The Double Jeopardy Clause does not forbid all multiple judicial outcomes. It forbids only second jeopardies “of life or limb.” Even without reliance on the “life or limb” text, it is self-evident that double jeopardy protection does not apply to all forms of blameworthiness. We would be justly surprised if told our tort suit for damages could not proceed because the tortfeasor had already pleaded guilty to a crime that manifested the same blameworthiness.
We would be surprised because the divergence between crime and tort can be traced at least as far back as Britton, whose treatise was published around 1290.1 Britton cautioned that a defendant acquitted as against private prosecutors is not necessarily “not guilty of what is laid to their charge” and thus should be prosecuted on “our” (the king’s) “behalf.”2 The modern rule permits both a civil and a criminal suit to proceed, regardless of whether the first proceeding ends in favor of or against the accused and regardless of whether the civil or the criminal case goes first.3
Unsurprisingly, locating the line between “civil” and “criminal” offenses is far from easy. A civil fraud prosecution may threaten liquidated damages that are hugely disproportionate to the loss caused by the wrongdoing; many regulatory offenses partake of both civil and criminal rationales. Reckless driving, for example, carries a penalty, typically a fine. Risk creation in driving is without doubt blameworthy, because of the potential harm to others, but is this double jeopardy blameworthiness?
Until 1997 the Court made quite a mess of the line between “civil” and “criminal”—once admitting that the “problem has been extremely difficult and elusive of solution.”4 In 1997 Hudson v. United States5 adopted a strategy that restores clarity and coherence to the civil/criminal dichotomy. Hudson manifests the legislative-prerogative thesis that underlies my entire double jeopardy account. But before I present the solution, I sketch the nature of the problem.
Though the Court has never seen the civil/criminal problem in terms of the literal meaning of the Double Jeopardy Clause, its “life or limb” language is a good starting place. As I argued in 1990, it seems unlikely that a civil judgment, even one that has elements of criminal punishment, could ever be characterized as “jeopardy of life or limb.”6 Thus, close attention to constitutional language suggests that the effort to decide which civil sanctions should be treated as criminal for double jeopardy purposes was wrongheaded from the beginning.
The “beginning” of this ill-fated venture was not the unanimous decision in United States v. Halper7 that the Court has now essentially repudiated.8 A hundred years earlier, the Court contemplated a robust double jeopardy crossover between civil sanctions and criminal penalties. In Coffey v. United States,9 the Court held that a criminal acquittal barred a civil suit brought on behalf of the government. The Court also discussed favorably a lower court case holding that a conviction for the criminal offense of conspiracy to defraud the United States barred the government’s civil suit to recover a penalty for the lost taxes.10
Coffey explained the earlier case: “The two alleged transactions were but one; and it was held that the suit for the penalty was barred by the judgment in the criminal case. The decision was put on the ground that the defendant could not be twice punished for the same crime, and that the former conviction and judgment were a bar to the suit for the penalty.”11 Oddly, Bishop did not mention Coffey when he concluded confidently in 1923 that, while double jeopardy applies to misdemeanors, it does not apply “to actions for the recovery of penalties.”12 A “life or limb” approach to understanding the scope of the double jeopardy protection agrees with Bishop rather than Coffey. But how are we to understand “life or limb”?
