Bowen v. Oregon: Why non-unanimous jury verdicts are Constitutional in criminal cases

A Rejoinder to the Amicus Brief filed by Samuel Gross, et al.

In Bowen v. Oregon, the Supreme Court of the United States is asked to reconsider its decisions in Johnson v. Louisiana and Apodaca v. Oregon (companion cases from 1972), finding that non-unanimous jury verdicts in criminal cases are consistent with the 6th Amendment of the Constitution. Some of the most respected scholars in the field of jury behavior -- political scientists, law professors, psychologists, sociologists and economists -- have banded together to submit an Amicus Curiae (friend of the Court) brief in support of the petitioner. That is, they all want unanimous verdicts to be Constitutionally mandated.

I always teach the Apodaca and Johnson cases in my class on the civil and criminal procedures of jury trials because the opinions are wonderful examples of arguments based on conjecture without empirical foundation. The majority argues that there is no evidence to suggest that jurors won't deliberate responsibly, regardless of the decision rule. The minority contends that there is no guarantee that the jurors will deliberate responsibly without unanimity. What we are left with is a disagreement about human nature, with neither side supported by any data. In 1972, the "optimists" won by a nose, with a jumble of separate opinions spread over two cases.

Gross and his colleagues recognize explicitly in their brief that the Court lacked empirical data in 1972, but point out that there have been many empirical studies on jury deliberation and decision-making over the past 37 years. As such, we no longer need to rely on conjecture. Let the data speak for themselves.

I hate the movie, "Twelve Angry Men." As drama, it is pretty compelling, but as a lesson in deliberation and jury behavior, it is the pits. My students all rally around the idea that unanimity is critical because otherwise Henry Fonda never would have been able to save that young man's life! It is truly ironic, then, that I find myself as a minority of one, disputing the conclusions of more than a dozen esteemed colleagues, some of whom are good friends. The problem is: they're just plain wrong.

They are, of course, correct that there has been a lot of empirical research on jury behavior over the past few decades. Much of the best work has been conducted by the very scholars who have signed the brief. Unfortunately, the data just don't tell us the story these folks wish they did. I will outline a few things here, but a complete rebuttal can be found in my 2000 article, co-authored with Warren Schwartz, published in the USC Interdisciplinary Law Journal, entitled "And So Say Some of Us...: What to do when jurors disagree." You can access the full article here.

The Rest of the World

We inherited our common law system from England, as did Canada and Australia. It is important to realize that England has been using a 10-2 decision rule from criminal trial since before the Apodaca and Johnson decisions. There has been no crisis of confidence in jury verdicts in England. No-one is crying out for the return of unanimity. In fact, with the hung jury rate rising to 0.7% (That's 7 hung juries for every 1000 cases), there have been calls to relax the decision rule even further. The states of Australia use a variety of decision rules and the couple retaining unanimity have seen recent public clamor to dump it in favor of 10 - 2 or 9 - 3.

Most other countries in the world use some form of "lay judges" for certain kinds of criminal cases. These folks usually sit in judgment with a small group of professional judges. None of these countries employ a unanimity rule. They usually employ some form of qualified majority that prevents the judges from "out-voting" the laymen. Japan just implemented jury trials for the first time since the Second World War. They use mixed panels (3 judges and 6 laymen) with a simple majority rule.

It is also worth pointing out that the courts that evaluate jury verdicts for "miscarriages of justice" and other decision-making infirmities make their decisions by simple majority rule. Military courts martial are conducted with a 2/3 majority rule. Similarly, when the Senate acts as a jury in impeachment proceedings, it uses a 2/3 majority rule. Let's not forget that George Bush won the 2000 Presidential election by a vote of 6 - 3. In light of the thousands of soldiers and civilians who have lost their lives as a result, it is hard to imagine any judicial decision with larger implications.

Does deliberation suffer without unanimity?

In light of the exceptionalism of the American unanimity rule, one might sensibly wonder,"What is everyone so worked up about?" Well, criminal defense attorneys are clearly concerned that conviction rates will go up. But what about the academics who signed this amicus brief? The rallying point for this group is that deliberations are superior when a unanimity rule is used than when a majority or supermajority rule is used.

One of the primary forms of evidence for this proposition is the post-study participant satisfaction survey. Mock jurors are asked how "satisfied" they are with their experience and how "confident" they are in the verdict chosen by their panel. Unsurprisingly, jurors in the unanimity setting express greater satisfaction and verdict confidence. Cognitive dissonance can be a wonderful thing. Since all of the jurors in the unanimity setting voted for the same verdict, by definition, they are, of course, inclined to report satisfaction with the result. To admit otherwise would be to impugn their own decisions to vote with everyone else. By contrast, a juror who was outvoted can prop up her own self-image by claiming that the deliberative process must have been flawed in some way. Such opinions can only be expressed in the non-unanimous setting. Hence, regardless of the objective quality of both the deliberations and resulting verdict, we would expect to see the differences in self-reporting that are regularly reported.

