The Yale Law Journal

January 2014

Vindicating Vindictiveness: Prosecutorial Discretion and Plea Bargaining, Past and Future

abstract. This Note explores the past and possible future of the doctrine of vindictive prosecution, which prohibits retaliation against a criminal defendant for the exercise of a legal right. It presents a new historical account of the doctrine’s accidental origins. It argues that a revitalized conception of vindictiveness may be relevant to current controversy and doctrinal innovation surrounding prosecutorial discretion and coercive plea bargaining. A rule prohibiting a prosecutor’s deliberate punishment of a defendant’s exercise of her right to trial may allow for substantive limits on prosecutorial discretion, but in a way that respects rather than undermines the right to trial.

author. Yale Law School, J.D. 2013. I thank James Dawson, Ben Eidelson, Michael Graetz, and Linda Greenhouse for their thoughtful contributions.


The day after the Internet activist and hacker Aaron Swartz committed suicide in January 2013, his family angrily described his death as “the product of a criminal justice system rife with intimidation and prosecutorial overreach.”1 The federal government had “contributed to his death,” they argued, by bringing an “exceptionally harsh array of charges” for his unauthorized downloading of materials from the academic database JSTOR.2

Subsequent commentary agreed that Swartz had been pursued overzealously. Larry Lessig passionately criticized “the absurdity of the prosecutor’s behavior,” arguing: “Somehow, we need to get beyond the ‘I’m right so I’m right to nuke you’ ethics that dominates our time. That begins with one word: Shame.”3 Many others with a variety of ideological leanings agreed.4

Some characterized the behavior of the U.S. Attorney’s Office as “vindictive.”5 The prosecutors’ approach at least outwardly appeared to be aimed at making the defendant suffer a severe price for conduct that hurt no one and resulted in no financial gain. In the ordinary sense of the word,6 the government’s conduct may well have been vindictive. But it almost certainly was not vindictive in the legal sense.7

Legal vindictiveness does not refer to a prosecutor’s generic ill feeling toward, or even his desire to harm, a defendant. Rather, as defined by the Supreme Court,vindictiveness means that a prosecutor has retaliated against a defendant for the exercise of a legal right, denying her due process.8 One might think, then, that pursuing more severe charges or a harsher sentence after a defendant exercises her right to a jury trial9 would constitute vindictiveness. But it doesn’t. The law specifically permits severely penalizing defendants for going to trial in an effort to induce a guilty plea—or, in the Court’s words, “openly present[ing] the defendant with the unpleasant alternatives of forgoing trial or facing charges on which he was plainly subject to prosecution.”10

Using charging discretion aggressively to pressure defendants into pleading guilty is exactly what the existing doctrine of vindictive prosecution permits. And this is, by and large, what prosecutors do.11 But this conduct is precisely what many people found reprehensible and “vindictive”12 about the government’s generally ordinary13 treatment of Aaron Swartz. The Swartz case therefore brings into relief how inadequate our existing legal vocabulary and doctrine are to address prosecutorial behavior that many intuitively find unfair and improper. Indeed, the legal concept of “vindictive prosecution” is an essentially useless analytic tool in its current form. It fails to capture much of the behavior that we might properly want the law to name and shame with that label.

The current state of affairs also invites reflection on whether plea bargaining ought to be more closely policed and, if so, how. In three recent cases,14 responding to contemporary scholarship about plea bargaining and coercive prosecutorial power, the Supreme Court has attempted to regulate the market for pleas.15 Its chosen route for doing so has been the defendant’s right to the effective assistance of counsel. That right safeguards the “fundamental fairness” of the proceedings, ensuring a level of reliability sufficient to sustain “confidence in the outcome.”16 The underlying premise of the Court’s recent intervention into plea bargaining, therefore, is that some bargained-for pleas may be “bad” or “false” outcomes unworthy of confidence. But it is not at all clear that imposing obligations on defense lawyers, retrospectively enforced on post-conviction review, is the best way to avoid “bad” bargains. The Court’s recent doctrinal innovation in this area should prompt us to ask which actors are best positioned to establish and enforce norms of what constitutes a “good” or “true” bargain.17

The inadequacy of the way the law currently talks about “vindictiveness” and the recent doctrinal and scholarly ferment about plea bargaining give rise to this Note’s two related projects.

The first is to explain why “vindictive prosecution” came to have the particularized, unusual, and ultimately unhelpful meaning that it does. To that end, I present new research about the origins and development of the doctrine of vindictive prosecution. Most scholarship on vindictive prosecution is relatively old and highly doctrinal.18 The story newly told here is, I hope, interesting in its own right. It may help to illuminate the unpredictable circumstances that shape Supreme Court doctrine and, as a result, the lives governed by it.19 More purposefully, I aim to dispel the notion that the law’s equation of “vindictive” with “retaliation for the exercise of a legally protected right, other than the right to a jury trial” ought to be seen as unimpeachable precedent. And, most important, I hope that excavating the development of the doctrine can help us understand why it did not work and how it might become relevant again.

The Note’s second project, then, is to rehabilitate the legal concept of “vindictive prosecution” in hopes of contributing to current debate and doctrinal development about prosecutorial discretion, plea bargaining, and excessive punishment. Intervening in the ongoing discussion and responding to the Supreme Court’s recent efforts, I offer a new proposal. I suggest that reviving some parts of the old vindictive prosecution doctrine while shedding some of its unnecessary strictures can produce a useful framework for policing discretion, fairness, and leverage in the plea bargaining process.

I argue that the idea of vindictive prosecution as retaliation for the exercise of a legal right was a poorly conceived accident from the beginning. It grew out of a case meant to be about something else, and it failed to negotiate the fundamental tension between encouraging plea bargaining and honoring the right to trial. Vindictive prosecution doctrine was unstable because it reflected a Court lurching from one pole to another—from glorifying process values with little regard for practical consequences, to protecting plea bargaining at almost any cost to the right to trial—without working to stake out a middle ground. I identify such a middle ground and attempt to revitalize the legal concept of vindictiveness in a way that negotiates the tension between pleas and trials. In particular, I argue that an updated vindictive prosecution standard prohibiting prosecutors from punishing the defendant’s exercise of the right to a trial as a wrong would be responsive to salient problems in today’s criminal justice system. It may not be an ideal solution, but it suggests that it is possible to address important and competing objectives—imposing boundaries on prosecutorial discretion, valuing the constitutional right to trial, and permitting efficient bargains—in minimally disruptive fashion.

This argument proceeds in two parts.

Part I tells the story of how vindictive prosecution doctrine came to be. Its erratic path reflects the fundamental tension between venerating the Sixth Amendment right to a jury trial and coherently regulating a system designed to discourage its exercise. Section I.A presents background on the practice and law of plea bargaining. Section I.B discusses vindictiveness doctrine’s origins in another context. Section I.C discusses the Court’s hasty and unthinking application of vindictiveness to prosecutorial conduct, while Section I.D explains its retreat.

Part II contends that the idea of vindictive prosecution could prove newly useful in regulating prosecutorial discretion while both genuinely respecting and formally venerating the right to trial. Section II.A discusses existing scholarship and doctrine addressing the regulation of the plea-bargaining market. I argue that the Court is moving toward, but has not yet effectively imposed, constraints on prosecutors’ ability to drive an exceptionally harsh bargain. Section II.B argues for the merits of a new standard prohibiting prosecutors from acting with the subjective intent to punishthe defendant’s exercise of his right to trial, rather than merely with the goal or effect of deterring it. A conclusion follows.

For a brief period during the 1970s, “it appeared that ‘vindictive prosecution’ claims would be both common and successful.”20 Today, such claims are “rarely made and even more rarely succeed.”21 But the history is worth understanding and the concept worth renewing.

I. the accidental doctrine and its underlying tensions

In 1974, the Supreme Court held that a defendant must be free to exercise his statutory and constitutional rights “without apprehension that the State will retaliate by substituting a more serious charge for the original one.”22 The Constitution requires that he be free of this apprehension even if the state has no “actual retaliatory motivation” in filing a higher charge.23

This holding seems to pose an obvious problem: If a defendant must not fear the possibility of higher charges in retaliation for his exercise of a legal right, how could a prosecutor possibly bring a higher charge if the defendant exercises his right to trial? How can the government use leverage to plea bargain?

In 1978, the Court recognized this problem, asserting that, “in the ‘give-and-take’ of plea bargaining,” there simply could be “no . . . element of punishment or retaliation so long as the accused is free to accept or reject the prosecution’s offer.”24 In unsatisfying fashion, it eviscerated the doctrine it had just created, effectively excepting the right to trial from those protected against prosecutorial retaliation.

This Part explains the history of vindictive prosecution doctrine. It begins with necessary background. Section I.A quickly explains the roots of plea bargaining law and practice in the United States. Section I.B then discusses the cases in which the idea of “vindictiveness” originated. In Section I.C, I tell the previously untold story of how the Court stumbled into the new doctrine of prosecutorial vindictiveness in hasty and ill-considered fashion, apparently due in large part to North Carolina’s bungled management of the key case. I then explain in Section I.D how the Court responded to the practical consequences of its decision by swinging back to the other extreme—expansively protecting plea bargaining while failing to engage intellectually with the pressure it placed on the right to trial.

The instability of the doctrine reflects the tension between enabling orderly plea bargaining and protecting the constitutional right to jury trial. The Court’s foundational efforts in this area of law failed to look for, let alone find, a middle ground to reasonably accommodate these competing interests.

