« Variable Transformation and Correct Parameter Estimations | Main | Guest Blogger: Richard Lempert »

04 August 2006


William Ford


My question at the end doesn't really have any kind of clear answer, but your question about how often a court would feel pressured to accept the empirical evidence sounds testable.

While it would be tough with judges (the ideal), rooms full of law students are easy to come by. Find out their prior views, provide the controlling legal standard, and then provide various amounts of evidence. Maybe have three different versions of the same quasi-fictional study, one with strong findings, one with weak findings, and one in the middle. See whose conclusions change.

It would be interesting to see how many people can be dislodged from their initial views in a case like Entertainment Software Association and how much evidence it takes to do so. (There must already be some studies of this sort?)


Rob Robinson

This doesn't really answer your question, but my sense is that courts are in the habit of creating standards for empirical evidence in constitutional cases that are so slanted in one direction or another that the result is almost foreordained. I'm thinking right now of the secondary effects cases, where despite some really terrible studies, the Court was quite willing to accept the findings. On more traditional First Amendment questions, though, I think the complexity of the relationships, the number of variables involved, and the difficulty in showing causation will always give unwilling judges enough cover to stay that the state has not met its burden.

This case also makes me think about David Faigman's piece "NORMATIVE CONSTITUTIONAL FACT-FINDING": EXPLORING THE EMPIRICAL COMPONENT OF CONSTITUTIONAL INTERPRETATION," 139 U. Pa. L. Rev. 541 (1991), where he makes a good case that the Supreme Court, when faced with pretty solid empirical evidence in _McClesky v. Kemp_ that murderers in Georgia who killed whites were far more likely to get the death penalty, changed the standard of what evidence was needed to show discrimination.

My distilled point (which applies to Daubert as well): how often would a court ever feel pressured to accept empirical evidence it does not want to accept? While the initial question--what evidence should be good enough--is an interesting one, I think this question is the more urgent one.

The comments to this entry are closed.


July 2014

Sun Mon Tue Wed Thu Fri Sat
    1 2 3 4 5
6 7 8 9 10 11 12
13 14 15 16 17 18 19
20 21 22 23 24 25 26
27 28 29 30 31    

Site Meter

Creative Commons License

  • Creative Commons License
Blog powered by Typepad