On Rigor, White Supremacy, and Occam’s Razor

A correspondent once took issue with my amplification of Alex Hanna’s indictment of Google as a racialized organization. They objected that Hanna’s accusations weren’t backed up “rigorously”. I’m publishing a version of my reply here because I may want to reference it in the future.

Our most “rational” professions have a long history of looking for the most convoluted explanations of obvious bias, and calling it “rigor”. A recent example: we know as a historical fact that “computer programmer” started as a female-dominated profession, until women were systematically pushed out in the 1970s and 80s. But that did nothing to keep logical, rigorous people from—by the time I got to the profession—advancing all kinds of explanations about how women were just “not natural programmers”. Anything except the simplest answer of all.

This sort of thing turns out to be everywhere. The very methods of statistical significance in scientific results were explicitly developed by eugenicists to validate their foregone belief that black people were closer to monkeys than white people. Nor were those men disappointed in their “rigorous” results, either: their methods—applied to the data they selected and accepted—confirmed their convictions.

And, of course, there is a vast amount of data around the mechanics of implicit white supremacy: things like people picking one identical resume over another because the person has a “white-sounding” name.

After discarding the false logic of over-complicated post-facto rationalizations, we’re left with this: if a “meritocratic” organization is overwhelmingly white in a society that is not, it means one of two things:

  1. They are artificially obstructing nonwhite participation (in ways that you and I may not be able to see)
  2. Non-white folks have less merit.

#2 is racism. #1 is white supremacy – a system in which white participants are treated as the default, and there’s extra friction for nonwhite participants. Any notional #3 one might propose violates Occam’s Razor.

To start with the incontrovertible data point of an organization which is dramatically less diverse than its surrounding society… And then to immediately search for anything other than the simplest and most common explanation (a system of implicit or overt white supremacy) is itself bias. Not rigor.