Resilience, Retribution, and Punitive Damages

Article - Volume 100 - Issue 6


Soerono Haryanto—a wealthy guest at a Marriott hotel—pointed a gun at Mohammed Saeed’s head, demanding that Saeed kneel and kiss his feet.[1] Saeed—one of the hotel’s employees—obliged.[2] Haryanto responded by threatening to kill Saeed while mocking him as a mere “servant.”[3] Not satisfied with threats, Haryanto held Saeed hostage, demanding one million dollars from the hotel in exchange for Saeed’s release.[4] After nearly an hour the hotel ultimately secured Saeed’s freedom.[5] But the ordeal left him shattered. He suffered severe physical and psychological injuries that forced him to leave his job.[6]

Saeed sued Haryanto.[7] A jury awarded him one million dollars in compensatory damages and two million dollars in punitive damages.[8] A Texas court of appeals upheld both awards.[9]

The case just described, Haryanto v. Saeed,[10] seems unremarkable in at least one way: it seems unsurprising that the court of appeals upheld the large punitive damages award. Haryanto seems like a paradigm case, a case in which awarding punitive damages to the plaintiff should seem perfectly appropriate. But, precisely because Haryanto is an “easy” case, it is a small scandal that the most widely accepted rationales for punitive damages fail to justify the practice of allowing plaintiffs like Saeed to keep any of their punitive damages. That is, tort law permits juries to impose on defendants who commit egregious wrongdoings extracompensatory damages and allows courts to award those damages to plaintiffs.[11] But the standard rationales for that first part—imposing these damages on defendants—do not justify the second—transferring that monetary award to plaintiffs.

This Article aims to both identify the problem more clearly and offer a solution. To preview, Part I describes the two dominant justifications for punitive damages—retribution and deterrence—and explains why they fail to justify an important aspect of punitive damages awards.[12] That is, they fail to justify delivering punitive damages awards to plaintiffs. Because these rationales are entirely wrongdoer-centric, they give us reasons to impose monetary penalties on certain defendants, but they are entirely silent on why the proceeds of those penalties should go to the plaintiff. As a result, it should make no difference, from the perspective of pursuing either deterrence or punishment, whether the entire two million dollars goes to Saeed rather than, say, a state treasury or to Saeed’s attorneys exclusively. The deterrent effect, if any, will likely be the same whether Saeed or someone else gets the proceeds. And Saeed’s abuser will be “punished” insofar as he is coerced by the state to pay the money, regardless of where that two million dollars goes. Part I also discusses other attempted justifications for punitive damages, including the idea that they cure shortfalls in compensatory damages, serve as an incentive for attorneys to file lawsuits, or serve as a substitute for revenge. The upshot of Part I will be that the practice of awarding punitive damages to plaintiffs rather than to state coffers lacks a coherent justification. Punitive damages look like unjustified windfalls.

Skeptical worries similar to those expressed in Part I have taken a toll on the practice of awarding punitive damages.[13] Over time, punitive damages have become harder to obtain.[14] But rather than joining the ranks of the critics, this Article takes these criticisms to motivate a better justification for punitive damages, one that tries to justify the extracompensatory gains to successful plaintiffs. To that end, and drawing lessons on the failures of extant theories, Part I concludes by setting forth several criteria that any justificatory theory of punitive damages should satisfy.

Part II supplies that justification. The key insight is to identify an ideal that is overlooked in theoretical discussions about remedies: the ideal of resilience. Subpart II(A) explains that, although retributive theories of punitive damages under-determine the form of punishment that punitive damages actually mete out, they succeed in highlighting an important expressive role for punishment. Subpart II(B) fills the gap by articulating the aforementioned conception of resilience. It interprets resilience as a normative ideal according to which the proper response to setbacks—including serious wrongdoings—is for the victim to emerge in some sense better than before that setback. Subpart II(C) introduces the moral foundation for punitive damages, which is called “Retributive Resilience.” According to this account, we should understand punitive damages as empowering victims to act retributively against their wrongdoers when they act with ill will, and to empower in a certain way: to make the wrongdoers pay, literally, to satisfy the victims’ interests in realizing the ideal of resilience. Retributive justice meted out by the plaintiff against the defendant, which makes the plaintiff better off at the wrongdoer’s expense, is instrumentally and intrinsically valuable. Subpart II(D) points out that punitive damages help to realize these values, which are difficult to actualize in the absence of law. Finally, subpart II(E) argues that the present account satisfies the criteria of theoretical adequacy set forth in Part I.

Part III takes a practical turn, suggesting ways that Retributive Resilience counsels reform. The chief reform acknowledges that punitive-damages jurisprudence is already heavily informed by the deterrence rationale. So Part III argues for divorce: officials should distinguish between retributive damages and deterrence damages. The former, when they are awarded, properly belong to the plaintiff and thus are not windfalls, even though they are extracompensatory by design and purpose. The latter, which are imposed strictly to deter either the defendant or other future defendants from committing serious wrongdoings, do represent windfalls, and may be safely siphoned off via split-recovery statutes or other redistributive mechanisms. The prospects for this divorce are discussed in subpart III(A). Subpart III(B) evaluates the factors that courts currently allow juries to consider in assessing punitive damages. Many of these factors, it is argued, make sense given this Article’s theory. But certain factors—including those that ask juries to consider the deterrent effect of punitive damages—do not properly play a role in assessing retributive damages.

Reforms aside, the primary aim of this Article is theoretical and justificatory. And this task is important. Lacking a compelling rationale for punitive damages matters. As Professor Gregory Keating remarks, “Our law . . . is theory laden and reflexive. What it is and what it becomes depend, in important part, on what we think it is and what we think it should be.”[15] If we have no compelling reason to permit plaintiffs to recover punitive damages awards—if punitive damages represent, as critics suggest, unjustified windfalls—then we should not mourn the ongoing erosion of punitive damages awards.[16] Opponents of punitive damages have won stunning victories in recent decades, both by successfully lobbying for split-recovery statutes (which tax punitive damages at extremely high rates), or by persuading the U.S. Supreme Court to severely limit these awards on Due Process grounds.[17] And some states simply ban punitive damages altogether.[18] But if opponents of punitive damages are correct, then these developments are to be celebrated.

This Article argues, to the contrary, that the erosion of punitive damages is to be lamented, limited, and to some extent, reversed. If plaintiffs are prevented from recovering them, individuals lose a crucial avenue by which victims may attain a constructive form of retributive justice that allows them to demonstrate—to themselves and others—their resilience. We should not allow a lack of theoretical imagination, and consequent failure by some commentators to understand why punitive damages exist, to destroy this crucial and time-honored mechanism for bouncing back better than before.

I. Existing Justifications for Punitive Damages

Punitive damages—also called “vindictive” or “exemplary” damages”[19]—have a long pedigree.[20] Plaintiffs are eligible to recover punitive damages only when the underlying tortious conduct by the defendant is especially egregious.[21] The mine run of negligence cases will not give rise to a viable claim for punitive damages.[22] Instead, punitive damages are typically appropriate only when “an individual’s actions are of such an outrageous nature as to demonstrate intentional, willful, wanton, or reckless conduct.”[23] This Article will use the umbrella term “ill will” to cover all of these punitive-damages-eligible misbehaviors.

Two other preliminary points. The first is that punitive damages provide extracompensatory relief.[24] That is, punitive damages represent money drawn from the defendant and awarded to the plaintiff in excess of compensatory damages.[25] Indeed, punitive damages are often available only upon prevailing on a claim for actual damages.[26] This extracompensatory feature of punitive damages leads courts and commentators to routinely characterize punitive damages as windfall gains to plaintiffs.[27] The second point is that punitive damages, as the name suggests, are genuinely punitive. This distinguishes them, at least in principle, from other forms of extracompensatory damages including, for example, statutory treble damages or aggravated damages.[28] The punitive dimension of the remedy helps to explain why some kind of heightened culpability—why some kind of ill will—is necessary to make punitive damages available.[29]

But why is there a need for a sound justification for punitive damages? Punitive damages have long had a bad reputation, having been described as “unsound in principle, . . . unfair and dangerous in practice,”[30] as an “anomaly,”[31] as a “ridiculous combination of lottery and pillory,”[32] and as a remedy that “flout[s] the boundaries separating crime from tort, public law from private law, and punishment from compensation.”[33] Commentators periodically argue that punitive damages should be banned.[34] Reasons vary. Critics contend that punishment has no place in civil disputes,[35] that it is inconsistent with corrective justice,[36] or that punitive damages are unconstitutional.[37] Some jurisdictions do more than criticize: they forbid punitive damages awards.[38] Short of outright prohibition, opponents of punitive damages have nevertheless persuaded the Supreme Court to limit the size of those damages,[39] place special procedural and evidentiary demands on plaintiffs seeking punitive damages awards,[40] and even claim that punitive damages awards have “run wild.”[41] Procedural reforms presuppose that punishment in general demands greater procedural safeguards than compensatory damages, a view endorsed by the Supreme Court.[42] A more recent set of reforms redirects portions of punitive damages awards from plaintiffs to public treasuries, a practice codified in so-called “split-recovery” statutes.[43] Opposition to punitive damages never fully goes away.

This should not come as a surprise. The main argument in this Part is an “internal” critique: a central aspect of punitive damages cannot be justified on its own terms because the most widely recognized rationales offered in favor of punitive damages—retribution and deterrence—fail to justify awarding punitive damages to plaintiffs, and so do little to stand in the way of that aspect’s demise. To preview the arguments, retribution and deterrence are wrongdoer-centric rationales that cannot justify awarding extracompensatory gains to plaintiffs. Compensation suffers several problems as a justification, not the least of which is its failure to cohere with the “punitive” aspect of punitive damages. And mixed theories inherit the problems of each of these proposals.

As subpart I(A) argues, the main rationales advanced in favor of them, deterrence and retribution, fail. They share a common shortcoming: they cannot adequately explain or justify transferring extracompensatory awards to the plaintiff. And retributive accounts often fail to justify allowing individuals to act retributively. Subpart I(B) argues that attempts to reconcile punitive damages with compensation likewise fail. Among other things, construing punitive damages as a form of compensatory damages effectively reads punitive damages out of existence.

A. The Main Rationales: Punishment and Deterrence

The most widely accepted justifications for punitive damages are the twin goals of punishment and deterrence.[44] As the name suggests, “punitive” damages may simply do what they say they do—provide the plaintiff with a vehicle for punishing the wrongdoer. As for why we should punish at all, one explanation holds that punishment condemns the wrongdoer.[45] But whatever the reason for punishing, the thought here is that criminal and civil punishment share the same goals. As for deterrence, the thought is that compensatory damages simply fail to provide an adequate incentive to deter people from engaging in certain manifestly undesirable behavior. Punitive damages may better deter antisocial conduct where compensatory damages will not suffice.

No doubt punitive damages can operate to punish and deter. But they cannot justify the practice of awarding plaintiffs punitive-damage windfalls, where “windfall” is understood to mean any damages award in excess of full compensatory damages.[46] Chief Justice Edward George Ryan, writing on the Wisconsin Supreme Court in 1877, struggled to understand why, “when the sufferer by a tort has been fully compensated for his suffering, he should recover anything more.” [47] Equally confusing, he continued, was “why, if the tortfeasor is to be punished by exemplary damages, they should go to the compensated sufferer, and not to the public in whose behalf he is punished.”[48]

Chief Judge Ryan was not alone among his contemporaries. In Murphy v. Hobbs,[49] decided in 1884, the Colorado Supreme Court bemoaned the fact that punitive damages appear to compensate “above and beyond” that which a plaintiff is entitled to receive “after being fully paid for all the injury inflicted upon his property, body, reputation, and feelings.”[50] The Washington Supreme Court emphasized in Spokane Truck & Dray Co. v. Hoefer[51] that full compensatory damages make plaintiffs “entirely whole,” compensate them for harms that “frequently border on the imaginary,” and therefore leave no further need for plaintiffs to “exact[] . . . [a] pound of flesh” through punitive damages.[52]

These nineteenth-century critics of punitive damages lost their battles.[53] As Professors John Goldberg and Benjamin Zipursky point out, “almost every state ha[s] allowed punitive damages to be awarded.”[54] But punitive damages skeptics may ultimately win the war. During the 1980s, after “a period of relative quiet[,]” hostility towards punitive damages flared up again.[55] This time the skeptics persuaded courts and legislatures to sharply restrict the availability and amount of punitive damages awards, as already noted.[56]

The tort reform movement may have peaked, but hostility towards punitive damages remains. Courts and commentators continue to voice concerns about their apparent status as windfalls, specifically. In 2013, for example, the Indiana Supreme Court observed that if “[t]he central purpose of punitive damages is to punish the wrongdoer and to deter him from future misconduct, not to reward the plaintiff and not to compensate the plaintiff[,]”[57] then punitive damages awards that flow from the defendant to the plaintiff seem inherently suspect because they appear to “result in an unfair windfall recovery for plaintiffs who have already been made whole through an award of compensatory damages.”[58]

The problem should be obvious. If punitive damages are justified solely in terms of deterrence and retribution, there is little reason to allocate any punitive-damage windfalls to plaintiffs rather than, say, the state,[59] or even to the plaintiffs’ attorneys to the exclusion of their clients.[60] Less often noticed is the deeper source of the problem, which is that both punishment and deterrence place the wrongdoer front and center, focusing exclusively on reasons for imposing extracompensatory damages on the wrongdoer rather than the practice of conferring a benefit on the victim.[61] To drive the point home, notice that we could further reform punitive damages to make sure that 100% of punitive-damages awards that plaintiffs currently receive could be redirected to public treasuries—or the “public” as Chief Judge Ryan suggests, above—without necessarily undermining punishment or deterrence.[62] Punitive damages would just as easily condemn the defendant, and the class of potential wrongdoers seems just as likely deterred regardless of whether the punitive damages are ultimately received by the public versus the plaintiff.[63] Indeed, if the goal of punitive damages is to punish and deter, using the money to help finance the state’s criminal justice system better comports with those objectives, given that the criminal justice system’s purported concern for punishing and deterring wrongdoers. So, wrongdoer-centric justifications seem incomplete at best.[64] In short, Chief Justice Ryan was correct to worry about why a fully compensated plaintiff rather than “the public,” or someone else like the plaintiff’s attorney, should reap those entire punitive rewards.[65] And to the extent we cannot come up with a sound justification for the extracompensatory feature of punitive-damages, perhaps courts should stop awarding more than compensatory relief to plaintiffs.

