Earlier this term, the U.S. Supreme Court considered Evans v. Michigan, a Double Jeopardy case. As in Blueford v. Arkansas, decided last year, the resolution of Evans turns on what ought to happen when a judge makes a mistake about state law and consequently takes a case away from the jury. Unlike in Blueford, however, the mistake in Evans helped the defendant. Seeing how the Court resolves this case will therefore tell us a lot about how principled the Court aims to be in applying its Double Jeopardy doctrine.
The Facts
Lamar Evans, the petitioner in the Supreme Court case, was charged with the state crime of burning a building other than a dwelling. Michigan law provides separately for the crime of burning a dwelling. The particular building that Evans stood accused of burning was apparently a vacant house.
During Evans’s criminal trial before a jury, after the close of the prosecution’s proof, the defense moved for a verdict of acquittal from the judge, on the theory that the Michigan law at issue required the prosecution to prove that the defendant had burned a building that was not a dwelling (which I will now call a “non-dwelling,” to save on words). On this theory, the defendant was entitled to an acquittal because the prosecution had failed to introduce evidence proving that the burned building was a non-dwelling.
The judge agreed with this interpretation of Michigan state law and accordingly concluded that, on the evidence presented by the prosecution, there was insufficient proof of the building’s status as a non-dwelling to allow for a verdict of guilty. The trial court accordingly granted the defense motion and directed a verdict of acquittal.
On appeal from the judge’s ruling, the Michigan Court of Appeals first rejected the trial court’s interpretation of the Michigan law in question. The law under which Evans was charged, according to the Court of Appeals, provides a lesser included offense relative to setting fire to a building that is a dwelling. In other words, if one sets fire to any building, then one is guilty of violating the law under which Evans was charged. If the government proved that the building was in fact a dwelling, then the defendant could be found guilty of violating an additional law (which prohibits setting fire to a dwelling). It was accordingly unnecessary for the prosecution to prove that the building to which Evans allegedly set fire was a non-dwelling.
An Analogy Clarifying the Logic of the Defendant’s Argument and Why It Was Flawed
To provide an analogy to the “dwelling” dispute, imagine a state law that prohibits the murder of a witness in a court case, which we will call “law A”. Imagine now a second law that prohibits the murder of an individual, whether or not he or she is a witness in a court case, a law that we will call “law B”. If the government is unsure about whether a particular murder victim would qualify as a “witness” under law A, it would likely prosecute the murder under law B. Suppose the defendant then brings a motion to the trial judge to direct a verdict of acquittal, on the ground that the government failed to prove that the victim was not a witness in a court case. The trial judge quite obviously should deny the motion, on the ground that legally, the prosecution need not prove that the victim was a non-witness in order for a jury to convict.
Additional Proceedings in the Michigan Case
After determining that the trial judge had erred in directing a verdict of acquittal, the Michigan Court of Appeals ordered a new trial. The defendant then appealed to the Michigan Supreme Court, which affirmed the Court of Appeals’ ruling.
Evans claims (and claimed in the state courts, as well) that a new trial would violate his right under the Fifth Amendment Double Jeopardy Clause (as incorporated by the Fourteenth Amendment Due Process Clause) to be tried no more than once for the same offense. The State of Michigan defends the Michigan courts’ rulings on the ground that a directed verdict of acquittal issued on the basis of the erroneous addition of a nonexistent legal element to the charged crime does not qualify as a verdict that terminates jeopardy and thus precludes retrial for the same offense.
Accordingly, in the oral argument before the Justices, the government and the petitioner argued over whether the judge’s directed verdict of acquittal does or does not qualify, under the Double Jeopardy precedents, as a final verdict of the sort that terminates jeopardy and precludes retrial.
In this column, however, I want to focus on a different Double Jeopardy case, one that helps raise a more fundamental question about the Court’s commitment to the neutral application of constitutional doctrine.
Blueford v. Arkansas
In Blueford v. Arkansas, the Supreme Court considered a case in which the judge had originally (and correctly) told the jury to consider and deliberate on the most serious offense first and then to move on to less and less serious offenses, sequentially, only after unanimously voting to acquit on the more serious, in turn. In the particular case, the defendant was charged with four degrees of homicide: capital murder, first-degree murder, manslaughter, and negligent homicide.
After some deliberation, the jury reported to the judge that it was deadlocked on the charge of manslaughter, and thus unable to reach a verdict. The judge asked the jury whether it had considered the lesser charge of negligent homicide. The foreperson, perhaps properly confused by the question, told the judge that the jurors understood their charge as precluding deliberation on a lesser charge unless and until they had unanimously voted “not guilty” on the more serious charge preceding it. The judge then told the jury to return to deliberation, which the jury did. After a short time, however, the jury reported that it remained deadlocked.
The defense then asked the judge to issue a verdict of acquittal (or ask the jurors to fill out verdict forms) on the first and second charges of capital murder and first-degree murder, to reflect the fact that the jury had voted unanimously to acquit the defendant on these two most serious charges. There could then be a mistrial on the two remaining charges, on which the jury had reached no verdict: manslaughter and negligent homicide.
The judge, however, refused the defense request, choosing instead to declare a mistrial, without modification.
When a new trial, with a new jury, began on the four charges, the defendant unsuccessfully moved to dismiss the first two charges on Double Jeopardy grounds. The Arkansas Supreme Court, in an interlocutory appeal, then affirmed the trial court’s refusal to dismiss the capital murder and first-degree murder charges. The U.S. Supreme Court affirmed the Arkansas Supreme Court’s decision, a decision with which I anticipatorily disagreed here. In the absence of an official and final verdict of acquittal, the Court deemed the trial judge’s mistrial necessary (and thus appropriate) and retrial accordingly permissible on all four charges.
