Posted by: CJ | December 1, 2009

I want a mechanism

I’ve been thinking about models and mechanisms recently because of behavioral genetics/evolutionary psychology. Mostly about how much I dislike pop genetics wedded to pop psychology. When I bother to coherently think about that hate, I have two big objections to those fields. And they’re common objections, so I claim no originality. The first is I don’t know why they’re asking the questions because it’s unclear to me what good the answers do. The second is that I don’t think their methods constitute science.

For the first issue, relevance, I just don’t get it. Suppose you do prove, definitively and conclusively, that some hot-button trait is gender-based, or race-based, or even just genetic. What do you do about it? If the methods used lack enough predictive power to be able to predict individual outcomes, then the information is only potentially useful for guiding policy in large institutions. Like government or large businesses. But as a nation we’re committed to equal opportunity, at least as much as possible. (Or, at the very least, no politician is going to run on a platform of screwing over those scientifically deemed incapable from birth.) So what’s the point? On the other hand, if the methods can predict individual outcomes, what would you do about it? If it turns out people from lower socioeconomic classes have children who are genetically incapable of performing truly great academic or artistic feats, do we deny them opportunities in the name of saving resources? The answer is still no, we don’t. Do we put more resources into their education in the hopes of changing the outcome? If that was a viable economic and political strategy then inner-city schools would not be so chronically underfunded. So I don’t get why anyone would care about the answers. They don’t help policy. (And this article at the NYTimes suggests test with individual level predictive power are unlikely, at least anytime soon.)

Of course, there’s also the potential answer that they hope understanding the causes could lead to cures. Or something. For treatment of diseases that would be wonderful, but that’s not what I hear about the most. Instead I hear people using “the latest science” to make their bigotry socially acceptable. The other possibility is genetic tampering with infants, which is, if anything, even more of a hot-button issue than behavioral genetics.

Now for the science aspect. In an older post I spent awhile wondering how I could always want a mechanism and yet not reject Newton. The answer I ultimately came to, based on Paul Gowder’s comment and more thinking, is that even though Newton lacked a specific mechanism there was an overarching model of physical reality that made a mechanism less pressing. Specifically, classical physics operates under the assumption that physical laws are universal in scope, unchanging, and deterministic. So one well-done experiment is equivalent to all possible experiments under all possibe conditions at all possible times. And casuality is much less slippery.

In contrast, the models in behavioral genetics and evolutionary psychology are much weaker. They are not deterministic. There are things the models do not claim to understand. (So when the numbers are crunched there are errors, things outside of the explanatory scope of the model.) They can make claims the laws their models embody are unchanging, but it’s very unclear that such laws really are. And, on top of all of that, there’s not usually a coherent mechanism. It’s just genetics. Or something. But the lack of a concrete mechanism makes it unclear what exactly the models claim. Just as important, to my mind, is the notion of error in the models. Unlike quantum mechanics, where if something is random then it’s truly random and outside the scope of potential human understanding, these models do not preclude understand the errors. So if someone deviates from what their genetics would predict, it’s entirely possible that you can understand why and explain it. It just requires looking at it from a different perspective, outside the genetic information the model uses.

All of which is to say I’m very unconvinced modern science has anything useful to say about gender, race, or genetic differences causing behavior differences in between two specific individuals. Maybe, MAYBE, between two specific large populations. But not between two specific individuals. And that distinction is so very sensitive, and very nuanced, that I’m very skeptical of genetic & biological approaches to understanding behavior in general.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


%d bloggers like this: