Wednesday, September 23, 2009

Criminal jury verdicts aren't really unanimous: Implications for your practice

The Worst Kept Secret in Criminal Law

We all knew it. Right? There are lots of reasons why individual jurors vote with their peers despite misgivings. They want to be liked. They want to go home. They are tired of being brow-beaten by the others. They don't want to disappoint the judge by hanging the jury. They just don't care. What we didn't know was just how often it happened -- until now.

The statistic previously cited in the debate about the desirability and/or practicality of unanimity was hung jury rates. You've read my discussion of hung jury rates in this very blog. The estimate of the national average is between 6% and 7% of criminal trials end in a hung jury. For felony trials in some California counties, the rate is over 20%. In Oregon, which uses 10-2 verdicts, the rate is below 1%. The problem, of course, is that the hung jury rate is a very poor proxy for what we really wanted to know: How often do juries remain "deadlocked" after completing deliberations? (where deadlock refers to a situation where a unanimous consensus about the right verdict has not been reached).

Nicole Waters of the National Center for State Courts and Valerie Hans of Cornell Law School have finally taken a step towards answering this important question. As part of the National Center for State Courts project on hung juries, the authors gave out post-trial surveys to almost 4000 jurors across 4 states and got back completed forms from almost 3500. This covered 367 trials. Among the questions they asked was: "If it were entirely up to you as a one-person jury, what would your verdict have been in this case?" It is important to appreciate that all of these verdicts were unanimous, in that all the jurors officially voted for it. Therefore, each "conforming dissenter" (Authors' term) is someone who voted for a verdict she thought was wrong.

The Non-Unanimous Bombshell

Remember that estimates of hung jury rates range around 7%. These are the cases with a dissenter who refused to conform. In the Waters and Hans study, 38% of juries contained at least one juror who disagreed with the general outcome of the case but voted for it anyway (conviction or acquittal). This is the most conservative measure of dissent. Juries are often faced with multiple charges against the defendant or a choice among lesser-included-offenses. So, it is possible that a defendant can be convicted, but not of the charge that a juror thought was most appropriate. Taking this into account, 54% of juries contained at least one juror who disagreed with the jury's verdict on at least one charge. That is, more than half of the juries contained at least one juror who voted insincerely. Nearly half of the juries (46%) contained at least one juror who disagreed with the verdict for the most serious charge facing the defendant.

According to the authors, only 82 of 351 dissenting jurors ultimately hung their juries. The article suggests that the average number of dissenters for cases resulting in a hung jury was just over 3. Therefore, we can tease out that about 25 of these cases resulted in a hung jury, or about 7% of the sample (consistent with previous estimates). By contrast, there were typically about 1.5 conforming dissenters on any case that returned a verdict (if there were any conforming dissenters at all). So, for every single hung jury, there were approximately 8 cases in which the purported unanimous verdict was a farce. When debating the merits of unanimity in the future, hopefully scholars and pundits can move beyond the misleading metric of hung jury rates and focus on the real problem at hand: How do we feel about a system that forces jurors to compromise their own values and judgments to conform to the majority will?

Making use of these results in your next case

While we wait for the Supreme Court and the state legislatures to figure out what to do with these revelations (probably nothing), you should consider them carefully as you prepare for future cases.

The first lesson is that very few criminal cases result in truly unanimous juries, even after prolonged deliberations. If you are a criminal defense attorney, you should always hammer away at the "personal responsibility" angle of jury service. Remind jurors that each has taken an oath to hear all the evidence and reach a conclusion based on her own wisdom and conscience. Emphasize that the Court would never want any juror to vote for a verdict she did not believe was correct. I would include language about being sympathetic to their plight: "Sometimes peer pressure can be strong. Sometimes it is just easier to go along with everyone else, to avoid conflict, to be cooperative. I recognize any sympathize with these instincts. But just remember that the Court has not tasked you with being accommodating or friendly or cooperative. The Court has tasked you with doing the right thing."

The authors of this study have examined some of the things that lead dissenters to hold their ground and hang the jury, rather than giving in to the majority. The first result, supported elsewhere, is that dissenters who favor acquittal are less likely to switch their votes than are dissenters who favor conviction. So, in a case with substantial support on both sides, an acquittal is much more likely than a conviction. A prosecutor needs to be certain that she will secure a very substantial majority for conviction in order to avoid a hung jury. By contrast, a defense attorney can secure an acquittal, even if a handful of jurors initially favor conviction.

Procedures matter. I have written in my column for Lawyers Weekly USA about the consequences of different methods of taking votes in the jury room. As I predicted in that piece, Waters and Hans discovered that juries that took secret ballots were more likely to have dissenters on them. The logic here is that a dissenter will worry less about sticking out like a sore thumb if she can register her objection in private. A secret ballot also allows potential dissenters to discover whether they have any allies on the jury. There might be three or four jurors who are reluctant to be the only one raising a hand in favor of the minority position. As a result, none of them do. A potential block of dissenters is never discovered. A secret ballot avoids this game of "chicken."

