Beyond Unconscious Bias & Sexual Harassment: a case study

A recent spate of all-male headlined conference announcements (note this one at Princeton here) made me reflect again on the mechanisms at work. Our friends at the GCC have been emphasizing the ways in which implicit bias and stereotype threat reinforce each other and, more recently, Brian Leiter has reminded us (rightly) that "sexual harassment is still the scandal of philosophy profession." Petitions (the call to action initiated by Martin Kusch and Mark Lance has almost a 1000 signatures; here is the official GCC petition; and an interdisciplinary one sponsored by Virginia Valian and Dan Sperber and friends) are not enough to solve these issues, of course. But I do not think that implicit bias, stereotype threat, and sexual harassment exhaust the pernicious mechanisms at work.

Recently, I wrote some young scholars who ended up hosting a resource-rich conference with a line-up of ten (mostly early career) males. (I wouldn't be surprised this particular conference also ends up being a special issue of a journal.) This surprised me because I had noticed the cfp and it explicitly mentioned blind review of abstracts, and had made no mention of invited speakers. Their response is illuminating because it shows how even selection designs that aim at quality can lead to unfair, gender-imbalanced outcomes without any malice or gender bias (I'll qualify this in a bit) at all:

From a call for papers, we have selected 7 papers (out of a total of 10 to be presented at this workshop). Possibly 7, but at least 5 (out of 24 submitted), abstracts have been submitted by women (judging on the basis of first names). All abstracts have been anonymously evaluated double blind by a pool of 30 reviewers (of which roughly half were authors that had submitted abstracts). The resulting rank order was arrived at by straight averaging, gender was not a factor in the decision. The group of authors selected remained the same irrespective of assigning weights (between 60/40 and 80/20) to reviewers’ scores or authors’ scores. We checked for gender only after the blog post [by the GCC] had become known to us. The best “female abstract” is probably on rank 11.

3 out of 10 talks are given by invited speakers. These had been planned before the call for papers. 2 of these are given  by scholars from [our home universities]. These authors had been asked before the call for papers, as we seek to give precedence to work originating [at our home universities]. Prof. X is the one invited speaker external to the [home universities.]

Now, notice that connected males have a much easier time than unconnected females (and males) to make it into the speaker line-up. So, this is really a case of the privileged helping their own, which, of course, tend to be boys. It also reminds us that for all the calls about folk focusing on quality, they never forget to help themselves first. That is, these privileged also make it clear that gender inclusiveness is an after-thought for them. Finally, I was struck by the confident claim that the referee pool ensured that "gender was not a factor in the decision." Well, if double-blind is really disinterested and blind, then, sure. But in this case half the referee pool had a non-trivial stake in the outcome; not to mention that in an age of pre-publication, electronic paper sharing (etc.), even the most noble of the well connected referees find it very hard to be genuinely blind.