Feminist Critiques of Scientific Methodology
And suggestions on how to improve it
When talking about feminist critiques of science, the focus tends to be on critiques of scientific institutions, e.g., sexual harassment on campuses, glass ceilings at universities, etc. What receives much less attention is feminist critiques of scientific methodology.
In this post I will go over some of these latter critiques, particularly the ones that focus on combatting “male bias” in science. Of course, there are also parts of science that will have female biases (and it goes without saying that feminism is about combatting those too), but given that women were, for the longest time, largely forbidden from even becoming scientists (and still are, in some parts of the world), and given that science as a whole is still male-dominated, it shouldn’t be too controversial to focus on the male bias in our scientific literature.
This bias comes in two forms: bias in data-collection, and bias in data-interpretation.
I don’t think the existence of such biases in data-collection is too controversial. I assume the scientists among us know of many stories where women were underrepresented —or even straight-up excluded— from scientific datasets. My personal favorite example is that time a feminist discovered that several evolutionary psychologists developed theories of the female orgasm, using only datasets of male orgasms. (Oh evo-psych, never change)
Bias in data-interpretation, on the other hand, might be a bit harder to visualize. So let me give you a famous example so you can see what the feminists are getting at:
Fertilization: A case of male bias in data interpretation
Traditionally, biologists adopted what is sometimes called the “Prince Charming model” of fertilization, wherein the process was treated as a hero story where a single, intrepid sperm battles through the hostile uterus, survives perilous challenges, defeats rival sperm, and finally claims the passive egg, which it penetrates and breathes life into.
This is similar to how men and women are often treated in literary works, with the men being active and heroic, while the women are passive trophies. Feminists argue that this imagery shaped descriptions of fertilization, even after it became clear this model is inaccurate since:
The egg is not passive; it actually actively selects a sperm and produces cell-surface projections that clasp the sperm and draw it inside.
Mammalian sperm cannot fertilize an egg immediately; they must undergo a process called “capacitation”. In this process the uterus is not an obstacle course for the sperm, but is actually an active participant that helps the sperm become ready for fertilization.
Photographs of 1 were published as early as 1895, yet their role was virtually ignored until the 1980s. Similarly, point 2 has been known since the 1950s, and yet the “Prince Charming model” was still the default way fertilization was presented. This suggests that gender stereotypes biased biologists’ interpretations; with “penetration” remaining the default verb, even though “engulfment” would be at least as accurate. The evidence was there, but male bias caused it to be ignored.
Combatting bias by increasing diversity
Feminists argue that to combat this bias, science should start incorporating feminist values. One way to do that, they argue, is by increasing diversity in research teams and peer review. This would help with the development of better scientific theories, because it increases the diversity of the theories we evaluate.
When a scientist generates a theory, they may do so for any reason (aesthetic reasons, political reasons, because it came to them in a dream…). Contemporary scientific methodology only concerns itself with testing theories, but what happens if there’s a systemic bias in which theories get generated?
Say a group of scientists have generated a couple theories which all have a male bias. When we start scientifically testing which one of these we should adopt, no matter how carefully we apply the scientific method, the one we will end up with is a theory that has a male bias. Science only helps us select the best theory amongst the available theories. To increase the quality of our theories we not only need to properly test our theories, we also need diversity in which theories get generated.
Feminist theory adoption
In science we often encounter the problem of underdetermination. This means that there are multiple theories that are equally supported by the available data. Which theory one should adopt in such a scenario is hotly contested amongst philosophers, but perhaps the most popular approach is to adopt the theory that has the most social utility.
For example, if one group of scientists present a theory that is totally inscrutable to almost everyone, while another group of scientists present an (equally empirical) theory that’s totally legible and allows engineers to start creating new inventions based on it, this approach champions adopting the second theory.
This is in line with the broader pragmatist school of epistemology, a school that says knowledge should be actually useful. If you (like me) also think we should adopt the theories that are the most socially useful, then adopting the more feminist theories (from the available evidence-based theories) would be a good idea in a world where we’re suffering from sexism. Do we live in such a world?
It sure seems that way…
Of course this doesn’t mean we should automatically pick the most feminist theory. There are other problems and biases too. You could, for example, copy-paste this entire blogpost and with minimal effort transform it into a blogpost about racial bias.
But it does suggest that we should overall err on the side of feminism.
I think the most pressing issue with regards to feminism in science is that sexism and sexual harrassment are sadly not that rare in science, especially in male dominated fields like physics. For example, I had a female friend doing a PhD whose lab partner told her that he wouldn't prepare samples for her unless she went for drinks with him. There is an excellent and very in depth youtube video from Angela Collier on the topic here: https://www.youtube.com/watch?v=8DNRBa39Iig.
It is clearly not good for science if half the population is discouraged from participation.
My apologies if I am misunderstanding you, but your argument for why we should pick the higher utility theory where multiple theories are equally likely to be true. Doesn’t strike me as very persuasive, at least if I’m understanding you correctly. In your example where one theory is very difficult for most people to understand and the other theory permits amazing inventions The reason you should go ahead with the attempted inventions is because it has 50% chance of the payout being great. If the difficult to understand theory suggest that attempting to make these inventions could cause a great catastrophe, it doesn’t make sense to just believe the theory under which the world would appear to be better. Thinking the world is as you wish it to be isn’t going to actually change it and might actively hinder your attempts to improve it by worsening your understanding. In generally appears to me that considerations of utility don’t make sense when deciding what’s true. Instead of when deciding what to do. At least if by considerations of utility, we mean, believing the theory under which the world would have higher utility. Indeed, this kind of practice might be, especially bad for achieving progress on your issue. Imagine where animal rights, people would be if instead of trying to determine whether some animals are conscious or not They adopted their views for ideological reasons since after all the evidence is generally compatible with both possibilities. It’s not a perfect analogy, since generally, the evidence is not exactly 50-50, but my point is that believing social theories because they sound feminist might actually hurt the feminist movement more because its advocates would be misinformed about reality and hence make mistakes while trying to achieve their goals. There are hypothetical situations where it’s useful to believe false things, but generally to determine whether you are in one of those situations, you have to know the truth to begin with. In any case, I don’t think this is very relevant to the discussion. Since in practice, the bigger concern with feminist science is that people might believe something because it sounds good rather than because it’s useful to believe as I think, usefulness and truth will generally point in the same direction. To be clear, I absolutely agree with you on things like having male dominated data sets and historical biases in interpretation, although I am unsure whether the bias persist to this day, at least in the scientific community in the west, where most science is conducted, as scientists in general tend to be way more liberal than the general population.