Cass Sunstein has an interesting post about what happened in Boulder and Colorado Springs when citizens got together in each city to deliberate about three controversial issues: affirmative action, a treaty to control global warming, and same-sex civil unions.
Basically, deliberation decreased diversity among the participants, and it increased their extremism. Citizens in Boulder became more liberal, and citizens in Colorado Springs became more conservative.
(Note: I read Sunstein's post to distract me from the paper I'm trying to write about Wendell Berry and liberalism, so I'm sure that's why the post made me think of . . . liberalism!)
These worrisome-at-first-glance results force us to interrogate our faith that deliberation is always a good thing. Are moderate positions at a deliberative disadvantage in comparison to extreme positions? Do people's pre-deliberation beliefs sometimes make deliberations more or less valuable? Is the "best" democratic political outcome more likely if we deliberate before voting, or if we're just polled without a lot of discussion amongst ourselves?
We still need a lot of clarification about just what the experiment measured (E.g., how was "extremism" defined?) Already though, it prods us to think about the "ought" and the "is" of liberal democratic theory. We sometimes talk about liberalism, in which autonomous agents deliberate with each other in freely-chosen associations, as both an idealized picture of the way society ought to be, and as a description of the way real democracies actually operate. We ought to act this way (some liberals say) because it's the best way we have of finding the most legitimate or the "truest" answers without violence and bloodshed, and we in fact operate this way (some other liberals say) because citizens actually do limit their deliberative appeals to "public reasons."
If the deliberation in this experiment were indeed of the liberal sort, does the fact that they promoted extremism count as an argument against the value of liberal deliberation? If these deliberations didn't conform to the liberal ideal, how anomalous are they? Does this experiment suggest that we simply don't deliberate reasonably as often as some liberals think we do?
(Ok; back to Wendell Berry...)