The earliest recorded use of “life or limb” appears in the Magna Carta, where it described trial by battle,13 a species of criminal trial in which the defendant’s life and limb were literally in jeopardy.14 Since trial by battle was, by the eighteenth century, a “long forgotten procedure of the dark ages,”15 the Framers almost certainly did not have the Magna Carta use in mind. Other evidence suggests that the Framers did not mean “limb” to be taken literally. Death replaced amputation and mutilation as the usual punishment for felons in England as early as the thirteenth century.16 Moreover, when debating the prohibition of cruel and unusual punishment, its opponents in the First Congress conceded that it would bar amputation. Questioning the wisdom of prohibiting cruel and unusual punishments, Samuel Livermore stated that “villains often deserve whipping, and perhaps having their ears cut off; but are we in future to be prevented from inflicting these punishments because they are cruel?”17 If Liver-more’s view of the scope of the Eighth Amendment was correct, there is no reason to believe the Framers would have implicitly signaled acceptance, in the Double Jeopardy Clause, of a type of punishment that they simultaneously forbade in the Eighth Amendment. The historical argument for taking “limb” literally is thus difficult to maintain, though Sigler implicitly accepts it.18
Moreover, to require a literal “jeopardy of life or limb” would have been a major restriction of the protection offered by Madison’s original language, which made “trial” or “punishment” the prohibited outcomes. As noted in chapter 2, the concern expressed on the House floor was not that the scope of protection was too expansive but, rather, that Madison’s language might deter an appellate court from reversing a conviction obtained in an unfair manner.19 During the debate on this point, Egbert Benson rephrased Madison’s protection as “no man’s life should be more than once put in jeopardy for the same offence.” This is virtually identical to Blackstone’s formulation of the “universal maxim of the common law of England, that no man is to be brought into jeopardy of his life more than once for the same offence.”20 Benson used “life” as the object of jeopardy, just as Blackstone did. But Benson did not intend to limit double jeopardy protection to “life” because, when seeking to describe the “humane intention of the clause,” he returned to Madison’s language, claiming that the intent was “to prevent more than one punishment.”21
Indeed, Benson moved to strike “or trial” from Madison’s formulation and thus to limit the prohibition to more than “one punishment.” In supporting that motion, Roger Sherman of Connecticut stated that “the courts of justice would never think of trying and punishing twice for the same offence.”22 Again, the focus seemed to be on punishment in general and not any specific form or gravity of punishment.
Benson’s motion to strike “or trial” from Madison’s language failed, and Madison’s original proposal was reported out of committee and later approved by the full House. But the Senate approved a Double Jeopardy Clause that echoed Blackstone’s universal maxim, with the expansive addition of “or limb” to Blackstone’s jeopardy of “life.” The Senate language was: “nor shall any person be subject to be put in jeopardy of life or limb, for the same offense.”23 For reasons forever lost to us, the conference committee agreed to language close to the Senate version.24
The final language, then, read “nor shall any person be subject for the same offence to be twice put in jeopardy of life or limb.”25 A fairly solid inference from this drafting history is that the Framers intended to expand the protection when they added “or limb” to “life” but that they did not mean “limb” to be taken literally. No one had insisted on limiting the clause to capital offenses, and it seems likely that the Framers simply wanted to broaden the protection to noncapital offenses by adding “or limb.” As Madison was a member of the conference committee that rejected his “punishment” language, there is no reason to believe he would have quietly agreed to a change that greatly reduced the protection of his original proposal.
Rejecting the literal meaning leaves “life or limb” unclear. One approach is to suppose that the Framers merely meant to reinstate Madison’s concept of punishment. This has been the Court’s working assumption for more than a century, but it did not prove a helpful concept until recently.
In 1873 the Supreme Court decided in Ex parte Lange that the Double Jeopardy Clause applied to all threats of criminal punishment—in effect, that a “jeopardy of life or limb” is the risk of a criminal penalty.26 This seems the best reading of Madison’s original language—“punishment,” in Madison’s formulation, probably meant “criminal punishment.”
But we do not know what “criminal punishment” is. Perhaps it had a self-evident quality in 1873. In the meantime, however, legislatures have created much new blameworthiness. On Michael Moore’s count, “a criminal code typically prohibits approximately 7,000 types of actions.”27 A substantial number of the new offenses are in the nature of regulatory or “strict liability” offenses that were unknown in 1873.
In Halper, the Court held that sixty-five counts of civil fraud constituted a “criminal” penalty that could not be sought following a criminal fraud conviction for the same fraud. The Court relied on the disparity between the money obtained by the fraud ($585) and the minimum liquidated damages that would attend conviction on all counts ($130,000) to conclude that the civil fraud prosecution was punitive and not remedial. Viewed as punitive, the “civil” proceeding was substantatively criminal and thus, a unanimous Court held, barred by the Double Jeopardy Clause.
