« Do Rankings Matter? | Main | "Yet the nature and effect of legal doctrine has been woefully understudied." »

21 April 2006


Rob Robinson

Judicial education (and publication of works such as the Federal Reference Manual on Scientific Evidence) are a good start. Even better: require some basic degree of scientific and methodological training (and i do mean basic) in law school.


I am a clinical neuropsychologist who regularly performs forensic consulting. I sometimes consult on Daubert challenges as they apply to psychological evidence. It must be difficult for judge to sift through scientific and technical evidence from so many different fields. The difficulty is the law has a completely different epistemology than that of scientist-practioners. Stare decisis for example is completely alien to scientists who spend their careers trying to severely falsify precedent. The evidentiary thresholds also differ. Scientists ask what variables have the strongest associations while judges only ask what evidence is probative. Under this scenario, two psychological tests used by opposing experts could have significant but still different validity coefficients, but even though one is scientifically more rigrous than the other, both would be admissible as probative under current FRE.

Having both sides agree on an expert rarely works in an adversarial system. In my experience, the risk is more wishy-washy experts get onto these lists, a recipe for more ambiguity. The best scientist-practitioners usually have strong points of view on specific issues based on multiple sources of evidence. Such strong positions almost always run contrary to one party's interests. Agreement on experts also strikes me as a legal posititvism theme, but I could be wrong on this.

I say give judges better generalist training in basic scienitifc reasoning or principles, especially in areas that frequently come up, let's say in epidemiology. Otherwise, continue to let them rule on borader probative issues. A special master doesn't sound like a bad idea either.

Jeremy Buchman

Defining the jurisdiction of specialized courts would appear problematic, if we're talking broadly about cases where Daubert and its progeny might be implicated, as opposed to specific subject matter. I'd prefer solutions that equip generalist judges to grapple with scientific and technical evidence.

Your suggestion concerning court-appointed expert witnesses has much to recommend it, but its impact will depend on whether judges choose to avail themselves of this option, which depends on whether judges believe that outside expertise will help them decide cases. Knowing why judges have been reluctant to appoint outside personnel would help here. Are they concerned that frequent appointments will be at odds with their beliefs as to the appropriate judicial role? Are they concerned that grappling with complex evidence, either directly or through auxiliary personnel, will leave them vulnerable to reversal on appeal? Do they feel that they can settle the legal issues without directly assessing the scientific or technical evidence? I anticipate that Ed Cheng's article will shed some light on this question.)

Expert witnesses would seem to have one important advantage over research done by judges or their clerks: expertise not only in scientific or technical methods, but also in particular application of those methods. Judges' clerks might have training in statistical methods writ large, but they are unlikely to understand how particular techniques or methodologies might translate into a specific discipline or issue area. Drawing on experts on a need-to-know basis allows for both kinds of knowledge to be captured, and giving both parties a say in their selection will provide a valuable check on experts with their own agendas.

The comments to this entry are closed.


December 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