Monday, July 17, 2006

Idealistic Pragmatist: Women in Parliament: unvoiced assumptions

Idealistic Pragmatist argues that women should have more representation in Parliament (Idealistic Pragmatist: Women in Parliament: unvoiced assumptions); strictly, the argument could be run for any minority group, so I'll just generalize over it. Here's the argument:
  1. "a successful representative democracy isn't just about choosing the best person in each individual constituency, but about making sure the resulting legislative body actually comes close to reflecting the makeup of the Canadian population."
  2. "a group of white men aren't going to be able to make adequate laws for a diverse population of women and minorities because they won't be able to wrap their minds around everybody's issues."
  3. (2) is true because "we all have biases we don't even know we have, and ... these biases affect the decisions we make, no matter how well-meaning we are."
  4. Therefore, Parliament should have more minority MPs.
I'll first note that (1) never actually gets defended in the post; however, (2) on its own is enough to make the conclusion follow (inductively, of course), so I'm happy to leave that aside. Is (3) actually enough to make (2) true? Unfortunately not.

It is certainly true that "we all have biases we don't even know we have, and these biases affect the decision we make, no matter how well-meaning we are." Cogsci has been riding that horse for years (although Kant got there first). But (2) doesn't follow from (3). (3) says that we have biases; (2) however goes further and says that these biases cannot be overcome. And it's the claim that the biases cannot be overcome that needs to be established -- I've never seen an argument for that which actually works.

Here's a possible argument (that doesn't work) to show the problem:
  1. We have biases naturally occurring as part of our mental make-up.
  2. Anything that occurs naturally (either inborn or developed) cannot be overcome.
  3. Therefore, these biases that are part of our mental make-up cannot be overcome.
This is deductively valid, but (2) here is blatantly false. I have naturally (through both inborn and developmental factors) no ability to lift a two hundred pound weight for more than a few seconds, at best. I could, however, affect my nature through training (or chemical enhancement) to give myself the ability to lift a two hundred pound weight for much more than a few seconds. Perhaps I shouldn't, but I could. I don't see why the same can't hold for the biases identified by cogsci: if we know they exist, and we can figure out how to counter them, then it follows that we can overcome them. So, assuming that there are biases against minorities (not unreasonable, and an empirical question), the appropriate thing to do is to look for ways to overcome them, not ways to cater to them.

Edit: Typo.


Idealistic Pragmatist said...

2) however goes further and says that these biases cannot be overcome.

That's not what I said at all; in fact, in the last long paragraph I say the exact opposite.

What I would claim, though, is that overcoming one's biases takes a great deal of effort (for evidence of that, I'd read through the long Washington Post Magazine article I linked to). You first have to come to realize that you have one (and if we all have biases we *don't know we have*, that in and of itself can be an ordeal), and then you have to do the work to overcome them. It would seem pretty unreasonable to expect that anyone--from your average Joe to your MP--is going to be able to take the time necessary to overcome every single bias they have. Therefore, any legislative body is simply going to be more likely to produce laws that work for everyone if it comes closer to being as diverse as the population.

ADHR said...

I'm sorry, but I don't see where in that paragraph you say any such thing. That paragraph seems to be about whether it's fair to expect people to overcome their biases, not whether the biases actually can be overcome. Are you referring to the example of the man whose wife is a Haitian immigrant? Certainly that suggests that some biases could be overcome, but it doesn't say that all biases could be overcome. Indeed, your comment now suggests that you think there is a class of biases that cannot be overcome, namely those that one is not aware of. (That does, however, defeat my interpretation of your point as "biases cannot be overcome"; I should have said that some cannot be overcome.)

Fair enough, but, then again, it's generally unfair to expect people to overcome errors that they are unaware they have made. Furthermore, if the individual who is supposedly biased has no idea that they have a bias, what evidence could there be that the bias exists? Surely evidence found through some third-party observation of this individual. But, if this were so, why wouldn't this information be made available to the individual? That is, any bias that cannot be known by the individual (and thus not overcome) must be a bias that is not knowable by anyone; and that there are such biases can never be shown.

So, putting that aside, let's accept that all the biases we're talking about as relevant to MPs are biases that could be overcome. You're suggesting that it's very difficult to overcome these biases. (I've actually read some of the cogsci literature on the subject, so I'm not too worried about what the WaPo says!) I'd agree that it certainly could be. I don't see why it's unreasonable, though, to expect that people who are applying for the job of helping to run the country should overcome at least the biases relevant to that position. (For example, if an MP is innately biased against foods that smell like popcorn, I don't think that's a bias that needs to be overcome.)

I also don't see what you mean by "laws that work for everyone". I would think that the purpose of the legislature is to produce good laws, as free from relevant bias as they can be. And I'm not convinced that a legislature composed of people who have worked to overcome their biases (and are open to overcoming more that might be revealed) is not better than a legislature whose composition seems to reflect the population's.

I should add (didn't put it in the post as it's a difficult general problem) that there's a general problem with the notion of "reflecting the population". Namely the problem of identifying which groups should be reflected, and which should not. The usual candidates are ethnic groups, religious groups, women, non-heterosexuals, and the disabled. The first problem is that these can cross-pollinate: there aren't just Barbadians, but female Barbadians, Catholic Barbadians, gay Barbadians, blind Barbadians, female blind Catholic lesbian Barbadians, etc. How much detail must there be in a group specification before it is detailed enough that we can stop and look for a representative member? The second problem is that we can add much more detail: hair colour, eye colour, height, weight (indeed, should obese people have representation in the form of an obese MP?), profession (should physicians be represented proportionally in Parliament), etc. Which are the irrelevant features that people could have, and why?

All in all, I think that the idea of a Parliament whose make-up reflects that of the population is not easier or more reasonable than a Parliament composed of people who have worked (and will continue to work) to overcome their biases (indeed, just the opposite seems to be true, given the difficulties of the reflection problems).