A. A Note on the Development of Plea Bargaining

To provide context for the argument that follows, this Section offers a very short primer on the development of American plea bargaining and the relevant law. The purpose is to make clear that the current plea-dominated system25 has deep historical roots and that plea bargaining’s practical importance should have been clear to the Supreme Court when prosecutorial vindictiveness first came before it.

Well before the Warren Court’s criminal procedure revolution, the rising crime rates of the 1960s and 1970s, or the increasing political salience of crime and public demand for law-and-order policies, plea bargaining was widespread in the United States.

John Langbein has explained that, through roughly the eighteenth century, the jury trial was a summary proceeding; as professional prosecution and the law of evidence developed in the nineteenth century, jury trials became resource-intensive affairs, and guilty pleas became an attractive alternative.26 By the early twentieth century, a plea was the most typical means of conviction. In Manhattan and Brooklyn, for instance, eighty-five to ninety percent of felony convictions were obtained by plea in the early 1920s.27 A number of cities established crime commissions and conducted studies in the 1920s, finding, to the “remarkable surprise” of many, that the overwhelming majority of convictions were obtained by plea.28 This system remained in place in the intervening decades. In 1967, the President’s crime commission similarly estimated that “as many as 90 percent” of defendants in some jurisdictions were convicted by plea.29

But, for virtually all of this time, plea bargaining was a kind of open legal secret, dominating the day-to-day administration of criminal justice without any formal recognition that it complied with the Constitution. In the late 1950s, it was very possible that the practice of negotiating to induce pleas might be declared entirely illegal. In 1957, a panel of the Fifth Circuit Court of Appeals held that plea bargains induced by prosecutorial promises of any kind were unlawful, only to be reversed by a three-to-two vote of the en banc court.30 In an unusual move that may have been designed to prevent the Supreme Court from pronouncing unfavorably on the legality of plea bargaining, the Solicitor General admitted that the defendant’s plea had been in error, making the case moot.31 Later, in 1969, the Court hardly seemed sanguine about the idea of negotiating pleas: “[A] plea of guilty is more than an admission of conduct; it is a conviction. Ignorance, incomprehension, coercion, terror, inducements, subtle or blatant threats might be a perfect cover-up of unconstitutionality.”32

In 1970, however, the Court explicitly acknowledged the “mutuality of advantage” that pleas offer the defendant and the government: the defendant “limit[s] the probable penalty” and can begin to serve his sentence immediately, while the state conserves “scarce judicial and prosecutorial resources” and “may more effectively attain the objectives of punishment” by imposing it swiftly.33 The Court conceded that plea bargaining’s prevalence and practicality did not necessarily make it constitutional—but then found it constitutional for essentially those reasons anyway. “[W]e cannot hold that it is unconstitutional,” the majority wrote in Brady v. United States, “for the State to extend a benefit to a defendant who in turn extends a substantial benefit to the State . . . .”34 Even though the defendant in Brady would have faced the death penalty if he did not accept the prosecutor’s offer, the Court found that the prosecutorial inducement was insufficient to compromise the “voluntary” and “intelligent[]” nature of the plea.35 Thus, alongside its companion cases limiting collateral review,36 Brady established that the use of prosecutorial leverage to extract pleas from defendants was fundamentally permissible.

There were some good reasons for the Court to finally bless plea bargaining in this period. First, the skyrocketing crime rates of the 1960s37 had produced more cases, and the emergence of crime as a high-salience political issue had led to an expansion of both substantive criminal law and policing, which, in turn, led to even more cases.38 And second, the expanded procedural protections recently granted to defendants increased the number of issues to be litigated at trial and made it more difficult and costly for the government to prevail.39 In Los Angeles, for instance, the average length of a criminal trial more than doubled between 1964 and 1968.40 As a result, with massive caseloads and higher costs—and without a commensurate injection of resources—plea bargaining became, even more so than in the past, the only way to obtain convictions and keep the system functioning.41

Thus, while plea bargaining had long been the norm, it was even more of a necessity by the mid-1970s. As the story of prosecutorial vindictiveness will show, however, the Court was hardly consistent in keeping its practical importance in mind.

B. The Idea of Vindictiveness

The seeds of prosecutorial vindictiveness doctrine were planted on Chief Justice Warren’s last day on the bench, when the Court handed down its decision in North Carolina v. Pearce.42 Justice Stewart’s opinion for the Court addressed two instances in which a judge had imposed a heavier sentence on a defendant who was retried for the same crime after successfully appealing his first conviction. Relying on the principle that the state may not impose unlawful conditions upon or penalize the exercise of constitutional rights, Pearce held that the Fourteenth Amendment’s Due Process Clause “requires that vindictiveness against a defendant for having successfully attacked his first conviction must play no part in the sentence he receives after a new trial.”43 Furthermore, the Court found that, “since the fear of such vindictiveness may unconstitutionally deter a defendant’s exercise of the right to appeal or collaterally attack his first conviction, due process also requires that a defendant be freed of apprehension of such a retaliatory motivation on the part of the sentencing judge.”44 Pearce therefore established a presumption of vindictiveness when a judge imposed a harsher sentence the second time around after a defendant successfully challenged his initial conviction.45

In follow-up cases, the Court declined to apply the Pearce rule where a different judge or jury imposed the harsher subsequent sentence.46 The Court reasoned that the second sentencing authority would have no “personal stake” in vindicating the initial decision by penalizing the defendant for appealing it, so there was no need for a presumption of vindictiveness.47 One such case, Colten v. Kentucky,48 addressed Kentucky’s two-tiered justice system for misdemeanors, which enabled a person convicted in a limited-jurisdiction inferior court to receive a de novo trial in a higher court. The Court similarly noted that there would be no particular reason for the second judge, who was simply making a fresh determination of guilt or innocence, to punish the defendant for his dissatisfaction with the first verdict.49 As such, a presumption of vindictiveness was unwarranted in a two-tiered system enabling de novo retrial as of right.

Pearce and Colten did not discuss whether the government’s desire to punish the defendant for seeking a new trial could violate due process. In fact, these cases seem to have left the question open. On the one hand, the state might have a “stake” in penalizing the defendant for trying to undo its success in the first proceeding. And, as Justice Marshall pointed out in dissent in Colten, the Pearce rule aimed to address the “inherent . . . danger” that the risk of a higher penalty could deter a defendant from exercising his right to appeal—a risk just as present from the government’s conduct as from the judge’s.50 On the other hand, nothing in Pearce or Colten spoke directly to prosecutorial behavior at all. And the government’s attempt to seek a higher sentence on retrial might not be “vindictive” in the same fashion as a judge’s imposition of higher sentence on retrial. It was the court’s prior decision—not the state’s—that the defendant had successfully appealed, so the judge would have a more personal motive than the prosecutor to get back at the defendant out of spite.

Against this doctrinal backdrop, the problems posed by as-of-right de novo retrials soon returned to the Court.

C. From Blackledge . . .

The best explanation for the doctrine of vindictive prosecution may be that Blackledge v. Perry,51 the case that created it, was simply not supposed to be about vindictiveness at all.

1. Waiver

North Carolina prisoner Jimmy Seth Perry had been convicted of misdemeanor assault for his role in a jailhouse fight and had received a six-month sentence from a court of limited jurisdiction that handled misdemeanor cases.52 Perry was entitled by North Carolina statute to a de novo trial in a higher court. When he filed his notice to seek one, the prosecutor charged him, on the same factual basis, with felony assault with intent to kill and inflict serious bodily injury.53 Perry pled guilty to felony assault and received a longer sentence.54

After he had pled to the felony, Perry filed a habeas petition, which the district court eventually granted on the basis of double jeopardy, which was at issue because of the two-tiered criminal court system. It found that “double jeopardy is involved when a defendant is subjected to prosecution for a greater offense upon trial de novo in a higher Court, after appeal from a lower Court.”55 The district court reasoned that the state should only have one initial opportunity to bring the higher charge, lest the initial misdemeanor trial function as “little more than a proving ground” for the state’s felony case.56 The state also argued that Perry had waived his constitutional protection against double jeopardy by pleading guilty, but the district court rejected this argument. Double jeopardy, it found, is a “fundamental right” that could not be implicitly waived by a guilty plea because it “goes to the power of a Court to try a person.”57 That is, double jeopardy could not be waived by a plea because it undermined the very jurisdiction of the court that had entered the plea. The Fourth Circuit summarily affirmed.58

Thus, when Blackledge arrived at the Supreme Court, it was almost entirely about double jeopardy, and the litigants and the Court were focused on the issue of waiver. North Carolina’s petition for certiorari presented only two questions for review, addressing double jeopardy and waiver.59 The Supreme Court had recently held that a defendant who pled guilty could only challenge the voluntary and intelligent character of his plea, and not any independent constitutional violations that may have preceded it, in a federal habeas petition.60 But there was a pre-existing split in the federal courts of appeals over whether double jeopardy could be waived like any other constitutional claim antecedent to a plea, or whether, as the Fourth Circuit thought, it was different because it went to the validity of the underlying indictment itself.61 Thus, the Court had a good reason to take Perry’s case: to determine whether double jeopardy claims fell within its new rule preventing collateral review of constitutional claims that were independent of otherwise valid guilty pleas.

The Court apparently took the case for this purpose. When the conference discussed the petition on October 1, 1973, six Justices voted to grant certiorari, presumably to vacate the Fourth Circuit’s summary order; the liberal group of Justices Brennan, Douglas, and Marshall voted to deny it.62 The only question, it seemed, was how far the Court should go in rolling back Perry’s successful double jeopardy claim. The Justices favoring certiorari seemed to be split over whether to remand on the waiver issue in light of new precedent, reverse the Fourth Circuit on waiver, or find that there had been no double jeopardy violation at all.63

The outcome seemed clear. One law clerk predicted:

This case is undoubtedly going to have to be vacated and remanded. The task of the conference will be to decide what instructions to give the lower courts on remand—simply to clarify the impact of a guilty plea [on waiver of constitutional claims] or to in addition speak to what the [district court] said about substantive constitutional issues.64

2. Stumbling into Vindictiveness

Ultimately, however, the Court declined to decide the double jeopardy questions. Applying Pearce, it instead sided with Perry on the ground that “the indictment on the felony charge constituted a penalty for exercising his statutory right to appeal, and thus contravened the Due Process Clause of the Fourteenth Amendment.”65

The conference did an about-face when it met on February 22, 1974, three days after hearing argument. Justices Stewart and White, who had been inclined to reverse on the issue of double jeopardy but divided over how to do so, now argued for affirming the Fourth Circuit by finding a due process violation.66 Justice Stewart described this as an “odd-ball” case, arguing that there had been no double jeopardy violation but that the defendant had a right not to be “hailed [sic] into higher court on an increased charge” because doing so would burden his right to appeal.67 Justices White, Blackmun, and Marshall agreed that the case should be resolved on due process rather than double jeopardy grounds.68

So why would the Court use Perry’s “odd-ball” case to extend the judicial vindictiveness rule from Pearce to prosecutorial conduct? It seems to have done so because North Carolina blundered into conceding the point at oral argument, which Assistant Attorney General Richard League handled for the state. In his handwritten notes from the bench, Justice Powell faulted League’s “hopelessly weak” presentation.69 League, who began by apologizing for the “shabby condition” in which his brief had arrived,70 seemed not to contemplate the possibility that the state was vulnerable to a due process argument on the ground that the higher charge had retaliated against Perry’s exercise of a legal right.

When Justice Marshall asked if Perry could have been indicted for a felony if he had not appealed, League answered that he could not have.71 Marshall immediately followed by asking, “So, because he appealed, he was indicted?” Remarkably, League answered, “I would say so, yes sir.”72 An audible snicker and an exclamation of “Thank you!” from an unknown party can be heard on the recording of the painful four-second silence that followed.73 Perhaps belatedly realizing the significance of his admission, League attempted damage control, saying, “But I would not attach to it perhaps the same significance to it as Your Honor. This well could have been an event where they tried to get it out of the way down below.”74 It is unclear what he meant by this second sentence; a new question changed the subject as he trailed off.75 Justice Marshall later read from the transcript of oral argument to support his position in conference that the Pearce vindictiveness rule applied to this case.76

Perry’s counsel advanced his vindictiveness claim in relatively cursory fashion in his brief, and only in the context of a larger argument about double jeopardy.77 The State essentially failed to discuss it at all; its one reference to Pearce was tangential.78 When offered the opportunity at oral argument to engage with Perry’s vindictiveness claim, League had only a weak response. League had been arguing that Perry had received a reduced sentence in exchange for his guilty plea to the felony charge, suggesting that his plea was knowing and intelligent and should have waived Perry’s constitutional claims. Chief Justice Burger then asked: “Does that bring you up against the Pearce case, about increasing sentences in any way?”79 League replied haltingly: “No, sir. I don’t think Pearce is applicable to this case, by virtue of what was said in the Colten decision that the possibility of vindictive punishment does not occur sufficiently within the two-tier system to warrant the imposition of the prophylactic rule in Pearce.”80

This was an accurate description of the holding in Colten, but it only applied to the possibility of vindictiveness by a different judge than the one who had initially heard the case before appeal.81 League evidently failed to grasp that the logic of Pearce could potentially be extended to encompass a prosecutor’s behavior.82 Justice Stewart did just that in his majority opinion, finding that “the opportunities for vindictiveness in this situation are such as to impel the conclusion that due process of law requires a rule analogous to that of the Pearce case.”83 Prosecutors who seek to conserve resources have an incentive to discourage persons convicted of misdemeanors from obtaining new trials.84 If they are permitted to increase the charge, they will be able to “insure that only the most hardy defendants” will exercise their right.85

And yet, just as League failed to realize that the Court might extend Pearce to prosecutors, the Court itself seemed not to recognize that this extension would implicate the practice of plea bargaining. Indeed, the majority only broached the topic of plea bargaining to explain why Perry had not waived his claim. Perry had been denied due process by the “very initiation of the proceedings against him,” the Court argued, so he could not forfeit his due process claim by pleading guilty to a charge that was unconstitutionally brought to begin with.86 Justice Rehnquist’s dissenting opinion, meanwhile, chided the majority for discouraging prosecutors from offering “plea bargains.”87 Permitting defendants to challenge guilty pleas on collateral review, he reasoned, would undermine the finality of pleas and therefore reduce prosecutors’ incentives to bargain for them.88 In a separate dissent that Justice Powell drafted but ultimately decided not to publish, he expressed a similar view that “the efficacious administration of justice demands that guilty pleas, made voluntarily and with the advice of counsel, be respected as a definitive resolution of antecedent issues.”89 But none of the discussion of plea bargaining dealt with the effect of the vindictive prosecution rule itself. The record simply reflects no awareness of the policy consequences of the decision.

The Court’s resolution of Blackledge was therefore not only unexpected, but also inadequate for failing to anticipate the difficulties it would soon create. We can only speculate whether the former caused the latter. Of course, there is a reason why the Court has the parties brief and argue the questions it considers, and we might reasonably hypothesize that more thorough consideration would have alerted the Justices to this issue.90 We might also wonder why the Court did not choose to set the case for reargument, which it occasionally does when it anticipates that it may decide the case on different grounds than the parties initially contemplated.91 There is no way to know whether better preparation would have helped the Court foresee the problems that plea bargaining would cause for its new extension of vindictiveness doctrine to prosecutors. But it seems fair to conclude that the Court cannot have improved the quality of its reasoning by abruptly changing course.

In the end, then, the Court’s decision in Blackledge created, in response to the contingencies of this “odd-ball case,” a new and undertheorized constitutional rule that the Court had not originally intended to develop. Justice Stewart explained from the bench: “We agree with the Court of Appeals that it was a violation of the Fourteenth Amendment for the State to up the ante after the respondent appealed his original misdemeanor conviction.”92 Left wholly unconsidered was whether it would also violate the Fourteenth Amendment to up the ante after the defendant went to trial.

D. . . . To Bordenkircher

Four years later, the Court confronted a case in which the prosecutor’s behavior seemed indistinguishable from that in Blackledge—except that the defendant had invoked a constitutional rather than a statutory right, which should only have strengthened his claim. Instead, the pendulum swung back to the other extreme, as the Court, faced with a conflict of its own making, gutted its protection of the right to trial in an effort to protect plea bargaining.

Blackledge generated a modest immediate response. The most extensive initial scholarly discussion focused on the holding that Perry, despite pleading guilty, did not waive his double jeopardy and due process claims.93 But Blackledge did lead to a wide variety of vindictive prosecution claims in the following years. Some closely followed its model, challenging a prosecutor’s decision to bring more severe charges after a defendant successfully appealed his initial conviction.94 Some shifted into the pretrial context, arguing that a prosecutor could not bring higher charges in an effort to achieve a particular outcome before trial.95 These attempts to bring a range of vindictive prosecution claims, and the fact that the Blackledge opinion offered little additional guidance on what constituted a “realistic likelihood of ‘vindictiveness,’”96 led some to conclude that the new rule of vindictive prosecution was “difficult to apply.”97

Paul Hayes was surely not the only defendant to claim vindictiveness when a prosecutor brought more or higher charges after he declined to plead guilty.98 Hayes had been indicted on a felony charge of forging an $88.30 check and faced two to ten years in prison.99 The prosecutor offered him a five-year sentence and threatened that, if Hayes did not plead guilty, he would be charged under Kentucky’s Habitual Criminal Act and subject to a mandatory life sentence because he had two prior felony convictions.100 Hayes refused the deal, the prosecutor increased the charge, and Hayes lost at trial and received a life sentence.101

Hayes’s federal habeas petition alleged that his “indictment and conviction under the Habitual Criminal Statute was vindictively sought by the Commonwealth of Kentucky in this case.”102 The district court was unpersuaded. It easily identified the wide-ranging and, from its perspective, undesirable practical implications of Hayes’s argument: “If prosecutors were precluded from seeking conviction of more serious offenses following the rejection by defendants of the opportunity to plead guilty to lesser offenses, the entire concept of plea bargaining would be effectively destroyed . . . .”103 Hayes had stumbled into a confrontation with the unconsidered policy consequences of Blackledge.

