Earlier this term, the U.S. Supreme Court held in Ramos v. Louisiana that the Sixth and Fourteenth Amendments require jury unanimity to convict a defendant of a serious crime. As it happens, virtually all of the states appear to have symmetrical conviction/acquittal vote requirements. That is, those that demand unanimity to convict (with the possible exception of post- Ramos Oregon) have also required unanimity to acquit. The Constitution, of course, speaks only to what it takes to convict. Making it harder to convict is a standard part of constitutional criminal procedure doctrine, developed to ensure that innocent people avoid incarceration. But making it more difficult to acquit is no express part of any constitutional requirement and could, if taken to an extreme, violate the rights of an accused.
No defendant would complain that it is too easy to acquit him and that the law should require additional acquittal votes before allowing a finding of not guilty. At the same time, though, there appears to be no reason to think there is a right to a less-than-unanimous jury for an acquittal. In this column, I will consider the policy question of whether it makes sense to do as at least forty-nine states now do and demand a unanimous jury before acquitting a defendant of a crime, requiring that the judge declare a mistrial if the votes fall short of unanimity for an acquittal.
Unanimity for a Conviction
The Supreme Court recently said that convictions require a unanimous jury, whether in federal or state court. As the Court had already, in Apodaca v. Oregon, said that federal convictions require a unanimous jury, the recent decision in Ramos extended that ruling to state criminal cases. An advantage of demanding unanimity to convict is that if the entire group is convinced beyond a reasonable doubt that the defendant is guilty, that would appear to make it more likely that the defendant really is guilty. If one or more jurors, by contrast, think the defendant might be innocent, then their uncertainty (or worse, their belief in his innocence) would seem to increase the odds of convicting and incarcerating an innocent person. In some sense, the jury unanimity for convictions requirement resembles the proof beyond a reasonable doubt standard in that both reject convictions that rest on something less than a fully persuaded jury made up entirely of fully persuaded jurors. One who believes the defendant is innocent is like an entire jury that is not convinced beyond a reasonable doubt. And a unanimity requirement for conviction helps induce people to deliberate with one another and thereby produce the benefits that come of having a group of diverse individuals determine guilt rather than an assortment of isolated individuals voting independently.
Unanimity for an Acquittal?
Why require unanimity for an acquittal, though? One answer is that even if the Constitution doesn’t worry about unjust acquittals, the people most certainly do. If some on the jury believe beyond a reasonable doubt that the defendant is guilty, then that increases the odds that the defendant truly is guilty, even if ten out of twelve jurors remain unconvinced of guilt. The evidence was apparently enough to persuade a couple of people of guilt beyond a reasonable doubt. We know as well that the case would not have been able to go to the jury if the judge did not believe that a reasonable person could conclude beyond a reasonable doubt based on the evidence that the defendant committed the crime. So all things being equal, the odds of guilt are rather high going in and become even higher when one or two jurors believe a conviction is warranted.
Given an elevated chance of guilt, the unanimity requirement for acquittal provides that if anyone is convinced of guilt beyond a reasonable doubt, then the odds of guilt are too great to justly allow for a full acquittal. The public, in other words, might wish to have confidence in the accuracy of its verdicts, whether those verdicts are guilty or not guilty.
Imagine for a moment that you had to persuade a jury of innocence beyond a reasonable doubt in order to get an acquittal. The jury would accordingly have to be quite close to certain of the defendant’s innocence to acquit, just as it would have to be quite close to certain of the defendant’s guilt to convict. This would presumably minimize type 1 errors (in which the jury wrongly identifies an innocent as guilty) as well as type 2 errors (in which the jury wrongly identifies a guilty person as innocent). But working in such a dual-error-reduction system, whether by erecting a high standard (beyond a reasonable doubt) or by demanding unanimity, leaves a residual category of cases: mistrials. If any failure to convict resulted in an acquittal, we would have no such residual category: if some of the jurors are unsure or if any number of jurors believe in guilt, but only by a preponderance and not beyond a reasonable doubt, the defendant walks away with an acquittal. The mistrials mean that at least some of the time, the prosecutor will take the criminal defendant to trial repeatedly. When there is neither a conviction nor an acquittal, the case may go to trial again. Is that an acceptable price to pay for avoiding unjust acquittals?
