Thursday 29 January 2015

Maclean's Racism Story: Am I a racist?

I'm finally getting around to this story. The Maclean's article: Welcome to Winnipeg: Where Canada's racism problem is at its worst, has pretty much gone viral and has spawned corrections, responses, counter-responses, and so on. I'm going to skip over most of that. Going back to the original article I'll outline my opinions on it and the race conversation in it's entirety.

First, the reason it took me so long to get to this article is that this entire conversation has long been uninteresting to me. Not because racism isn't an issue in Canada, or is an important issue to me, but because the issue is stale. I feel like I've been on all sides of all arguments, and I just don't see any tangible progress being made. When all options are on the table, and a real honest conversation is taking place, I'd happily take an active role in the process. That time is not now. Articles like this help reinforce that.

Maybe I'm mis-defining what stereotyping means, but I use it as a 'judge a book by it's cover' type of term. Now, I've never understood why this term carries a negative connotation. If you go into a book store. Do you grab a book at random, read the entire thing through, then decide if you should buy it? No, you gather as much information as you can from the covers and form a judgements of their viability.

Based on those judgements of that cover, you might be able to make a selection. But if your requirements are more stringent, you take the information you've gathered and flip open the jacket and read the summary. Perhaps you read the jacket reviews, or perhaps you quickly flip to amazon and read some customer reviews. The point is, you gather information every step of the way to a level sufficient enough to fill your criteria. There is no objective 'right' level of scrutiny required, but simply, a subjective 'right' level given time and personal preferences.

My point is that useful information can, and should be made, when little knowledge is at hand. An honest discussion about it's use and how to improve upon how we upgrade our priors would be much more useful than simply calling it racism and wiping our hands of it.

On to the article.

My first problem: The use of polling 

I clicked through the CIIM link, but after three clicks of trying to find the study that the figures were drawn from, I lost interest. How do you rely on figures and not make them easy to access?

Citing that, 'one in three Prairie residents believe that "many racial stereotypes are accurate" or that, '52 per cent of Prairie residents agree that Aboriginals' economic problems are "mainly their fault."' as proof of racism seems to carry a pretty significant explanation. We don't get it.

Let's consider racial stereotypes. Thought experiment: random white person vs. random asian person; math contest... who would you bet on? (full disclosure: as a half white and half asian person I can make this comparison). One more? Random white person vs. random asian person; driving contest... who do you bet on?

Now, the failure to update your bet upon acquiring further information, may stem from prejudice or from poor gambling skills, whatever motivates that that decision (to me) would determine if a person is racism or not. But to suggest that holding an opinion based racial identity as 'racist', kind of dilutes the term to the point of being worthless. This is where I believe discussions find themselves now.

If someone called me up and said: do you think most racial stereo types are true, I'd say, yes I'm a pretty effective stereo typer. Am I racist? Depends how you define it. I make a pretty qualitative difference, and as a good bayesian, I update my priors as often as possible and as objectively as possible. Simple race based information is largely useless next to good info.

So, perhaps asking about what the author believes to be racial stereotypes, specifically would be useful. And then asking what how that stereotype impacts their view of a random person would be useful. But a vague yes or no to a vague question about a vague topic? Useless (this of course assumes that is how the data was collected, which the article didn't really present readily, and in the internet age, that seems inexcusable).

In terms of the 'their fault' result. First, what are Aboriginals' economic problems? I really hope a more specific set of questions were asked, but again, I don't care enough to dig through website links. Likely, this pushes another vague concept: 'economic', and reports a yes or no answer.

Life is a sum of our choices. Not graduating high school? A choice. Not going to University? A choice. Doing illegal activities and running the risk of a criminal record? A choice. These are all significant factors in determining economic outcomes. I'd answer yes, it is 'mainly their fault' with the caveat that these choices are ultimately up to the individual but they stem from a complex web of interconnected occurrences both involving individual choices and non-individual surroundings and influences. Those surroundings, largely out of the control of the individual will often pre-dispose them towards making choices, if they don't maximize the probability of economic successes then the blame does not solely fall on the individual. Ultimately, a choice is a choice. But there is a lot going on in those choices. Again though, the implication is I'm a racist.

I'll even leave the 'hearing negative comment' poll alone (did they ask about intent?).

My Second Problem: Causation vs. Correlation

Drawing causal inferences from observed outcomes is difficult in almost any circumstances. That's why one handed economists are so often longed for (... but on the other hand). But that exists for a reason. Simply taking an out come, and observation, seeing the correlation and chalking the observation up as the causal factor is troubling even with a decently flushed out theoretical underpinning.

