I’ve read the infamous Google document, and I found it a mixed bag, though I don’t think the guy should have been fired for it. (That said, I have no idea about his history with Google). As reader Coel pointed out in a comment in the discussion thread on the Google fracas, there’s a much better article on the Slate Star Codex by Scott Alexander, “Contra Grant on exaggerated differences“, which makes the case that disproportionate outcomes, like a different proportion of men versus women in different professions, might reflect at least in part a difference in preferences based on psychological differences between men and women. The point is not that there’s no gender bias in the tech industry (my techie friends whom I trust say that there is), but that the disparity in representation might not completely reflect sexist barriers to entry but also different preferences (Alexander discerns no differences in abilities). That is, even in the absence of gender bias, and with free entry and equal opportunity for women, you might still not get the desired 50/50 male/female ratio in any field, much less technology, since representation reflects abilities, interests, and bias.
I can’t evaluate the literature he cites, nor do I know the evolutionary psychology literature about psychological preferences, so I can’t evaluate that either. Nor can I say anything about to what extent differences in preferences reflect evolved genetic differences (which of course don’t constitute “unchangeable differences”) versus culturally instilled differences. But nearly everyone agrees that there are differences in both behavior and preference between men and women, whatever their source. (I’m pretty sure, for reasons I’ve discussed before, that a lot of the differences in sexual behavior and preference between men and women are based on evolution.) My view is that we need to guarantee equal opportunity for the sexes and people of different ethnicities. But that’s a hard thing to do, for it involves detecting and eliminating biases that can affect children at an early age, making sure all school systems are equal in quality and in the opportunities they afford different kids, and so on. And that will take decades. In the meantime, some form of assuring diversity seems to me desirable. (Even that’s hard, for what are we aiming for: should people in all professions be represented in the same proportions as their groups occur in the general population?)
Too, the data on preferences and abilities are not completely consistent; different studies give different results. That’s not surprising given that we’re dealing with human behaviors and studies that differ in populations and methodologies. But Chanda Prescod-Weinstein, (a research associate in particle physics at the University of Washington, a philosopher of science, and editor of the online literary magazine Offing, has written a piece in Slate saying that the Google study and the sexism she sees it reflecting is inherent in the nature of science. Her piece, “Stop equating ‘science’ with truth,” was published just yesterday and already has over 1800 comments (I haven’t read any). I was surprised at this, for she’s not indicting sexism for the Google study, but the nature of science itself, which, she says, not only finds false evidence for sexism and other inequities, but acts as an imprimatur for bias:
It is impossible to consider this field of science without grappling with the flaws of the institution—and of the deification—of science itself. For example: It was argued to me this week that the Google memo failed to constitute hostile behavior because it cited peer-reviewed articles that suggest women have different brains. The well-known scientist who made this comment to me is both a woman and someone who knows quite well that “peer-reviewed” and “correct” are not interchangeable terms. This brings us to the question that many have grappled with this week. It’s 2017, and to some extent scientific literature still supports a patriarchal view that ranks a man’s intellect above a woman’s.
. . . Science’s greatest myth is that it doesn’t encode bias and is always self-correcting. In fact, science has often made its living from encoding and justifying bias, and refusing to do anything about the fact that the data says something’s wrong.
I think she’s wrong on several counts here. First of all, science is a methodology, not a body of data, as that data and the “facts” it uncovers are provisional—even though some of those facts, like the observation that DNA is normally a double helix, are unlikely to be proven wrong in the future. Of course science can reflect and buttress the biases of its practitioners, but that’s why we have things like peer review, statistics, replication, and so on. And, in fact, I don’t think modern scientific literature “supports a patriarchal view that ranks a man’s intellect above a woman’s.” You might be able to cherry-pick different studies that show that, but you can find other studies that show the opposite, or that women’s intellects are superior to men’s. Such a statement seems insupportable anyway since the sexes rank differently in different areas of endeavor, regardless of the cause.
I’m not sure what Prescod-Weinstein means by saying that science “refuses to do anything about the fact that the data says [sic] that something’s wrong”, but if she means science refuses to correct flawed data showing inequities, she’s wrong. In his book The Mismeasure of Man, Steve Gould went after Samuel Morton’s data on skull volume showing that non-Caucasian skulls had smaller volumes, arguing that it was a case of fudged data reflecting Morton’s racist bias. A reanalysis showed that Morton’s measurements were actually accurate, and I think people may still be working on this. Assertions of female inferiority based on data have been roundly attacked by scholars like Cordelia Fine. There are now plenty of women academics who are well placed to analyze data on sex differences, and they’re doing so. But if people use scientific data, or distort it wrongly to reflect their own preconceptions, that is not the fault of science. Yes, science can “encode” bias, and yes, it sometimes takes a long time to correct, but I deny that the nature of the enterprise is not self correcting. After all, some of the important tools in the toolkit of science are doubt, replication, and constant questioning—not just of yourself, but of others. If science can’t dispel data that supports bigotry, what can?