Halper was an easy case; the minimum statutory damages were 220 times the amount obtained by fraud. It is difficult to view the prosecutor’s use of the civil statute under these circumstances as anything other than punitive. But as Stanley Cox has noted, “Halper’s assumptions and implications are far-reaching.”28 Indeed, “[i]f Halper’s message is taken seriously, many civil and criminal prosecutions could fall, because under current statutes, both civil and criminal prosecutors often try to punish the same conduct.”29 Defense lawyers made Halper-based arguments that many administrative sanctions were double jeopardy punishments—for example, driver’s license revocation;30 other license suspensions,31 including the suspension of a funeral director’s license;32 civil banking penalties;33 and the eviction of tenants.34 It is even possible, under a robust reading of Halper, for a wrongful death tort judgment to be viewed as a criminal punishment, thus undermining the parallel tort/crime systems that have been in place for centuries.35
Yet another example of the “far-reaching” implications of Halper is Department of Revenue of Montana v. Kurth Ranch,36 where the Court split 5-4 over whether a tax on the possession of illegal drugs is a criminal penalty barred by criminal conviction for the same drug possession. The Kurth Ranch majority concluded that the tax was punitive and thus a criminal penalty. The penalty was being collected in a second proceeding that was, the Court noted in passing, “the functional equivalent of a successive criminal prosecution.”37 Thus, despite clear legislative intent to impose the tax in addition to the criminal sanction, the Double Jeopardy Clause required that the tax “be imposed during the first prosecution or not at all.”38
Scalia used Kurth Ranch as an opportunity to recant his original support for the Halper concept, arguing in dissent that a civil proceeding could never be functionally criminal unless it was actually criminal. Scalia here recovers the historical rationale reported by Bishop: double jeopardy extended to felonies and misdemeanors, but not to actions “for the recovery of penalties because these are not criminal proceedings.”39 Though Bishop gave no test for “criminal proceedings,” the Court developed a seven-part test in Kennedy v. Mendoza-Martinez.40 The issue in Kennedy was whether deprivation of citizenship was a criminal punishment that required the procedural protections of the Fifth and Sixth Amendments. The Court held that it was, relying on seven factors. One factor was the sole basis for the Court’s test in Halper: whether the sanction “will promote the traditional aims of punishment—retribution and deterrence.”41 Others included whether “the sanction involves an affirmative disability or restraint, whether it has historically been regarded as a punishment, whether it comes into play only on a finding of scienter,” whether “an alternative purpose to which it may rationally be connected is assignable for it, and whether it appears excessive in relation to the alternative purpose assigned.”42
Most of these factors are helpful to Scalia’s claim that the seven-part Kennedy test would not characterize a tax proceeding as criminal.43 A tax is not a “restraint,” nor is it an “affirmative disability” in a punitive sense. Taxes have not been historically regarded as punishment, and the duty to pay taxes arises without regard to criminal intent (“scienter”). There is an alternative purpose that is “rationally connected” to the tax (various governments heavily tax alcohol and cigarettes). The only Kennedy factors in favor of treating the tax proceeding as a criminal trial are that the tax serves traditional criminal law goals (retribution and deterrence) and is applied to conduct that is already a crime. On balance, though, the Kennedy factors suggest that a tax proceeding is not a criminal proceeding.
This would also be the result from applying a legislative-prerogative view of the role of legislative intent in double jeopardy analysis. Recall Justice Washington’s 1820 view that the legislature can have as many penalties as it wants.44 If my legislative-prerogative principle is right, then Stanley Cox is wrong that double jeopardy provides limits on government’s ability to have two (or more) penalties for the same conduct. Because the Montana statute explicitly contemplates the tax being imposed following criminal sanctions, the tax statute is not the same legislatively authorized offense as the criminal statute. Nancy King has argued that Kurth Ranch strikes a discordant note when compared to Dixon (the case that overruled Grady).45 Dixon used an inference from the Blockburger test to permit two avowedly criminal trials for the same conduct. Yet in Kurth Ranch, the explicit legislative intent to create separate offenses was ignored, and a tax was held an impermissible criminal penalty.
This seems truly odd. Explicit legislative intent to create cumulative sanctions cannot justify a tax, but implicit intent can justify a prison term. Indeed, to suggest that a proceeding to collect taxes places a defendant’s “life or limb” in jeopardy is a linguistically startling proposition. There is no risk of incarceration and, therefore, no sense (metaphorical or otherwise) in which the taxpayer’s limbs are in jeopardy.