On appeal, the Sixth Circuit disagreed with the district court’s policy analysis. While Kentucky “urges that the entire concept of plea bargaining will be destroyed if prosecutors are not allowed to seek convictions on more serious charges if defendants refuse to plead guilty,” Judge McCree wrote, “[w]e do not agree.”104 The court of appeals held that a prosecutor may “offer a defendant concessions relating to prosecution under an existing indictment,” i.e., drop some charges or reduce the sentence, in the course of plea negotiations.105 By bringing the initial indictment, however, the prosecutor has made a “discretionary determination that the interests of the state are served by not seeking more serious charges.”106 If he were to increase the charges after plea negotiations broke down, therefore, “a strong inference is created” that he did so out of vindictiveness, since he had already determined that the lower charge was appropriate.107

Thus, the stakes of the case quickly crystallized. In light of the Court’s recent Brady line of decisions, which had finally blessed plea bargaining as not only practically necessary but legitimate,108 could the practice be sustained if prosecutors could not penalize defendants’ exercise of the right to trial? And, if not, would it make a meaningful constitutional difference if prosecutors initially brought higher charges and then enticed the defendant to forfeit his right in exchange for a reduction, rather than adding higher charges to retaliate against the defendant for exercising it? Or, as Kentucky bluntly and pragmatically framed the issue:

This case involves a current bargaining practice used in plea discussions. . . . [T]he realit[y] of plea discussions involving charges unbrought but legally susceptible of being brought is that it is entirely appropriate, legally and constitutionally, for the prosecutor to offer the accused not to seek indictments on the additional charges for a plea of guilty to a charge already brought.

The inevitable effect of plea bargaining is to discourage the assertion of the Fifth Amendment right not to plead guilty and to deter the exercise of the Sixth Amendment right to demand a jury trial.109

Thus, Bordenkircher represented a collision between “two separate and inherently conflicting lines of case law”—prohibiting prosecutorial vindictiveness and establishing the legitimacy of plea bargaining.110 But, given the far broader practical significance of the latter, and the former’s utter failure to account for it, the real question was not which would prevail, but how completely the Court would bend vindictive prosecution doctrine to meet the needs of plea bargaining.

The answer: almost entirely. By a five-to-four vote, the Court rejected the Sixth Circuit’s distinction between offering to reduce a charge if the defendant accepted a plea and threatening to increase a charge if the defendant refused a plea. “As a practical matter,” Justice Stewart explained, these two sequences were “no different.”111 Justice Blackmun, joined by Justices Marshall and Brennan, endorsed the Sixth Circuit’s sequencing rule in dissent. Arguing that “[p]rosecutorial vindictiveness in any context is still prosecutorial vindictiveness,”112 Justice Blackmun conceded that, practically speaking, it may “make[] little difference how this case, now that it is here, is decided.”113 Aggressive prosecutors might simply bring harsher indictments up front. Nonetheless, Justice Blackmun thought it preferable as a policy matter to “hold the prosecution to the charge it was originally content to bring and to justify in the eyes of its public.”114

Justice Powell, meanwhile, struggled deeply with the case, which he found “terribly unjust,”115 and waffled on his position. He initially criticized the Sixth Circuit’s opinion on policy grounds, fearing that it would compel prosecutors to indict for the maximum possible penalty, creating a larger gap to bridge with defendants and thereby inhibiting plea bargaining.116 At conference after argument, he sided with the majority and voted to reinstate Hayes’s conviction.117 He wrote in his handwritten notes that he agreed with Justice Stewart, who had commented that “[p]lea bargaining, by definition, involved ‘threats’ as to consequences of not accepting a proposal. There may be a point (deception, fraud) beyond which prosecutor may not go. But not here.”118 Ultimately, however, while noting that he “agree[d] with much of the Court’s opinion,”119 Justice Powell dissented separately and argued that the prosecutor in this “exceptional” case had acknowledged vindictiveness.120 Because it could be inferred from the prosecutor’s initial charging decision that he had “deemed it unreasonable and not in the public interest” to subject Hayes to a life sentence, this was the rare instance in which the scales of plea bargaining had become “so unevenly balanced as to arouse suspicion.”121

The Bordenkircher result has been discussed by many commentators and strongly criticized by some of them.122 It is not my purpose to review the criticism here, and the propriety of the habitual offender statute itself is another matter entirely. But a couple of Bordenkircher’s particular shortcomings demonstrate just how wildly the Court swung the pendulum back in favor of the practical imperatives of plea bargaining, while thoroughly discounting the value of the trial right and the need to insulate its exercise from retaliation.

First, the majority’s assertion that “in the ‘give-and-take’ of plea bargaining, there is no . . . element of punishment or retaliation so long as the accused is free to accept or reject the prosecution’s offer”123 simply does not withstand analytical scrutiny. For one thing, Perry was just as free not to appeal as Hayes was not to plead guilty. And the defendant’s ability to make a voluntary choice cannot explain why there is no element of retaliation. That two parties are engaged in a give-and-take negotiation does not preclude one party from attempting to punish the other for walking away from the table. The chance that both parties might benefit, moreover, has no logical bearing on whether one might act with retaliatory motive. It simply does not follow from the fact that plea bargaining offers a “mutuality of advantage”124 that there can be no “element of punishment or retaliation” by the prosecutor. What the Court really seems to have meant is that, because plea bargaining offers a mutuality of advantage to both parties, there is no reason to police it for punishment or retaliation. Plea bargaining is a permissible practice, so the prosecutor’s desire to achieve a plea is a permissible motive for his discretionary charging decision. The concern for the defendant’s legally protected right that animated Blackledge was simply absent from the equation four years later.

Second, Bordenkircher underhandedly rejected Blackledge’s basic view that a defendant must be able to exercise his rights free from the fear of retaliation, whether or not the prosecutor has an actual retaliatory motive.125 Justice Stewart reinterpreted the doctrine to “emphasize[]” that due process is not violated simply by the “possibility that a defendant might be deterred from the exercise of a legal right,” but rather by the “danger that the State might be retaliating against the accused.”126 This is plainly not what Blackledge held.127 Without explicitly acknowledging it, then, and without openly attempting to reconcile the competing imperatives of protecting plea bargaining and vindicating the importance of the right to trial, the Court essentially abandoned the doctrine it had established several years before.

II. bringing vindictiveness back

So far, this Note has explained the accidental origins of vindictive prosecution doctrine and—perhaps as a result of those origins—the doctrine’s failure to strike, or even seek, a balance between the practical necessity of plea bargaining and the constitutional value of the jury trial. This underlying tension between trial and plea remains vital today. Now, however, the pressing question is not how to permit and recognize plea bargaining, but how to constrain and regulate it. In this Part, I argue that the idea of vindictive prosecution, understood in a way that more closely reflects what we ordinarily mean by vindictiveness, could prove newly useful.

Section II.A explains recent doctrinal innovation about pleas. It argues that the Court is seeking more and more to regulate plea bargaining and is justifying its actions on the ground that trials are increasingly irrelevant. It has come to view the terms of a “standard” plea to a given offense as the criminal justice system’s “true” outcome. Nonetheless, the Court has acted against a backdrop of doctrine and constitutional values that still presume the importance of the right to trial. As a result, the Court has only been able to regulate plea bargaining indirectly and has not imposed any substantive limits on prosecutorial discretion itself.

Section II.B proposes and defends a standard that would prohibit prosecutors from acting with the intent of punishing a defendant’s exercise of his right to trial as a wrong. Enforced through an evidentiary presumption, such a standard would allow for some substantive regulation of prosecutors’ charging discretion. But it would do so for the sake of protecting the right to trial, not on the basis of its insignificance. Among other advantages, then, this revitalized vindictive prosecution doctrine would partly reconcile the competing values that the old vindictive prosecution doctrine could not accommodate.

A. Regulating Pleas in an Administrative System

The Supreme Court has recently come around to the view that plea bargaining “is not some adjunct to the criminal justice system; it is the criminal justice system.”128 The Court has consequently shown a greater inclination to regulate plea bargaining,129 beyond the minimal requirements of the colloquy in which the trial judge formally ratifies the previously negotiated agreement.130 In doing so, however, the Court has relied on a body of law treating pleas as the exception even as it now declares pleas the rule. As a result, recent developments in constitutional criminal procedure have only been able to regulate plea bargaining through indirect means.

As I explain below, the Court has crafted these new rules of criminal procedure on the assumption that there is something like a “standard” or “correct” plea offer that a defendant ought to receive. It has deployed this assumption to fairly modest effect, entitling a defendant to this “standard” deal if his lawyer fails to inform him when he has been offered it or properly advise him about whether he should accept it. But the Court has imposed no obligation directly on the prosecutor to offer the standard deal—or even something that remotely approximates it—in the first place. As I will go on to argue in Section II.B, however, the assumption of the “standard” offer could serve as the basis for a more robust rule that would constrain bad prosecutorial behavior and reduce excessive discretion.

The Court began its innovation in Sixth Amendment doctrine with its 2010 decision in Padilla v. Kentucky,131 holding that a defendant was denied the effective assistance of counsel when his attorney failed to advise him that his plea would result in his removal from the United States. That holding broke new ground in making clear that a criminal defense attorney must attend to a conviction’s collateral consequences to fulfill her obligations to her client.132 But it also reflected a new willingness to tailor new procedural protections specifically to plea bargaining.133

The 2012 companion cases of Missouri v. Frye134 and Lafler v. Cooper135went further in imposing procedural order on the bargaining process. In Frye, a defendant’s attorney had failed to communicate plea offers that, if accepted, probably would have resulted in a lower sentence than the defendant ultimately received after accepting a later, harsher offer.136 In Lafler, the defendant rejected a lenient plea offer on the faulty advice of his lawyer and ultimately received a much longer sentence after being convicted at trial.137 In both, the Court held that the defendant’s Sixth Amendment right to counsel had been violated and that the defendant was entitled at least to the possibility of resentencing and perhaps to the benefit of the earlier offer.138

Lafler and Frye rely on an implicit idea of a “standard” plea deal that a defendant who engaged in certain conduct ought to be offered. As a logical matter, it makes little sense to require that a defendant be given the benefit of an erroneously forgone plea offer unless he was, in some sense, “supposed” to receive that deal rather than the harsher one he ultimately got. The normative force of the claim that plea bargaining “is the criminal justice system” is that a typical plea made in the ordinary course of business—not the sentence imposed by a judge after a jury trial—is the true outcome that the system aims to produce.139

Indeed, both cases explicitly rely on an understanding that the normal, ordinary-course plea offer is the outcome that the defendant should receive. Frye held that, to establish that he had been denied the effective assistance of counsel, a defendant would need to show a reasonable probability that neither the prosecution nor the trial court would have later prevented him from accepting or entering the plea offer he never received.140 It “should not be difficult” for courts to evaluate such claims by defendants, the Frye majority explained, because “in most jurisdictions prosecutors and judges are familiar with the boundaries of acceptable pleas and sentences.141 In other words, because they know what a standard plea deal looks like, local trial courts are in a position to determine whether they would have later rejected the plea that was never communicated to the defendant. If the offer the defendant never heard about looks like a reasonable offer, taking into account any intervening circumstances, then the defendant may be entitled to it.

Yet, even as they legitimize and operationalize the idea that a defendant is “supposed” to receive a “standard” offer in this way, the recent plea-bargaining cases say nothing about whether or when prosecutors must actually make such offers. Precedent and competing systemic values prevent the Court from creating a substantive entitlement to a standard plea deal. Indeed, “there is no constitutional right to plea bargain” at all.142 Judges also have the discretion to reject even “standard” negotiated pleas.143 Both of these may seem like uncontroversial points, but both depend on the central proposition—purportedly rejected by Lafler and Frye—that plea bargaining is the aberration and not the norm. If one really believes that plea bargaining is the criminal justice system, it is difficult to understand why a defendant would not have a right to plead.144

The Court’s inability to create a substantive entitlement to a standard deal stems from the more general principle that prosecutors have “broad discretion” in choosing whether and how to prosecute a case.145 But this view similarly presupposes that trials, not pleas, are the criminal justice system. Precedent makes clear that prosecutorial decisions are “particularly ill-suited to judicial review” because they depend on the government’s priorities and its assessment of subjective factors, like “the strength of the case,” that courts are not “competent” to review.146 This logic may be motivated partly by separation-of-powers concerns,147 but it is also substantially prudential.148 Regardless, the notion that prosecutorial charging is unreviewable presumes some kind of judicial review later in the process—that is, it presumes that trial (or at least a more searching inquiry into the bargained-for plea than a rote colloquy) is the norm. It would be even more unwise and equally offend the separation of powers for the judiciary to review neither the charging decision nor the determination of guilt.149 One can only countenance the unreviewability of prosecutorial discretion by assuming some subsequent opportunity to review whether the defendant’s conduct actually satisfies the charge filed.

The result is that our current law is an uneasy blend of half-measures: sometimes it sees pleas as the “true” outcome, and sometimes trials. As a result, it governs the day-to-day administration of criminal justice only through indirection, and it imposes virtually no legal limits on prosecutorial discretion itself.150 A prosecutor may not single out a person and charge him on the basis of his race,151 of course, but even then, it is nearly impossible for a defendant to prove that a prosecutor has done so.152 The common law has long recognized the tort of malicious prosecution, but a prosecution is not malicious unless the defendant first establishes that it was resolved in his favor.153 Where a prosecutor has probable cause and no discriminatory motive, her freedom of action thus remains unfettered.

B. Toward a New Standard: Vindictiveness-as-Vengeance

I have just argued that constitutional criminal procedure increasingly recognizes that plea bargaining is the criminal justice system and must be regulated. In doing so, it reflects an emergent understanding that there is such a thing as a “normal” deal for particular conduct in a given jurisdiction, and that a defendant generally ought to get that deal. But because of contrary precedent, genuine esteem for the constitutional ideal of the jury trial, and a “governing ideology [that] does not admit” that the prosecutor is really making an administrative determination of guilt,154 the Court is unable to impose direct limitations on plea bargaining that would completely normalize it as a substitute for trial.

I now argue that the idea of prosecutorial vindictiveness might help to regulate the market for plea bargains within these legal and normative constraints. There may be better ways to address the problems of excessive discretion and coercive plea bargaining, including lowering and standardizing sentence lengths and eliminating overlap between criminal statutes to reduce the government’s ability to choose between them. In the absence of such systemic reforms, however, the idea of prosecutorial vindictiveness is conceptually useful precisely because it is roundabout. A rule against vindictive prosecution, properly understood, would directly regulate plea bargaining not at the cost of further entrenching pleas as the rule and trial as the exception, but for the sake of protecting the constitutional right to trial.

My claim is that a different standard prohibiting prosecutorial vindictiveness could do much of the work the original was meant to do, and could be reconciled with the pervasiveness of plea bargaining, if it defined vindictiveness more narrowly and more sharply. Specifically, prosecutors could be barred from acting with the subjective intent of punishing, as a wrong,the defendant’s exercise of a legally protected right, including and especially the right to trial. I call this new way of understanding prosecutorial vindictiveness “vindictiveness-as-vengeance.”

This Section elaborates and defends vindictiveness-as-vengeance in three steps: first, identifying the conduct it aims to prevent; second, describing how it might be enforced through an evidentiary presumption; and third, arguing why it is a good idea.

1. The Standard

Before proceeding further, it is important to explain what kind of behavior the vindictiveness-as-vengeance standard would and would not proscribe and, relatedly, how it differs from the rule of Blackledge.

The prosecutor would not be prohibited from taking actions that have the incidental effect of discouraging the defendant from exercising her legal right. Nor would the prosecutor necessarily be prohibited from taking actions intended to discourage the defendant from exercising the right to trial. Such deterrence, of course, is precisely what offering a defendant a lesser sentence if he pleads guilty is trying to accomplish, and it is what gives rise to the problem I am attempting to solve. Rather, as I will explain, vindictiveness-as-vengeance would inquire more deeply into the reasons behind the prosecutor’s effort to discourage the defendant from exercising the right to trial. It would ask whether, going one step further up the chain of reasoning, the government sought to deter the exercise of the right to trial on the basis of permissible or impermissible motives.

Blackledge, by contrast, uses the term “retaliation” to describe what prosecutors are not permitted to do.155 The important difference is not the choice of words, but the underlying idea of causation that the term “retaliation” captures. Under the “antiretaliation provision” of Title VII,156 for instance, it is unlawful for an employer to take adverse action against an employee “because” she filed a complaint of discrimination or otherwise participated in enforcement proceedings.157 An employer’s “retaliatory motive” inheres in the causal connection between the employee’s protected conduct and the employer’s adverse action.158 It would be irrelevant to a Title VII retaliation claim that the employer did not feel wronged by the employee’s protected action, or view it with opprobrium, but simply wanted to save money on such claims in the future. The action would still be retaliatory since it was taken “because” the employee had engaged in statutorily protected conduct.159

This is precisely the kind of retaliation with which Blackledge was concerned. The subjective intent behind the prosecutor’s decision was essentially irrelevant. In fact, there was “no evidence that the prosecutor . . . acted in bad faith or maliciously in seeking [the] felony indictment.”160 The problem was one of causation: as Justice Marshall said at oral argument, “because he appealed, he was indicted.”161 As applied to the plea-bargaining context, the Blackledge rule would reach cases where, because the defendant went to trial, he received a higher sentence. Indeed, because the proscribed government conduct did not depend on any particular prosecutorial mens rea, Blackledge went further and also protected defendants from the “apprehension” of retaliation.162

Of course, it might make sense to proscribe retaliation in this way. There is good reason for the law to prohibit adverse actions taken becauseof the exercise of protected rights, regardless of the particular state of mind underlying the causal connection between the protected right and the adverse action.163 But this simply doesn’t work if one takes as given that plea bargaining exists and that the law allows it, for plea bargaining necessarily “threatens to inhibit exercise” of the right to trial.164 Therefore, vindictiveness-as-vengeance covers only a subset of cases in which the government’s plea-bargaining tactics deter the defendant’s exercise of the trial right. These are the cases in which the government sees the defendant’s exercise of his right as a wrong warranting punishment.

But what does it actually mean for the prosecutor to think this way? My claim is that there might be a genuine constitutional difference between two prosecutorial states of mind. The first, permissible mindset respects the defendant’s constitutional right to trial while acknowledging that it must frequently be traded away to satisfy the imperatives of the system. One might imagine a prosecutor with this mindset explaining herself to a defendant as follows: “You have the right to take this case to trial. I will seek a significantly increased penalty if you do. I encourage you to take the deal that is in your own best interest (relative to the alternative) as well as the government’s.” The second, impermissible mindset is captured by one prosecutor who explained his approach to plea bargaining as follows:

I’ll give you a deal if you don’t bust my ass. You start taking a bunch of depositions, filing a bunch of motions—fuck you.

This system is overloaded as it is. Most of these people know if they’re guilty or not. . . . If you hold out[,] . . . if you don’t recognize what you’ve done and try to get through here with a little bit of facility, then I’m going to try to bust your ass.165

On this view of vindictive prosecution doctrine, then, vindictiveness means something more like what it usually means: the prosecutor is “disposed to seek revenge” for a perceived wrong.166 Vindictiveness inheres not in the fact that the defendant’s exercise of his right causes the prosecutor to bring a higher charge, but in the prosecutor’s subjective view that the defendant’s exercise of his right is a wrong that warrants a higher charge.

Under vindictiveness-as-vengeance, it makes sense to say—as the Bordenkircher Court tried, unpersuasively, to do—that no vindictiveness typically arises in the plea bargaining context. For there are any number of other legitimate state interests recognized by the law that may be the underlying motives for the prosecutor’s decision to increase charges because the defendant goes to trial. These might include conserving resources; sparing witnesses from inconvenience, emotional trauma, or retaliation; obtaining the defendant’s cooperation against others; hastening a conviction to obtain some desired collateral consequence like a restraining order, a resignation from office, or a civil forfeiture; or letting someone who has already served enough time in pretrial detention go home. The prosecutor is still retaliating against the defendant for exercising his constitutional right to trial. But she acts in furtherance of some other legitimate goal, rather than punishing the defendant’s exercise of the right for its own sake.

Yet the reader might wonder whether what I am characterizing as the vengeful “fuck you” motive is truly distinct from a more legitimate prosecutorial motive. One could argue that even the prosecutor who threatens to “bust [the defendant’s] ass” if the defendant does not cooperate is, at bottom, motivated to move the defendant through the system expeditiously to save public resources. In other words, even if it is true that the prosecutor sees the defendant’s exercise of the right to trial as a wrong to be punished, he may only think it is a wrong because trials waste time and money. On this view, the inquiry into motive will identify a permissible basis for the prosecutor’s behavior if it simply goes one step deeper into his reasoning.

This is possible, of course, but it isn’t necessarily true. Perhaps instead the prosecutor’s higher-level motivation is ego: “Defendants who refuse to accept the outcome I assign them should be punished for doubting my authority.” Perhaps it is competitiveness: “I’m here to defeat the other side, and the defendant’s noncompliance makes me want to beat him that much more.” Perhaps it is actually a certain sense of justice: “This defendant is clearly guilty, and if he refuses to own up to it, he ought to pay a commensurate price.”167 Or perhaps there is no higher-level motivation for this conduct at all, and the prosecutor is just reflexively acting on internalized norms of office culture that trials are bad and defendants who pursue them are to be punished. More likely, these motives and others probably overlap and blend together. Identifying a person’s ultimate motive for acting may be impossible, or that ultimate motive—e.g., maximizing happiness—may be too general to be useful. Yet the law often calls for rough and reductionist judgments of motive anyway to assign social meaning to relevant conduct.168 That a desire to punish the defendant and a desire to save money can be difficult to separate does not mean they are analytically the same or should be treated that way.

A separate but related question is whether, assuming the impermissible and permissible prosecutorial states of mind are conceptually distinct, an external observer can tell them apart. After all, if the “fuck you” mindset is impermissibly vindictive, a prosecutor who adheres to it will probably not say so to a defendant and will certainly not say so to a court. And, even if he does, the statement is susceptible of multiple interpretations. This is a difficult but fairly conventional proof problem whose solution is essential to vindictiveness-as-vengeance.

2. The Evidentiary Presumption

It is difficult to determine any actor’s intent with certainty, and most people who act with an intent the law proscribes probably avoid bragging about it to the authorities.169 But, borrowing from other areas of criminal law and criminal procedure that pose the problem of evaluating an actor’s subjective intent170—and, indeed, maintaining the basic structure of the old vindictiveness doctrine171—an evidentiary presumption might help.

Regardless of whether a prosecutor bargains by charging a higher crime first and offering to drop, or by charging a lesser crime first and threatening to add, there will be a disparity between the charge and sentence offered for a plea and those sought at trial. This difference is the trial penalty.172 Given that prosecutors and defense counsel (especially public defenders) are repeat players, one would expect the trial penalty for any common criminal act in a given jurisdiction to fall within a standard range over time.173 Indeed, in many and probably most jurisdictions, actors in the system are well aware that there is a “standard plea offer” for any common offense and offender profile.174 Where there are no formal standard offers, repeated interactions between prosecutors and defense attorneys tend to generate stable outcomes over time.175 By implication, the “standard offer” carries with it a normal trial penalty: the difference between the terms of the standard deal and the exposure that the defendant will face at trial if he rejects it.

In this context, a prosecutor’s unreasonably excessive deviation from the jurisdiction’s normal trial penalty might give rise to a rebuttable presumption of vindictiveness-as-vengeance. This presumption recognizes that there are only a limited number of legitimate motives that would justify such a deviation. Thus, when the plea offered to a particular defendant entails a substantially harsher trial penalty than the average deal, there is a reasonable chance that the prosecutor has acted on an impermissible motive. The presence of this reasonable chance shifts the burden to the prosecutor, who must rebut the presumption of vindictiveness-as-vengeance by showing a legitimate motive for the plea offer that exerts uncommonly strong pressure on the right to trial. Although the presumption will not be perfect, it is a probative (and easily administrable) tool for determining whether the prosecutor has acted with a proscribed state of mind.176

When a plea offer is structured to create unusually severe pressure to take the deal, it raises the question why, exactly, the prosecutor wants this particular defendant to suffer these consequences. Maybe she has a good answer. Perhaps the charges are exceptionally complex, for example, so a trial would be more costly than usual. Perhaps the evidence points strongly to the defendant’s guilt but the circumstances pose an unusual risk of jury nullification. Perhaps the defendant has information that would be unusually valuable to the government if he cooperated. But perhaps there is no good explanation, in which case the best inference left is that the prosecutor is trying to punish the defendant—“I’m going to bust your ass”—for the perceived wrong of going to trial.

The reader may wonder why the prosecutor’s legitimate interests in conserving resources and promoting cooperation won’t be sufficient to defeat the presumption in virtually any case. It is crucial to recognize that, while those are good reasons why trial penalties exist at all, they aren’t sufficient reasons for an abnormally largetrial penalty in a particular instance. Where a standard bargain exists, enforced by repeat-player interactions over time, it is by definition a deal that tends to satisfy most defendants’ desire to avoid the harshest possible penalty while providing some beneficial savings to the government. Of course, not every guilty defendant will accept that offer, for any number of reasons. But if the government makes an offer with an unusually harsh penalty, it is necessarily not doing what it does to encourage cooperation and efficiency in the ordinary case. Why not? Perhaps there is some relevant attribute of this defendant or this case that means he needs to be presented with an unusual plea offer that is structured to be more coercive—but perhaps not. Either way, reciting the government’s interests in saving time and money will be logically insufficient to explain the unusual nature of its conduct. Because the evidentiary presumption only attaches in cases where the most common permissible motives do not apply, we can be more confident that the presumption will be a useful proxy for the impermissible motive of vindictiveness-as-vengeance.

In fact, the Court has deployed an evidentiary presumption in the related context of civil claims for retaliatory prosecution, in which a former defendant sues an official for inducing the government to prosecute him in reprisal for protected conduct.177 Because “it would be unrealistic to expect a prosecutor to reveal his mind[set],” a plaintiff asserting a retaliatory prosecution claim must plead and prove that there was no probable cause to pursue the charges against him.178 Unlike the presumption I propose, the presumption in Hartman is chiefly meant to address the issue of causation—whether the prosecutor “would not have pressed charges” but for the protected conduct.179 It nonetheless suggests that it may be workable to use an evidentiary presumption based on a rough reasonableness standard180 to identify illegitimate reasons for prosecutorial behavior.

This is not to say that the evidentiary presumption is perfect. It might occasionally chill legitimate exercises of prosecutorial discretion and creativity, for example, although this would suppose that the presumption is vigorously enforced and that prosecutors are not fully confident in their ability to rebut it even when the law is properly on their side. This concern should not be disregarded, but its likely consequences are not severe: prosecutors restrain themselves from using the fullest possible leverage in some cases, so a few more defendants reject pleas and go to trial, and perhaps a few more at the margins walk.181 There might also be a concern that the presumption could incent prosecutors simply to increase the standard trial penalty, thus obviating any need to deviate upward from it. But that response would let the tail wag the dog. The government has a host of other important concerns at play, including fairness, consistency, enforcement, incarceration costs, and electoral politics. If these have dictated an equilibrium bargain, discarding it to maximize flexibility in the exceptional case would be strange.182

To be clear, I do not mean to suggest that judges would engage in statistical analysis of conviction and sentencing outcomes in order to apply this evidentiary presumption. Because I am proposing a broad reasonableness standard rather than a bright-line rule—no fixed percentage of deviation from the usual trial penalty counts as “unreasonable”—I envision the analysis as more discretionary and impressionistic. Most sentencing courts should be able to draw on their experience in the jurisdiction to assess whether a proposed trial penalty is unreasonably excessive. And, of course, the parties would presumably inform the court about relevant outcomes for other similarly situated defendants.

To see how the presumption might work in practice, consider the case of Shane Guthrie, arrested in Gainesville, Florida, for “beating his girlfriend and threatening her with a knife.”183 Guthrie was initially offered a plea deal that would have resulted in two years in prison plus probation.184 Based on the facts of his alleged conduct, let us assume that this offer would have required him to plead guilty to aggravated assault, a third-degree felony punishable by a maximum of five years in prison.185 When Guthrie turned down the deal, the prosecutor (having previously threatened to do so, one assumes) filed additional charges that would have resulted in a mandatory life sentence on conviction at trial.186 At the time, Guthrie was twenty-four years old, so a life sentence would likely mean about fifty years in prison.187 Thus, the prosecutor imposed a trial penalty of roughly twenty-five times the prison term he had initially deemed to be a sufficient punishment. This is severe enough that a reasonable observer might suspect the prosecutor had acted in bad faith.

Under the rule I propose, the question is whether the prosecutor imposed this penalty for permissible reasons of efficiency or impermissible reasons of spite. If Guthrie’s trial penalty were unusually and unreasonably steep, it would tend to show that the prosecutor acted on motives other than her usual imperatives to save money and get convictions. To challenge the additional charges,188 therefore, Guthrie would need to make a prima facie showing that the twenty-five-fold trial penalty was outside the normal range for similar conduct in the jurisdiction.189 He might do so by drawing the court’s attention to the outcomes of similar cases in the public record, and perhaps by submitting affidavits from attorneys with relevant knowledge of bargaining practices in the jurisdiction. The court would also presumably bring its own experience and common sense to bear.190 Whether the court would compare Guthrie’s case to others filed in the same year or in several recent years is an open question; I think it suffices to say for now that the court should use a reasonable period of comparison.191

If the trial court found that the trial penalty the prosecutor imposed on Guthrie was unreasonably excessive, the burden would shift to the prosecutor to justify it. He could point to some unique factor about Guthrie or his case that warranted an unusually steep trial penalty. Or he could argue that the relevant comparison was not to other serious instances of domestic violence generally, but other instances of domestic violence involving pregnant victims, or recidivist offenders, or some other salient fact, and explain that these cases had been treated similarly. Ultimately, if the court found that the prosecutor had failed to offer a plausible explanation for his conduct, it would hold that the trial penalty violated Guthrie’s due process rights.

In a case like Guthrie’s, the appropriate remedy is straightforward enough. If it is before trial, quash the indictment on the additional charge (here, by hypothesis, kidnapping) and allow the prosecution to proceed only on the lesser one (aggravated assault). If Guthrie has actually gone to trial and suffered the penalty, vacate the kidnapping conviction, vacate the sentence in its entirety, and remand to the trial court for resentencing consistent with the Constitution.192

I do not envision, however, that a defendant who pleads guilty after being threatened with a vindictively steep penalty would have any direct remedy.193 Vindictiveness-as-vengeance prevents the government from actually imposing vindictively excessive trial penalties, thus undermining the credibility of its promises during the bargaining process to do so. If defendants have a reasonable possibility of defeating or successfully appealing unreasonably excessive trial penalties when they are actually imposed, then prosecutors will derive less bargaining leverage from threatening them. In the aggregate and over the long haul, then, limiting the price the government can extract for going to trial would combat the excessive and abusive use of leverage that coerces defendants into bargains they do not want to accept.

3. The Merits

But, even if the vindictiveness-as-vengeance standard is workable, why is it a good idea? I offer four basic reasons, which I broadly categorize as practical, doctrinal, expressive, and conceptual. I address these in turn.

a. Practical

The first and simplest reason why vindictiveness-as-vengeance is desirable is that it might provide real help to defendants who have been subjected to truly objectionable government conduct, even if it only applies in a limited set of cases. Take the case of Kevin Ring, a former Washington lobbyist who was one of twenty-one defendants charged with corruption in the Jack Abramoff scandal.194 Unlike most of the defendants, who cooperated and pled guilty, Ring went to trial; the first ended in a mistrial, and the second produced a conviction.195 Though most of the other defendants served no prison time, the government recommended a sentence of seventeen to twenty-two years for Ring, prompting Judge Huvelle of the district court to remark, “That’s a pretty big penalty for exercising a constitutional right.”196 The government invoked a sentencing enhancement against Ring that it had not invoked against any of the other defendants, even though his conduct was similar or less serious.197

Judge Huvelle ultimately rejected the government’s recommendation and sentenced Ring to twenty months.198 The facts are not precisely analogous, since we do not know what deal Ring rejected or what alternative to pleading guilty the other defendants faced. But this does look a bit like vindictiveness-as-vengeance. With the other defendants already convicted and sentenced, and thus with no need for Ring’s cooperation, there would seem to be relatively few good reasons for the exceedingly steep trial penalty the government sought. It is at least plausible that the government was acting out of spite to punish Ring for persisting in going to trial, and a court may not have been convinced by any permissible motives the government asserted for its conduct. In a pretrial challenge to the government’s filing of the sentencing enhancement, or on appeal after receiving the higher sentence, Ring might have had a good vindictiveness claim under the standard I am proposing.

b. Doctrinal

Even if the idea is appealing, though, a critic might point out that it seems to come out of left field. The Court could craft many sensible rules to address this problem, but this one appears to have no particular warrant in the Due Process Clause or any other established constitutional principle. The second normative justification for vindictiveness-as-vengeance, however, is that it is reasonably well grounded in the law of due process.

As an initial matter, the standard I propose stakes out a middle ground between Blackledge and Bordenkircher. As I pointed out above,199 the Bordenkircher majority’s position that plea bargaining can involve no element of punishment or retaliation because it is mutually advantageous to both the prosecutor and defendant simply does not make sense. Vindictiveness-as-vengeance would retreat from that stance by recognizing that, in a subset of cases, the prosecutor may act vindictively even though it would still advantage the defendant to accept the plea. And yet, most of Bordenkircher’s underlying rationale would remain undisturbed. Vindictiveness-as-vengeance would not dispute that, for example, “by tolerating and encouraging the negotiation of pleas, this Court has necessarily accepted as constitutionally legitimate the simple reality that the prosecutor’s interest at the bargaining table is to persuade the defendant to forgo his right to plead not guilty.”200 Nor would it challenge Bordenkircher’s prevailing view that there is no meaningful difference between increasing charges after the defendant goes to trial and decreasing charges after the defendant pleads guilty. The prosecutor may still “openly present[]” the defendant with the “unpleasant alternatives” he faces.201

At the same time, vindictiveness-as-vengeance would undermine, for purposes of the right to trial, Blackledge’s command that a defendant may not be made to fear a prosecutor’s retaliatory motive even where there is none.202 And, as I have explained, it would shift the gravamen of the harm from causation to impermissible motive. But it would still allow the most fundamental point of Blackledge—that prosecutorial punishment of the exercise of a legal right is itself impermissible—to encompass one of the rights most central to the criminal process: the constitutional right to trial.

Of course, one might fairly point out that the first half of this Note was devoted to criticizing the combination of Blackledge and Bordenkircher as unwieldy and unreasoned, so perhaps splitting the difference between the two is no great achievement. Moreover, that a due process violation may turn on the prosecutor’s mens rea seems at odds with our modern understanding of due process as a means to ensure the accuracy and reliability of government procedures.203 As such, the skeptic might plausibly argue that the proposal I am advancing, even if it broadly accommodates both Blackledge and Bordenkircher, has little to do with due process properly understood. Indeed, vindictiveness doctrine has always been opaque about the underlying vision of due process it instantiates.204

I would freely acknowledge that my claim that due process prevents vindictiveness-as-vengeance is more consistent with a vision of due process that is chiefly concerned with dignitary values rather than accurate results. To some extent, this vision of due process might be an outmoded, pre-Mathews one, recalling an era when due process was described as “[r]epresenting a profound attitude of fairness between man and man, and more particularly between the individual and government”—a “feeling of just treatment” rather than a vehicle for producing correct outcomes.205 But I would resist the notion that these normative underpinnings of due process have entirely eroded. It would be unthinkable, for instance, to hold that the quintessential due process requirements of notice and a hearing could be satisfied by offering them to someone other than the affected party without consent, even if we knew that the third party could better represent the affected party’s interests and produce a truer outcome. The most fundamental principles of due process are still hard to explain without some reference to the dignitary interests that we evoke when we say someone “had his day in court.”206

Perhaps for this reason, there are circumstances in which a criminal defendant’s due process rights do depend on whether an actor had a proscribed mental state. For instance, it violates the defendant’s due process rights for the prosecutor to knowingly elicit false testimony207 or to fail to correct testimony he knows is false.208 This knowledge requirement cannot be justified solely on the grounds of the accuracy or reliability of the outcome. False testimony will equally distort the factfinding process regardless of the prosecutor’s intent in offering it. If the purpose of the rule were simply to deter the government from offering testimony likely to undermine the accuracy of the proceedings, it would make far more sense to dispense with the knowledge requirement and prohibit the government from offering testimony that a reasonable person would believe to be false.

I would suggest instead that this requirement may be justified in part by a particular normative understanding of the way the criminal justice system must treat a defendant: by engaging in a good-faith inquiry into his culpability rather than a rigged stampede to take away his liberty. As the Supreme Court has explained:

[The requirement of due process] embodies the fundamental conceptions of justice which lie at the base of our civil and political institutions. It is a requirement that cannot be deemed to be satisfied by mere notice and hearing if a State has contrived a conviction through the pretense of a trial which in truth is but used as a means of depriving a defendant of liberty through . . . the presentation of testimony known to be perjured. Such a contrivance by a State to procure the conviction and imprisonment of a defendant is as inconsistent with the rudimentary demands of justice as is the obtaining of a like result by intimidation.209

For similar reasons, a defendant’s due process rights are violated when a judge has become so “personally embroiled” in an argument with the defendant or his lawyer as to make the judge “unfit to sit in judgment.” Because “the concern of due process is with the fair administration of justice,” there must be no possibility that the sitting judge is bringing to bear anything other than “the impersonal authority of law.”210

A proscription against vindictiveness-as-vengeance would similarly embody the Due Process Clause’s underlying commitment to meaningful truth seeking in which the defendant (or other person facing a state-imposed deprivation) maintains some control over her conduct of the proceedings. For the state to punish the criminal defendant’s exercise of the right to trial as a wrong offends this scheme: rather than simply try to persuade the defendant that it is not in his interests to assert the right to trial, it denies that the defendant is empowered to claim dignified adjudication at all.211

c. Expressive

The third advantage of the approach I advocate is that it conforms reasonably well to the sort of governmental conduct that we might intuitively wish to stigmatize as vindictive. The fit between the ordinary and legal meanings of vindictiveness will never be perfect, and there is no particular reason why it should be. But the old vindictiveness doctrine, which says that the retaliatory “fuck you” approach to plea bargaining is by definition not vindictive, seems to do a particularly poor job of capturing the social meaning of that particular practice. If one believes that the proscription on certain government behavior serves an expressive function for prosecutors, defendants, or the public,212 then it seems appropriate to recognize plainly vengeful behavior as “vindictive.”

And there are many good reasons to so believe. Without digressing into the expansive debate over law’s capability to shape social and institutional norms,213 it suffices here to note that many have persuasively argued that law’s affixing a label to particular conduct affects public response to that conduct. Antismoking campaigns pointing out the deception of tobacco advertising may cause young people to associate smoking with being duped rather than being cool.214 Labeling sex-related teenage cyberbulling as gender discrimination may make its victims likelier to report it and others less likely to tolerate it because of the deeply engrained stigma that our society now attaches (at least formally) to things called “gender discrimination.”215 Laws prohibiting same-sex couples from adopting children may reinforce harmful stereotypes that gay men are a danger to children.216

On this theory, shaming prosecutorial vengefulness as “vindictive” rather than tolerating it as “legal” may help to convince observers of and participants in the criminal justice system that a defendant’s right to trial is worthy of respect. If one believes that trials are a valuable mechanism for adducing truth and vindicating a defendant’s dignity, as I do, then it is desirable in itself for more members of the public to think that defendants shouldn’t be punished for going to trial. It may also be good for the legitimacy of the criminal justice system for the public to believe that defendants are not in fact punished for going to trial, though it is not clear that any such legitimacy would actually be earned from truth.

Moreover, the expressive content of the label “vindictiveness” may itself shape prosecutorial behavior for the better. There is a rich literature discussing ways to encourage and reinforce more ethical prosecutorial conduct,217 and I do not intend to intervene in the conceptual debate about how best to alter conduct in general. The rule I propose both alters incentives (by threatening to overturn vindictively obtained convictions and sentences) and alters meaning (by reframing what was permissible as shameful). I simply wish to observe that a rule that prompts prosecutors occasionally to ask, “Am I punishing the defendant’s right to trial as a wrong?” may help them avoid doing so by making them conscious of their own behavior even if they do not fear any particular sanction.218

d. Conceptual

The fourth and, in my view, most important benefit of vindictiveness-as-vengeance is conceptual. My approach is helpful because it offers a way to justify some substantive regulation of prosecutorial bargaining behavior without entirely disregarding other important constitutional values and precedent. Rather than, say, policing prosecutorial behavior for duress in the formation of the bargain,219 this approach regulates prosecutorial bargaining behavior for the sake of protecting the right to trial.

Recall the moves the Court makes in Lafler and Frye.220 It explains that it wants to regulate plea bargaining because the right to trial is largely irrelevant, yet it is does so indirectly (and perhaps incoherently) because it doesn’t really foreswear the right to trial at all. The right to trial lurks just beneath the surface as a powerful constraint on the Court’s freedom of action. It is the reason, or at least a prerequisite, for deference to prosecutors’ broad charging discretion and the lack of a right to a plea in the first place. These, in turn, are the reasons why the Court cannot create a substantive entitlement to the standard deal that it thinks the defendant is generally supposed to receive. So, by purporting to consign the right to trial to irrelevance but not fully doing so, the Court ends up regulating plea bargaining only in roundabout fashion, through the conduct of defense lawyers.

The beauty of a rule against prosecutorial vindictiveness-as-vengeance is that it exalts and actually protects, rather than writes off, the right to a jury trial. But, in doing so, it actually achieves more substantive limits on prosecutorial behavior in plea bargaining than the Lafler-Frye indirection does. Vindictiveness inheres in intentionally punishing the defendant’s exercise of the right to trial. The need to prevent vindictiveness justifies a rule against excessive trial penalties. This, in turn, creates a de facto substantive entitlement to a plea deal that does not deviate too terribly from the standard one without good reason—precisely what Lafler and Frye shy away from doing. In this way, a revitalized vindictive prosecution doctrine can be a foot in the back door to direct judicial review of prosecutorial charging decisions in a partly administrative system of criminal justice.221

I concede that this sort of legal innovation may seem rather unlikely, and I do not aim to suggest that the Court would or could create such a rule in a single case tomorrow. Among other practical problems, because the usefulness of vindictive prosecution doctrine is currently so limited, it is somewhat difficult to envision the posture of a case that would let the Court move in this direction even if it wanted to. In this dramatically changing area of the law, however, “unlikely” is a relative term. It is hardly clear that presuming a prosecutor’s unreasonably excessive trial penalty to be vindictive is more unlikely or more destabilizing than, say, overturning a voluntary plea because counsel failed to advise the defendant that he might be deported as a collateral consequence.222

More generally, however, my arguments here are intended to stake out a different conceptual space in the ongoing discussion about regulating plea bargaining through constitutional criminal procedure. Many supporters of the recent doctrinal innovations have adopted a view that might be broadly described as: “Because the constitutional ideal of the jury trial is a dead letter, we need to regulate plea bargaining.”223 Many of those resisting the Court’s recent moves have taken a view that might broadly be characterized as: “We ought not directly regulate plea bargaining at the risk of further eroding the constitutional ideal of the jury trial.”224 Yet, as this Note’s discussion of vindictive prosecution has shown, a third view is possible and perhaps preferable: plea bargaining is amenable to regulation precisely because it is the alternative to the constitutional ideal, which we wish to remain relevant, of the jury trial. This simple point, while perhaps obvious on reflection, seems to have been lost a bit in the crossfire. This Note ultimately aims, then, to reassert it when it may be helpful in smoothing the transition from the criminal procedure of trials to the criminal procedure of pleas. Honoring the trial as the “24-karat test of fairness”225 and developing a more robust law of plea bargaining may actually be more compatible goals than we have recognized.


In the wake of the Aaron Swartz case, there seems to be newfound public momentum for constraining excessive prosecutorial discretion and making plea bargaining less coercive—or at least some newfound public interest in the subject.226 The question becomes where to go from here.

The cleanest and most attractive solutions to the problem may also be the most far-reaching: eliminating overlapping provisions of substantive criminal law, reducing overall sentence lengths, promulgating stricter and more objective charging guidelines within the executive.227 These aren’t new ideas. Indeed, they are old, independently good ideas. Their merits are well known, but their effectiveness in constraining discretion, reducing coercion, and thwarting quick-and-easy pleading may be precisely what prevents them from being implemented. In any event, they seem well beyond the competence of the judiciary as it develops new law in this area.

So, this Note has instead suggested an effort to square a persistent doctrinal circle. The history of vindictive prosecution makes clear the problem. That a prosecutor could commit a due process violation by penalizing the exercise of a legally protected right was a somewhat accidental and ill-considered notion from the start. As its impracticality became obvious, it was undone almost as quickly as it originated. The law was unstable because it made no serious attempt at an accommodation between the practicality of pleas and the constitutional value of trials.

Our law remains caught uncomfortably between trials and pleas. Because plea bargaining is viewed as legitimate and because the system would grind to a halt without it, preventing prosecutors from forcefully deterring trials would be a non-starter. Because our governing ideology maintains that the criminal justice system ought to be meaningfully adversarial, the law should honor rather than undermine the right to a jury trial. I have attempted to offer one possible way out. By preventing the government from intentionally punishing defendants who exercise the right to trial with an unreasonably steep penalty, we can distinguish between denigrating the right to trial and merely discouraging its exercise. In doing so, we can achieve modest substantive constraints on charging discretion.

The idea of vindictive prosecution has the capacity to respond to some of what seems objectionable about a system that uses the threat of astronomical penalties to extract guilty pleas from defendants who would otherwise exercise their constitutional right to a jury trial. The government probably needs to be able to make it prudent for the defendant to accept a plea. But the government probably ought not be able to tell the defendant that it is wrong, indeed impossible, for him to go to trial—a true, terrifying “offer he can’t refuse” that strips him of agency and denies him the possibility of dignified adjudication.228 Somewhere between these two states of prosecutorial mind is a line of constitutional significance.