statement seems to be spatial abilities, such as the ability to rotate 3-dimensional objects in one’s mind. This ability may be relevant in some areas of engineering, but it’s not clear why it would matter for coding. Thus, the large gender gap in coding (and in tech in general) cannot be explained as resulting to any substantial degree from differences in ability between men and women.
2. Gender differences in interest and enjoyment of math, coding, and highly “systemizing” activities are large. The difference on traits related to preferences for “people vs. things” is found consistently and is very large, with some effect sizes exceeding 1.0. (See especially the meta-analyses by Su and her colleagues, and also see this review paper by Ceci & Williams, 2015).
3. Culture and context matter, in complicated ways. Some gender differences have decreased over time as women have achieved greater equality, showing that these differences are responsive to changes in culture and environment. But the cross-national findings sometimes show “paradoxical” effects: progress toward gender equality in rights and opportunities sometimes leads to larger gender differences in some traits and career choices. Nonetheless, it seems that actions taken today by parents, teachers, politicians, and designers of tech products may increase the likelihood that girls will grow up to pursue careers in tech, and this is true whether or not biology plays a role in producing any particular population difference. (See this review paper by Eagly and Wood, 2013).
Our verdict on Damore’s memo: Damore is correct that there are “population level differences in distributions” of traits that are likely to be relevant for understanding gender gaps at Google. Even if we set aside all questions about the origins of these differences, the fact remains that there are gender differences in a variety of traits, and especially in interest/enjoyment (rather than ability) in the adult population from which Google and all other tech firms recruit.
They’ll have a followup post on the memo’s claims about the value of viewpoint diversity.
I’ve reread the infamous Google memo by James Damore, and my opinion is about the same: it’s a mixed bag insofar as it makes some weak biological/evolutionary claims about male versus female preferences, and it could have used some citations (but of course there’s lots of literature to cherry-pick, and that would have made it into a paper, not a memo). Damore seems to take observed sex differences in behavioral traits like “neuroticism” to argue, implicitly or explicitly, that differences in psychology or ability are biological differences instilled in our ancestors by natural selection. He doesn’t consider that some part of these differences, or even the bulk of them, could be cultural—due to socialization and biases—and therefore should not be taken as “evolutionarily hardwired”. And even “evolutionary hardwired” differences can be susceptible to cultural change. Further, Damore’s argument that these differences are “universal and therefore genetic” is not only a priori illogical (nearly everybody in the world is religious, but does that mean we have a gene for it?), but I even doubt that every society has been surveyed to show the universality of sex differences in psychology, preference, or ability.
That said, I think the memo makes points worth considering, has been grossly misrepresented by people who attacked it, and likely led to Damore’s firing simply because he violated the Regressive Leftist dictum that there are no biological, or even existing, psychological differences between men and women, and therefore differential representation must be due to sexist bias leading to failure to hire or promote. But, as I’ve said, there are three reasons for disproportionate representation: different abilities, different interests, and unequal opportunity (sexism, cultural expectations, socialization, etc.). The simple observation of gender disproportionality doesn’t single out only one of these factors as responsible, especially given we know that there are, on average, psychological differences (whatever their cause) between men and women that could lead to difference for preferences in what they want to do.
What Damore was trying to say, I think, was not to indict women as having lower abilities in tech professions, but to suggest that the nature or culture of tech is such that it leads women to be less interested in the field. Nor did he deny the existence of sexism. He was making points for discussion that were considered taboo, and therefore he was fired. After all, he did say that he favored diversity, but suggested that present ways of achieving it may be less than optimal if it reflects more than sexism. And he suggested ways to increase diversity. At the very least, Google’s employees could have discussed his ideas, some of which have considerable merit (see this article by Scott Alexander for a reasoned discussion of differential average preferences). But Google was subject to an anti-Damore campaign, their managers seem amenable to ideas of the Regressive Left, and so they let Damore go. Damore was shamed and fired simply for expressing an opinion worthy of consideration. This is the demonization by the Regressive Left that we’ve come to recognize.
Yesterday I posted about a Slate article written by a woman physicist who thought that Damore’s argument showed that the very structure of science was unreliable, and played into right-wing tropes, so that science itself was responsible for discrimination against minorities. You would have expected such a piece in the truly regressive Salon, but not in Slate, and not by a scientist! Well, a Salon article appeared on Tuesday, and it basically makes the same claim: the Google document cites pseudoscience that justifies sexism, and is one more bad example of “biological determinism” used to buttress the status quo. (Never mind that Damore says he favors gender diversity in the workplace, but thinks there are better ways to achieve it—ways that recognize differences in psychology between men and women. Note again that he does not say that women are less able than men to do the work at Google.)
The Salon article is “The ugly pseudoscientific history behind that sexist Google manifesto“, and is by Keith A. Spencer, a Salon writer whose scientific training appears to be a B.A. in astrophysics/English at Oberlin (double major) and then subsequent work in the humanities and writing ever since (he also has a master’s degree in literary and cultural studies from Carnegie Mellon). Although I’m not a credentials monger, perhaps Spencer’s lack of biological training is shown in the way he refutes Damore’s “pseudoscience”: his refutation relies on a single book published in 1984: Not in Our Genes, by Richard Lewontin, Steven Rose, and Leon Kamin (henceforth LRK). I am well familiar with that book, as the first author was my Ph.D. supervisor, and I have to note two things. First, The book not a dispassionate review of the literature: the authors wrote it because they were committed to dispelling biological determinism, and were certainly diehard opponents of evolutionary psychology, then called “sociobiology”. You cannot count on that book to be an objective review of the literature, as it’s a polemic. It should not have been used by Spencer as an authoritative refutation of gender differences.