When law gets that far removed from common usage and common understanding, it cries out for correction.46 David Rudstein would meet me halfway; he argues that a civil proceeding is not a jeopardy and thus the state could always bring a civil proceeding first and then a criminal one.47 The state could not bring a punitive civil suit following a conviction, on Rudstein’s view, because he accepts Halper’s assessment that dual punishments are barred irrespective of legislative intent.
But Peter Henning has demonstrated the incoherence of the Halper rationale.48 There is so much wrong with Halper that Henning’s critique consumes twenty-eight pages and is too detailed to summarize adequately here. Suffice to say that Henning agrees with me that a civil penalty is simply not a criminal punishment for purposes of the Double Jeopardy Clause. Moreover, Henning asks, what does proportionality have to do with double jeopardy? Double jeopardy in no other context depends on measuring the quantity of the penalty; rather, it is a “categorical prohibition” of “repeated trials and unauthorized sanctions.”49 Henning’s critique is powerful, in my view leading to a legislative-prerogative thesis: if Congress wanted both a civil penalty and a criminal conviction for the same act of fraud, that is the end of the double jeopardy analysis.50
If my legislative-prerogative thesis is right, Halper must be wrong. And the Court in 1997 explicitly rejected the Halper analysis, using legislative prerogative as the basis for the decision.
The issue in Hudson v. United States51 was whether the Double Jeopardy Clause barred criminal prosecution of defendants for “essentially the same conduct” that had led to “monetary penalties and occupational disbarment.”52 The government initially assessed civil money penalties against defendants for making bank loans in an improper manner. Defendants agreed to pay fines of $12,500 to $16,500 and not to “participate in any manner” in future banking affairs without written consent of the government. Defendants were later indicted for three criminal charges: conspiracy, misapplication of bank funds, and making false bank entries.
Though I find the Hudson analysis a vast improvement over Halper, Hudson is itself unconventional. As Justice Stevens correctly noted in his concurring opinion, the Court could have avoided the Halper issue by holding that the criminal offenses were not the same Blockburger offenses as the civil offenses—each required proof of an element that the others did not.53 For some odd reason, the Court seems to forget or confuse the same-offense requirement when applying the Double Jeopardy Clause to civil sanctions. There was, for example, no reference to “same offense” in either Halper or Kurth Ranch.
Confusion was also manifested in Kansas v. Hendricks when the Court remarked obscurely: “The Blockburger test … simply does not apply outside of the successive prosecution context.”54 This statement is flatly wrong. Blockburger itself was a single-trial case, and the Court has cited and relied on Blockburger many times for its presumption about congressional intent to impose cumulative penalties in a single trial.
Perhaps the Court meant in Hendricks that Blockburger does not apply when the issue is whether a civil sanction is a double jeopardy punishment. If this is what the Court meant, it is enigmatic and puzzling. Remember that the issue is whether the double jeopardy multiple-punishment doctrine forbids a civil sanction after a defendant has already been convicted of a criminal offense. But the only way the Double Jeopardy Clause can bar the civil sanction is if the two offenses are the same offense. Otherwise, the Clause by its literal language does not apply.
That the Court seems to forget or confuse the same-offense requirement when applying double jeopardy across the civil-criminal divide probably speaks volumes about the lack of conceptual fit. Justice Stevens is right in his separate Hudson opinion that offenses must be tested by Blockburger before there is any concern about multiple punishments arising in the civil and criminal spheres. So tested, the offenses in Hudson would not have imposed multiple punishments for the same offense, and the convictions could have been affirmed without mentioning Halper.55
But conceding Stevens’s technical point, it is clear from the other opinions, especially that of Justice Scalia, that the Court wanted to finish the job of overruling Halper that it had begun the year before in United States v. Ursery.56 Ursery held that civil forfeiture is not a double jeopardy punishment (the items forfeited were a house, in one case, and the proceeds of drug transactions, in the other). The Court stressed the inherently remedial nature of forfeiture57 and, more important, read Halper narrowly.58 Ursery was correctly decided. As Stefan Cassella noted prior to Ursery, “For two hundred years, it was unquestioned that civil forfeiture did not implicate the Double Jeopardy Clause.”59 Cassella recommended returning the law to its historic understanding, as Ursery did.
