June 17, 2009

Cass Sunstein, Extremism, and Judicial Panels

My ex-thesis advisor Cass Sunstein (ex because he got hired away by Obama in January) has a new book out – Going to Extremes: How Like Minds Unite and Divide – and there’s a pretty good review in Slate. I haven’t read the book, but it looks like the gist of it is similar to a paper he published this winter with Ed Glaeser: when groups of like-minded people discuss an issue together, they tend to come to a more extreme conclusion than they would have reached individually. Glaeser and Sunstein call this “Credulous Bayesianism,” reasoning that people in such groups draw support from their peers’ views, but don’t account for the fact that their peers may all be biased in the same direction.
I was interested to see that Sunstein seems to be extending his reasoning to heterogenous groups too: “If you bring the two clashing sides together, they don't find middle ground any more than like-minded people do,” Christoper Caldwell writes in the review. “Each side digs in.” That would seem to go against the grain of Sunstein’s work on three-judge federal appeals panels, which I wrote my thesis on. The very fact that Democratic or Republican appointees “cast more ‘extreme’ votes when they are in the majority than when they are not” means that judges who hold a minority view on a heterogenous panel are more likely to join the majority opinion anyway than to write a dissent. That doesn’t mean they actually change their minds, but it does mean that they go along with the majority rather than “digging in.”
Judicial panels aren’t an archetypal “deliberative group,” of course: the norm of judicial collegiality discourages dissents (as does the sheer number of cases appellate judges decide). Furthermore, dissenting judges can usually support their position with prior precedents, so the skewed dynamic of Credulous Bayesianism can’t really operate. The book seems to tie together a bunch of different studies beyond judicial panels (by Sunstein and others), and I’ll be interested to see how he connects the dots to make his overall points.

No comments:

Post a Comment

Note: Only a member of this blog may post a comment.