Second, the book is outdated. It is now 33 years old, and a considerable literature has accumulated since then. Not one thing is cited from that literature save in support of the absence of two sexes (see below)—Spencer just emits quote after quote from that book. And he uses it to refute three assertions that, he claims, Damore makes—at least implicitly:
Men are better at certain fields like engineering. He quotes LRK’s argument that there are no differences between the sexes in various tests on things like reading, vocabulary, and so on. Now I’m not familiar with the literature on differential abilities on the job, but there are certainly extensive data about sex differences seen on tests of spatial abilities, verbal and reading skills, and so on. And that data shows that there are differences, whatever the cause. Pity that Spencer didn’t look this up, but he had to crank out that ideological piece for Salon quite quickly.
Spencer also notes that sex differences could well be based on socialization and not genetics. That is true, and it’s a point I make above, having noted yesterday that to equalize the playing field, we must make sure that young boys and girls aren’t told that they are better suited for one thing or another, or pushed toward one type of behavior or another, or directed to different studies. That said, I think there’s no doubt that some of these differences are manifest before infants can even be socialized, and are also seen in other primates. This is clearly the case for sexual behavior, for instance. The reason men are more promiscuous and females more choosy is based at least in part on evolution.
Hormones make us who we are. Again, to refute this, Spencer exclusively cites the polemic LRK book, but he’s not successful. What he says is this:
Another of Damore’s persistent claims is that the “biological differences” between men and women “have clear biological causes and links to prenatal testosterone.” This myth, too, has been debunked.
“Insofar as sex differences are determined by hormones, they are not a consequence of the activities of uniquely male or female hormones, but rather probably of fluctuating differences in the ratios of these hormones and their interactions with target organs,” Lewontin, Rose and Kamin write. In other words: there isn’t hard science that shows that “testosterone = drive for leadership.”
But this is weaselly. Yes, both men and women have estrogen and testosterone, but the balance between these hormones has substantial effects on secondary sex characteristics, including appearance, aggressiveness, and other behaviors. There is certainly hard science showing that “testosterone titer is correlated with aggressiveness”, and, seriously, does Spencer think that aggressiveness has nothing to do with drive for status or control?
Note, though, how LRK weasel on this well known result, attributing it not to hormones but to their “fluctuating ratio” and “their interactions with target organs.” That does not refute the claim that hormones cause differences in behavior. We know that injecting biological women with testosterone makes them exhibit more “masculine” secondary traits and behaviors, while the opposite holds with men having low testosterone. To deny that hormones have nothing to do with behavior shows a deep desire to avoid reality in favor of ideology.
Career choices prove biological difference. Spencer attacks Damore’s claim that the innateness of biological differences is proved by the fact that different professions are differentially “gendered” in different places (i.e., LRK note that there were relatively far more physicians in the Soviet Union). That certainly shows that social factors can affect these proportions, but it doesn’t say that there’s no biological underpinnings to them. After all, we know well that cultural change can overcome evolution; if it couldn’t, we wouldn’t have people choosing not to have kids. But what Damore is claiming is that given equal opportunity, psychological differences and (to a lesser extent) differential ability may still lead to disproportionate outcomes in the gender composition of professions.
Spencer, however, is correct in saying that universality says almost nothing about genetic determinism of a trait (though it doesn’t say nothing: virtually all humans have two legs and go to sleep, and that’s certainly coded in our genes), and Damore should have stayed away from that.
In the end, though Spencer places the Google memo in “the dark history of biological determinism”, lumping Damore in with early racists, with sexists, and with transphobics. (Spencer asserts that “the mere idea of two sexes may be a myth”, which is deeply misleading, for the supporting evidence he gives is the usual argument about the presence of intersexual individuals or other “intermediate cases.” Yet these constitute a very, very small percentage of all humans. Those intermediates occur in flies, too, and also at a low frequency, but no geneticist says that “the idea that there are two sexes of fruit flies is a myth.”)
In his last paragraph Spencer goes Full Weasel: hastening to assure us that Damore is not advocating eugenics (what? it’s not even relevant to his memo!), but adding that his arguments might lead to it:
Damore doesn’t go as far as to advocate for eugenics, to his credit; he argues that men and women have differences that need to be respected, and rather doesn’t say men are “superior” — even if in his estimation, men are predisposed to be traditionally successful by our current social metric. Still, biologically deterministic arguments like his can easily slip into eugenicist doctrines of yore.
The last sentence is simply a “Chicken Little” way to dismiss Damore’s arguments without engaging them. “We can’t discuss this because it could slip into eugenics.” That would be laughable if we didn’t have someone using that argument to stifle discussion about how to achieve diversity.
I conclude that Spencer’s article is misguided, for it is a Regressive Leftist piece that relies on a single outdated polemic to make its case. If Spencer wanted to refute Damore’s arguments, he should have done a bit more research. But regardless, unless there’s something I don’t know about Damore’s other activities at Google, it is shameful that he was fired for writing this memo.