A proponent of the deter-and-punish orthodoxy has a ready response. A seemingly obvious reason to award punitive damages to plaintiffs is to provide them with an incentive to bring a claim, especially in cases where compensatory damages may not suffice to adequately incentivize filing a lawsuit.[66] On this view, punitive damages are merely bounties allocated to plaintiffs for bringing especially pernicious behavior to light.[67] Notice that the bounty conception of punitive damages is consistent with the view that punitive damages also serve to punish and deter. The bounty is just a reward that provides plaintiffs and their attorneys the incentive to bring suits, over and above whatever incentives might inhere in compensatory damages alone.

This response is dubious. Primarily, bounty rationales do not explain why the bounty-hunter plaintiff must have been injured by the defendant to obtain the punitive-damages “windfalls.” Bounty hunters, or qui tam plaintiff analogues, need not show that defendants have infringed their rights before becoming eligible for their bounties.[68] Accordingly, the bounty rationale is consistent with transferring the entirety of punitive damages to plaintiffs’ lawyers to the exclusion of plaintiffs themselves.[69]

There are highly sophisticated versions of each of these theories.[70] A rich literature in law and economics construes punitive damages as a way to achieve optimal deterrence, where otherwise legal norms would go underenforced.[71] These theorists worry about underdeterrence where compensatory relief would fail to motivate plaintiffs to bring claims and where wrongdoing is difficult to discover.[72] Despite its sophistication and influence, this perspective—which tends to view private law as simply public policy by other means—still views the plaintiff as merely a private attorney general or bounty hunter, which once again renders mysterious the need for plaintiffs to establish that they have suffered personal injuries. And when optimal deterrence becomes the main aim, this distorts our understanding of punitive damages. For example, optimal-deterrence theorists argue that the reprehensibility of misconduct should be irrelevant for assessing punitive damages.[73] In other words, there’s nothing necessarily punitive about punitive damages.

As for retributive theories (or at least broadly deontological theories), Marc Galanter and David Luban provide a sophisticated example, arguing that punitive damages correct mistaken claims about the wrongdoer’s superiority over the victim.[74] Serious wrongdoers that manifest degrading or demeaning attitudes towards their victims effectively send the message that the wrongdoer may do as she pleases with or to the victim, as though the wrongdoer were superior to the victim.[75] Punishment in general, and punitive damages in particular, aims to refute that message by empowering the plaintiffs to defeat their oppressors.[76] Because inflicting this harsh treatment is incompatible with the claim of moral superiority—superiors do not have to submit to harsh treatment by their inferiors—defeating wrongdoers operates to refute the message.[77]

Like Galanter, Luban, and others, this Article will accept that certain wrongdoings—i.e., those expressing ill will—manifest mistaken value judgments about the wrongdoer’s superior worth relative to the victim.[78] But even if we think that this suffices as a justification for empowering plaintiffs to punish their wrongdoers, their theory ultimately suffers from a familiar shortcoming: it fails to explain why courts should order defendants to transfer windfall damage awards to the plaintiff rather than the state or entirely to the plaintiff’s attorney. Still, there is something correct about a retributive approach to punishment and punitive damages in particular. We will return to the topic later.[79]

B. Compensation and Mixed Theories

Punishment and deterrence fail to supply adequate justifications for punitive damages because they’re wrongdoer centric. Compensation is sometimes offered as an alternative rationale. The idea here is that punitive damages deliver no windfall at all; they instead represent relief that stands in for difficult-to-monetize losses including harms to dignity or pain and suffering.[80] The strategy behind this family of proposals is to deny, in effect, that punitive damages are truly extracompensatory. There is nothing “extra” about them. Compensatory damages aspire to make victims “whole.”[81] So do punitive damages on this family of views.[82] They are simply compensatory damages under a different label.[83] Because various versions of this view have been ably and extensively criticized elsewhere,[84] this Article’s discussion of the approach will be brief.

There are at least two ways of construing punitive damages as compensatory. The first focuses on the ways in which punitive damages compensate the individual tort claimant with the aim of repairing harms done to her, while the second construes punitive damages as a way of compensating for social harms. Both turn out to be recipes for eliminating rather than justifying punitive damages.

The strongest version of the first kind of theory goes something like this: Punitive damages compensate for intangible moral injuries, insults, or other similar indignities not implicated by run-of-the-mill negligence claims.[85] Michigan courts purport to adopt this view, asserting that “[t]he real purpose [of exemplary damages] is to compensate the plaintiff for the injuries he has suffered.”[86] Consider the Michigan Supreme Court’s commentary in Kewin v. Massachusetts Mutual Life Insurance Company.[87] According to the Court, exemplary damages—which most courts view as synonymous with punitive damages[88]—serve an exclusively compensatory purpose.[89] Specifically, exemplary damages try to compensate “for the humiliation, sense of outrage, and indignity resulting from injuries maliciously, willfully and wantonly inflicted by the defendant.”[90]

The Michigan approach nicely fits certain facts about punitive damages. For example, it explains why they are normally not available in ordinary negligence actions, which do not involve the kind of malicious, willful, wanton misconduct necessary to trigger punitive damages.[91] The approach, like all punitive-damages-as-compensation approaches, also seems self-justifying. If punitive damages are simply compensatory damages under another label, then whatever justifies compensatory relief in general also presumptively justifies this kind of extension of compensation to dignitary harms.

But the logic of this approach also simply reads the “punitive” aspect of punitive damages out of existence. In most jurisdictions, exemplary damages are just another name for punitive damages. Sometimes Michigan courts follow this common usage. In Peisner v. Detroit Free Press, Inc.,[92] for instance, a court of appeals explained, “In Michigan, exemplary or punitive damages are recoverable as compensation to the plaintiff, not as punishment of the defendant.”[93] But this leaves us with the incoherent result whereby “punitive” damages are nonpunitive—a confusion noted by a concurring judge in the Peisner decision.[94]

Nor is this simply a labeling issue. Despite the hiccup in Peisner, Michigan courts generally eschew the label of “punitive damages” because they have abandoned punitive damages.[95] This is the logic of the position that holds that punitive damages are compensatory damages. Construing punitive damages as merely compensation for dignitary harms explains their punitive nature away rather than embracing it—and is consistent with abolishing punitive damages altogether.[96] But my theoretical aim in this Article differs: to see whether we can avoid the conclusion that punitive damages ought to be eliminated, a conclusion that seems preordained given common rationales provided for them. The Michigan approach embraces that end rather than resisting it.

A second approach also construes punitive damages as a form of compensation. This one characterizes them as compensation for social harms inflicted by actors on a community, perhaps even beyond the class of litigants named in a class action.[97] The view seems appealing, for example, in cases of corporate misconduct not easily pinned on any single individual but which nevertheless betrays or signals some form of malice and which may affect many members of a community. And the social-harms view arguably preserves a sense in which punitive damages provide extracompensatory gains to individual plaintiffs to the extent that they end up recovering for harms inflicted on that broader community.[98]

But this framework faces insuperable difficulties. By construing punitive damages as a way of shoring up compensatory shortfalls inflicted on the public, the account ignores or renders mysterious the fact that certain kinds of egregious behavior are necessary to trigger punitive relief.[99] That is, it is puzzling why this goal would be limited to cases of serious misconduct—something akin to malicious, wanton, or other egregious behavior—sufficient to trigger the availability of punitive damages.[100] Societal damages can happen as a result of a discrete act of corporate negligence as well yet punitive damages would not normally be available in such cases.

More to the present Article’s point, shoring up compensatory shortfalls fails to justify or explain the practice of awarding punitive damages to plaintiffs. Indeed, Professor Catherine Sharkey has used this observation to help justify the practice of siphoning off punitive damages for public use via split-recovery statutes.[101] And there is no logical stopping point to this view, which would entail that most if not all punitive damages should be reallocated to the state or shared between the state and the plaintiff’s attorneys, to the exclusion of the plaintiffs themselves. After all, if we accept that punitive damages operate, and should be seen to operate, as compensation for injuries beyond those suffered by the plaintiffs themselves, then the state, standing in on behalf of the community affected by the defendant misconduct, should recover on their behalf as well.

More might be said on behalf of, and against, theories that interpret punitive damages as compensatory damages under either the first or second approaches.[102] But a complete discussion of these views will not be undertaken here. This Article aims instead to present an alternative that tries to vindicate punitive damages, even assuming for the sake of argument that they represent extracompensatory recoveries. Compensatory views do not even try.[103]

Most of these difficulties flow from a common source. While deter-and-punish theories are entirely wrongdoer-centric, compensatory theories overcorrect by becoming wholly victim-centric. But simply combining victim-focused and wrongdoer-focused rationales will not solve the problems. This kind of hybrid account inherits the problems of each of its components.[104] Notice, for example, that eliminating punitive damages altogether is consistent with mixed theories. Let the criminal law work its punitive magic independent of civil recovery. Or make the defendant pay a fine directly to the plaintiff’s attorney or to the treasury.[105] And let a public system of, say, disability insurance fill any compensatory gaps that remain after the court awards full compensation. Mixed justifications cannot solve the deeper problem of justifying plaintiff windfalls, which requires justifying not just awarding an extracompensatory benefit to the plaintiff and imposing punitive fines on the defendant. What needs justification is the relationship between wrongdoer and victim: the plaintiff windfalls awarded by punitive damages come at the expense of the wrongdoer.[106] Mixed justifications do not account for this structural feature.

So the dominant justifications for punitive damages don’t work. The problem is deep, even when supplemented by attempts to construe punitive damages as bounties. The most prominent goals of punishment and deterrence are instrumental justifications defined independently of the underlying structure of punitive damages, focusing entirely on justifying extracompensatory costs imposed on the wrongdoer. Collecting punitive damages from defendants makes sense in a wrongdoer-centric model. But then we lose sight of why defendants must pay those damages to the plaintiff, let alone why the resulting windfall has a punitive character.

C. Revenge-Based (or Revenge-Adjacent) Rationales

Another set of possible justifications has drawn attention from tort theorists but has received less explicit attention in courts. Individual theories included in this set vary greatly, but they draw inspiration from the idea that punitive damages are a vehicle of, or substitute for, revenge or retaliation. To understand this family of views, it is helpful to situate it within a venerable tradition in legal theory that understands the whole system of civil justice as explained and perhaps even justified by reference to the social goal of replacing practices of private revenge and retaliation with civil forms of relief.

Referring to this tradition, Oliver Wendell Holmes, Jr. writes:

It is commonly known that the early forms of legal procedure were grounded in vengeance. Modern writers have thought that the Roman law started from the blood feud, and all the authorities agree that the German law begun that way. The feud led to the composition, at first optional, then compulsory, by which the feud was bought off.[107]

Some contemporary writers, to a greater or lesser degree, defend versions of this view as both an explanation and justification of tort law and private law more generally. These proponents of revenge-substitute views or closely adjacent views can arguably be found in the work of civil recourse theorists John C.P. Goldberg and Benjamin Zipursky and those inspired by them.[108] Descriptively, Goldberg and Zipursky argue that tort law empowers victims to demand fair redress from individuals who have wronged them.[109] Normatively, Goldberg and Zipursky argue that both the U.S. Constitution and Lockean political morality require the state to recognize individual rights to demand this kind of redress, as compensation for giving up their rights to use self-help (including perhaps retaliation) to enforce their rights.[110] Punitive damages are consistent with this picture to the extent that they are tantamount to allocating to an individual a “right to be punitive” towards those who seriously mistreat them.[111]

Responding to the civil recourse theorists and embracing rather than distancing himself from revenge (as they do),[112] Scott Hershovitz argues that both revenge and civil litigation share the same constitutive aim: to thwart messages conveyed by unchecked wrongdoings, like the messages that victims do not matter and that they are persons for whom it is permissible to wrong.[113] Like Goldberg and Zipursky, Hershovitz presents his view as justifying tort practice while explaining its structure.[114] And like Goldberg and Zipursky, Hershovitz interprets tort practice in part to explain why compensatory damages are not the lone remedy available in torts cases.[115] Others argue in a similar spirit, observing that nothing about punitive damages is particularly alien to the idea of tort law once we understand tort law as empowering plaintiff–victims to demand satisfaction from their defendant–wrongdoers.[116]

These views have been criticized on numerous grounds.[117] But for present purposes notice that, to the extent that these views construe tort law as providing a civil substitute for revenge or retaliation, they risk overgeneralizing, since not all wrongdoings—not even all legal wrongdoings—intelligibly call for revenge or retaliation.[118] Holmes recognized this point long ago. Once “compensation” took on a role as “the alternative to vengeance,” Holmes opined, “we might expect to find its scope limited to the scope of vengeance,” which in turn is limited to cases in which “a feeling of blame” would be appropriate.[119] Wrongdoings for which vengeance and blame are fitting responses are limited, thought Holmes, and “can hardly go very far beyond the case of a harm intentionally inflicted,” famously remarking that “even a dog distinguishes between being stumbled over and being kicked.”[120] Holmes concluded that revenge-substitute views failed to explain or justify civil liability for behavior other than intentionally imposed harms or similarly egregious wrongdoings, suggesting that negligence claims, the largest class of tort claims, cannot be explained or justified by those views.[121]

That said, even if one is convinced that civil-recourse theory overgeneralizes and looks too much like revenge, this might seem like a virtue if the theoretical goal is to explain and justify the availability of punitive damages. After all, demands for punitive damages are typically available only when the underlying tort reflects malice or other mental states that indicate a serious wrongdoing against the plaintiff.[122] And this set of wrongdoings is precisely the set of claims that fits best with revenge-based theories. Indeed, Benjamin Zipursky has argued persuasively that punitive damages reflect a “right to be punitive” that has always been a part of tort law.[123] Anthony Sebok, another civil-recourse-inspired theorist, goes further in arguing that punitive damages simply are a form of state-approved revenge.[124] So the apparent proximity of civil-recourse theory to revenge may turn out to be a virtue in trying to justify punitive damages, even though it’s a potential vice in trying to make sense of modern structures of tort law on the whole.

Perhaps surprisingly, a minority of jurisdictions in the United States already draw explicit connections between revenge and punitive damages, with some in effect following Sebok in construing punitive damages as a form of revenge, one that allows “the wronged plaintiff to take his revenge in the courtroom and not by self-help.”[125] Most courts among this minority, however, follow the more genteel approach that recognizes that punitive damages can serve as a substitute for revenge.[126] The Supreme Court of Appeals of West Virginia, for example, repeatedly asserts that “[p]unitive damages provide a substitute for personal revenge by the wronged party.”[127] And West Virginia is not alone. Courts in the District of Columbia, New Jersey, and at least one federal court have suggested that punitive damages serve as a substitute for revenge.[128]

Regardless of whether characterized as a substitute for revenge or another means for pursuing it, revenge-based justifications might still seem implausible. Revenge is thought of as “anti-social”[129] behavior rooted in a desire to harm another person—and thus is supposed to be “utterly irrelevant to the modern tort system.”[130] But according to some theorists, the depravity of revenge is precisely what justifies punitive damages as a substitute for it,[131] with courts likewise asserting that punitive damages involve “safeguarding the public peace”[132] or “diverting the plaintiff’s desire for revenge into peaceful channels.”[133] Punitive damages represent quite literally a “civil” and regulated alternative to unchecked payback, while offering a sense of finality to disputes that is hard to come by in the midst of self-perpetuating cycles of revenge. And as noted above, this kind of justification is implied in Holmes’s discussion of the origins of civil justice: to displace violent self-help.[134]

Notice that these revenge-based theories have a structural advantage over the prevailing instrumentalist rationales of deterrence and punishment. Recall the structural problem: deterrence and punishment rationales are entirely wrongdoer-centric, and as a result, fail to explain or justify the role of the plaintiff’s recovery. There is, in other words, no essentially bipolar relationship between punishment and deterrence of the wrongdoer, on the one hand, and the victim of the wrongdoing on the other. The plaintiff, as a victim of wrongdoing, plays at best an accidental role.

Revenge-based justifications, by contrast, are not exclusively focused on the wrongdoer at the exclusion of the victim. Instead, revenge-based views essentially take into account both victim and wrongdoer, recognizing that they stand in a normatively significant relationship. Put differently, revenge is a response that is intelligible only by reference to a victim–wrongdoer relationship, real or alleged. Revenge is taken by someone who claims to have been wronged against a person who has allegedly committed the wrongdoing, just like a victim of a tort brings an action against the tortfeasor.[135] By incorporating the concept of revenge—either in itself or as a substitute for it—the revenge-based rationale for punitive damages provides a clear structural advantage over the retribution–deterrence orthodoxy.

Revenge-inspired views nevertheless face problems, normative and structural. The normative problems are obvious and have already been telegraphed. A pure revenge-based view—which identifies punitive damages as a form of revenge—will strike many as normatively undesirable. And perhaps with good reason. Revenge and retaliation should not be dignified in courts of law, one might think, or anywhere else for that matter.[136] Viewing punitive damages as a substitute for revenge seems preferable. After all, the story that courts and commentators tell here emphasizes that providing a civil outlet for vengeance is precisely what justice calls for.[137] But even revenge “substitutes” may end up improperly validating revenge as a practice by recognizing it as an impulse worthy of accommodation—insofar as they are seen to have the same “point” of recognizing impulses unworthy of recognition.[138]

Ultimately, this substantive concern may turn on empirical questions that I cannot settle here.[139] That is, if proponents of revenge-substitute views are correct that civil litigation generally, and punitive damages in particular, are practically necessary in the modern world to avoid cycles of violent retaliation and revenge from taking root, this very well seems like a potentially strong justification for civil litigation and for punitive damages—even if, indirectly, punitive damages end up “dignifying” the desire for revenge. Dignifying the instinct would be a necessary evil.

But there is a deeper and by now familiar problem with trying to justify punitive damages on the basis of revenge, one built into the normative structure of revenge: extant revenge-based rationales cannot justify the extracompensatory payments to plaintiffs. The constitutive aim of revenge, and presumably its substitutes, is at least in part to inflict some kind of harm on the wrongdoer.[140] In the state of nature this harm may be violent. As we move towards a system of civil justice, we tame vengeful impulses by channeling them into a civil substitute. But the telos must remain the same to provide an adequate substitute.[141] Accordingly, some measure of damages is imposed on the wrongdoer in order to sate the victim’s disposition to seek revenge. But nothing about this account requires that the plaintiff obtain the damages collected from the defendant as a windfall, as opposed to siphoning off 100% of that amount to the state, perhaps in order to help finance its justice system. So, after all is said and done, unorthodox views grounded in revenge or substituting for it—though preferable to orthodox views discussed above[142]—face one of the same and most fundamental problems facing orthodox ones.[143]

D. A Test: Identifying a Good Justification for Punitive Damages

The shortcomings of these rationales suggest criteria by which we can evaluate proposed rationales for punitive damages. A successful theory accomplishes at least the following tasks: First, it will explain or at least cohere with the sense in which “punitive damages” are indeed punitive. Second, and relatedly, the theory will reflect the fact that punitive damages are available only when defendants commit especially pernicious wrongdoings. These criteria likely interrelate because punitive reactions are justified, if at all, only in response to morally culpable mental states. Mere negligence is not enough for punitive damages.[144]

The dominant justifications modeled on punishment and deterrence satisfy these two criteria adequately. After all, both deterrence and retribution remain dominant justifications of criminal punishment, so it hardly seems a stretch to apply those theories to punitive damages. But the dominant justifications failed with respect to a third criterion: they must justify extracompensatory damages awarded to the plaintiff in particular. Fourth, and relatedly, the justification must justify the fact that those damages come from the defendant. Philosophers of tort law in other contexts have called criteria three and four jointly the requirement of “bipolarity” and have used this structural feature of litigation to criticize instrumentalist theories of tort law in general.[145] This Article repurposes this critique to provide tests for adequacy that any normative theory of punitive damages must satisfy in order to avoid the pitfalls faced by the dominant theories.

II. Punitive Damages as Retributive Resilience

The main theories of punitive damages canvassed in Part I fail because they try to justify either imposing harsh treatment on a wrongdoer or healing the plaintiff’s wounds, or because they present an unstable mixture of these goals. These approaches fail because they take the core case in favor of punitive damages to be an instrumental one in the narrow sense; that is, they identify some goal external to the underlying bipolar legal form according to which a successful plaintiff obtains extracompensatory relief from the defendant. Justifying punitive damages requires explaining and justifying that distinctive form of punitive action, according to which a plaintiff demands extracompensatory relief from the defendant.

This Part provides a new account, which will be called “retributive resilience.” To preview: Punitive damages empower victims to act punitively against their oppressors by requiring them to finance those victims’ resilience interests—i.e., their interests in bouncing back better than before the wrongdoing—if they so demand. Insofar as this is a form of punishment, the magnitude of relief must be proportional to the gravity of the wrongdoing. This proportionality principle acts as a constraint on the awards that may be given to the plaintiff at the wrongdoer’s expense.

The explanation and justification of this account builds in stages. Subpart II(A) revisits a retributivist theory of punishment, which justifies retributive action in terms of its expressive value. But because this account does not explain why punitive damages should be transferred to plaintiffs, subpart II(B) begins to fill the gap by articulating an ideal of resilience, according to which a proper response to serious wrongdoings involves the victim transforming the wrongdoing into an opportunity to emerge stronger or better off than before the wrongdoing. Subpart II(C) puts the pieces together: punitive damages should be understood as empowering victims to inflict punishment in the form of a demand to satisfy the victim’s resilience interests. Subpart II(D) shows that the punitive damages practice solves practical problems that retributive justice—taken alone—creates, including the fact that retribution rarely produces resilience. Finally, subpart II(E) concludes, explaining how the resulting theory satisfies the desiderata set forth in subpart I(C), above.

A. Revisiting a Version of Retributivism[146]

Consider why the state might permit individuals to act retributively. Marc Galanter and David Luban, briefly introduced in Part I, argue that punitive damages aim to empower victims to defeat their wrongdoers.[147] To make sense of this claim, notice that serious wrongdoings express certain false value judgments about the wrongdoers’ own superiority to their victims.[148] Such claims include, for example, “‘I count but you do not,’ ‘I can use you for my purposes,’ or ‘I am here up high and you are there down below.’”[149] Galanter and Luban build on the work of Jean Hampton, who had argued that “[w]hen someone wrongs another, she does not regard her victim as the sort of person who is valuable enough to require better treatment.”[150] “By victimizing me,” continues Hampton, “the wrongdoer has declared himself elevated with respect to me, acting as a superior who is permitted to use me for his purposes. A false moral claim has been made.”[151]

Haryanto v. Saeed illustrates Hampton’s point.[152] Haryanto held Saeed hostage at gunpoint and threatened to kill him.[153] Haryanto felt entitled to use Saeed for his own purposes (e.g., demanding one million dollars from the hotel).[154] Then, invoking a degrading ritual, Haryanto demanded that Saeed kneel before him and kiss Haryanto’s feet.[155] Coercing Saeed to comply with this demand served only to express Haryanto’s contempt for Saeed and Saeed’s inferiority, while attempting to demonstrate Haryanto’s dominance and superiority.[156] Adding insult to injury, Haryanto mocked Saeed by insisting that he was worthy of being “no more than a servant.”[157] “Demeaning” understates Haryanto’s gunpoint degradation of Saeed.

If malicious wrongdoing expresses the victim’s inferiority, punishment, according to Galanter, Luban, and Hampton, is supposed to undermine that message by defeating the wrongdoer.[158] Hampton asserts that, on her brand of retributivism, “the false claim [must be] corrected.”[159] She continues: “[t]he lord must be humbled to show that he isn’t the lord of the victim,” by “caus[ing] the wrongdoer to suffer in proportion to my suffering at his hands.”[160] Meting out punishment involves “master[ing] the purported master,” “showing that he is my peer,” and in turn ensuring that the wrongdoer’s “elevation over me [the victim] is denied[] and moral reality is reaffirmed.”[161] Less cryptically, punishment is a form of accountability that reaffirms the equal moral worth of the victim and wrongdoer, contravening messages to the contrary.[162]

Mere compensation is not necessarily enough to signal this equality. Compensatory relief alone risks sending the wrong message when the wrongdoing reflects ill will. In the context of punishment, “[i]f the punishment is too lenient,” write Galanter and Luban, this “implicitly ratifies the view that the victim is the sort of person it is all right to treat badly.”[163] For example, citing Hampton, Galanter and Luban remark that “if sentences for forcible rape are low, the legal system is expressing a contemptuous view of the value of women relative to men.”[164] And when damages awards are too low, they express contempt towards the victim and validate the wrongdoer’s contemptuous behavior.

But the problem is not merely that compensation will be too low in proportion to the wrongdoing. After all, compensatory damages can be substantial.[165] The problem with awarding compensatory damages alone concerns what message this communicates when wrongdoings that are different in kind (because of their ill will) from mere accidents are treated no differently than mere accidents, which in turn may “express” nothing more than absent-mindedness or evince little more than the actor’s acting “stupidly.”[166] Making only compensation available in such cases risks leaving that message unmitigated, and worse, suggests that wrongdoing involving ill will is no different in kind from mere negligence.[167] Analogously, it is like calling a homicide “manslaughter” rather than “murder” even if the ultimate sentences turn out the same.[168] Suggesting that Haryanto’s conduct towards Saeed is morally no different in kind from an accident likewise risks ratifying Haryanto’s message about Saeed’s inferiority.[169] By failing to demarcate the distinctive aspect of the wrongdoing—that is, by failing to make available a distinctive remedy that is responsive to the ill will that marks out this wrongdoing as distinctive from a mere “moment of carelessness”[170]—courts risk reinforcing rather than mitigating that false message of inferiority. Only wrongdoing that manifests ill will calls for correcting this message of the victim’s inferiority.[171] Accordingly, a retributive response is called for only when the wrongdoing manifests ill will.[172]

So compensation will not normally suffice to mitigate the degrading messages communicated by serious moral wrongdoings. The idea that serious wrongdoings express or imply this type of message—that serious wrongdoings, in Jean Hampton’s words, essentially “demean” victims—has proven influential.[173] And it will be taken for granted here. But as discussed earlier,[174] the view struggles to justify punitive damages by itself because it fails to explain why punishment—which consists of harsh treatment of the wrongdoer—ought to take the form of transferring extracompensatory relief to the plaintiff rather than the state. We need another ingredient.

B. Resilience and Its Reasons

Retributive theories of punitive damages are incomplete because they do not account for the form that the punishment should take. This is where resilience steps in. Given the reasons for behaving resiliently—what we might call “resilience interests”—it is simply fair to require that wrongdoers finance that resilience. That is, it will be argued that punitive damages are fair because they allow victims to not only act punitively towards their wrongdoers (for the reasons set forth above), but because nobody is more fairly required to pay for what will be called the “resilience interests” of their victims upon their demand. Filling in the details of this argument will require a detour into the conception of resilient action presupposed here, as well as a detailed look into some of the reasons we have for acting resiliently.

1. An Ideal of Resilience.—

Talk of “resilience” is commonplace—and increasingly so.[175] Several definitions exist, most of which focus on the capacity of individuals to persist in the face of adversity.[176] But unlike a common usage of the term, “resilience” as used here does not refer to a trait, capacity, or virtue.[177] Instead, “resilience” refers to a feature of an action or undertaking in which, in response to a setback, one successfully makes one’s situation meaningfully better than before that setback.[178] In other words, the result of one’s undertaking is to make oneself better off, in some meaningful sense, than before a setback. The undertaking itself is also somehow understood by the relevant agent or community as being undertaken in response to that setback.

Resilience, so understood, was a major theme of Joseph R. Biden’s campaign for the presidency.[179] Biden asserts, for example, that after the COVID-19 pandemic, “it’s not sufficient to build back, we have to build back better.”[180] Elsewhere he remarks, “I know times are tough, but I want you to know you’re not alone. We’re all in this fight together. And together, we’ll emerge stronger than before.”[181] President Biden is not alone, either. The Global Facility for Disaster Reduction and Recovery (GFDRR) independently adopted this slogan in a report called Building Back Better: Achieving Resilience Through Stronger, Faster, and More Inclusive Post-Disaster Reconstruction.[182] The report “explore[s] how countries can strengthen their resilience to natural shocks.”[183] The United Nations likewise endorses this ideal of resilience in its Sendai Framework for Disaster Risk Reduction, listing “Building Back Better” as a priority in responding to disasters.[184]

Beyond the result of becoming better off, the manner in which that outcome is achieved matters. To realize the ideal of resilience, the person who has suffered the setback must somehow author or at least participate in making that result happen. Suppose that Tim loses his house in a hurricane. If a charitable organization comes by and builds a better house in a safer location less vulnerable to destruction, this might accomplish the goal of coming back better. But unless Tim played some role in rebuilding, Tim has not himself realized the ideal of resilience—he has not acted resiliently—even though he is better off than before the disaster.[185]

The discussion that follows, however, sets aside cases like pandemics or natural disasters. Instead, we focus on a subset of cases involving resilience in the face of wrongdoing. These cases raise distinctive normative and evaluative issues. But the basic ideal—that the victim must in some way make it the case that they have emerged better off than before—remains the same. So when a victim is wronged, the victim’s realization of the ideal of resilience would mean that he assumes some responsibility for emerging from the setback not just in as good a position as he was before the wrongdoing, but emerging in some sense better off than he was before that wrongdoing.

Whatever the reasons we have for acting resiliently in general,[186] there are special reasons to do so after suffering a wrongdoing. Consider a speech given by Mayor Bill de Blasio in 2016, heralding the return of federal agencies as leaseholders in One World Trade Center.[187] Referring to the 9/11 terrorist attacks on the World Trade Center, de Blasio remarked:

And it’s a day to remember – and the Secretary [Jeh Johnson] always invokes that point – we won’t allow ourselves to be terrorized. Well, the terrorists attacked this nation, they attacked New York City, in particular. And I often say – it was not just a physical attack, it was an attack on our values. It was an effort to undermine our democracy. It was an effort to make us retreat from what we believe. And had we gone into a defensive crouch, had we as a nation decided we had to retreat and retreat, it would have been sadly a victory for a terrorist. But there was such resolve at the federal level, state level, city level. We’d rebuild here. We would bring back all of the strength that we associated with this location, including now the presence of our federal government. We would show that the terrorists achieved none of their aims. In fact, they strengthened our resolve. And today is one of those days that puts an exclamation point on that idea.[188]

At the same event, Jeh Johnson, then-Secretary of the Department of Homeland Security, echoed de Blasio’s sentiment, “Terrorism cannot prevail if we refuse to be terrorized,”[189] adding, “It is a sign of our determination to move forward and to come back stronger than ever before. Literally out of the ashes, we have rebuilt stronger and taller.”[190] Mayor de Blasio and Johnson were not cribbing from the same script. Officials routinely intone that the proper response to wrongdoings is to demonstrate resilience not just by restoring what has been lost but to return in some sense better than before.[191]

This rhetoric is not merely political theater. Ordinary citizens often seek to reframe tragedy in terms of personal growth. After voicing some “concerns” about returning to work after the bombing of the Murrah Federal Building in Oklahoma City, Robert Roddy interpreted returning to work as something of an act of personal growth and resilience: “I once had my concerns about returning to a federal building and I fully realize the new federal building is not indestructible. However, it is stronger and better than before—like all of us.”[192]

Resilience is also a major underlying theme in Susan Brison’s groundbreaking work, Aftermath, in which she combines philosophical reflection on “violence and its aftermath” with a first-person narrative of the process of rebuilding her sense of self after having survived a horrific rape and attempted murder.[193] Long after the physical damage healed, Brison recounts struggling with depression and post-traumatic-stress disorder.[194] Much of the book describes how suffering from trauma shattered her sense of self, tries to explain what this “shattering” meant, and recounts the process of trying to piece together a life.[195] As part of that process, Brison participated in a rape survivors’ support group, whose facilitator—herself a survivor—asserted, “You can never be the same. But you can be better.”[196]

Brison initially bristled at the claim.[197] She’d previously loved her life, and felt that much of what she’d loved—including her sense of security—had been lost forever.[198] But ultimately Brison came “to agree,” with the caveat that the “better” life would not mean “having a life that’s more coherent, in control, [or] predictable.”[199] That would be forever lost. Concretely, Brison extols the benefits of learning a new skill and in particular self-defense, one of several new “skills and insights” she gained,[200] crediting her instructor for helping her learn that she could be in some sense “tougher than ever.”[201] That Brison encountered this puzzling aspiration while in therapy should not be surprising: the ideal that each of us should try, as difficult as it might seem, to transform from agents who merely survive trauma into agents who thrive is a cornerstone doctrine of Cognitive Behavioral Therapy.[202]

2. Why Try to Realize the Ideal?—

But why should victims try to emerge better than before? Some reasons are trivial and perfectly general, in the sense that they apply even when the setback does not involve any wrongdoing. Consider the view that setbacks are sunk costs.[203] If so, those that suffer them should not rest content with getting back what has been lost because that, at best, irrationally honors what has been before.[204] Bouncing back better is simply another way of saying we ought to maximize expected utility, which is little more than a dictate of practical rationality.

The risk-reduction literature, including the Sendai Framework cited above, zeroes in on precisely this sort of practical concern.[205] And with good reason. Rebuilding after a natural disaster should not replicate unnecessary vulnerabilities.[206] But reducing resilience’s value to nothing more than means–end rationality or expected-utility theory is a mistake. Resilience after wrongdoing carries normative and evaluative significance beyond encouraging victims to behave rationally. Bouncing back stronger has, for example, expressive and narrative significance. The 9/11 attack on the World Trade Center had an obvious symbolic importance that extended beyond attempting to destroy particular buildings.[207] Recall that Mayor de Blasio construed the 9/11 attack as not only an attack on “physical” facilities but also an “attack on our values.”[208] The attacks represented a wrongdoing that sent a message. They were, according to de Blasio, “an effort to undermine our democracy” and “an effort to make us retreat from what we believe.”[209] Whether de Blasio correctly captured the specific message is contestable.[210] But the fact that the terrorist attacks have a symbolic aspect is uncontroversial.[211] And like all serious wrongdoings manifesting ill will, they surely tried to send some hostile message—e.g., “Beware: I may do what I please to you; you are weak and vulnerable.”[212] Although the precise content of the message that terrorist attacks send may be contestable, the symbolic dimension of terrorism is especially salient.

The analysis of Hampton-style retributivism given above finds a parallel in the case of resilience. Hamptonian retributivism construes punishment as defeating the wrongdoer, and in turn, mitigating the wrongdoer’s subordinating messages.[213] But defeating wrongdoers is consistent with mutually-assured destruction—i.e., other-defeat is compatible with self-defeat or pyrrhic victory. Resilience, however, complements the retributive message by communicating that not only has the wrongdoer been defeated, the victim has emerged victorious, as demonstrated by becoming better off (in some sense) than before. Mutual destruction this is not. While retributive action allows the victim to “say” something like “I [the victim] have defeated you [the wrongdoer],” resilient action complements this message by saying something like “Nor have I been defeated; I have prevailed.”

We can illustrate the resilience side of the equation by returning to Mayor de Blasio’s speech. He and others who understood the imperative to “build back better” understood that the response must be tailored in a way to deprive the terrorists of “victory.”[214] So, concluded de Blasio, rebuilding had to show that the terrorists “achieved none of their aims.”[215] This required not only “bring[ing] back all the strength that we associated with this location”—through repair—but also showing the terrorists that they succeeded only in “strengthen[ing] our resolve”[216]—i.e., that the terrorists’ aims were positively counterproductive and ultimately self-defeating.

We can put the point in another way. Repudiating the terrorists’ aims and messages required a response capable of operating effectively on a symbolic level, providing what philosopher Hilde Lindemann calls a “counterstory.”[217] “Counterstories,” according to Lindemann, “take a story that has (for the moment at least) been determined, undo it, and reconfigure it with new significance.”[218] They allow individuals to repurpose and reinterpret the past—including past wrongdoings—in a way that resists harmful stories that others try to impose on us.[219] So, if the narrative that the terrorists tried to impose on us was one of the weakness of the United States, then the most fitting response, to that end, is a counterstory according to which the terrorists ultimately rendered the targeted community stronger. But realizing that counterstory required actually making the community stronger and for the community itself to author that resurgence. The relevant counterstory, in other words, required responding resiliently.

We can generalize the point. Implicit in this idea of a “fitting response” is that by responding to wrongdoings, victims can send messages of their own by authoring their own narrative counterstories in response to wrongdoings. If wrongful conduct aims to oppress me or expresses the thought that oppressing me is appropriate, then one of the most forceful ways of responding may be to demonstrate through my flourishing not only the failure to accomplish those oppressive aims or messages but also to show that doing so is wholly counterproductive, as if to say: “you aimed to harm me or diminish me or suggested that doing so is okay, but not only did I survive, I am now thriving and better than before.”

Having the ability to tell plausible counterstories is not just a feel-good response. Counterstories impact how others perceive us, and in turn, how others may treat us. That is, counterstories have expressive potential. Among the things that the terrorists tried to accomplish was to undermine the perceived strength of the United States, with the attacks laying bare its vulnerability.[220] Countering that message is important to set a precedent and to signal to the relevant community to which one belongs that one is not as vulnerable as the terrorist would suggest.[221] Second, when these pernicious narratives take hold, individuals internalize them.[222] Doing so is demoralizing.[223] Individuals understand themselves in terms of the narratives that they adopt about their lives,[224] including how they respond to wrongdoings and what those responses reveal about themselves. Pernicious narratives about us that we accept damage our self-respect, which harms our ability to plan and execute our agency in the world—the very reason why Rawls singled out self-respect, or at least the conditions of self-respect, as “perhaps the most important primary good.”[225] After all, if one does not have self-respect—if one fails to see one’s own worth—then it will be very difficult to motivate oneself to formulate or execute the type of commitments constitutive of a person’s life.

To summarize: There are at least two reasons why bouncing back better is not merely about maximizing expected utility. Doing so empowers victims to “write” a counterstory capable of resisting the harmful narratives that wrongdoers impose on their victims. By realizing resilience, victims can create new narratives that in part signal to others how we may be treated (mitigating messages signaled by the wrongdoer) while also demonstrating to the victims themselves that they are competent agents capable of emerging stronger from setbacks (in the service of self-respect). By contrast, merely getting back to the place where the victim started before the wrongdoing occurred seems, at best, a partial and inadequate response to egregious wrongdoing involving ill will. Merely restoring the victim to the status quo seems inadequate to the task of creating a counterstory by “undo[ing] it, and reconfigur[ing] it with new significance.”[226]

C. Acting Resiliently by Acting Retributively

Although retributive justifications of punitive damages are incomplete, we are now in a position to complete them by relying on the conception of resilience highlighted above. Here is a summary of the account, called “Retributive Resilience”:

Punitive damages embody a practice that empowers victims to act punitively against their oppressors by requiring them to finance those victims’ resilience interests—i.e., their interest in bouncing back better than before the wrongdoing—if those victims so demand. Insofar as punitive damages are a form of punishment, the magnitude of relief must be proportional to the gravity of the wrongdoing. This proportionality principle acts as a constraint on the awards that may be given to the plaintiff at the wrongdoer’s expense.

Given the victim’s reasons for acting resiliently, what we might call the victim’s “resilience interests,” it is fair for a victim to demand assistance from the wrongdoer—or at least a wrongdoer who visits ill will on the victim—to satisfy her interests in returning better than before the wrongdoing, if she so demands. After all, demonstrating one’s resilience in this way can be costly. Revitalizing downtown Manhattan, if that is how “building back better” is interpreted, required billions of dollars.[227] This raises the question of who should pay. In the case of the 9/11 attacks, the United States federal government defrayed many of the costs of repair and building back “better.”[228] A victim compensation fund also distributed federal funds raised for victims of the attack.[229] But wrongdoers are more fairly required to pay, upon demand.

1. The Basic Argument for Retributive Resilience.—

An initial argument for retributive resilience appeals to basic fairness. Despite possible alternatives for satisfying the victims’ resilience interests (say, through the public fisc), nobody is more fairly required to pay compensation for the victim’s resilience interests than the wrongdoer. Nobody is more fairly required to pay for the costs of satisfying Saeed’s (the victim’s) resilience interests than Haryanto (the wrongdoer), if Saeed so demands. Nobody other than the terrorists are more fairly required to pay for the costs of rebuilding Manhattan better than before, to the extent that those funds could be confiscated. So, insofar as the victim of a wrongdoing manifesting ill will has special reasons to bounce back better—special resilience interests—wrongdoers may be fairly required to pay for those costs, if the victim so demands.[230]

Indeed, given the victim’s reasons to act retributively in response to wrongdoings, and given that retributive theory underspecifies the form of retribution, punitive damages seem well situated to empower plaintiffs who have suffered egregious wrongdoings to simultaneously act punitively and resiliently. They act punitively insofar as they extract extracompensatory “fines” from their wrongdoers only when their wrongdoings express ill will, and they act resiliently insofar as they emerge from litigation better off—at least financially and in principle—than before the wrongdoing itself. By satisfying demands for punitive damages, the legal practice unites resilience and retribution under one roof.

At this point, however, a worry emerges: that interpreting punitive damages as retributive resilience provides nothing more than a new kind of “mixed theory” of punitive damages. This suggests that the account suffers from the same defects associated with mixed theories.[231] Recall that some mixed theories of punitive damages try to combine compensatory and retributive goals.[232] The problem was one of instability: mixed accounts, even if they could explain why it was desirable to compensate plaintiffs, and even if they could explain why punishing defendants by imposing extracompensatory fines on them is likewise desirable, could not explain why transferring the extracompensatory relief to the plaintiff and from the defendant was distinctively justifiable. In other words, recall, we could simply punish the wrongdoer by relying exclusively on criminal conviction and offer a state-controlled victim compensation fund.

But the present account does not face the same problem. True, it represents a “mixed” theory in the limited sense that the goals of compensating the victim (beyond making plaintiffs “whole,” to satisfy their resilience interests) and punishing the wrongdoer are both attained. So, the account portrays punitive damages as instrumentally valuable to the extent that both those goals are worth achieving. But attaining these goals does not exhaust the value of punitive damages according to retributive resilience. The relationship of accountability, according to which the plaintiff acts retributively against the wrongdoer by making the wrongdoer help the victim to realize her resilience, is itself non-instrumentally valuable. More specifically, empowering victims to benefit from acting punitively against their oppressors is constitutively valuable, insofar as it instantiates a valuable relationship between the wrongdoer and victim.

There are several ways to explain the relevant relationship. As a form of mutual accountability, for example, retributive resilience embodies a form of relational equality.[233] Assuming that a victim’s demand that oppressors submit to retributive resilience is an appropriate one, the present account also counts as a form of corrective justice, broadly construed, which concerns empowering victims to force wrongdoers to rectify their wrongdoings.[234] The distinctive contribution of the present account is its ability to justify the transfer of extracompensatory relief from the defendant to the successful plaintiff as an essential feature of that relationship.

2. Locating Other Value in Linking Retribution and Resilience: Constructive Retributivism and Constructing Counterstories.—

Retributive resilience is non-instrumentally valuable in more easily overlooked ways. To see how, an example from Frederick Douglass’s Narrative of the Life of Frederick Douglass will help.[235] Douglass describes how, after receiving a brutal whipping of at the hands of “slave breaker” Edward Covey, Douglass fought back quite literally, drawing blood from Covey and giving him “entirely the worst end of the bargain.”[236] Douglass continued:

This battle with Mr. Covey was the turning-point in my career as a slave. It rekindled the few expiring embers of freedom, and revived within me a sense of my own manhood. It recalled the departed self-confidence, and inspired me again with a determination to be free.[237]

In fighting back against his wrongdoer successfully, Douglass reaffirmed his own worth.[238] This episode turns out to be vital to Douglass—a “turning-point”—supplying him with a renewed “self-confidence,” one that motivated him “with a determination to be free.”[239] Douglass also credits his victory with ensuring that Covey “never laid the weight of his finger upon me in anger” thereafter.[240] And ultimately, though not immediately, Douglass gained his freedom.

We can draw several lessons. Most obviously, Douglass’s victory over Covey successfully communicated to Covey that Douglass is not to be trifled with. Retaliation sometimes serves this function; it signals not only to one’s wrongdoers but to other future wrongdoers that one is not permitted to be treated as the wrongdoer has treated him. Douglass successfully deterred Covey. Less obvious, however, is that Douglass’s retribution helped him to realize resilience. Recall that realizing the ideal requires the victim to in some way participate in or author the result of becoming better off.[241] Here, Douglass authored the outcome at least in part because he personally fought back against and defeated Covey. His “betterment” consisted in Douglass’s successfully executing his plan to secure his own freedom. But this beneficial result was made possible because defeating Covey rekindled his motivation and self-respect, which allowed Douglass to see himself as having a life worth living and a life worth pursuing.[242]

The fact that Douglass’s emerging better off comes as a result of his punishment of Covey is non-instrumentally valuable. The claim here is not merely that Covey’s suffering—as the wrongdoer—is intrinsically good or deserved in the way that some retributivists insist.[243] Nor is the claim simply that retributive action corrects certain false claims about the wrongdoer’s superiority, as explained above.[244] Instead, the claim is that there is something non-instrumentally valuable about a victim’s emerging better off as a result of his acting punitively against his wrongdoer.

There are two reasons why. First, in virtue of the victim’s emerging better off as a result of the retribution, resilience improves the value of the retribution even if retributive action taken alone would be justified. That is, even if a retributivist rejects the expressivist line endorsed by Galanter and Luban,[245] the quality of the retributive justice meted out improves when the retribution is meted out in a constructive way. Retributive justice becomes transformative and beneficial, not simply inert justice that “hangs in the air,” at best. Retribution untethered from further good downstream consequences may accomplish a good (punishing those who deserve it) but no good for anyone, akin to Kant’s infamous call for killing the last prisoner on the island.[246] Linking retribution to resilience improves the quality of retributive justice itself by ensuring it makes the victim better off in some meaningful way.[247]

But this first reason does not obviously represent an improvement over the mixed theory criticized above, though retributive resilience would still have a comparative advantage by showing why extracompensatory relief is justified. After all, one could claim that making up compensatory shortfalls “improves” the quality of the retribution as well. But there is another reason why linking retributive justice and resilience is non-instrumentally valuable: as discussed before, resilience has an expressive value that complements and improves the narrative, expressive, or symbolic components of certain forms of retributive theories. Following Galanter and Luban, retributive justice aims to correct false messages about the inferiority of the wrongdoer vis-à-vis the victim.[248] Retribution here represents the wrongdoer’s defeat at the victim’s hands, aiming to correct the message about the wrongdoer’s superiority to the victim. But retribution alone is relatively silent with respect to the victim; it does not say whether the victim’s victory was pyrrhic or involved the victim’s self-defeat.

Resilience fills in the details. What resilience makes possible is a counterstory according to which the victim emerges stronger, guaranteeing that not only has the victim defeated the wrongdoer through retribution but also that the victim’s retribution has not weakened him. Indeed, the victim has emerged better off as a result of seeking justice. Insofar as counterstories “redefine [one’s] past . . . undo it, and reconfigure it with new significance,”[249] and insofar as having these counterstories is crucial for victims attempting to reconfigure their lives after wrongdoing, they are constitutively valuable. Narratives like this one—whether expressed to third parties or not—are constitutive of who we are as persons.[250]

We saw this value in Douglass’s retaliation against Covey. Douglass himself referred to his defeating Covey as a “turning” point[251] that enabled him to emerge better off and, in effect, to realize his own counterstory. By punishing Covey and ultimately escaping to freedom as a consequence of that defeat, Douglass “redefine[d] [his] past . . . und[id] it, and reconfigure[d] it with new significance.”[252] We can elaborate the particular counterstory in many ways. For example, by punishing Covey, Douglass transformed Covey from malefactor to, in effect, benefactor. Or: by punishing Covey, Douglass ultimately and effectively forced Covey to help him, thereby transforming Covey’s ill will into something ultimately valuable to Douglass. Or: Douglass not only rightfully punished Covey but ensured that the punishment was constructive, insofar as it contributed to Douglass’s emerging better off. However the story’s details are filled in, it is made possible by Douglass’s resilient behavior, and is valuable for the reasons resilience is valuable in general.[253]

To generalize, acting retributively against the wrongdoer may contribute to the victim’s resilience.[254] When this occurs, the punishment helps the victim realize a counterstory that mitigates the claim of the victim’s inferiority.[255] Victim resilience that flows from that punishment allows them to transform their wrongdoers into benefactors as part of that narrative.[256] So in addition to realizing retributive values and the values of resilience, integrating the two carries special value that is not reducible to its parts.[257]

D. How Punitive Damages Help Victims Realize Retributive Resilience

To recap: resilience requires someone who has suffered serious setbacks to, in some meaningful sense, come back better than before the setback.[258] Sometimes those setbacks come in the form of wrongdoings, which impose certain narratives on or express degrading messages about the victim. Countering those narratives or mitigating those messages involves developing a counterstory, which may require victims to defeat or punish the wrongdoer.[259] And when victims punish their wrongdoers, they sometimes are able to realize resilience, insofar as doing so allows them to emerge better than before the wrongdoing. When punishing the wrongdoer serves the victim’s resilience interests, this improves the value of the retributive justice meted out while simultaneously providing grist for a narrative according to which the wrongdoers’ defeat provided a turning point in the victim’s life.

