Ben Cobley criticizes the "ideological feminism" of Laurie Penny and others:

The idea there is a system perpetuating male privilege seems strange...This theory of patriarchy would be an almost perfect example of what Karl Popper called 'pseudo-science' if only its advocates claimed any sort of scientific backing. Instead feminists use the term as a free-standing fact that does not need serious evidential or logical support. The existence of patriarchy and the structures and systems which support it have to be taken on trust.

I'm not sure about this. I suspect that Laurie's ideas can be translated into scientific terms.

One thing we know about societies is that they are path-dependent; even quite distant history shapes behaviour decades later. For example, the extent of anti-Jewish pogroms in the 14th century predicted (pdf) the extent of Nazism; variations in the prevalence of US slavery are correlated (pdf) with variations in the black-white educational gap today; tax avoidance is widespread in Greece because it was seen as a patriotic duty under Ottoman rule. And so on. It is therefore surely plausible that today's societies contain at least some traces of their past sexism.

There are (at least) four inter-related mechanisms through which this might be so:

- The just world effect.People develop theories to justify inequality. It could be, therefore, that ideas of differences between men and women developed in response to inequalities between them, as an attempt to rationalize those inequalities. Such ideas mean that misogyny is tolerated.

- Stereotype threat and the Pygmalion effect. The Stanford prison and Oak School experiments show that people tend to live up (or down) to labels that are assigned to them even randomly; for example, children who are randomly labeled intelligent go on to do better on IQ tests than those not so labeled, and "guards" quickly become dominant whilst "prisoners" become submissive. Isn't it likely that a similar thing happens to genders - that (some) men conform to masculine stereotypes of ambition, dominance and violence and (some) women to stereotypes of docility? Empirical research shows they do. Laboratory experiments have found that women can be primed to do badly on "masculine" tasks, and research shows that when women are partially freed from their stereotype by attending all-girl schools, they choose more "masculine" subjects and become more competitive.

- Minority behaviour matters. Violent behaviour by some men can drive many women out of the public sphere. And this process can be self-perpetuating because...

- Role models matter. If we see people like us doing something, we'll be inclined to do it ourselves. The mere fact that many professions are dominated by men can therefore cause them to remain so dominated, because younger men are attracted into them and women deterred. As Akerlof and Kranton say:

In every social context, people have a notion of who they are, which is associated with beliefs about howe they and others are supposed to behave. These notions...play important roles in how economies work. (Identity Economics, p4)

What I'm saying here is that "patriarchy" might be a shorthand term for the fact that gender inequalities, like much else in the social sciences, are an emergent process. They arise not because of a conscious conspiracy, but as the unintended result of mechanisms. Perhaps, then, the concept of patriarchy is more scientific than the "not all men" trope.