The question remaining after Ursery was whether the retrenchment on Halper would be limited to civil forfeiture. This is where Hudson comes in. If one were willing (as the Court has been) to blithely ignore the same-offense requirement in these civil/criminal cases, the Halper mode of analysis gave the Hudson defendants a good chance to use the civil penalties to bar the criminal trial (indeed, the district court ruled that the monetary penalties constituted double jeopardy punishment). Halper defined punishment as a sanction that “cannot fairly be said solely to serve a remedial purpose, but rather can only be explained as also serving either retributive or deterrent purposes.”60 This definition does not screen out many trivial offenses. Most regulatory offenses serve deterrent purposes. The fine for double parking, for example, hardly serves a remedial function; its purpose is to deter double parking. While license revocations serve a remedial purpose, do they not also serve as a deterrent and as a punishment?61 As Peter Henning noted, Halper’s broad definition of punishment created “enormous uncertainty” about whether civil and criminal sanctions could ever be imposed for the same conduct.62
The Court in Hudson agreed that Halper’s analytical structure was flawed and “ill considered” and had “proved unworkable.”63 It adopted a legislative-prerogative presumption: the legislature’s preference for a civil or criminal “label” for the sanction will control unless there is the “clearest proof” that the sanction is punitive despite the legislative intent to the contrary. This legislative preference can be express or implied. Hudson makes clear that the legislature can have both a civil and a criminal penalty, in the absence of the “clearest proof” that the civil penalty is actually criminal.64
The rebuttable nature of the Hudson presumption is inconsistent with my robust legislative-prerogative thesis but is unlikely to have much pragmatic effect, given the requirement of “clearest proof” that a penalty civil in form is substantively criminal. By adopting the seven-part Kennedy test that Justice Scalia endorsed in his Kurth Ranch dissent, the Court has made it difficult for defendants to show the “clearest proof” that a sanction is punitive. Most of the Kennedy factors are not satisfied by penalties less severe than incarceration.
Risk of incarceration is also consistent with a “life or limb” understanding of when double jeopardy applies. We can begin with history. Early English law permitted defendants to purchase emendment (atonement) for some crimes by pecuniary compensation to the injured party (the bot) and to the king (the wite).65 But emendment was not available for serious crimes.66 The rationale underlying this principle was that “by the gravest, the unemendable, crimes, a man ‘forfeited life and member and all that he had.’”67 Although the definition of which crimes could be emended varied over time,68 it included only serious offenses. Thus, the “life or limb” concept in the Double Jeopardy Clause could well have referred to crimes for which a defendant forfeited “life and member,” not in a literal sense but in the metaphorical sense that denied him the possibility of paying compensation in lieu of “punishment.”
Halper misconceived what is unique about crimes when compared to other statutory prohibitions. The question is not whether a particular law seeks to deter conduct—indeed, contract law also seeks to deter certain kinds of conduct—but whether that conduct is deserving of condemnation. Condemnation manifests retribution (deserved punishment), rather than deterrence.
Figuring out which statutory penalties fall on the condemnation side of the line is no easy matter unless we have a mechanical test. In determining the roughly analogous question of whether Sixth Amendment cases involve a “criminal prosecution,” the Court has drawn two mechanical lines. On the issue of whether a trial is a criminal prosecution for purposes of the right to a jury trial, the Court explicitly used the “gravity” of an offense as part of its analysis. The Court noted that the “penalty authorized by the law of the locality may be taken ‘as a gauge of its social and ethical judgments’ of the crime in question.”69 Based on this criterion, the Court drew a distinction between “serious” crimes, which required a jury trial, and “petty” offenses, which did not. The Court drew the “gravity” line for jury trials at imprisonment of six months.70
But six months’ imprisonment is not a good “life or limb” line. In the jury trial context, the dividing line was based not on whether conduct deserved condemnation but, rather, on whether the crime was “serious.” By beginning with the idea that some crimes (petty ones) did not require a jury trial, the Court was dividing the world of crime into two parts. But it seems likely that all crimes manifest condemnation. Perhaps a better marker for crime versus noncrime would be whether the conduct is punishable by a jail sentence. While the line between incarceration and other sanctions might not always prove a true measure of condemnation—some states might authorize a few days in jail for trivial offenses—there is value in having a clear, easily administered line.
This suggests that a better Sixth Amendment analogy is the right to counsel, rather than the right to a jury trial. In Argersinger v. Hamlin,71 the Court held that incarceration created a right to appointed counsel. The Court later identified as the “central premise of Argersinger … that actual imprisonment is a penalty different in kind from fines or mere threat of imprisonment” and then remarked that this premise “is eminently sound and warrants adoption of actual imprisonment as the line defining the constitutional right to appointment of counsel.”72
The Argersinger rule is backward-looking: judges may not incarcerate a defendant who has been denied the right to counsel. But in the double jeopardy context, the rule must be forward looking. Jeopardy implies risk, without regard to the actual outcome. The prohibition of double jeopardy is intended to prevent the second exposure to risk. For this reason, the Court has held that an interlocutory appeal must be granted a defendant who raises a double jeopardy objection to a second trial.73 Thus, the “mere threat of imprisonment” should justify treating an offense as a “life or limb” offense.
Incarceration seems qualitatively different from penalties that do not include incarceration.74 Moreover, incarceration captures what is unique about criminal punishment—its retributive, unemendable nature. Fines resemble civil law outcomes in tort, but incarceration is uniquely criminal and uniquely lacking in remedial or compensatory goals. In holding the Double Jeopardy Clause applicable in juvenile delinquency hearings, a unanimous Court noted that the consequences of these hearings included both the stigma associated with having committed a criminal act “and the deprivation of liberty for many years.”75 Thus, whether one seeks to define “criminal punishment,” as the Court has done, or attempts to equate double jeopardy blameworthiness with condemnable conduct, as this chapter has done, the end result should be the same: the Double Jeopardy Clause protection is triggered only if both offenses authorize incarceration.
If either offense does not authorize incarceration, the second trial or punishment cannot be a second jeopardy of life or limb and is therefore permissible. If both offenses authorize incarceration, that is not, of course, the end of the inquiry; there is still the issue of whether the two offenses are the same offense. That is why Kurth Ranch is doubly wrong: it not only missed the nonjeopardy status of the tax proceeding but also held that the two proceedings were for the same offense, despite the legislative intent to authorize both.
Does my legislative-prerogative thesis mean that Halper was wrongly decided? Not necessarily. The prosecutor’s use of a civil offense in Halper arguably manifested retribution rather than deterrence. But, if so, he was using the civil fraud statute in a way not intended by the legislature. There is no reason for a legislature to enact a civil fraud and a criminal fraud statute unless the two have different purposes. The self-evident difference is that criminal fraud serves retribution and deterrence, while civil fraud serves the goals of compensation and deterrence.
Read this way, Halper suggests that “life or limb” blameworthiness can occur in two ways. One is that a statute creates “life or limb” blameworthiness by proscribing conduct that deserves condemnation, signaling this purpose, I have suggested, by authorizing incarceration. But if Halper is right—and the result is intuitively satisfying—prosecutors can use a statute that does not proscribe “life or limb” blameworthiness in a way that condemns the conduct and seeks retribution for it. When prosecutors use civil statutes in that manner, the Double Jeopardy Clause should apply.
This second “life or limb” usage is necessarily a case-by-case determination for which no easy rule exists. The heavy presumption should be that no proceedings involve jeopardy unless brought under a statute that authorizes incarceration. Exceptions would be, in the words of Halper, “the rare case” where “the civil penalty sought in the subsequent proceeding bears no rational relation to the goal of compensating the Government for its loss.”76 In this very narrow category of cases, it would seem that prosecutor has abused the intent of the legislature by using a non-jeopardy statute to achieve retribution. Limited to this very narrow category of “life or limb” application of a nonjeopardy statute, Halper can be defended.
