In 1973, the Ninth Circuit somewhat famously said, in United States v. Barnard, that "the jury is the lie detector." The ghosts of George Orwell and Franz Kafka probably issued a collective sigh of relief. Yes, Barnard only dealt with a court's denial of a defense offer of a psychiatrist and psychologist as "experts" to testify to the credibility of a cooperating prosecution witness. But the likes of Orwell and Kafka would have well known how wide the "expert" doors would have swung open, if the supposed skills that experts -- and the interpretations they might be allowed to offer -- would be permitted to replace the God-given common sense of an impartially-selected jury. Experts, indeed!
We cherish the exclusive province of juries to determine truth or falsity. So much so that when Justice Thomas, writing, in 1998, for the Supreme Court in United States v. Scheffer, upholding the ban on lie detectors in military courts (and thereby, effectively, all federal courts), went way beyond the fact that polygraph had not reached a sufficient state of reliability to be admissible. Justice Thomas went further to say that by its very nature, polygraph evidence may diminish the jury's role in making credibility judgments. The old argument: the jury will give too much weight to the opinions of the polygraph examiner, clothed as he is in scientific expertise, and at times offering a conclusion about the ultimate issue in the trial.
But isn't that a realistic concern? Not for Justice Kennedy, writing for four Concurring Justices. Nor for Justice Stevens writing in dissent. For them, though articulated differently, the "excessive reliance" concern was a "tired" argument that reflects a "distressing lack of confidence in the intelligence of the average American."
So, what does this all mean? It seems to mean that a significant number of those Justices believed -- perhaps the Court's current composition does too -- that as long as the evidence is sufficiently reliable scientifically, we shouldn't worry that the jury will be too influenced by a testifying "scientist," even if that testimony goes to the ultimate issue of guilt or innocence. So, if there were to come a time that lie detector evidence were as reliable as, say, fingerprint or DNA testing -- put aside the "integrity" of the collection process (raised so well in the O. J. Simpson case) -- presumably a strong number of Justices would be quite fine with lie detector experts basically being game, set, match for criminal prosecutions, right?
Bear in mind, if an unimpeachable lie detector expert could testify to his results in a one witness versus another case, determining that the victim was truthful and the accused was lying, or vice versa, he essentially would decide the case instead of the jury. Left for the jury would, then, only be the limited role of pronouncing a verdict.
All this said, the current blogosphere would surely have Orwell and Kafka quaking. A little-noticed story in the Wall Street Journal on March 6 ("Could Robots Replace Jurors?") recognizes the possibility that soon enough robots -- just imagine -- will be able to tell if someone is lying. Using "text analysis software" -- whatever that is -- European researchers have compiled transcripts of court hearings where witnesses and defendants were later found to have made false or deceptive statements. They used these exemplars to learn patterns in the testimony. The system, apparently, for now at least, only picks out false statements 53 percent of the time, and true statements 75 percent of the time. For now. One can assume, however, that the "science" will progress rapidly.
So, assume robotic science does progress to near certainty. Will we ever want a robot standing in the corner of a courtroom, with a jury essentially being asked to follow its decision making, articulated through the robot's expert analyst as to what they should ultimately decide? Accepting as correct the Justices Kennedy and Stevens notion that juries are quite able to follow a judge's legal instructions and not "blindly" accept an expert's outcome-determinative testimony, is this something we really want? Do we want the value of an independent jury of humans subsumed by an artificial construct, however independent it will be?
It's also worth noting this: Assume a battery of the world's finest lie detector experts. And assume they are all independent, but in agreement in a particular case. John Doe, criminal defendant who allegedly killed his girlfriend Josephine Jones, is polygraphed by each. Doe, however, suffers from severe delusional psychosis -- and these lie detector experts don't have the professional capacity to test for psychosis. Doe, however, claims to be Napoleon Bonaparte. When asked during the polygraph examination(s) if he is Napoleon, he answers "yes," every time. No perspiration, no blood pressure changes, no nothing. They each find "no deception" -- meaning, he is "truthful" in saying that he's Napoleon. So much so that you almost expect that he was also defeated by the British on June 18, 1815 at Waterloo. And, that's the truth!
Needless to say, there will always be an indispensable, overarching role for the jury in the American courtroom. As long as judges know how to run their courtrooms, there should never be a realistic concern that juries will not follow a judge's instructions to carefully weigh the evidence presented to them -- expert and lay alike.
Should we ever reach the day when robots become as reliable as DNA or fingerprint evidence, trials will likely become shorter and more focused on robotics. If the goal of a trial is "truth" (and there is a genuine question about that), shouldn't we want scientists to help maximize that goal in the courtroom, as long as Twelve Honest Men (and Women) and True, still hold the ultimate fate of the litigants in their independent hands?
Bring it on Hal!