Scientific rationality is one of the foundations of western civilization. The discovery of the natural laws behind the useful work done by a diesel engine, the electron clouds propagating through conductors, and the modus operandi of a virus, has given the geopolitical upper hand to North America and Western Europe. So it comes to no surprise that many have attempted to use a similar methodology to uncover the fundamental laws that regulate the human world.
Although this rationalist method of parsing the human world is intimately coupled with the spirit of anglo-saxon capitalism, with economic marginalism (e.g. Hayek, Samuelson, etc.) being its first coherent expression – there has been a recent, growing rationalist movement that attempts to bring this perspective to the culture wars. Some examples of these platforms are the popular blog Slate Star Codex, and the publication Quilette. An important animus behind this upsurge is a reaction against the sociological theories of the Left, such as structural theories on gender and racial discrimination. Many of these rationalists, instead postulate that the perceived empirical disparities (e.g. gender wage gap, racial inequality, lopsided gender and racial ratios in STEM etc.) between races and genders are connected to biological-essentialist variables such as sexual reproductive strategies, or the differences in IQ among races.
It’s hard for me to discuss in a completely detached and charitable manner these theories, because of my ethnicity, leftist leanings, and utter contempt for Vulcan-wannabe dudes with shitty STEM degrees. However I will try to use peer-reviewed articles that are popular among them in order to argue that ultimately, their “rationalist” methodology is fundamentally wrong. The outline of my argument is as follows: (i) the only thing these papers demonstrate is an empirical correlation not causation. (ii) The reason why they cannot demonstrate causality is that the problems they are dealings have many variables that are extremely hard to isolate. (iii) Because of the large epistemic uncertainty in the casual links, politics become unavoidable. (iv) Because of politics, this rationalist project collapses, and their vulcan-like rationality becomes a political ideology amongst others.
A good example is the often cited paper by Schmitt et al. The main thesis is that personality differences between women and men seem to widen in more gender equal countries. The paper finds a moderate correlation between personality sexual dimorphism and gender equality. However, what is generally referred is one of the conclusions, which argues that personality dimorphism is not enforced by stringent policing in gender equal countries. Rather, gender equality lets sexual dimorphic traits diverge into their natural equilibrium. In other words, free societies let women and men express their intrinsic, gendered personality traits that are a function of darwinian processes.
I’ve seen many “rationalist” sources refer to this paper either explicitly or implicitly. It’s seen as one of the most powerful attack against feminist points, such as how certain gendered disparities, like lopsided ratio of women in some STEM fields, or the lack of females in certain leadership positions, are product of sociological and structural factors such as socialization and sexual harassment. The “rationalist” argues that policies aimed at making certain fields like STEM more sexually diverse, or increasing the number of women in leadership positions, are misguided and potentially counterproductive. Very recently, a study also showed that percentage of women in STEM fields seems to actually decrease as a function of equality, where in relatively unequal country such as Algeria, about 41 percent of STEM workers are women. This study seems to vindicate the previous study of Scmitt et al.
I am not going to question the methodology behind these studies, but I feel necessary to point out that quantifying things like “personality traits” and “gender equality”, and also aggregating them, is probably not trivial and riddled with assumptions. However, even without questioning the methodology, and taking at face value these empirical relations, the papers at most demonstrate the existence of empirical correlations and nothing more. One could try to hypothesize a multiple of causes, including a biological-essentialist link, but ultimately, these studies only demonstrate a correlation between two empirical measurements, and nothing else. This is the old adage of correlation does not imply causation. Here is a very funny site showing all sorts of spurious correlations, such as the relationship between suicides by strangulation and government spending on science. A better way to understand this problem is to imagine a situation where two variables are correlated: A and B. There are actually four plausible causal explanations for this correlation: (1) A causes B, (2) B causes A, (3} A and B are caused by some variable C, or (4) the correlation is only a spurious coincidence. Therefore an empirical correlation, while an important result in itself, is not sufficient proof to establish causation.