Another metric that is often summoned to support unanimity is the length of time that a mock jury spends deliberating. In most studies, juries using unanimity deliberate for longer than do juries using some form of supermajority rule. But this begs the question: "Is all deliberation good deliberation?" Imagine a twelve person jury that spends two hours carefully reviewing all the evidence and testimony, establishing a timeline they can all agree to, and coming to a mutual understanding of what their instructions are asking them to decide. They then take a vote and discover that 10 jurors have been convinced of the defendant's guilt beyond a reasonable doubt and two have not. The two "holdouts" are somewhat more mistrustful of the police officer's testimony and seem to have a somewhat more demanding conception of reasonable doubt. So far, everyone on the jury has behaved admirably. Information has been shared, opinions expressed and conclusions drawn after careful consideration. In order to reach a verdict, however, the 10 majority jurors will have to convince the other two to change their votes. Through browbeating, exasperation and sometimes personal attacks, such a majority will usually get the holdouts to acquiesce. This process usually takes a while, so "deliberation" has taken longer. But what has been accomplished? Two jurors have been forced to vote against their consciences, civility has been sacrificed and the public has been lied to. Is this "better" than letting the jury hand down a 10-2 verdict?

Are unanimous verdicts really unanimous?

One of the studies often used to support the "unanimous deliberations are superior" hypothesis is by Nemeth (cited in the amicus brief). He compared six-person juries operating under unanimity with those operating under a 4-2 rule (He carefully constructed the juries so that half of each sample started with a 4-2 majority for acquittal and the other half started 4-2 for conviction). What no-one (but me) seems to focus on is that the self-reporting of participants after the study indicates that virtually none of the unanimous verdicts were the result of actual unanimous consensus among the jurors. Consider the table I made from Nemeth's data.



To underscore the important point that a unanimous verdict does not necessarily reflect unanimous consensus, we turn to a very recent study conducted by Valerie Hans, one of the signers of the amicus brief. In her article with Nicole Waters, entitled "A Jury of One: Opinion Formation, Conformity and Dissent on Juries," ( 6 (3) Journal of Empirical Legal Studies, 513 (2009)), the authors reveal that roughly one-third of jurors in actual criminal cases admitted that they actually disagreed with the verdict they had voted for. That is, on average, a unanimous jury verdict in a criminal case was actually only 8 - 4.

Finally, let's remember that, every time a defendant is retried after a hung jury, the resulting verdict has previously been rejected by at least one juror who heard the case. Otherwise, there wouldn't have been a hung jury in the first place. Who's to say that the jury that could reach consensus was more representative, more attentive, smarter or more deliberative than the one that couldn't reach consensus. We also know from the data outlined above that the second jury might just have had jurors more willing to compromise their principles to generate a unanimous verdict.

This begs another obvious question: "If unanimity is a big fat lie, why should anyone want to preserve it?" Perhaps the general public wants to be lied to in this way. Perhaps there is comfort in not considering the implications of sending people to prison (or worse) despite reservations of some jurors. Maybe we don't really want to know how the sausage is made. Even so, it is hard to imagine this as a legitimate rationale for the scholars who penned the amicus brief in support of Bowen's position.

The Future of Unanimity

Unanimity in criminal jury verdicts is what polite lawyers refer to as a "legal fiction." As such, I don't see it as being worth preserving. I would much rather have majority verdicts, handed down by jurors who are strongly encouraged to stick to their principles and vote their consciences. As I have written in both scholarly papers and elsewhere, the elimination of the unanimity requirement would also support the elimination of peremptory challenges, a practice that disenfranchises many Americans from the jury system. It is interesting to note that many of the scholars who signed the amicus brief in support of Bowen's case have also called for the curtailing of peremptory challenges. Eliminating peremptory challenges would create more fully deliberative juries because they would better represent a fair cross section of the community. More voices would be heard. In addition, such a reform would dispense with the need for the ruling in Batson v. Kentucky and its progeny, prohibiting peremptory strikes made along racial lines. Batson, as Marshall predicted in his concurrence more than 20 years ago, has been largely ineffective at reducing overt and subconscious bias in the exercise of peremptory challenges. Non-unanimous verdicts, coupled with truly randomly selected juries, would eliminate this problem once and for all.