| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
LETTER |
Correspondence: Requests for reprints should be sent to Gio Batta Gori, DSc, MPH, The Health Policy Center, 6704 Barr Rd, Bethesda, MD 20816 (e-mail: gorigb{at}msn.com).
A sea of ink in a recent Journal supplement1 contains much criticism of the Supreme Court’s Daubert opinion. Most of this criticism is based on the perception that the opinion imposes unattainable scientific standards of absolute proof that would impede justice. Yet Daubert aims only to ensure that testimony proffered as scientific is not raw ipse dixit opinion. It does so by offering 2 primary criteria: that an issue be testable by the scientific method and that an error rate be stated. Corollary but not essential questions refer to scientific consensus and peer-reviewed publication. Given further that Daubert ’s guidelines are not binding on the courts, and that science is not about absolute truths but about accruing evidence, what could be so vexatious about such elementary criteria? The trouble is that Daubert frustrates those who follow a millenary and artful tradition of posturing and gives preference to the fact-finding model of the scientific method.
If Daubert appears wanting, it is because it misses clarifying what the scientific method is all about. In fact, a clear statement of the method is difficult to find, because philosophers have muddled its definition to the point that the National Academy itself gives only an opaque account, despite claiming the method is intuitive.2 And intuitive it is, for to be of general application the method can only address the evidentiary logic of any and all experiments, and not the manifold complexities of their execution and underlying theories. A modest proposal for a concise statement of the scientific method is that any otherwise ethical scientist, regardless of the theoretical and material intricacies of an experiment, ultimately seeks to warrant (1) the authentic identity of what is being measured, and that the error of measurement is sufficiently small; (2) that circumstantial variables unrelated to the experiment are accounted for and do not interfere with the outcome; and (3) that the outcome is reliably reproducible.
Scientists would agree to some flexibility in these requirements, but they would also agree that outcomes are not validated and remain hypothetical if these requirements have not been met. Scientists would also grant that what is considered valid today may be superseded tomorrow. Still, rational action can be driven only by today’s evidence, and therefore those simple criteria would complement Daubert in assisting the courts, which must decide there and then what may be admissible evidence. They cannot wait for tomorrow. To reject the transparent fairness of these criteria would not speak well of motives and intentions. Indeed, by any reckoning, Daubert is an advancement of justice and of fairness in civilized discourse. Pace its opponents, Daubert is here to stay.
References
1. Am J Public Health. 2005;95(suppl 1).
2. Responsible Science. Ensuring the Integrity of the Research Process. Vol 1. National Academies Press, Washington, DC. 1992. Also available at: http://www.nap.edu/catalog/1864.html. Accessed October 14, 2005.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