The issue of causation is very deep and has lead to centuries old discussions in the sciences and philosophy. For example, the Scottish philosopher David Hume argued that there is no logically consistent way of assuming causation from correlation. However, my argument isn’t really as absolute, but more practical in an everyday sense. Studies, like the ones I referred above, deal with problems that are too multivariate to convincingly establish a biological argument by just a mere correlation. In the case of physics and the hard sciences, causation is usually proven through experimentation that isolates all the irrelevant variables, or if lab experiments are not possible, through computational simulations where all the important variables and physical laws are plugged into a computer code.
In the case of other “softer” sciences, such as bio-informatics, social sciences, etc. that deal with complex, multivariate problems that cannot be dissected by controlled experiments, the important variables are isolated through statistical techniques that try to take take into account all the relevant parameters. For example, here is a very easy to understand paper that argues against the book of IQ and the Wealth of Nation, by disproving the idea that that some biologically detemined lower IQ of the “non-white” races leads to underdevelopment in their respective countries, by using a simple multiple regression analysis that takes other variables beyond IQ into account. Furthermore, in many cases, especially studies with political consequences, even sophisticated statistical techniques are not enough to establish causation beyond reasonable doubt, given that there is always the possibility of unknown variables not being accounted for. A famous example is the history of the cigarette-lung cancer link, where it took decades of different types of studies, from lab experiments, to questionnaire based correlations, to establish a causal link. This weakness was obviously abused by tobacco conglomerates, but the point is that even the scientists hired by these tobacco companies at some point began to accept the validity of the evidence, since various research trajectories triangulated into the cancer-cigarette connection.
Now lets go back to the previous statement on how sexual dimorphism in gender egalitarian countries implies an inherent, biologically hard-wired tendency that makes men in average more interested in engineering than women. This causal link is almost completely impossible to establish beyond doubt, at least with the known experimental and scientific techniques. This problem is incredibly much more complex than the subject of the cancer-tobacco link. This complexity arises due to the existence of many social variables interfacing with the career choice of women that are extremely hard to take into account. For example, it is obvious that in the most gender, unequal limit, there wouldn’t be almost any women in engineering jobs (e.g. England in the 19th century)! It is only in today’s particular configuration that this correlation seems to be valid, which already shows the existence of hidden socio-economic variables that affect these studies.
The inability to establish causal links beyond reasonable doubt in many socio-economic problems (e.g. economics) is actually well defined mathematically. For example, in the case of mathematical physics, the Holy Grail for all these vulcan-like rationalists, the problems that can be solved are extremely limited in scope. Poincare showed in the late 19th century the exact solution for the trajectories of more than two interacting bodies is mathematically non-integrable. In the 1960s, Lorenz discovered that despite the sophistication of computers, many multivariate problems, such as the one of simulating the weather, become intractable after a certain point due to chaos. These uncertainties are not even a matter of not properly accounting for all relevant variables, but are embedded in the mathematical structure. So it is quite arrogant to argue with confidence that a couple of mere empirical correlations are enough to disprove the lived experience of many female students and STEM workers, which point at discouragement from peers, lack of role-models, unwelcoming workplaces etc.
Given the existence of large amount of noise, chaos, and “hidden” variables in socio-economic systems, there cannot be a pure rationalist and “scientific” way of tackling these problems. The existence of this epistemic uncertainty therefore gives rise to politics in a much more stronger sense, than when dealing with simpler, “mathematical’ problems. Therefore the cry of “centrists”, “classical liberals”, “rationalists”, Jordan Peterson, etc. of feminism, leftism, etc. being ideological is a case of pot calling the kettle black. Given the epistemic opacity of socio-economic problems, this claim of rationality is simply a bed-time story – a shallow aesthetic consideration for soul-less logical chopping and boring prose. Instead, they have agendas, not unlike the “irrational” leftists and feminists. In fact, if I were uncharitable, I could claim it isn’t reason that animates them, but some burning resentment for women, minorities, feminists etc. invading their spaces. I am not ashamed of admitting my own agendas as well, and that’s why this blog is explicitly partisan, and not written in the spirit of some shitty analytic philosophy paper.
If you liked this post so much that you want to buy me a drink, you can pitch in some bucks to my Patreon.