| Yes, Richard Dawkins the famous sociobiologist. Today I was perusing his new book A Devil's Chaplain. One of the essays, available on-line, is called "Trial by Jury".
Dawkins argues that juries are notoriously unreliable, and that the group is swayed by a few opinions. He wonders how much independent juries would agree with each other, if they were allowed to observe the same trial, but not allowed to talk. A good test, Gordon Tullock tells me it once was done (he claims the two juries disagreed one-third of the time, I cannot document this). Dawkins argues we should compare jury trials to judges by using an agreement test. Give two judges (juries) the same trial, don't let them talk, and see how much they agree. The system that yields the greater agreement then ought to be the better system, according to Dawkins. If you don't believe me, he writes: "Whichever system, Trial by Jury or Trial by Judge, yields the higher score of agreements over a number of trials is the better system and might even be accredited for future use with some confidence." This seems wrong to me. Parties that agree more often are not necessarily the better truthseekers. To provide a simple counterexample, let us say that the general citizenry was intolerant and thought that accused defendants were always guilty. The two juries would then agree more than the two judges would, but of course this doesn't mean that juries are the better system. A disagreement means that one of the two parties are wrong, but you don't maximize chance of rightness by looking for the mere fact of agreement. Overall agreement is a good signal about a system when that system is usually right in the first place. Which would mean that the initial problem was not enormous. It is not such a good signal when decisionmakers wear universal and common blinders. I'm all for a critical scrutiny of the jury system, but let's judge it by fair standards. -- Posted by Tyler Cowen to The Volokh Conspiracy at 9/14/2003 08:31:01 PM Powered by Blogger Pro |
- [The Volokh Conspiracy] Richard Dawkins favors judges over jur... Tyler Cowen
- Tyler Cowen
