On Sept. 30, David Bernstein of George Mason U. Law had this article in the Wall Street Journal (subscription) bemoaning that "despite far-reaching reforms, junk science still plagues American courtrooms." Prof. Bernstein considers the standard of Daubert v. Merrell Dow Pharmaceuticals, 509 U.S. 579 (1993) to be a "strict test" which results in the exclusion of junk science allowed in by the tests used in many states, including most of the largest states.

In today's WSJ, there is a responsive letter by Elliott Foucar, M.D. of Albuquerque noting that the problem is not so much the test used but rather that scientific judgments are being made by judges who very often lack even the most minimal qualifications for making them.

I am inclined to agree with Dr. Foucar. The decline of core curricula in America's colleges means that people can often earn bachelor's degrees without even an introductory course in the sciences, and then they go on to law school, become lawyers, and become judges without any exposure to science other than what they pick up in the course of their cases. A motion to exclude scientific evidence often means a crash course for the judge, but no matter how diligent or dedicated the judge may be in this self-education, a crash course can never be a substitute for years of study. Dr. Foucar suggests that judges consult unbiased experts of their own, rather than relying on the competing testimony of experts-for-hire, and that professional boards scrutinize the testimony of the hired experts for ethical violations.

Those steps would be worthwhile, but I would go a step further. Trial judges should be able to certify scientific questions of evidence to a special science court, composed of a judge and several scientists who are knowledgeable in the general field but noncombatants in the particular controversy. Let scientists who propose new techniques and claim wonderful results for them testify before a panel of people who really understand the science, and let them decide what is reliable enough for proof in a court of law, using a strict standard.