Sua Sponte: A Judge Comments

No one would disagree that “junk science” is bad and real science is good. But how do those simplifications translate into courtroom practice? Daubert drafted trial judges into a corps of untrained, ill-equipped, frontline soldiers for scientific reliability. As Justice Rehnquist presciently bemoaned in dissent, Daubert unnecessarily foisted on judges “the obligation or the authority to become amateur scientists.” Daubert v. Merrell Dow Pharm., Inc., 509 U.S. 579, 600–601 (1993). Because critical evaluation of scientific ideas is not within most judges’ intellectual wheelhouse, it’s unsurprising that judges “do science” inadequately. The fault, though, does not lie solely with my colleagues. By erecting the 11 “reliability criteria” Dr. Lorandos identifies, the trilogy has so complicated the process of sorting the junk from the sound that it’s no wonder judges fall back on personal predilections or other unscientific motivators. Yes, the criteria are supposed to be illustrative, not determinative. Often, however, judges pick and choose the governing factors only after deciding which experts to believe. And that appellate courts review these decisions only for an abuse of discretion increases the likelihood of inconsistent, arbitrary, and just plain wrong-headed decision making.

Unlike the trilogy, science values simplicity. Perhaps apocryphally, Albert Einstein is said to have advocated that “a scientific theory should be as simple as possible, but no simpler.” The trilogy adopts the opposite approach, compelling judges to consider complex inquiries ranging from an expert’s methodology, to the validity of testing data, to the content of peer-reviewed literature. Lawyers typically address the first with hired guns and the last with a document dump, filing hundreds of pages of literature for the judge to read, comprehend, and insightfully fold into an enlightened opinion. Who knows which tentacle of the trilogy’s multipart test a judge will grab and hold?

In a scientific study, too many variables may spoil the soup, risking results that are imprecise or erroneous. Yet, the trilogy encourages judges to employ junk variables in rooting out junk science. For example, in a medical malpractice case involving a surgical error, why should a judge concern herself with an absence of controlled experiments when such experiments would be ethically forbidden? Why insist on epidemiological studies where the numbers of an affected population are not large enough to justify the costs? Because the trilogy constructs a juridical gate with so many different locks, judges lose sight of the big picture: Is the expert’s testimony relevant, helpful, and reasonably well grounded in science? If so, let the jury decide who and what to believe.

After saddling trial judges with the role of scientific inquisitors, the trilogy commands that appellate courts review their decisions most deferentially. The trial court may have gotten large pieces of its analysis wrong, but we must uphold a decision resting on some discernible logic. This means that conflicting admissibility conclusions reached by two judges in the same courthouse should be affirmed by the same appellate panel, as long as neither decision reeks of blatant error. Good science, it seems, is more evanescent than tangible. Most appellate judges share Dr. Lorandos’s view that judges are woefully unschooled in science. We may be experts in the rule against hearsay, but when it comes to p-values, epidemiology, and tumor biology—not so much. And the fix does not include a standard of review that immunizes all but the most egregiously incorrect decisions.

When it comes to forensics, Dr. Lorandos hits the nail on the head. Nowhere is scientific integrity more critical than in criminal cases, where lives and futures hang in the balance. Yet, lawyers and judges who practice in that arena routinely assume that the state’s science is valid. Judges are supposed to be gatekeepers, but when counsel provide no hints that an expert’s testimony is unreliable, the predisposition to unquestioningly admit the prosecution’s evidence is strong. The resource disparity baked into the system partially accounts for this reliability vacuum. By the time a case reaches appeal, it’s too late for a panel of skeptical judges to insist sua sponte on a Daubert hearing. Perhaps one remedy is to mandate that the prosecution demonstrate forensic reliability in all cases well in advance of trial, and to allocate funding for defense challenges. By highlighting the flawed scientific evidence historically and currently tainting our system of justice, Dr. Lorandos has opened the gate to an important discussion of much-needed reform of the trilogy.