I wanted to pose a question that relates directly to the issue of empirical work and the legal profession. Cases are increasingly comprised of technical and scientific issues. Judges are assigned the difficult task of determining whether evidence of this nature should be admissible. Although Daubert provides greater guidance for judges in evaluating such evidence, the judge is still left with the unenviable - and largely unreviewable - task of make this determination. And in many cases, the admission/exclusion of evidence largely determines the substantive outcome (e.g., toxic torts, voting rights, medical malpractice).
Judges are limited in how they can address deficiencies in their inability to evaluate such evidence. They can hire special masters, but evidence suggests that they rarely exercise this option. Judge can engage in independent research, but it is arguably contrary to the adversarial system to do so. [By the way, for an excellent article describing these issues and drawing upon an original interviews with state judges, see Ed Cheng's INDEPENDENT JUDICIAL RESEARCH IN THE DAUBERT AGE, forthcoming in Duke Law Journal].
Thus, I am interested in your thoughts as to what, if anything, the judicial system should do to address this issue? Should expert witnesses be drawn from a pool of experts common to both parties? Should judges be allowed to engage in independent research (with disclosure to both parties)? Should there be specialized courts?