I was wrong: about “positive discrimination”

I’ve always believed that it is a Very Bad Thing to discriminate between people based on group membership, especially when the groups are things people don’t choose for themselves (race, gender, sexuality, class, nationality, etc). Apart from the obvious damage to the people who get discriminated against, it’s bad for the discriminators who miss out on potential recruits/customers/friends for no good reason.

I used to think that this belief was a simple matter and all its corollaries obvious. Discrimination is bad, so we must not discriminate, end of story. If I’m judging people by the content of their character and not the colour of their skin, I must be doing my part to fulfill Dr. King’s dream, right? By the same token, surely it’s bad to ever consider group memberships at all? In fact, wouldn’t it be better for recruiters (colleges, employers, etc) to just never collect information about race or gender? A decade ago I would have wholeheartedly agreed with all of those, but I’ve learned that there’s more to it.

There were many steps along the path, but three data points stand out in my mind as particularly relevant.


The first was a conversation at the 2004 Artificial Life conference. As is always the case with academic conferences, conversations with fellow researchers were at least as interesting as the presentations, papers and proceedings we were officially there for, and we didn’t exclusively talk about work. I had a particularly interesting conversation with a fellow grad student who happened to be from South Africa, and at some point the topic of  Black Economic Empowerment came up.

At the time, I took the Economist’s standard line on the subject, which was to argue that it’s not only intrinsically undesirable to consider race in things like employment decisions, but also a bad idea for practical reasons. The pragmatic arguments were along the lines that people would always find ways to fiddle the system, all the benefit would accrue to the subset of black people who were already relatively well off, and that as a temporary solution it was just storing up problems for when it would eventually be phased out.

My new friend responded, fairly, that I just didn’t understand how crippling the legacy of Apartheid had been. In essence his point was that Apartheid had so restricted black South Africans’ access to education and wealth that they couldn’t now afford to educate their own children in turn. He was adamant that non-discrimination alone would never get enough people out of that sort of poverty trap, and without an active program of fixing that legacy Apartheid might as well still exist. I wasn’t immediately persuaded, but as you can see his argument made quite an impact on me, and it started me on the journey towards thinking about equality as a matter of actively breaking poverty traps rather than just passively getting out of the way.


The second has been watching my wife’s career development. We met in grad school, and at some point I was informed that I was dating the entire population of female American grad students in the computer science department, which was easy given that it consisted of her. She did rather well, getting her Master’s on schedule and lining up an offer of a good job which was how we ended up in Seattle. Since then she’s continued to succeed professionally in the IT industry, which I probably don’t need to tell you is heavily male-dominated.

On the face of it, this might sound like a good argument for the simplistic view of non-discrimination: get direct discrimination out of the way, and a smart woman gets to prove her ability and reap the rewards. But what I’ve also seen is how different her experience of work has been when she’s been the only woman in her team, versus not being alone, versus having at least one woman in the management chain. Even as the big picture of her career progress has involved substantial success and seems to have been a good experience along most of the way, there have (of course) been obvious bumps in the road. Some of those had clearly been caused by working with men who had so little experience relating to knowledgeable technical women that they either dismissed them or were made deeply insecure by them.

It’s clear to me that if my wife had always been the only woman in her team, her experience would have been at best a lot less positive, if not negative enough to have driven her out of the field. It’s also clear from how well she’s done without the burden of being the sole woman on a team that this is not down to any lack of ability on her part.


The third was what I learned from one of the first classes I organised for Sustainable Seattle: Dialog about Race: The Power of an Illusion. This is in general a pretty eye-opening workshop about both the history of racial inequality in the US and the mechanisms by which it’s still perpetuated today even though direct discrimination has been banned for a couple of generations. The specific point that really got me was a claim that almost the entire difference between white and black American childrens’ performance through school could be explained by the huge difference in wealth between the races. As far as I know, this claim is well supported by data; certainly wealth seems to be the dominant variable in determining which if any college a child will go to.

There’s a sliver of good news in that finding: it implies that direct discrimination no longer dominate’s black childrens’ lives in the way it did in Dr. King’s time. But that’s not going to be much comfort to the average black parent, knowing that even if they themselves have the resources to get their kids through college, most of their peers won’t, and most of their peers’ children will fail not because of the content of their character, but because of the colour of their skin. A degree of indirection doesn’t make this any less true; to go back to the beginning of Dr. King’s speech, it remains true after another 50 years that:

…the life of the Negro is still sadly crippled by the manacles of segregation and the chains of discrimination. One hundred years later, the Negro lives on a lonely island of poverty in the midst of a vast ocean of material prosperity….

It would even have been better news if income had turned out to be the mediating factor, because at least that could have been fixed in one generation, whereas the wealth gap is the cumulative effect of many generations of closed doors.


These were the biggest things, the ones that I’m directly conscious of having moved my thoughts. Along the way there were others, like seeing the burden of unicornism and imposter syndrome, the number of my leads & opportunities that have come in through informal networks, and becoming more and more conscious of just how ferociously our culture tries to shove everyone into a convenient pigeonhole and fights their attempts to leave it.

Over time this set of things convinced me that even if every single person in the world could somehow be persuaded overnight to never again discriminate against anyone because of their group membership, the problems caused by such discrimination would persist. Non-discrimination alone can’t address a lack of resources, role models and peers. Left alone, this problem might resolve itself over a few generations. I’m not convinced it would, at least not without other momentous changes in how we organise society, because poverty traps can be so utterly vicious even without the help of any -ism. But even if it would, why should the intervening generations suffer? What have they done to deserve that? Without a good answer to that question, I see a burden on everyone who has privilege in a given situation—and especially those who control access to financial resources—to actively level the playing field for those who don’t.

This is not to say that the problem is easy to solve, or that well-intentioned efforts to solve it don’t routinely have unintended consequences. I’m still very wary of rigid quota systems in particular. I also think anything that looks at race independent of poverty,  gender independent of childrearing & domestic labour allocation, or nationality independent of language & class, is doomed to fail in a very specific way: by having most of its intended benefit captured by a small, privileged subset of the group it’s meant to help. And I still believe at least one thing I thought 10 years ago: that much of the work of erasing long-established inequality has to be done by and through state schools, which can serve as a great leveller if they’re consistently good.

None of this is a reason not to try. “This terrible problem in the world is difficult to fix” is always reason to work harder at getting it right; never to give up and condemn billions of people to the flames of withering injustice.

This entry was posted in I was wrong and tagged , , , , , , . Bookmark the permalink.