Monday, March 11, 2013

Thoughts on Feminism

There was a recent comment thread in Marginal Revolution (original post here) in which a guy (who goes by the handle Tom West) gave a very illuminating perspective on modern feminism and feminist politics that I felt compelled to share. I've reproduced the thread at the end of this post, so you can read it there first or find it on MR first.  It's one of the first few comments.  Either way, I really want to just jump into discussion here.

Let me preface all of this by saying that as the proud carrier of a Y chromosome with an expressed SRY region, I've never fully understood the underlying drivers for the modern feminist movement - nor will I.  My introduction to modern feminist politics was the furor over Larry Summers' resignation in the early spring of 2006.  Summers' speech didn't seem at all inflammatory to my eyes.  To me, the budding scientist and researcher, what Summers suggested was (among other things) an extraordinarily simple, plausible testable hypothesis: that there exists some innate aspect women and men's intellectual capabilities that, statistically, predisposes either gender to have some level of over-representation in certain fields.  Other things Summers mentioned were also proven statistical correlations, including that women with a certain level of education in the United States tend to marry men of an equal or greater level of education.

Yet, all of my feminist friends - even ones that were normally lukewarm about advocacy - seemed up in arms about the speech.  I wondered why, and didn't understand, not least because I felt that the statements were very properly given with caveats and an emphasis that the correlations were a necessary but not sufficient condition for causation. Worse still, when I tried to defend Larry Summers' propositions as reasonable*, I was attacked - not criticized, but full-on ad hominem attacked - by people I considered my friends.  In later years, in the rare instances the topic was brought up, the same situation repeated itself; eventually I learned to avoid the topic altogether.

All this, and other instances, led me to negatively categorize the behavior of the modern feminist movement as hypersensitive and too willing to allow ideology to trump any empirical evidence, especially if the hypothesis in question suggested any innate differences between men and women (beyond the physical, of course). 

None of it quite made any sense until I read the refreshing explanation from the MR comment thread.  "Tom West" sums it up quite succinctly;

Feminism, in general, seems aghast at even acknowledging a lot of innate gender differences.
Perhaps. But lets face reality. The fight for equality is a political battle on both sides. And most of us realize that acknowledging that ‘nature’ has any part in current inequality has historically and will continue to be used as a weapon to go *far* beyond that nature to try and force all women (and men) into roles that they might only statistically tend towards. Man likes to categorize, and we are continually insisting that any statistical correlation above 0 must be 1.

So, if you know that acknowledging that a given ratio might actually be 0.4 and 0.6 will be turned into a very politically effective argument that the ratios must be mandated as 0 and 1, you can be forgiven for claiming the ratio is 0.5 and 0.5 despite some evidence to the contrary. It also doesn’t help that in quite a number of cases, it turns out the accepted wisdom is pretty much wrong, and once we conduct experiments to eliminate the cultural component, results between genders are far closer than original experiments indicated.
Suddenly, a lot of things started to make sense.  It makes sense, for example, that if I start talking about innate gender differences, I am suddenly mistaken for one of the gender-roles-are-hardwired crowd.  I am aware that statistical differences in the intellectual ability of genders are smaller than is commonly believed, and that even these statistical likelihoods does not change the likelihood of extraordinary individuals existing from a group that might have a lower aptitude on average.  In fact, I have firsthand experience (by virtue of my college) of extraordinarily talented individuals of both genders in roles seen as the preserve of the other.  But my arguments largely appear the same as those who don't have this understanding, and routinely use "nature" arguments to pigeonhole women into their traditional roles.

That, in a sense, explains the visceral reaction that "nature" arguments tend to get, even from people like me that acknowledge the problems that the feminist agenda wants to solve.  One of two misunderstandings will happen when someone with a well-grounded, empirically valid understanding of innate gender differences pitches an argument to a feminist: either a subtle distinction or clarification is lost in conversation, or, more likely, the argument will be functionally identical to one made by some clod who thinks that the underlying relative correlations with ability are closer to 0.9 to 0.1 than 0.5 to 0.5.  Either way, the feminist assumes (not unjustifiably) that the underlying assumptions are sexist, and then proceeds to chew the other guy out.

The case of a cautious and well-versed proposer being chewed out is a good example of a type I error, in which the feminist response assumes that the proposer has wrongheaded biases where they do not exist.  Unfortunately, the reason that it's so hard to dissuade people from making the assumptions that lead to type I errors in my case, and probably Larry Summers' case as well, is that most of the time the assumption is valid.  Most people who make the innate differences argument are drawing fantastically wrongheaded conclusions from equally fantastically wrongheaded estimates of what the true statistical differences in ability are.  These estimates are informed not by familiarity with the literature but by cultural and social biases.  In other words, the type I error occurs because most people who propose this position just use the possibility of natural variation in intelligence between genders as a vehicle for their own preconceived notions of gender roles.  A good portion of these people will also deny the existence of outliers who can perform just as well in non-traditional gender roles as they could in ones seen as the preserve of the other.  Either way, it's disgusting and all-too-frequently the case.

As Tom West put it in a later response:
... my personal experience with teachers (teachers!) who took a certain Harvard president’s words as “proof” that women can’t do math made me pretty sensitive to the reality that people work hard to misinterpret tiny factoids to make them into giant edifices that can have profound effects on many, many others. It’s certainly left me sympathetic to those who want to err on the side of assuming more equality of nature, especially given the uncertainty in any claims as to what’s nature and what’s culture.)
So like Tom West, I'm now much more sympathetic to feminist movement's tendency to discredit or attack gender-based intellectual ability claims.  I have personally been hurt by the calculus that goes into making that decision, and at least partially because of that I disagree with the strategy.  But I also recognize that introducing subtlety into arguments allows the opposition to claim that feminists actually agree with them, and agitate for the hell of it, or some other bullshit argument.  Refuting that claim takes careful presentation of facts, and sound bytes count for far more than facts in politics.

* I have to emphasize here that my working assumption is that intellectual ability is roughly equal until proven otherwise.  The values suggested by literature - even with statistically significant results - all point to differences that are too small in most categories of intelligence to justifiably change that assumption.

Full source thread after the jump: