Thursday, March 7, 2013

Reweighting the Pacifist's Dilemma Table

Has any group of people gotten together to figure out ways to reweight the quadrants of Steven Pinker's Pacifist's Dilemma grid from The Better Angels of Our Nature?

In the final chapter of the book, Pinker introduces what he calls the Pacifist's Dilemma. It's a twist on the classic Prisoner's Dilemma from game theory, in which two prisoners are set up to either compete or cooperate. If Prisoner A sells out Prisoner B, A goes free while B serves a year (and vice versa). If both confess, they each serve three months. If both stay silent, they each serve only one month. So clearly, they would be best off if both stayed silent, but the temptation to defect is strong. The dilemma is usually shown as a four-quadrant table like this:

In the Pacifist's Dilemma, the four comparable quadrants are labeled as follows:

As you can see, the penalty for pacifism in the face of aggression is extreme (-100), much less than the one for meeting aggression with aggression (-50). And the "costs to the victim (-100) are vastly disproportionate to the benefits to the aggressor (10)" (p. 679).

The question is, How can the basic assumptions of the Pacifist's Dilemma be changed or reweighted to make peace and prosperity a more likely outcome? This is what Pinker argues has been happening over the course of history, resulting in the decline in violence.

This seems like a job for public policy, so I'm thinking of a gathering of policy wonks and behavioral economists, plus Stewart Brand, Maggie Koerth-Baker, and Steven Johnson. They'll have to remember it all takes place in a context of global warming and peak oil.

It's the biggest of big picture questions. How do we start answering it?

No comments: