Stripping people of their liberty or life is one of the most solemn tasks exercised by liberal democracies. Yet, bizarrely, the American criminal-justice system pays too little regard to the factual accuracy of its verdicts.
Consider the steady trickle of exonerations, which occur at a rate few observers would have predicted not long ago. The recently launched National Registry of Exonerations lists the details of 989 exonerees since 1989, and the Innocence Project reports on 300 convicted inmates who have been exonerated based on DNA testing alone. (An additional group of 1,170 defendants have been exonerated in 13 “group exonerations” that followed major police scandals.) No doubt, the actual number of false convictions is much higher.
A stronger sense of the marginalization of truth can be gleaned from the way the criminal-justice system operates. An astounding 95 percent of felony convictions are consummated not in open court but through plea bargains struck between prosecutors and defense attorneys in backroom dealings. Not unlike in other walks of life, the bargains struck are determined primarily by the strength of the negotiators’ cards, which do not always correspond to the merits of their positions. Thus, guilt or innocence can readily fall by the wayside. The power dynamic of plea bargains is skewed also by the current regime of excessive punishment, broad prosecutorial discretion, and the parties’ informational asymmetry.
Yet even when the pageantry and protocols of the criminal process are extended to their fullest, there are considerable reasons for concern about the accuracy of the outcomes. That’s because the operative objective of the process is not the accuracy of convictions but adherence to various procedural rights. In principle, those procedures are designed with an eye toward accuracy. But in the rough and tumble of the adversarial process, they can easily steer the proceedings off course.
As is typical in procedure-centric regimes, form tends to trump substance, and bureaucratic considerations can overwhelm the putative values of the system. Nowhere is the topsy-turvy conception of justice manifested more poignantly than in the U.S. Supreme Court’s 20-year-long indecision about how to handle cases of convicted inmates who make a compelling demonstration of innocence (Herrera v. Collins, 1993). Justice Antonin Scalia recently reminded us that the court “has never held that the Constitution forbids the execution of a convicted defendant who has had a full and fair trial but is later able to convince a habeas court that he is ‘actually’ innocent” (In re Davis, 2009). This logic brings to mind the creature in the Beatles’ film The Yellow Submarine that violently sucks away every object it encounters, until its vacuuming beak accidentally sucks in its own foot and ultimately its whole self, disappearing into oblivion.
But the most serious threat to the integrity of criminal verdicts is the uncertain nature of the evidence on which they rely. Criminal investigations yield an unknown mix of accurate and inaccurate evidence, and the adjudicative process is ill equipped to distinguish between the two. Moreover, the typical opacity of the investigative record deprives all actors of valuable information about the investigative procedures used and the exact sources of the evidence they produced. It also tends to conceal evidence that was not contained in the formal charges.
The problems with the integrity of the evidence are patently demonstrated by the postmortem examinations performed on false-conviction cases. They are also manifested by the legal-psychological research that examines how witnesses, victims, detectives, lawyers, judges, and jurors perform their roles in the criminal-justice process. Extensive research shows that performance on even the simplest of tasks in the process can be exceedingly complicated and nuanced. Importantly, erroneous evidence can be induced by the investigative and adjudicative procedures themselves, even when those involved are executing their duties honestly and diligently. In bettering the system, then, our focus should be on those procedures.
Studies show, for example, that when people are asked to recall the particulars of criminal events, the witnesses can be led to report facts that were suggested to them by the interviewer, facts that they merely imagined, and facts that are compatible with the dynamic of the interview. The witnesses’ confidence in both true and false memories can readily be boosted or inhibited by the interviewer.
Lineup studies show that about one of every three times that witnesses pick out a suspect, they pick an innocent filler. That error rate can readily be exacerbated by the manner in which the lineup is conducted—for example, by the wording of the instructions given to the witness, the selection of fillers, and the method of presenting suspects to the witness, as well as by cues communicated from the administrator, however unintentionally and unnoticeably.
The experimental research points toward two key recommendations. First, the accuracy of evidence could easily be enhanced by adopting best practices like using appropriate protocols for interviewing witnesses, and double-blind and sequential, rather than simultaneous, lineups. Second, the transparency of the evidence could be enhanced by videotaping every encounter with potential witnesses and sharing the records with all of the parties involved.
Such advances would minimize both unjust prosecutions and frivolous defense claims. While these reforms cannot guarantee that juries will always draw the correct inferences, providing jurors with the best available evidence—and reducing their exposure to misleading evidence—should better equip them to achieve that goal.
A judicious incorporation of current psychological research into our criminal-justice process would serve as a key component in the restoration of accurate fact finding as its core objective. Ideally, the push for reform should be spearheaded by the Supreme Court, which is the ultimate authority on individuals’ liberties and the only body that has effective jurisdiction over the multitude of law-enforcement agencies and state courts. Unfortunately, in January of this year, eight Supreme Court justices flatly rejected that course of action (Perry v. New Hampshire). At the same time, we are witnessing promising reforms arising from states including New Jersey, North Carolina, Illinois, and Wisconsin, as well from local jurisdictions, such as Dallas County, Tex., and Santa Clara County, Calif.
The rest of the country would do well by following these entrepreneurial jurisdictions and reinstating the punishment of the guilty and the freeing of the innocent as the paramount mission of the American criminal-justice process.
Dan Simon is a professor of law and psychology at the University of Southern California. This essay is adapted from his new book from Harvard University Press, In Doubt: The Psychology of the Criminal Justice Process.