But we are left with several problems. Notice that the causal connection between Covey’s punishment and Douglass’s betterment seems the exception rather than the rule. In other words, rarely does retributive conduct seem to guarantee the victim’s betterment. A commonplace criticism of retribution is that, even if justified, “satisfaction” earned by punishers is fleeting and ends up helping no one. If anything, the thought continues, retributive responses do more harm than good. So, although punishing Covey did in effect help Douglass once Douglass beat him up, these results are by no means guaranteed.

There is another problem. Douglass’s ability to realize the ideal of resilience also seemed precarious. His conduct, morally justified though it was, risked incredible harm to himself. The only reason legal officials did not punish Douglass for beating Covey, surmises Douglass, was that Covey wanted to protect his reputation as a slave breaker.[260] Publicizing the fact that Douglass had overcome Covey would destroy that reputation, which depended on the perception that Covey could dominate the slaves under his control.[261] Absent these unique circumstances, there was no guarantee that by punishing Covey, Douglass would realize the ideal of resilience.

Solving these problems further justifies the practice of punitive damages. To see how, imagine that, rather than defeating Covey, Douglass had managed—in the dead of night—to confiscate enough money from Covey to cure the harms that he had suffered or a fair amount to approximate those harms. What’s more, imagine that Douglass had confiscated more money than needed to cure those harms. Suppose instead that he confiscated enough money to finance his own escape, secure his freedom, and begin a new life as a free man. Enough money, in short, to emerge better than before the wrongdoing.

In some ways, this way of realizing resilience through punitive action might have been less viscerally satisfying for Douglass. But in other ways confiscating Covey’s assets would be a more reliable and perhaps even more enduring way of becoming better off. The imagined confiscation is less contingent on the downstream causal impact that defeating Covey had on Douglass’s psychological makeup. Instead of Douglass’s emerging better than before being caused by his punishment of Covey—which, again, seems highly dependent on Douglass’s motivational wiring and contingencies that led Douglass to go unpunished in return—in our hypothetical Douglass made himself better than before the moment he safely secured sufficient extracompensatory assets from Covey. At that moment, Douglass secures his betterment in virtue of punishing Covey. Douglass’s making himself better off financially—and thereby realizing the ideal of resilience—is Covey’s punishment. Realizing the ideal is, in other words, constitutive of that punishment.

How this all relates to punitive damages should be obvious. The extracompensatory gains extracted from wrongdoers at the expense of their victims realize the ideal of resilience. Acquiring extracompensatory relief itself counts as being made, in a real sense, “better off”—albeit financially—than before the wrongdoing. Although becoming better off financially does not guarantee that the victim will be better off all things considered, it does guarantee being better off financially—which is no small thing. And being able to inflict this form of punishment is valuable for all the usual reasons that realizing the ideal is valuable in the aftermath of wrongdoing. Perhaps most importantly, being made better off in this way—at the wrongdoer’s expense—empowers victims to truthfully claim a counterstory according to which they transformed their malefactors into benefactors by obligating the wrongdoers to make them better off than before. This transforms, symbolically, the wrongdoing itself into an opportunity for personal growth, reconfiguring the significance of the wrongdoing in the process. And the fact that the victim is made better off at the wrongdoer’s expense shows that the wrongdoer’s punishment is constitutive of the victim’s betterment.

In addition to realizing the ideal of resilience and, in turn, empowering victims to tell a valuable counterstory, notice something else. Punitive damages offer a kind of improvement over Douglass-type punishment of Covey. As already noted, much could have gone wrong in securing his satisfaction against Covey—even in our imagined case where Douglass confiscates Covey’s assets. Again, nothing about physically defeating Covey or confiscating his money guarantees a good outcome. Douglass could have been caught after the fact. These are practical obstacles to realizing resilience through punishment. Punitive damages help to solve them. For one thing, once the extraction of punitive damages is complete and delivered over to the plaintiff, there is a sense in which the plaintiff is concretely better off: financially. So, in winning punitive damages over Haryanto in court, Saeed’s punishment of Haryanto seems much more likely to concretely realize the ideal of making himself better off than before the wrongdoing.

We can put things more precisely. Douglass’s physical punishment of Covey causally contributed to his becoming better off—by re-activating the motivation he needed to ultimately secure his freedom—while Saeed’s punishing Haryanto was constitutive of Saeed’s becoming better off because the damages inflicted on Haryanto are the same thing as Saeed’s financial betterment. This is important. Punitive damages represent an improvement over self-help not just because of the familiar ways that juridical self-help enforcement of rights improve upon extrajudicial self-help—e.g., preventing harms associated with self-help, including violence and its escalation. Punitive damages improve on extrajudicial self-help because they better secure the underlying value of resilience-through-punishment. And this value is better secured because the relationship between the wrongdoer’s punishment and the victim’s becoming better off is transformed from being causal (and hence subject to precarity) to being constitutive (and hence guaranteed). Put differently, in winning punitive damages rather than exercising some sort of extrajudicial punishment, the wrongdoer’s punishment goes from potentially leading to or causing the victim’s betterment (e.g., in the Douglass case), to being essentially the same thing, as embodied in extracompensatory transfer from wrongdoer to victim (e.g., in Haryanto).

E. Does the Account Pass the Test?

Does this theory pass the test set forth above at the end of Part I? First, will the theory preserve the “punitive” aspect of punitive damages, and second, will it explain why punitive damages are available only in cases where the wrongdoing giving rise to the lawsuit reflected ill will? Each element is satisfied. Punitive damages aim, as the name suggests, to empower victims to inflict punishment on their wrongdoers.[262] Second, these punishments are appropriate only in cases involving ill will. Although there are a host of difficult problems that emerge in the philosophy of criminal law in explaining these relations more precisely, this essay need not wade into these waters.

The preceding two criteria are satisfied by a number of theories of punitive damage. But the present theory provides an edge over competitors with respect to the third and fourth criteria because it explains and justifies the practice of awarding at least some of the extracompensatory damages to the plaintiff in particular and explains why those same damages must come from the defendant. The present theory satisfies these criteria. When punitive damages are awarded, this means that the plaintiff has inflicted punishment on the defendant in a way that realizes the ideal of resilience for the plaintiff. Specifically, the plaintiff punishes the defendant by pressing her into service, financing the plaintiff’s financial betterment. Because securing the plaintiff’s betterment is the defendant’s punishment, this explains why the extracompensatory relief is not an unjust windfall: it represents rightful payment by the defendant to the plaintiff. So, apart from being grounded in the values associated with resilient action and retributive justice, the account of punitive damages offered here avoids the pitfalls that face the main theories of punitive damages accepted by courts today.

III. Recommendations for Reform

The theory of punitive damages presented above tries to satisfy several criteria for adequacy, criteria reflecting the inability of extant theories to account for certain structural features of bipolarity and the substantive extracompensatory aspect of punitive damages. But it is unlikely that any account of punitive damages, short of being ad hoc or trivial, will fit every facet of legal practice governing punitive damages.[263] That said, we need not necessarily bemoan a lack of fit between a theory and extant practice. Indeed, we should expect some lack of fit if a theory is going to perform any normative work.[264] So long as the lack of fit is not fundamental—such as the lack of fit between deter-and-punish instrumentalism and the fundamental structure or substance of punitive-damages practice—we should be heartened, in fact, if a theory without that defect provides useful guidance at the margins of the practice in question.

With those methodological remarks in mind, subpart III(A) takes on the challenge faced by the fact that punitive damages are widely understood to serve a deterrence function. To the extent that my proposal aims to explain and justify certain features of punitive damages, the widespread understanding of punitive damages’ role in deterring wrongdoers should be explained or explained away. This Article urges separation: officials should take care to distinguish between “genuinely” punitive damages and deterrence damages. Subpart III(B) evaluates the factors that courts allow juries to consider when calculating punitive damages. Given the new theory of punitive damages presented here, it is argued that to better facilitate the “divorce,” courts should not instruct juries to consider certain factors because they are irrelevant to the core of punitive damages. Separate instructions should be given under the label of “deterrence damages,” which seek special and/or general deterrence.

A. Distinguishing Retributive from Deterrence Damages

The present account of punitive damages rejects deterrence as a rationale. But as the Supreme Court has remarked, “Regardless of the alternative rationales over the years, the consensus today is that punitives are aimed not at compensation but principally at retribution and deterring harmful conduct.”[265] And this makes a practical difference. Following the Supreme Court’s lead, courts charged with determining whether punitive-damages awards are unconstitutionally excessive “must consider the goal of deterrence.”[266] So to the extent that punitive-damages practice is shaped and informed by the deterrence rationale, the theory that this Article presents will seem fundamentally at odds with punitive-damages practice.

Given this possibility, and as a matter of “fair labeling,”[267] this Article recommends sharply distinguishing between “retributive” damages and deterrence damages. The theory presented in this Article explains why retributive damages properly belong to the victim and why they are not windfalls after all. That is, the chief accomplishment of the present theory is to justify the practice of retributively transferring to the plaintiff a core amount of extracompensatory damages from the defendant. This justification, in turn, prima facie protects the plaintiff; these damages properly belong to them. Failing to account for this transfer is where standard defenses of punitive damages fall short.

By contrast, damages imposed on wrongdoers for the purpose of deterrence do not in principle “belong” to the victim. To the extent those damages are transferred to the plaintiff, they remain windfalls. In turn, to the extent that legal reform siphons this amount off to the public coffers, via split-recovery statute or some other means, this Article presents no objection to that practice.[268]

To facilitate this proposal for reform, jury instructions should carefully distinguish between retributive damages and deterrence damages.[269] Each form of damages should receive its own line item, and courts should give different guidance for each form of relief. More specifically, the factors that juries should consider in imposing retributive damages should differ from the factors that they should consider in imposing deterrence damages. For an illustration on how those factors might differ, consider more specific reforms below.

B. Reforming the Factors That Juries Consider

In assessing punitive damages, juries are typically tasked with both assessing whether the defendant’s misconduct warrants punitive relief, and if so, how much to assess. The theory of punitive damages offered in this Article is consistent with the first task but may not be consistent with all aspects of the second.

The whether-to-award question is comparatively straightforward, insofar as punitive damages are available only when the misconduct is egregious, infected with malice, or otherwise demonstrates ill will.[270] This task, although not always easy, differs little from the kind of inquiry juries face when determining whether a defendant committed a crime with the requisite culpable mental state.

The how-much-to-award question is trickier for juries to answer. And jury instructions provide little help. In some jurisdictions, judges provide little guidance, affording juries broad discretion.[271] Other jurisdictions simply list the policy objectives that this Article has criticized,[272] while still others provide juries with more detailed instructions. But the how-much-to-award question is constrained, in the first instance, only by open-ended lists of factors to consider, with an eye towards punishing and deterring the plaintiff.[273]

The first reforms recommended by this Article do not solve the practical problem facing juries in determining how much by way of punitive damages to award. More modestly, this Article proposes editing the list of factors that courts permit juries to consider in coming up with a number. To get a grip on some of these factors, consider the following jury instructions upheld on appeal in Texas. The instructions contained five factors for jurors to consider:

‘EXEMPLARY OR PUNITIVE DAMAGES’ means an amount that you may in your discretion award as punishment of the wrongdoer and as a warning and an example to prevent the wrongdoer and others situated like him from committing like offenses and wrongs in the future. In determining the amount you may consider [1] the nature of the wrong, [2] the character of the conduct involved, [3] the degree of culpability of the wrongdoer, [4] the situation and sensibilities of the parties concerned, and [5] the extent to which such conduct offends a public sense of justice and propriety.[274]

Beyond these factors, juries in other jurisdictions have been instructed to consider [6] the wealth of the defendant.[275] The thought motivating this factor is that punitive damages must be large enough to present a genuine sting to deep-pocketed malefactors.[276] Some jurisdictions also allow juries to consider whether the award will be enough to [7] provide an adequate incentive for plaintiffs to bring claims against egregious wrongdoers[277] and [8] deter other would-be wrongdoers.[278]

Many of these factors are compatible with the justification of punitive damages offered in this Article. Applying the theory laid out above, this Article recommends that juries impose punitive damages to make plaintiffs sufficiently better off in proportion to the gravity of the wrongdoing—i.e., the worse the wrongdoing, the more severe the damages.

But because these damages are taken from the defendant and given to the plaintiff to demonstrate the plaintiff’s prevailing over the defendant, and because remedy is a punitive one, it is constrained, like all punitive action, by a principle of proportionality.[279] This naturally explains, in calculating punitive damages, why jury’s may properly consider [1] the nature of the wrong, [2] the character of the conduct involved, [3] the degree of culpability of the wrongdoer, [4] the situation and sensibilities of the parties concerned, and [5] the extent to which such conduct offends a public sense of justice and propriety.

All of these considerations represent, more or less, specifications of factors relevant in assessing proportionality. For example, other things being equal, a highly culpable, highly harmful wrongdoing will call for more compensation than a reckless though accidental low-harm conduct. And arguably all of these factors encourage jurors to make just such an assessment. Even [6], the wealth of the defendant, may matter in calculating damages since a thousand-dollar-punitive-damages award may not qualify as genuinely punitive to a multi-billionaire because it will fail to signal the plaintiff’s victory unambiguously.

Certain other factors that juries are sometimes asked to consider, however, cannot be easily justified in light of the principle of retributive resilience. Jurors should not consider either [7] plaintiff incentives or [8] deterrence of others in calculating punitive damages. The theory presented here rejects both bounty-based rationales for punitive damages (as in [7]) and deterrence rationales (captured by [8]). Those are the wrong kinds of reasons to consider when trying to calculate the appropriate amount of damages to award plaintiffs at the defendant’s expense.

Apart from culling considerations from the list, this Article’s theory may suggest adding a surprising new one: the victim’s wealth. Take our billionaire. It is unclear what amount of money could make the multi-billionaire better off, in a significant respect, if the defendant were a day laborer, without imposing a disproportionately large harm on the laborer. So the billionaire might not need as much to secure expressive defeat against the defendant in such a case, subject to the principle of proportionality. Ability to pay might serve as both a sword and shield.

These reform proposals are tentative. They tinker at the margins. And they are merely illustrative, since jurisdictions vary in the kinds of factors that they permit jurors to consider when tabulating punitive damages. But what they do illustrate is that this Article’s theory can perform independent normative work.


This Article has taken seriously the problem of justifying punitive damages—and in particular, the alleged problem of windfalls—and attempts to justify their essentially extracompensatory nature. The normative logic of the dominant theories of punitive damages is that that would be lost if plaintiffs stopped receiving punitive damages. After all, plaintiffs do not need windfalls and have no special reason to receive them. Before we continue along in one of those directions, however, plaintiff “windfalls” should receive a fair hearing. And, within constitutional constraints, extracompensatory, punitive relief serves a normatively important but overlooked purpose. It demonstrates to victims that, through their actions, they can recover from serious wrongdoings better than before—and conscripts their wrongdoers into service towards that end. Victims can demonstrate to themselves and the broader community their own resilience. In short, by showing how punishment can relate to an ideal of resilience, this Article’s theory of retributive resilience locates an essential role for the extracompensatory dimension of punitive damages, and to that extent has considerable advantages over competitor rationales.