This article takes as given that 'racism' is both an observable and casual factor in social and economic outcomes without explaining the linkages. I'm not even saying it isn't a factor here, I'm saying that it's a lazy and uninteresting way to make an argument.

I remember back during my undergrad hockey playing days (at St. Lawrence University) one of the deans was giving the atheletes a lecture on racial sensitivity and, using the hockey team as an example, said (paraphrased): as a black man, if I were on the hockey team, I'd have to wonder where the diversity is.

Now, my brother (also on the team) and I are very much visible minorities and were both laughing. The implication that the outcome: no black folks on the team. Equated to a bias: against black hockey players. Seemed laughable to us then, and still to me now. Do you know how many black kids played hockey in Canada when I was growing up, probably about the same as there was asian kids (like myself, who was often 'stereotyped?' as being aboriginal). You know how many black NCAA hockey players there were when I was in college? Probably about the same as there was asian NCAA hockey players when I was in college. IS that racism? Or were there more good white hockey players to choose from? Is it racism to have the people best suited for an outcome, get that outcome? Depends how you define it I suppose.

When it comes to more serious issues like health, education, criminal, and economic outcomes the implication that correlation equates causality along these simple racial lines isn't laughable, but it should be met with a pretty healthy dose of skepticism and context.

Murdered teens, elevated incarceration rates, elevated poverty rates, these are outcomes that are used to show us that racism persists. For racism to be the cause shouldn't we find a better control group? What about other races with comparable levels of education, from comparably broken homes, and people in as high of risk factor settings.

Certainly, proportionally aboriginals find themselves in higher risk situations, but addressing these situations instead of assessing how racist the outcomes are seems like a different conversation all together. Am I mistaken here?

The author clearly states that a number of risk factors are more prevelant in the aboriginal communities, but doesn't offer an explanation about why it's simply 'racism' to blame.

If the risk of getting murdered as a teenage girl goes up in the presence of: growing up in a broken home, travelling alone in a city, substance abuse (self inflicted and the presence of it), and sex trafficking, is race the dominant factor? My assumptions are clear, that these factors do increase the risk of ending up murdered. But to conclude that race is the dominant factor you'd have to assume that race is the primary cause of these factors. I'm not even saying that this is false. I'm just saying that it should have been explored in this article.

Were these outcomes perpetrated on the victims by people with racist intent? I haven't heard if racially motivated people perpetrated the murders, abuses, or neglect. If that is so, I am all for addressing that. This article doesn't flush any of that out.

Instead the author observes the correlation (not causation) between race and these risk factors. Then points to these risk factors and the victim. Then draws the line between race and the victim; without justifying the other steps. Outcome. Correlation. Conclusion. Outcome. Correlation. Conclusion.

It just seems lazy.


Look, Racism is a despicable act that needs to be addressed whenever, and wherever it rears up. But I don't think that outcomes falling disproportionately on a race equate racism. Racism, to me, is when opinions, beliefs, and MOST importantly actions and policy can only be justified on racial terms. I abhor this absolutely (or sex, or religion, or sexual orientation) but that isn't what the author is after.

If school funding is the problem (as implied, but my knowledge on the topic is basically nil), go after whoever is in charge of that and insist that they justify it without using race (is it net tax flows? is there something in the treaties?). If they can't? It might be racism.

If native deaths are being neglected, go after the investigators and insist that they justify it without using race (neglect across the board? higher incidence rate?). If they can't? It might be racism.

Is this unreasonable? I don't know, it doesn't seem so to me.

Policy addressing these 'racial' issues need not be justified in racial terms. If you're disadvantaged, we'll help. If you're in a high risk situation, we'll help. Income redistribution? I'm actually for it (despite my largely libertarian bend), but I don't understand why race need be involved in that discussion. Same rules for everyone makes sense to me.

If you're comparing social outcomes of two groups wouldn't the obvious place to start looking be how each group is represented legally and making them the same? I don't get it, and I'd love for someone to fill me in. This article doesn't do that.

And this is largely why I simply abstain from this whole mess of a conversation, and why nothing will change. I'm usually told it's a legal issue with treaties and all that good stuff. That may very well be so. I don't know. I don't care. The whole conversation shifts from looking for the best possible outcome, to justifying ones priors. I'll leave that to politicos and the lawyers. I don't find that interesting. So I move on.

The bottom line is that there is a huge discrepancy in outcomes that I'd love to play a part in addressing. I just don't think all of the options are on the table. They aren't. But broadly painting outcome differences as simply 'racist' doesn't seem helpful. Writing articles as sloppy as this doesn't seem helpful either. Everyone gets to reaffirm their priors and the conversation sits exactly where it's been.

No comments:

Post a Comment