Error Deflection

Because it’s come up a few times, and because I feel like a more life-based post is in order, I’d like to hit on a topic that’s come up before: even and especially when making especially important decisions in politics, or the law, how do we choose between two courses of action?

The concept I use to address some of these questions is “error deflection,” a term borrowed from one of my favorite law school professors. It starts by accepting that we act (always) on the basis of imperfect information, and that, especially with hard decisions, mistakes will often result. (“Bad facts make bad law.”)

Accepting the possibility of error, we change the question from “what should I do?” to, “accepting that I may be wrong, how would I like to be wrong?” Mistakes often have consequences, but they’ll always have different consequences. Choices effect results, in success or in error.

This has the effect of changing a question from one about possibilities, to one about values. For example, many choices are intrinsically about risk: do you preserve the status quo, or try to make it better? Error deflection in these cases is all about you. Do you take chances to build a better world? Or is what you have worth defending, and too valuable to potentially lose?

Others, especially in politics, reduce to questions about worldview. The presumption of innocence is nothing more than a conscious choice to always deflect the risk of error in a verdict towards liberty over safety. If we’re going to be wrong, we’d rather free a murderer than jail Valjean. Similarly, the liberal case in national security chooses to deflect error towards the open society. We’d rather take a 1% risk of a terrorist attack than accept the fact (or, 100% likelihood) of a society that profiles on the basis of race. And we don’t believe the ethical calculus of the presumption of innocence alters just because the magnitudes of risk increase across the board. True, freeing a terrorist is a horrifying prospect; but torturing an innocent man, and keeping him from his family until the “cessation of hostilities” is pretty bad too.

The concept works especially well with murky probabilities. Pretend that anthropogenic global warming is a 50% theory — it’s equally likely, in other words, that mankind is or isn’t affecting the climate. We shouldn’t be paralyzed by indecision, because the question of whether to do something about it, once recast, is easy. A 50% chance of human annihilation is worse than a 50% chance of trying to avoid it and failing, no matter how much it costs Exxon. (This is the case I made a few years ago. I’m just updating it for new readers.)

And, error deflection can be romantic! Should you call the girl? Well, it depends. Would you rather know, or always wonder? (And here’s a song all about error deflection: Fires in France, “Love is Strong.”)

Finally, what about the cases where success is impossible? Well, if you have nothing to lose, you might as well try. Sometimes you have to roll the hard six.

(Photo credit to this person.)

%d bloggers like this: