Limits on the Retrial

Limits on the Retrial

Once the earlier analysis shows that a second trial is permitted, two questions remain: what the government may pursue in it, and what the defendant may face if it ends in conviction. The general rule allowing retrial after a successful appeal rests on the idea that a defendant who attacks a conviction seeks to nullify it, so that the slate is wiped clean and no trial is treated as having occurred.¹ That idea has a natural boundary. It explains why the government may retry what the defendant set aside, but it gives the government no reason to reach a charge the first jury already resolved in the defendant's favor, and it gives no authority to make the defendant pay for having appealed.

The two limits in this section therefore have separate logic. The first, implied acquittal, protects a decision the first jury actually made, and its mechanism is finality. The second, the restraint on heavier sentences, protects the defendant's willingness to appeal, and its mechanism is the removal of a deterrent. Both operate after the retrial question has been answered in the government's favor, which is why they come last in the analysis.

Implied Acquittal

When a jury is instructed on a greater offense and a lesser included offense and returns a verdict of guilty on the lesser alone, the verdict operates as an implied acquittal of the greater offense.² The leading illustration is Green v. United States, a murder prosecution in which the first jury convicted on the lesser degree and the conviction was later reversed for trial error. The government could retry the defendant on the lesser offense, but any attempt to reinstate the greater charge was barred, because the first jury's verdict had already closed that charge off.³ Both outcomes appear in a single case, which is the useful feature of Green: the same reversal reopens one offense and leaves the other shut.

Two different judgments came out of the first trial, and the defendant's appeal reaches only one of them. The verdict on the lesser offense is a conviction, and by appealing it the defendant asks to have it treated as a nullity, which is why that offense may be tried again.⁴ The implied acquittal of the greater offense is a separate judgment. The jury had the greater charge before it and the power to convict on it, and by choosing the lesser it is treated as having decided that the greater was not proved. The defendant never asked an appellate court to disturb that decision, so the appeal leaves it standing.

The alternative shows why the rule has this shape. If a successful appeal reopened the greater charge, every defendant convicted of a lesser offense would have to weigh a meritorious appeal against the risk of facing the more serious charge a second time, and some would give up the appeal. The prosecutor would also gain a second attempt at a charge the first jury had declined to find. Treating the lesser verdict as final on the greater removes the deterrent from the defendant and the second attempt from the government.

The inference does not depend on the jury alone. If the trial court refuses to instruct on the more serious charge but instructs on the lesser one, the verdict on the lesser offense is likewise treated as an implied acquittal of the greater.⁵ Sometimes the government retries the greater charge anyway, and the later outcome then depends on what the second jury did. In Morris v. Mathews, the defendant was convicted of the barred greater charge at the second trial, and the Court held that the appellate court could reduce the conviction to the permissible lesser charge.⁶ In Price v. Georgia, the improper second trial on the greater charge instead produced a conviction on a lesser offense that was not itself barred, and a new trial was required because prosecuting the accused on the higher charge may have prejudiced the jury.⁷ When the conviction is for the barred offense, the court can trim it back to the lesser charge; when it is for a lesser offense, there is nothing to trim, and the possible prejudice from the higher charge requires a fresh trial.

The implied-acquittal inference also depends on the jury having returned a verdict. In Blueford v. Arkansas, the jury was told to consider each lesser offense only after finding the defendant not guilty of the greater ones, and it reported that it was unanimous against capital murder and first-degree murder but deadlocked on manslaughter. The jury kept deliberating after that report and finally announced that it was hopelessly deadlocked, and the Court held that retrial on the greater charges was not barred, because the reported votes were not final verdicts and remained open to reconsideration.⁸ Green and Blueford differ in the one fact that matters. The Green jury returned a verdict on the lesser offense while the greater was before it, and the Blueford jury never returned any final verdict, so the earlier discussion of deadlock reports controls there.

Students lose this issue when they treat the reversal as a fresh start on the original indictment, since the appeal reopens only what the conviction covered. On an exam, ask whether the first jury returned a verdict on a lesser offense with the greater offense before it. If it did, cap the retrial at the lesser offense, and if it deadlocked and returned nothing final, the government may pursue the charges on which no verdict was reached.

Sentencing After Retrial

The second limit concerns what the defendant may receive, and it applies even when the retrial is proper and the charges are correctly framed. In North Carolina v. Pearce, the defendant was convicted of assault and sentenced to 12 to 15 years, and the conviction was later reversed because an involuntary confession had been admitted. On retrial he was convicted again and received an eight-year term, which, added to the time he had already served, produced a longer total sentence than the first one.¹⁰ The Court did not forbid every increase. It held that a more severe sentence may follow retrial so long as it rests on objective information occurring after the initial conviction and is not simply a penalty for having appealed.¹¹

Each half of that standard answers a distinct danger. The first is retaliation by the sentencing judge: a judge whose first trial was reversed has a motive, which may operate without conscious intent, to answer the appeal with a heavier term. The second danger falls on the defendant's decision whether to appeal. A defendant who knows that a successful appeal could be met with a longer sentence will weigh a meritorious claim against that risk, and some will let the conviction stand, which defeats the purpose of having an appeal. Requiring that any increase rest on objective information forces the judge to justify it with something other than the appeal. The rule leaves room for higher sentences, because a judge who learns new facts about the defendant is entitled to take them into account, and what it screens out is an increase that has no explanation except the appeal.

Three later decisions refine the standard by showing where the danger of vindictiveness is larger or smaller:

• Texas v. McCullough allows a court sentencing after retrial to consider any event committed before the original sentencing, including a conviction, so the range of acceptable justifications is wide.¹²
• Chaffin v. Stynchcombe holds that when a jury imposes the second sentence, the jury's independence minimizes the possibility that vindictiveness will influence the result, since a jury has no personal stake in the earlier reversal.¹³
• Alabama v. Smith holds that the presumption of vindictiveness does not apply when the first sentence rested on a guilty plea and the higher sentence follows a trial on the merits, because a trial puts far more about the offense before the judge than a plea record does.¹⁴

Where a judge imposes a heavier sentence after a successful appeal and none of these refinements applies, the risk of retaliation is at its highest, and the sentence must be justified by objective information.

The pair holds the trial, the judge, and the numbers constant, so the only variable is the reason on the record. Students lose this issue when they treat a heavier sentence after retrial as automatically forbidden, since the standard polices the basis for the increase and does not cap the sentence itself. On an exam, ask first who sets the second sentence and whether the first sentence followed a guilty plea. If a judge sets it after a first sentence that followed trial, require an objective reason that has nothing to do with the appeal.

Implied Acquittal

Once the earlier analysis shows that a retrial is available after a conviction is set aside for trial error, the next question is what the government may put before the second jury. A first verdict can settle more than its words say. When a jury has been instructed on a charged offense and on a lesser offense included within it, and returns a verdict of guilty on the lesser offense alone, that verdict operates as an implied acquittal of the greater offense.¹⁸ The acquittal survives the defendant's successful appeal of the lesser conviction, which means the only offense open on retrial is the lesser one.¹⁹

The lesser-offense option itself comes from Rule 31(c), which allows a jury to find a defendant guilty of an offense necessarily included in the offense charged, of an attempt to commit the offense charged, or of an attempt to commit a necessarily included offense that is an offense in its own right.²⁰ That option is what gives the verdict its meaning. A jury holding both the greater and the lesser charge that returns the lesser has chosen between them, and the choice would make no sense if the jury believed the greater had been proved, so the law reads the silence on the greater charge as a rejection of it.²¹ The rejection then carries the finality that every acquittal carries. A second trial after an acquittal, however mistaken the acquittal may have been, would let a government with vastly superior resources wear the defendant down until even an innocent person might be convicted.²²

In Green v. United States the jury was instructed that it could find the defendant guilty of first or second degree murder, and it returned a verdict of guilty of second degree murder while remaining silent on the first degree charge.²³ The Court gave that verdict the same effect as an express finding of not guilty on first degree murder, so when the second degree conviction is set aside on the defendant's appeal, the retrial is limited to second degree murder.²⁴ The result can look inconsistent with the general rule that a successful appeal wipes the slate clean, since a conviction that is set aside is treated as though no trial ever occurred.²⁵ The reconciliation is that the first trial produced two separate outcomes. The defendant's appeal attacks the conviction, and only that outcome is nullified. The implied acquittal of the greater offense is a different outcome that no one has asked a court to disturb, so it stays in place and the greater charge stays closed. Reading the rule the other way would also expose a defendant with a meritorious appeal to the risk of facing the more serious charge a second time, a risk that would discourage the very appeals the system relies on.

The inference does not depend on the jury being the one that declines the greater charge. If the trial judge refuses to instruct on the more serious offense but instructs on the lesser one, a conviction of the lesser offense is likewise treated as an implied acquittal of the greater.²⁶ The charging document can frame the point as well. Where an indictment charges murder and manslaughter in the alternative for a single death, the jury can convict of only one, so a manslaughter verdict acts as an acquittal of murder, and after a successful appeal only manslaughter may be retried.²⁷

The pattern this rule now prevents is visible in Palko v. Connecticut. A jury that could have convicted of first degree murder returned a verdict of second degree murder, and the defendant was sentenced to life in prison.²⁸ Connecticut obtained a reversal on appeal for legal error, a second trial followed, and the new jury returned a verdict of first degree murder and a sentence of death.²⁹ The Court rejected the double jeopardy challenge because the Fifth Amendment guarantee was held not to bind the states, so Palko is useful here for its facts and not for its holding.³⁰ Those facts match the pattern Green addresses: a second degree verdict from a jury free to convict of first degree closes the first degree charge, and a first degree conviction at the retrial is the outcome Green's reasoning forecloses.

A guilty plea to a lesser included offense does not carry the same inference.³¹ The implied-acquittal rule depends on a factfinder who weighed the greater charge and chose the lesser. A plea has no such step, because no jury or judge has considered the evidence on the greater charge, and entering the plea does not cause jeopardy to attach on the greater offense.³² Nothing has yet been decided in the defendant's favor, so nothing is closed.

If the government retries the barred greater offense anyway, the remedy depends on what the second jury did. In Morris v. Mathews the defendant was convicted of the barred greater offense at the second trial, and the Court held that the appellate court could reduce the conviction to the permissible lesser offense.³⁵ In Price v. Georgia the second jury instead convicted of a lesser offense that was not itself barred, and a new trial was required because prosecuting the defendant on the higher charge may have prejudiced that jury.³⁶ The difference lies in what the verdict contains. A conviction of the greater offense necessarily rests on findings that cover the lesser one, so the court can trim it, while a verdict on some other lesser offense may have been shaped by the improper higher charge, and only a fresh trial can remove that possible taint.

The inference also requires that the first jury actually return a verdict. In Blueford v. Arkansas the jury reported that it was unanimous against capital murder and first-degree murder but deadlocked on manslaughter, then kept deliberating and finally announced that it was hopelessly deadlocked. The Court held that retrial on the greater charges was not barred, because the reported votes were not final verdicts and remained open to reconsideration, as the earlier discussion of deadlock reports explained.³⁷ The Green jury returned a verdict on the lesser offense while the greater was before it, and the Blueford jury never returned any final verdict, which is the single fact that separates the two outcomes.

Students lose this issue when they reason from the label lesser included offense alone, without asking whether a factfinder actually returned a verdict on the lesser offense with the greater offense before it. A guilty plea supplies no such verdict, and neither does a jury that reports its votes and then deadlocks. On an exam, identify what the first proceeding actually produced. If a jury convicted on the lesser offense after being given the choice, cap the retrial at that offense, and if nothing final was returned on the greater charge, the government may pursue it.

Sentencing After Retrial

Once a retrial is permitted and the charges are properly framed, a separate question remains: how much punishment may follow a second conviction. The Double Jeopardy Clause by itself places no ceiling on that sentence. In North Carolina v. Pearce the Court reasoned that the power to retry a defendant whose first conviction was set aside carries with it the power to impose any legally authorized sentence on reconviction, whether or not it exceeds the first one.³⁹ The explanation is that the defendant's own appeal nullified the original conviction, so the unserved portion of the first sentence will never be served and the second sentence punishes a new conviction.⁴⁰ That reasoning shows why a longer sentence is not automatically barred, and it also shows why a different safeguard is needed to keep that freedom from being used against defendants who appeal.

The safeguard appears in the facts of Pearce itself. The defendant there was convicted of assault and sentenced to 12 to 15 years, and the conviction was later reversed because an involuntary confession had been admitted. On retrial he was convicted again and received an eight-year term, which, added to the time he had already served, left him with a longer total sentence than the first.⁴¹ The Court held that a more severe sentence after retrial may stand so long as it rests on objective information that arose after the initial conviction and is not simply a penalty for having appealed.⁴²

Each half of that standard answers a different danger. The first is retaliation by the sentencing judge, who has just been reversed and may be tempted, consciously or not, to answer the appeal with a heavier term. Pearce's footnotes record that a retaliatory motive would be extremely difficult to prove in any single case and that longer sentences on reconviction were far from rare.⁴³ Because a judge's motive is so hard to prove, a safeguard that asks for an objective reason on the record works better than one that asks the defendant to prove retaliation. The second danger falls on the defendant's choice whether to appeal at all. A defendant who knows that winning a reversal could cost additional years has reason to leave a meritorious claim unpursued, and that result would hollow out the appeal as a remedy. Requiring objective information as the basis for any increase leaves judges free to respond to genuinely new facts about the defendant while taking the appeal itself off the list of permissible reasons.

Three later decisions show where the danger of retaliation is larger or smaller, and each one changes how closely a heavier sentence is examined:

• In Texas v. McCullough, a court sentencing after retrial may consider any event committed before the original sentencing, including a conviction, so the pool of objective justifications open to the judge is broad.⁴⁴
• In Chaffin v. Stynchcombe, a jury that imposes the second sentence is independent enough to minimize the possibility that vindictiveness will influence the result, since jurors have no stake in the earlier reversal.⁴⁵
• In Alabama v. Smith, the presumption of vindictiveness does not apply when the first sentence rested on a guilty plea and the higher sentence follows a trial on the merits, because a trial puts far more about the offense before the judge than a plea does, and that difference supplies an explanation for the higher term other than the appeal.⁴⁶

Where a judge imposes the second sentence after a first sentence that followed a trial, none of these refinements helps the government, the risk of retaliation is at its highest, and the heavier sentence must be justified by objective information.⁴⁷

The judge, the verdict, and the numbers are identical in both versions, so the justification on the record is the only fact that moved the answer. That is the same test Pearce applied, and it is why the analysis can be run in a fixed order.

Students lose this issue when they assume that the second sentence cannot exceed the first. The double jeopardy guarantee imposes no such limit, and the only restraint is the one aimed at retaliation, which examines the reason for an increase and leaves the length open when objective information supports it.⁵⁰ On an exam, a longer second sentence is therefore the start of the analysis, and the answer depends on the sentencer, the form of the first sentence, and the stated justification.