Purposes of the Bar

Purposes of the Bar

The text of the Clause says little about how far the guarantee reaches, and the debates of the founding generation shed little light on the question, so courts read the Clause in light of the purposes it serves.¹ The account below therefore comes before the mechanics. It treats the purposes as one connected story about the power of the State, the finality of a verdict, and the interest a defendant acquires in the jury that has been chosen.

The oldest strand is finality. At English common law a defendant counted as having been put in jeopardy only after a conviction or an acquittal following a complete trial, and a primary function of that rule resembled the one res judicata serves in civil cases.² Res judicata is the civil doctrine that once a court has finally decided a dispute between parties, they may not litigate it again. The common law carried the idea into criminal cases through the maxim that no one ought to be twice vexed for the same cause, and it enforced the maxim through pleas of former acquittal and former conviction.³ A verdict is the point at which the legal system has finished speaking about a charge, and a system that let the charge be reopened at will would never produce an answer anyone could rely on. The civil analogy has a limit, however. In civil litigation finality saves court time and spares parties redundant suits, whereas in a criminal case it chiefly shields the defendant from prosecutorial oppression.⁴ That difference leads to the second strand, which concerns what the State can do with its resources.

In Green v. United States the Court described the underlying idea as one deeply ingrained in Anglo-American law: the State, with all its resources and power, should not be allowed to make repeated attempts to convict an individual, because each attempt subjects the person to embarrassment, expense, and ordeal, forces a continuing state of anxiety and insecurity, and raises the chance that even an innocent person will be found guilty.⁵ Benton relied on that passage when it held the Clause fundamental enough to bind the States, a subject taken up in a later section.⁶ The last harm on the list follows from an imbalance in the contest. The prosecution needs to win once, the accused must prevail every time, and each added trial costs the defendant far more than it costs the government. A rule allowing unlimited attempts would therefore convict, sooner or later, some people the evidence never justified convicting. The same reasoning explains a result that can look odd at first: an acquittal bars a later prosecution for the same offense even when the acquittal appears mistaken, and the government cannot obtain a new trial by appealing an acquittal it thinks erroneous.⁷ As Justice Brennan restated the point in dissent in Dowling v. United States, allowing a second trial after an acquittal, however mistaken, would create an unacceptably high risk that the government could wear the defendant down.⁸ The system accepts the occasional wrong acquittal as the price of denying the State repeated tries.

A related strand concerns the prosecution's own incentives. Barring the retrial of an acquitted defendant keeps the State from rehearsing and improving its presentation through repeated attempts until one succeeds.⁹ A first trial teaches the prosecutor which witnesses faltered on cross-examination, which theory the jurors doubted, and what the defense will say. If the State could use that knowledge at a second trial, the first would become a practice run, and the prosecutor would have little reason to put the best case forward at the outset. The working principle is that the prosecution gets one chance to try a defendant.¹⁰ A concurring opinion in North Carolina v. Pearce describes the same concern in three settings: harassment by a multitude of accumulated trials, retrial after an acquittal in the hope of a different result, and retrial after a conviction in the hope of a heavier penalty.¹¹

The second scenario in that pair rests on a rule the next sections develop: a dismissal before the jury is sworn leaves the State free to refile within the limitations period.¹² Before reaching it, Green itself shows how the strands operate on one set of facts. The jury in that case could have convicted Green of first-degree or second-degree murder, and it found him guilty of second-degree murder without convicting him on the first-degree charge.¹³ The court of appeals reversed the second-degree conviction for lack of evidence and sent the case back for a new trial, where the State prosecuted him for first-degree murder. The Court held that the second trial put him in jeopardy twice for the same offense.¹⁴ The same opinion confirms that a defendant whose conviction is set aside on his own appeal may be tried again for the offense he appealed.¹⁵ What separated the two charges was how the first trial had treated them. Green's own appeal put the second-degree charge back in play, whereas the first-degree charge was one the jury had declined to convict on and his appeal had not touched. Trying him on it again would have handed the State exactly the second attempt the bar exists to prevent.

The last strand protects something the common-law rule never recognized. Under that rule, a jury discharged without a verdict left the defendant open to a second trial, because no one could say that the defendant's life or limb had been in jeopardy.¹⁶ Modern law starts the protection earlier, and the reason is the defendant's interest in having the trial finished by the tribunal that was chosen. In Crist v. Bretz the Court called this a "valued right to have his trial completed by a particular tribunal" and treated it as the foundation of the rule that jeopardy attaches when the jury is empaneled and sworn.¹⁷ The interest is easiest to see by asking what the State could do without it. Selecting a jury takes effort and strategy on the defense side. If the prosecution could end the trial whenever the panel seemed inclined to acquit and begin again before a fresh panel, it could discard decision-makers it disliked until it found some it preferred. Green names that danger directly, observing that the rule prevents a prosecutor or judge from forcing a second prosecution by discontinuing the trial when it appears the jury might not convict.¹⁸ The section on when jeopardy attaches returns to this interest, because it is the anchor for the sworn-jury rule.

Because the Clause is read through its purposes, it does not reach situations where none of them is at stake. One federal court of appeals faced a district court that had replaced a single juror before any witness testified. The replacement exposed the defendant to no more embarrassment, expense, anxiety, or ordeal than seating an alternate at the start would have done, and nothing suggested prosecutorial tactics lay behind it. Holding that the first trial had ended at that stage would also have defeated the public interest in giving the prosecution one full opportunity to convict.¹⁹ The State is entitled to one complete chance, and the defendant is entitled to be spared a second, so the Clause marks the line between those two interests. The same account guides the next two questions. On incorporation, Benton asked whether the bar against repeated prosecution is a fundamental part of the American scheme of justice, and it answered that question by relying on these purposes.²⁰ On attachment, Crist treated finality, protection from the harassing exposure of a criminal trial, and the interest in a chosen jury as having combined to produce the federal rule.²¹