Moreover, Halper does not have to fall in order to take down Kurth Ranch. The difference between the two is that the prosecutor in Halper used a nonjeopardy statute in a way contrary to what the legislature intended. In Kurth Ranch, the legislature explicitly contemplated imposition of the tax following criminal sanctions. There was, then, no reason even to reach the issue of how to characterize the sanction. On Michael Moore’s insight, two offenses are the same (or different) regardless of whether prosecuted together or separately.77 On my legislative-prerogative view, clear legislative intent to punish cumulatively always creates different offenses.78 Combining these insights, I argue that the tax should have been upheld in Kurth Ranch, whether or not it was viewed as a “life or limb” penalty. As Justice Washington noted, the legislature can have as many penalties as it wants; in this book, I add only that we must have good reason to believe the legislature intended more than one penalty. The statute in Kurth Ranch provides this assurance, and the Court should have stopped at that stage of the analysis.
The civil/criminal issue, like all others connected with my double jeopardy account, ultimately turns on legislative intent. If, for example, the legislature provides that it does not want both a “civil” and a “criminal” sanction to apply, no one would argue that both should apply. In this context, to apply both would be to punish without legislative authorization, which, as Scalia points out, is a violation of due process.79 The harder case is that in which the legislature states that it wants both sanctions to apply—the Kurth Ranch situation. On a legislative-intent account of double jeopardy, however, clear legislative intent means that both sanctions can be imposed. Why would intent to punish not be given effect in the same way as intent not to punish?
If the legislature intends both sanctions, the Court has held that they cannot be for the same offense when imposed in a single trial, and the civil/criminal question is irrelevant. The only way in which my legislative-prerogative account is substantively different than the Court’s double jeopardy doctrine is that the Court continues to suggest that the Double Jeopardy Clause may somehow operate independent of legislative intent in the context of successive prosecutions. Because criminal and civil sanctions are very unlikely to be imposed in a single proceeding, the civil/criminal issue takes on the character of successive prosecutions and thus raises the issue of whether successive prosecutions should be treated differently for purposes of the “same offense” analysis.
The Court’s labored seven-part test that can rebut the Hudson presumption would be totally unnecessary if “same offense” had the same meaning in successive prosecutions that it does in single trials. If that were true, Missouri v. Hunter would mean that the legislature can have as many penalties for the same conduct as it wants. Thus, on my account, there would be no possibility to rebut the Hudson presumption that the legislature can authorize both a civil penalty and a criminal punishment.
Drawing the statutory “life or limb” blameworthiness line at authorized incarceration easily resolves the civil/criminal dichotomy. Because civil statutes do not authorize imprisonment, they can never create “life or limb” blameworthiness unless the prosecutor abuses the civil nature of an offense, as in Halper. This argument is fully consistent with Ex parte Lange’s holding that “life or limb” means “criminal penalty.”
This argument is also consistent with the constitutional text. If “life or limb” is 1792 shorthand for criminal penalties that could not be avoided by paying a fine (unemendable penalties), it would include offenses that are crimes for purposes of the Sixth Amendment “criminal prosecution” requirement. Prosecutions for crimes punishable by incarceration seem likely to meet the Kennedy test for “criminal prosecution.” Incarceration would involve an affirmative restraint and would have historically been regarded as a punishment. And, in all likelihood, any crime that authorizes incarceration would require some form of criminal intent (scienter) and would promote retribution and deterrence over remedial goals.
Limiting double jeopardy protection to offenses that proscribe “life or limb” blameworthiness has practical advantages. The more serious the offense, the more likely that the prosecutor’s office is on notice of the existence of the charge and pending prosecution. If the prosecutor goes ahead with a prosecution for a less serious offense—but one that authorizes incarceration—the state can be charged with having made an election of that offense rather than a more serious one.80 But if the first “prosecution” is for a civil offense or a regulatory offense that fails to provide for incarceration, no bar would arise to a criminal prosecution, thus avoiding the unseemly result of a traffic fine barring a trial for vehicular homicide.81 Judges naturally recoil from applying the Double Jeopardy Clause to permit minor traffic offenses to substitute for serious felonies.82
Reading “life or limb” as a floor for double jeopardy blameworthiness is conceptually satisfying; “life or limb” is an explicit part of the Double Jeopardy Clause and, understood as referring to offenses that authorize incarceration, it is consistent with the line the Court has drawn in defining the right to counsel in “all criminal prosecutions.” Now that we have a floor for double jeopardy blameworthiness, the next task is to understand when the legislature intended distinct double jeopardy blameworthiness and thus different offenses.