Had the trial court correctly understood Arkansas law to preclude deliberation on manslaughter absent unanimity in favor of acquittal on capital and first-degree murder, it might have given the jurors verdict forms on which they could have formalized their acquittal of the defendant on the first two charges. But the trial judge revealed potential confusion about his own correct instructions by asking the jury foreperson for the vote on negligent homicide, even after learning of the jury’s having deadlocked on manslaughter.
In the Evans trial, the trial court erred in the opposite direction. Instead of disregarding the jurors’ affirmative findings (for the defendant), as the judge did in Blueford, by refusing to give the jurors verdict sheets, the trial court in Evans refused to honor the jurors’ proper role by taking the case away from the jury (despite the sufficiency of the evidence) and by reaching a verdict (for the defendant) on its own.
If the U.S. Supreme Court were most concerned about substance, rather than form, then Blueford would have won and Evans would lose. Why? Because Blueford actually persuaded a unanimous jury to vote to acquit him of two offenses. He accordingly ought to have been able to obtain closure on those two charges, and thus should have faced a new trial only on the remaining two charges, on which the jury had either deadlocked or not yet deliberated. The Court in Blueford, however, elevated form over substance, and ruled that since there was no official verdict on any charge, it was fine to retry Blueford for everything. After all, who knows what the jury would have done after the additional brief deliberation, if asked to deliver a verdict? One of them might have changed his or her mind.
In Evans, by contrast, there was a verdict. The judge, understanding the elements of the offense incorrectly, weighed the evidence and found that the prosecution had failed to prove beyond a reasonable doubt the “non-dwelling” status of the place to which the defendant had allegedly set fire. Indeed, the judge found that the evidentiary failure was so decisive that no reasonable juror could believe that the evidence on the “non-dwelling” issue was sufficient to prove guilt. The judge accordingly directed a verdict of acquittal.
If substance mattered more than form, then the state courts in Michigan made the right choice, by rejecting the defendant’s claim that the judge’s verdict of acquittal should bar retrial. Absent the trial court’s legal mistake, that court would certainly have allowed the jury to deliberate, because the evidence—on the law—evidently was sufficient to go to the jury. Therefore, the prosecution is entitled to have a jury actually evaluate the government’s evidence and reach a verdict.
Yet if the Court is committed to form over substance in this area, then it ought to maintain that commitment without regard to the outcome. This means that in Blueford, where there was technically no verdict, the defendant had no recourse at all to the Double Jeopardy Clause, notwithstanding his having endured a trial and having managed to persuade a unanimous jury of his innocence on the two most serious charges. But it also means, here in Evans, that when there is a verdict of not guilty that is rendered by a fact-finder (here, the judge) after the prosecution has presented its case, the Double Jeopardy Clause ought similarly to bar retrial, despite the fact that the jury should have been allowed to deliberate (and to deliberate with proper instructions).
Innocence Versus Closure
Why would I bring up innocence at all in this column?, readers may wonder. In one respect, a defendant’s innocence or guilt is irrelevant to the Double Jeopardy Clause. The Clause protects a defendant from being forced to undergo more than one trial for the same offense. The goal is to provide closure for the defendant, and correspondingly to prevent the government from oppressively re-litigating a criminal case over and over until it receives the verdict that it wants.
In many cases, we could easily imagine that the reason the government might want to retry a case is that (a) the first jury acquitted against the evidence (i.e., it engaged in jury nullification and refused to apply the law to the facts); (b) the first jury acquitted because some extremely compelling evidence of guilt that later surfaced was not yet available during the trial; or (c) the first jury acquitted because the prosecutor did a terrible job and thereby lost a case that ought to have been won.
In all three of these sorts of cases, by hypothesis, we may well have a guilty defendant who therefore deserves to be convicted and punished. Yet the Double Jeopardy Clause, as understood by the Supreme Court, precludes any retrial—even a retrial at which the truth might truly have the opportunity to surface and receive its day in court.
There is, however, a feature of Double Jeopardy that is sensitive to innocence. When we ask one jury to determine what happened, and we apply a very demanding standard (guilt beyond a reasonable doubt), we have a decent chance of getting at the truth. On the other hand, if an eager prosecutor can keep presenting the evidence to different juries, one after another, until a jury finally convicts, then that option significantly diminishes the likely accuracy of the ultimate verdict. Indeed, if the sample is large enough, it will likely eventually be possible to find twelve people to say “Guilty,” no matter how innocent the defendant actually is. That is simply the law of large numbers. For the defendant to have to endure trial after trial is therefore oppressive in part because it increases the chances that an innocent person, who was properly acquitted, will in the end be convicted.
To the extent that we care about the innocence feature of Double Jeopardy, we may be especially unhappy with a doctrine that contains Blueford (in which a unanimous vote of not guilty on two charges is followed by a retrial—and the possibility of conviction—on those same two charges) and that also contains my proposed outcome in Evans (in which a possibly-to-probably guilty defendant is protected from ever having any fact-finder properly consider the evidence against him on the crime with which he was charged, as correctly defined).
Yet I am still drawn, in considering the Supreme Court’s dilemma in Evans’s case, to hope that the Court protects Evans from a retrial. Unlike Blueford, Evans received a verdict of acquittal, so the story should end there. To deny both Blueford and Evans access to the Double Jeopardy Clause would be to do something that I would regard as worse than elevating form over substance. It would be to elevate outcome over everything else. The Double Jeopardy Clause has to mean something, so that people—legislators and attorneys, for example—can know how to avoid its pitfalls in the future. And for that to happen, we must have a doctrine that is consistent, even if that consistency gives us a substantively undesirable outcome in this individual case.