I have previously discussed in The Jury Box the importance of "straw poll timing" for the nature of jury deliberations. Juries that immediately take a vote find themselves embroiled in "verdict-driven" deliberations, while those who delay voting spend more time in the generally superior "evidence-driven" mode of deliberations. After a vote is taken, jurors tend to fall into camps, with spokespersons emerging for each. As a result, fewer jurors participate in discussion and jurors tend to get locked into their views. The Waters and Hans study shows a small but significant effect of early polling of jurors. When a jury takes an early vote, it is more likely that at least one juror will disagree with the final verdict. This is probably because an initial dissenter is more likely to feel "outvoted" than "convinced" if the deliberations devolve into a discussion of how to get everyone to vote the same way -- rather than getting everyone to agree.

If you have a case in which you are advocating what you fear will be the "unpopular" position (thought legally defensible), you would be advised to encourage jurors to vote by secret ballot. While you cannot instruct the jury to vote in a particular way, keep in mind that they have no explicit instructions about procedures. As such, a subtle suggestion in closing can be effective. "I want you all to keep an open mind. As you tear up those little slips of paper to cast your ballots, I hope that you will consider everything my client has said on the stand..."

It is much harder to find a way to suggest that jurors take an early straw poll. In fact, many judges now recommend to juries that they not be in a hurry to take a vote, first spending a fair amount of time just sorting through the evidence. Remember that an early vote is a two-edged sword. While it might cause your supporters to harden their positions a bit, it also reduces their abilities to sway the votes of jurors on the other side.

One important question is how volatile are juror evaluations of cases. That is, do they generally change their minds during a case? The answer to this question is a bit unclear. Over 62% of jurors indicated that they had changed their minds at least once during their case. But what does a juror mean when she says she "changed her mind"? Given that 40% of conforming dissenters admitted to changing their minds during deliberations, as compared to 25% of holdouts (those who hung their juries), I fear that many respondents equated changing their votes with changing their minds. The percentage of jurors who reported changing their minds at some stage prior to deliberations is closer to 25% (difficult to pinpoint from the reported data). Some of these respondents are undoubtedly reporting an earlier change as a compensatory mechanism for cognitive dissonance -- they don't want to admit to themselves that they were bullied into changing their votes during deliberations. That said, it would seem a conservative estimate that about one in five jurors really does change her evaluation of the case at some point in the proceedings.

The lesson here would seem to be that a case is not really won or loss during jury selection, as many cynics would argue. There are a substantial number of jurors who can be convinced to re-evaluate their initial positions during a trial. So, pay close attention to your trial technique. Make sure your case is coherent and clear. Finally, be sure to hit the high points in closing to reinforce the items that might have gotten jurors rethinking their views.

This brings us to the question of jury selection. It turns out not to be so critical to select a jury fully comprised of your supporters. These juries seem few and far-between. Most cases result in at least some form of disagreement among jurors that must be resolved during deliberations, either through a genuine exchange of ideas or, more likely, the conversion of conforming dissenters. This reinforces something that trial consultants like me have been trying to explain to clients for years. When evaluating a prospective juror, it is important to consider not only whether the juror is likely to be sympathetic to your case, but also what role the juror is likely to play on the jury. If the juror turns out to favor the other side, is she going to be a holdout or a conforming dissenter? I can't always tell if a prospective juror is going to be pro-prosecution or pro-defense, but I usually have a good sense about whether she is likely to be a leader or a follower, whether she seems more eager to please the court or those around her, whether she relishes the opportunity to be a contrarian. These characteristics are critical and must be explored during jury selection.

OK. This post is really, really long. That's because I think these results are really, really important. Among other things, they emphasize the importance of relying on data, not conjecture. Those who presumed that unanimous verdicts were produced by unanimous juries are just dead wrong. What else might they be wrong about? Do you really want to rely on conjecture the next time you need to pick a jury, or prepare a witness, or select a trial theme? There is no substitute for good research. OK, there's my quick sales pitch.

Please comment on this post! Let's see if we can stir up some trouble.


  1. At this point, i totally agree with you that there is no option other than good research.

  2. Anonymous4:41 PM

    While your comment, that cognitive dissonance affects jurors into believing that they did not change their vote because they were bullied into it, is valid, it only scratches the surface.

    A holdout juror in the midst of deliberations will face a tremendous amount of psychological pressure to conform and resolve it through cognitive dissonance. The pressure will be great to rationalize that the other jurors are correct and conformity will occur mostly out of the holdout's pressure to please the other jurors.

    Ironically, this psychological pressure is similar to that faced by a suspect being interrogated by law enforcement which can result in false confessions as well as genuine ones.

    The jury community will ostracize the holdout much the same as the prosecution attempts to ostracize the defendant in the jury's eyes.

    The only way a juror can successfully maintain their position is the same way the average citizen can prevent uttering a false confession:
    do not reverse your position and stop talking after your position has become ignored.

    A note to the judge stating that your position is irreversible due to the fact that the prosecution did not prove its case to your satisfaction should then be sent.

    The question is: do most jurors have the psychological stamina to withstand the constant pressure to conform during deliberations, especially after sitting through a guilt innocence trial filled with spin from the opposing attorneys? After looking at many verdicts that went sideways and resulted in miscarriages of justice, I fear the answer is no.