In fact, it looks as if Dr. Prescod-Weinstein is guilty of just what she accuses others of, for she wants scientific educations to become politicized, and to teach people that science is biased in exactly the way she thinks it’s biased:
Most saliently in the context of the Google memo, our scientific educations almost never talk about the invention of whiteness and the invention of race in tandem with the early scientific method which placed a high value on taxonomies—which unsurprisingly and almost certainly not coincidentally supported prevailing social views. The standard history of science that is taught to budding scientists is that during the Enlightenment, Europe went from the dark ages to, well, being enlightened by a more progressive mindset characterized by objective “science.” It is the rare scientific education that includes a simultaneous conversation about the rise of violent, imperialist globalization during the same time period. Very few curricula acknowledge that some European scientific “discoveries” were in fact collations of borrowed indigenous knowledge. And far too many universally call technology progress while failing to acknowledge that it has left us in a dangerously warmed climate.
Much of the science that resulted from this system, conducted primarily by white men, is what helped teach us that women were the inferior sex. Racial taxonomies conveniently confirmed that enslaving African people was a perfectly reasonable behavior since, as Thomas Jefferson put it, black people were “inferior to the whites in the endowments of body and mind.” Of course, this apparent inferiority never stopped Jefferson from repeatedly raping his wife’s half-sister, Sally Hemings, herself a product of rape. Jefferson is remembered as a great thinker, but when one reads his writing about race, it becomes immediately evident that rather than being much of a scientist, he was a biased white supremacist who hid behind science as a shield.
Actually, she’s talking not about science, but the history of science, and I, for one, didn’t get any of that during my science education. As for global warming, that phenomenon was uncovered by scientists, and now everybody knows about it. To call science responsible for global warming is just wrong. The technology that gave rise to global warming is a product of science, was often used for financial gain by greedy people at the expense of the environment, and, in fact, people didn’t know about the phenomenon until fairly recently. It is science that has found it out, scientists who are alerting us to its dangers, and if anything can fix it, it’ll be the scientific method. And yes, science was used to buttress racism, but believe me, while slave owners might have justified their actions post facto with bogus science, the institution of slavery was not built on scientific data. Science can be used to support many bigoted views (although, contrary to some, Hitler didn’t use Darwin to prop up his genocides), but that’s the fault of the bigots, not science. It’s simply unfair to use Thomas Jefferson’s relationship with Sally Hemings as a weapon against science. You might as well use the existence of the Nazi gas chambers as an indictment of both architecture and chemistry.
There’s more, too, but I’ll let you ponder this:
The problem is that science was just the shield he needed in the 18th century, and unfortunately, it seems that it continues to function that way today. In other words, pseudoscience has always been a core feature of post-Enlightenment scientific knowledge and it remains that way because scientists refuse to integrate contemporary science, technology, and society studies research into university curricula. And so too many of us get out of school and end up in a world where we are suddenly forced to grapple with the reality of how science, in practice, is not as objective as we hoped. Enough of us have heard a man, sometimes the president of our college, sometimes our research adviser, express the view that women’s brains “just work differently” and “aren’t suited to technical skills” the way men’s are. Nonbinary people don’t exist, and transgender people are de-normalized in these narratives. Women of color listen to white women normalize Europe as the birthplace of scientific intelligence while telling us that our curly hair isn’t professional-looking. Senior men who we would hope could be mentors turn out to be our sexual harassers, and with some frequency, senior women tell us to suck it up and lean in, rather than helping us.
Here she’s dragging in examples not of science misused, but bigotry and oppression that have nothing to do with science, like disapprobation of curly hair and sexual harassment. It’s almost as if Prescod-Weinstein wants to pin all the issues of social justice on science, even if they have nothing to do with it. But that won’t stick; by and large, scientists these days are nearly all sensitive to how their findings might be perceived. After all, the guy who produced that Google document was fired. If Google, a science-based firm, really supported bigotry, he would have been promoted. The bulk of Prescod-Weinstein’s piece seems to me to reflect an anti-science bias infecting certain elements of the Left (it of course infects the Right, too, but Prescod-Weinstein and I are both Leftists).
But she and I do agree on one thing. As she says in her last paragraph, we won’t know for sure about differences in abilities and preferences—and how they produce outcomes— until the playing field is fully leveled, and to me that means equal opportunities. So (with the exception of the term “bro,” which I consider sexism—the equivalent of calling a woman a “babe” or a “chick”), I agree with the “experiment” she proposes at the end of her essay:
Google bro would argue that we ought to consider the possibility that white women and racial minorities simply produce lower-quality work, which is why we struggle to be recognized as competent knowledge producers. It’s time to turn the tables on this debate. Rather than leaning in and trying endlessly to prove our humanity and value, people like him should have to prove that our inferiority is the problem. Eliminate structural biases in education, health care, housing, and salaries that favor white men and see if we fail. Run the experiment. Be a scientist about it.