The Problem with Retrials
In one sense, having to redo the trial until we get it “right” might sound like it makes sense. If we were doing our math homework, for example, and we got a question wrong, our teacher might ask us to do the problem over again, whether our answer was too high or too low relative to the correct one. Trying the same case over again, however (with the same evidence, unless something new developed), is quite different from having to do a math problem over again until we get it right. It is, instead, like approaching people on the street, one by one, to resolve a dispute: whether you are right or I am right about what some third person said when she spoke out of her window (a fact that we can never know for sure). You say she yelled “one,” and I say she yelled “ton.” If lots of people say “I heard one” or “I heard ton,” then we seem to be approaching an accurate result. But if people are split on the question and we keep on asking more and more people until we have a two-person majority for either you or me, then it would appear that we are not getting closer to the truth as we count votes. If people could not agree unanimously on either guilt or innocence this time around, it would seem, likewise, that getting a unanimous jury the second time around reveals only the fact that eventually, you can find twelve people to agree on one side or the other, no matter what the truth might be. Another way of saying this is that unanimity means less when it is one unanimous jury out of two or even three jury trials.
In response to this problem, a prosecutor might say that the burden of persuasion actually intervenes to reduce the impact of repeat trials. Because the burden is as heavy as it is, a prosecutor who fails to achieve a conviction (because one or more jurors voted to acquit) might choose not to prosecute the case again. One might hope that only in cases in which the prosecutor believes the jury was confused or otherwise represented a group of outliers, would the prosecutor try the case again with a new jury. For purposes of the defendant’s liberty, then, the mistrial would prove to be nearly the equivalent of an acquittal.
In one respect, of course, a prosecution ending in mistrial is distinct from an acquittal. People think of an acquittal as a sort of finding of innocence. It is not, of course. A jury could believe the defendant more likely than not (though not by proof beyond a reasonable doubt) committed the crime, and it would still have to vote “not guilty.” But if you told someone “I was charged with a violent crime, but the jury said not guilty,” that statement would be quasi-accurate and have the sound of vindication to it. A mistrial is different. If you were to say, “I was charged with a violent crime, but the jury could not agree on whether to say guilty or not guilty,” few people would read that statement as a “not guilty” conclusion. You would have to go through the rest of your life with a finding by a jury that you might or might not be guilty of the crime of which you were accused. Is that unfair? Well, if you are innocent, then it seems unfair, but then any negative consequences attaching to a false accusation would seem unfair because the accused is innocent. If you are guilty, however, then it seems more than fair. You committed a serious criminal offense for which you may deserve a very harsh sentence. Instead, because it is better that ten guilty go free than one innocent party suffer, you walked away with no serious punishment. Once in a while, someone might think ill of you, but if you are guilty, then they are right to think so. And if you have trouble finding a job or making friends because of the ambiguous finding that a mistrial represents, then once again, you deserve that and a lot more if you are guilty of a serious crime.
The government has no constitutional rights and therefore has no entitlement to avoid an acquittal in the absence of unanimity (or in the absence of a heightened standard of proof). But the people whom the government represents care about type 2 errors. They believe that guilty individuals deserve to be punished. Increasing the odds that this will happen is, to the minds of most, a good thing. Requiring unanimity as a prerequisite to acquittal appears to serve this avoiding-type-2-errors objective, particularly given prosecutors’ frequent decision not to retry cases that they failed to win the first time. Perhaps it is desirable for society to keep an eye on a person who could be neither convicted nor acquitted. Though perhaps less important than ensuring that an innocent person avoids punishment, it is still very important that a guilty person receive some of what is coming to him. If the courts fail in ensuring that this happens, after all, society might resort to other methods, as it once had to do in the absence of a sovereign and with far less accuracy and fairness.