  1. .Haryanto v. Saeed, 860 S.W.2d 913, 917 (Tex. App.—Houston [14th Dist.] 1993, writ denied).
  2. .Id.
  3. .Id.
  4. .Id.
  5. .Id.
  6. .Id.
  7. .See id. (“Mohammed Saeed . . . brought suit against Soerono Haryanto . . . alleging false imprisonment, negligence, gross negligence, a terroristic threat, assault, reckless conduct, and intentional infliction of emotional distress.”).
  8. .Id.
  9. .Id.
  10. .860 S.W.2d 913 (Tex. App.—Houston [14th Dist.] 1993, writ denied).
  11. .At least in part. So-called “split-recovery” statutes exist in some states that effectively tax punitive damages at high rates. For a discussion of these statutes, including a description of split-recovery statutes in eight states, see Catherine Sharkey, Punitive Damages as Societal Damages, 113 Yale L.J. 347, 372–89 (2003). Split-recovery statutes are relatively recent developments. Part of the justification for these statutes flows, it seems, from the normative problem identified in this Article: that the dominant rationales for punitive damages—punishment and deterrence—cannot justify allocating any amount of punitive damages awards to plaintiffs.
  12. .Id. at 356–57, 386–89.
  13. .See infra Part I.
  14. .The most recent, salient chapter in the story of erosion is found in The Dutra Grp. v. Batterton, 139 S. Ct. 2275 (2019), which shut the door on punitive damages arising from seaworthiness claims. Id. at 2287.
  15. .Gregory C. Keating, The Theory of Enterprise Liability and Common Law Strict Liability, 54 Vand. L. Rev. 1285, 1289 (2001). Although in context Keating is discussing the debate between strict liability and negligence as the preferred conception of tort law, it is no less true for punitive damages that “[b]ecause our law is shaped by our conceptions of it, our conceptions of it—and our Restatements of it—matter.” Id.
  16. .For evidence of continued erosion, see The Dutra Grp., 139 S. Ct. 2275.
  17. .See infra Part II.
  18. .See infra Part II.
  19. .John C.P. Goldberg, Anthony J. Sebok & Benjamin C. Zipursky, Tort Law: Responsibilities and Redress 536 (4th ed. 2016).
  20. .See, e.g., Silkwood v. Kerr-McGee Corp., 464 U.S. 238, 255 (1984) (“Punitive damages have long been a part of traditional state tort law.”); Wilkes v. Wood (1763), 98 Eng. Rep. 489, 498–99 (KB) (recognizing that exemplary damages serve compensation, punishment, and deterrence functions); Theodore Sedgwick, A Treatise on the Measure of Damages 517 (5th ed. 1869) (observing that courts award exemplary damages for wrongdoers’ “gross fraud, malice, or oppression”); Clarence Morris, Punitive Damages in Tort Cases, 44 Harv. L. Rev. 1173, 1176 (1931) (describing the practice of awarding punitive damages as “centuries old”).
  21. .Ward Farnsworth & Mark F. Grady, Torts: Cases and Questions 552 (2d ed. 2009) (“Punitive damages usually are sought in cases where the defendant has committed gross misconduct, typically with a culpable state of mind.”).
  22. .See, e.g., Camillo v. Geer, 587 N.Y.S.2d 306, 309 (App. Div. 1992) (New York Law provides for “punitive damages only for exceptional misconduct which transgresses mere negligence”); James B. Sales & Kenneth B. Cole, Jr., Punitive Damages: A Relic That Has Outlived Its Origins, 37 Vand. L. Rev. 1117, 1138 (1984) (“Most courts refuse to impose punitive damages for ‘negligent’ conduct.”).
  23. .Bannar v. Miller, 701 A.2d 232, 242 (Pa. Super. Ct. 1997); see also Camillo, 587 N.Y.S.2d at 309 (explaining that punitive damages are appropriate when a “wrongdoer has acted ‘maliciously, wantonly, or with a recklessness that betokens an improper motive or vindictiveness’ . . . or has engaged in ‘outrageous or oppressive intentional misconduct’ or with ‘reckless or wanton disregard of safety or right.’” (citations omitted)).
  24. .John C.P. Goldberg & Benjamin C. Zipursky, Recognizing Wrongs 168 (2020) (“Punitive damages are by definition extra-compensatory.”).
  25. .Id.
  26. .1 Linda L. Schlueter, Punitive Damages § 6.1(D)(2) (7th ed. 2015) (“A vast majority of jurisdictions hold that as a predicate for an award of punitive damages, a plaintiff must establish actual injury and be entitled to an award of at least nominal damages.” (citations omitted)); see also id. § 6.1(D)(3) (indicating that some jurisdictions also require plaintiffs to win compensatory damages).
  27. .See, e.g., Weinberger v. Est. of Barnes, 2 N.E.3d 43, 48 (Ind. Ct. App. 2013) (“Punitive damages have long been criticized for a number of reasons, chief among them that such awards result in an unfair windfall recovery for plaintiffs . . . .”); Transportation Ins. Co. v. Moriel, 879 S.W.2d 10, 17 (Tex. 1994) (“Although punitive damages are levied for the public purpose of punishment and deterrence, the proceeds become a private windfall.” (citation omitted)); Dunn v. Hovic, 1 F.3d 1371, 1396 (3d Cir. 1993) (Weis, J., dissenting) (“[P]unitive damages are a windfall. They do not reimburse losses that plaintiffs have suffered, but provide an amount over and above that necessary for fair compensation.”); Rosener v. Sears, Roebuck & Co., 168 Cal. Rptr. 237, 242 (Cal. Ct. App. 1980) (“Such damages constitute a windfall, which, though supported by law in proper cases, creates the anomaly of excessive compensation which makes the remedy an unappealing one.” (citation omitted)); Schlueter, supra note 26, § 2.2(A)(2) (noting that some object to punitive damages “[b]ecause the plaintiff already received full compensation, [therefore] any additional award is merely an unjustified windfall for the plaintiff”); Ernest J. Weinrib, Corrective Justice 97 (2012) (“Punitive damages do not restore to plaintiffs what is rightfully theirs, but instead give them a windfall.”).
  28. .See Cieslewicz v. Mut. Serv. Cas. Ins. Co., 267 N.W. 2d 595, 599–601 (Wis. 1978) (distinguishing statutory multiple damages awards from common law punitive damages); Peissig v. Wisc. Gas Co., 456 N.W. 2d 348, 351–52 (Wis. 1990) (same); Allan Beever, The Structure of Aggravated and Exemplary Damages, 23 Oxford J.L. Studies 87, 94 (2003) (distinguishing aggravated damages from punitive damages).
  29. .See Dan D. Dobbs, Paul T. Hayden & Ellen M. Bublick, Hornbook on Torts § 34.4, at 862 (2d ed. 2000) (“Courts have traditionally agreed that punitive damages can be awarded only when the tortfeasor causes harm by conduct that is ‘outrageous’ or ‘that constitutes an extreme departure from lawful conduct’ and that is motivated by or evinces an antisocial mental state as well.” (citation omitted)).
  30. .Spokane Truck & Dray Co. v. Hoefer, 25 P. 1072, 1075 (Wash. 1891).
  31. .Rosener, 168 Cal. Rptr. at 242 (observing that punitive damages create an “anomaly of excessive compensation which makes the remedy an unappealing one”); Morris, supra note 20, at 1176 (“The doctrine [of punitive damages] is widely denounced as an ‘anomaly.’”); Theodore Olson, Some Thoughts on Punitive Damages, Manhattan Inst. (June 1, 1989), https://‌‌/html‌/some-thoughts-punitive-damages-5670.html [https://‌‌/
    Z96F-YSCZ] (“Punitive damages are an anomaly in our civil justice system because they import the public function of criminal punishment into what is otherwise thought of as a primarily private system of restorative justice.”).
  32. .John Gardner, What Is Tort Law For? Part I. The Place of Corrective Justice, 30 L. &
    1, 48 (2011).
  33. .Goldberg, supra note 19, at 537 (reporting, without endorsing, common skepticism about punitive damages).
  34. .See, e.g., Sales & Cole, supra note 22, at 1165 (“Neither right nor justice, neither fairness nor equity, and neither the vitality of the tort reparations system nor economic considerations warrant the survival of this outdated doctrine.”); John Dwight Ingram, Punitive Damages Should Be Abolished, 17 Cap. U. L. Rev. 205, 223–24 (1988) (“The doctrine of punitive damages is unwise, unjust and outmoded, and it should be abolished.”).
  35. .See Morris, supra note 20, at 1176 (repeating the argument “that punishment is not a proper object of the civil law; that it is the office of the criminal law to punish; for one who has committed a public wrong has committed an offense against the state which alone should have the power to inflict penalties . . . .”); Beever, supra note 28, at 106 (“Private law has a function—compensation. On the other hand, the role of criminal law is to punish. Hence, exemplary [punitive] damages properly belong to criminal rather than private law.”).
  36. .Ernest J. Weinrib, supra note 27, at 97 (“Punitive damages are inconsistent with corrective justice for reasons both of structure and of content.”).
  37. .Martin H. Redish & Andrew L. Mathews, Why Punitive Damages Are Unconstitutional, 53 Emory L.J. 1 (2004).
  38. .See, e.g., Vincent v. Morgan’s La. & T.R. & S.S. Co., 74 So. 541, 545 (La. 1917) (finding no support for an individual bringing an “action for damages to combine therewith a criminal prosecution for the punishment of the defendant in the interest of the community” and no “good reason why there should be”); Burt v. Advertiser Newspaper Co., 28 N.E. 1, 5 (Mass. 1891) (“Vindictive or punitive damages are never allowed in this State.”); Boyer v. Barr, 8 Neb. 68, 72–75 (1878) (listing, and agreeing with, authorities that disallow punitive damages awards); Spokane Truck & Dray Co. v. Hoefer, 25 P. 1072, 1075 (Wash. 1891) (“[W]e believe that the doctrine of punitive damages is unsound in principle, and unfair and dangerous in practice, and that the instruction of the court on the subject of punitive damages was erroneous.”). At least one of these jurisdictions has nevertheless adopted a wrongful-death statute that allows for limited punitive-damages awards. Aleo v. SLB Toys USA, Inc., 995 N.E.2d 740, 753 (Mass. 2013) (“The Massachusetts wrongful death statute permits an award of punitive damages where the decedent’s death was caused by the ‘malicious, willful, wanton or reckless conduct of the defendant or by the gross negligence of the defendant.’” (quoting Mass. Gen. Laws Ann. ch. 229, § 2 (West 2000))).
  39. .See BMW of North America, Inc. v. Gore, 517 U.S. 559, 562–63, 585–86 (1996) (finding that “a $2 million punitive damages award” was “grossly excessive” and “transcend[ed] the constitutional limit”).
  40. .See Pac. Mut. Life Ins. Co. v. Haslip, 499 U.S. 1, 18 (1991) (noting that “general concerns of reasonableness and adequate guidance from the court when the case is tried to a jury properly enter into the constitutional calculus” of a punitive-damages award).
  41. .Id.
  42. .Cf. id. at 18–23 (holding that a punitive-damages award did not violate the Due Process Clause partly because the state provided substantial procedural protections to defendants before awarding punitive damages against them).
  43. .See, e.g., Weinberger v. Est. of Barnes, 2 N.E.3d 43, 48 (Ind. Ct. App. 2013) (“A number of states have enacted statutes similar to I.C. § 35–51–3–6 [which reallocates 75% of punitive damages awards to Indiana’s treasury], which have come to be known as split-recovery or allocation statutes.”).
  44. .Exxon Shipping Co. v. Baker, 554 U.S. 471, 492 (2008) (describing these twin goals as the “consensus”); BMW of North America, Inc. v. Gore, 517 U.S. 559, 568 (1996) (“Punitive damages may properly be imposed to further a State’s legitimate interests in punishing unlawful conduct and deterring its repetition.”); Kemezy v. Peters, 79 F.3d 33, 34 (7th Cir. 1996) (Posner, C.J.) (“The standard judicial formulation of the purpose of punitive damages is that it is to punish the defendant for reprehensible conduct and to deter him and others from engaging in similar conduct.”); Moskovitz v. Mt. Sinai Med. Ctr., 635 N.E.2d 331, 344 (Ohio 1994) (“[W]e reiterate that the purpose of punitive damages is to punish and deter.”).
  45. .A classic statement belongs to Joel Feinberg, The Expressive Function of Punishment, 49 The Monist 397, 403 (1965) (“If we reserve the less dramatic term ‘resentment’ for the various vengeful attitudes, and the term ‘reprobation’ for the stern judgment of disapproval, then perhaps we can characterize condemnation (or denunciation) as a kind of fusing of resentment and reprobation.”).
  46. .Sales & Cole, supra note 22, at 1165 (“Considering the expanded and virtually unlimited access to compensatory damages, punitive damages simply provide a windfall to the plaintiff, penalize the innocent consumers or society, and unnecessarily sap the vitality of the economy upon which society is totally dependent.”).
  47. .Bass v. Chicago & Nw. Ry. Co., 42 Wis. 654, 672 (1877) (Ryan, C.J., concurring).
  48. .Id.
  49. .Murphy v. Hobbs, 5 P. 119 (Colo. 1884), superseded by statute, 1889 Colo. Sess. Laws 64–65.
  50. .Id. at 122, as recognized in Courvoisier v. Raymond, 47 P. 284, 286 (Colo. 1896) (“Before the passage of [the 1889] act the question [of the permissibility of punitive damages] was one upon which the courts disagreed, but the statute has now settled the practice in this state.”).
  51. .25 P. 1072 (Wash. 1891).
  52. .Id. at 1074.
  53. .Goldberg & Zipursky, supra note 24, at 168.
  54. .Id.
  55. .Goldberg, supra note 19, at 537.
  56. .Id.
  57. .Weinberger v. Est. of Barnes, 2 N.E.3d 43, 48 (Ind. Ct. App. 2013) (quoting Crabtree ex rel. Kemp v. Est. of Crabtree, 837 N.E.2d 135, 139 (Ind. 2005)); see also Pacific Mut. Life Ins. Co. v. Haslip, 499 U.S. 1, 19 (1991) (“It is true, of course, that under Alabama law, as under the law of most States, punitive damages are imposed for purposes of retribution and deterrence.”).
  58. .Weinberger, 2 N.E.3d at 48.
  59. .Dunn v. HOVIC, 1 F.3d 1371, 1396 (3d Cir. 1993) (Weis, J., dissenting) (“[T]here is no compelling reason why injured but fully compensated plaintiffs should receive punitive awards. The aims of retribution and deterrence can be accomplished by making punitive damages payable to the state.”); E. Jeffrey Grube, Note, Punitive Damages: A Misplaced Remedy, 66 S. Cal. L. Rev. 839, 844, 850–52 (1993) (arguing that deterrence does not justify paying punitive damages to plaintiffs and that damages should instead go to the state because society as a whole also experiences harm from the egregious behavior); Schlueter, supra note 26, § 2.2(A)(2), at 31 (articulating without endorsing the argument that “[t]hese additional damages are imposed in the form of a criminal fine which normally would be paid to the state instead of the plaintiff.”).
  60. .Grube, supra note 59, at 856. Another thought is that charities should receive the windfalls. Dede W. Welles, Charitable Punishment: A Proposal to Award Punitive Damages to Nonprofit Organizations, 9 Stan. L. & Pol. Rev. 203, 205 (1998).
  61. .One scholar who does notice the problem is Maria Guadalupe Martinez Alles. See Maria Guadalupe Martinez Alles, Moral Outrage and Betrayal Aversion: The Psychology of Punitive Damages, 11 J. Tort L. 245, 246–47 (2018) (observing that leading theories of punitive damages render the tort victim a contingent participant in the practice).
  62. .Bass v. Chicago & N.W. Ry. Co., 42 Wis. 654, 672 (1877) (Ryan, C.J., concurring).
  63. .A proponent of the deterrence rationale will quickly add that if the state rather than the plaintiff were given punitive damages awards, this would diminish the incentive for bringing claims in the first place, which in turn would undermine the deterrent effect of the awards. Two responses: First, the present point is simply that nothing about the deterrence rationale requires the plaintiff to get paid. An alternative or complementary regime that would reward plaintiffs’ attorneys rather than the plaintiffs themselves would counteract this concern, since private plaintiffs’ attorneys would still have the incentive to litigate on behalf of their clients. See Grube, supra note 59 at 856 (arguing that the state should pay attorneys to argue the case for punitive damages). Second, even if this arrangement would reduce the number of claims brought by plaintiffs, it is not clear that highly variable punitive damages awards provide much more incentive beyond far more tractable compensatory damages. In other words, compensatory damages do plenty to provide the requisite incentive, and it’s not obvious how much more incentive punitive damages awarded to the plaintiff would supply at the margins in a world where they would be awarded to attorneys rather than their clients.
  64. .This is not the last word in favor of punishment or deterrence. But I’m not sanguine about the possibility that these justifications can account for plaintiff windfalls in any satisfying way. Indeed, some writers on punitive damages are so committed to punishment or deterrence as justifications for punitive damages that they concede that certain windfalls are unjustifiable, arguing in support of legislative and judicial efforts to sharply constrain punitive damages rewards, in part by siphoning off parts of those rewards to public coffers. As it stands, this Article will say little about these efforts. For a more systematic criticism of deterrence rationales, see generally Anthony J. Sebok, Normative Theories of Punitive Damages: The Case of Deterrence, in Philosophical Foundations of the Law of Torts 313, 321–28 (John Oberdiek ed., 2014).
  65. .Bass, 42 Wis. at 672.
  66. .See, e.g., Dan Markel, Retributive Damages: A Theory of Punitive Damages as Intermediate Sanction, 94 Cornell L. Rev. 239, 307 n.224 (2009) (“Under the scheme here, the civil system would . . . mak[e] a bounty available for plaintiffs as part of the retributive damages structure.”).
  67. .Benjamin C. Zipursky, A Theory of Punitive Damages, 84 Texas L. Rev. 105, 130 (2005) (describing punitive damages as “in part like fines collected by the bounty hunters who prosecute tort cases”). Zipursky ultimately argues that punitive damages have a dual aspect, which also includes a “right to be punitive.” See generally id. at 151. The present Article complements this view by providing deeper justifications for this claim.
  68. .Marvin v. Trout, 199 U.S. 212, 225 (1905) (explaining, in describing qui tam actions, “[s]tatutes providing for actions by a common informer, who himself had no interest whatever in the controversy other than that given by statute, have been in existence for hundreds of years in England, and in this country ever since the foundation of our Government.” (emphasis added)).
  69. .There is a worry about conflicts of interest that might arise in this arrangement, though this concern kicks the can down the road by raising the question of whether the rules governing conflicts could or should not also be relaxed. See Model Rules of Pro. Conduct r. 1.8(i) (Am. Bar Ass’n 2020). I thank Scott McCown for raising this point.
  70. .For a thoughtful and complex “mixed” view, see Markel, supra note 66. Special problems exist for retributive theories. Most saliently, retributive theories bear the burden of explaining why imposing harsh treatment as a response to serious wrongdoings is justified in general. If retributivism is conflated with the view that making wrongdoers suffer is intrinsically good, and further, it is right for an authority to impose those sufferings on wrongdoers, then retributivist theories face the same double burden: arguing that harsh treatment is good and that allowing the plaintiff to impose it is somehow right or permissible. But skeptics disinclined to accept retributivism in general will likewise be disinclined to accept it in the context of civil relief. See Victor Tadros, The Ends of Harm 1, 18 (2011) (offering powerful arguments against retributivism). My own account, presented in Part II below, will be “retributive” in the broad sense that it tries to explain and justify the permissibility of empowering plaintiffs, in certain cases, to punish victims in the form of extracting extracompensatory relief without regard to deterrence. But it might not be retributive in the precise sense of seeing the suffering of wrongdoers as intrinsically good. For a recent discussion of how to distinguish between retributive and other theories of punishment, which construes retributivism more narrowly than I do, see Joshua Kleinfeld, Reconstructivism: The Place of Criminal Law in Ethical Life, 129 Harv. L. Rev. 1485, 1525–34 (2016).
  71. .A touchstone piece in this literature is A. Mitchell Polinsky & Steven Shavell, Punitive Damages: An Economic Analysis, 111 Harv. L. Rev. 869 (1998).
  72. .Mathias v. Accor Econ. Lodging, Inc., 347 F.3d 672, 677 (7th Cir. 2003) (Posner, J.).
  73. .Polinsky & Shavell, supra note 71, at 875. For criticism of deterrence theories of punitive damages that are explained in terms of incentivizing litigation or forcing disclosure of useful information, see John C.P. Goldberg & Benjamin C. Zipursky, Torts as Wrongs, 88 Texas L. Rev. 917, 961 n.220 (2003) (“On this theory, one should never see an award of punitive damages in cases of tortious conduct causing substantial harms, nor should courts permit punitive damages in cases of open and obvious misconduct. The law allows punitive awards in both kinds of cases.”).
  74. .Marc Galanter & David Luban, Poetic Justice: Punitive Damages and Legal Pluralism, 42 Am. U. L. Rev. 1393, 1432 (1993).
  75. .Id.
  76. .Id.
  77. .See id. (stating that punishment “reassert[s] the truth about the relative value of wrongdoer and victim.”).
  78. .See Jeffrie G. Murphy, Forgiveness and Resentment, in Forgiveness and Mercy 14, 25 (Jeffrie G. Murphy & Jean Hampton eds., 1988) (saying that moral injuries are “symbolic communications” in which a wrongdoer tells a victim “I count but you do not”); Jean Hampton, Forgiveness, Resentment and Hatred, in Forgiveness and Mercy, supra, at 35, 44 (quoting Murphy approvingly); Pamela Hieronymi, Articulating an Uncompromising Forgiveness, 62 Phil. & Phenomenological Rsch., at 529, 546 n.27 (same); Scott Hershovitz, Tort as a Substitute for Revenge, in Philosophical Foundations of the Law of Torts, supra note 64, at 86, 93 (same).
  79. .See infra subparts II(B)–(D).
  80. .See Kemezy v. Peters, 79 F.3d 33, 34 (7th Cir. 1996) (Posner, C.J.) (suggesting punitive damages are appropriate in such cases and would “assure full compensation without impeding socially valuable conduct”).
  81. .See id. (“Compensatory damages do not always compensate fully.”).
  82. .Schlueter, supra note 26, § 2.2(B)(1), at 35 (arguing that awarding punitive damages on this theory “makes the plaintiff whole because the plaintiff is fully compensated”).
  83. .Richard W. Wright, The Grounds and Extent of Legal Responsibility, 40 San Diego L. Rev. 1425, 1431 (2003) (“Properly understood and administered, punitive damages in tort law also compensate for discrete private injuries.”). The view under discussion above is that punitive damages are justified when necessary to ensure tortfeasors internalize the negative externalities that they impose on the society as a whole rather than on just the victim. See, e.g., Ciraolo v. City of New York, 216 F.3d 236, 245 (2d Cir. 2000) (Calabresi, J., concurring) (“A more appropriate name for extra-compensatory damages assessed in order to avoid underdeterrence might be ‘socially compensatory damages.’”). This variation on the punitive-damages-as-compensatory-damages theme renders mysterious windfall gains on the plaintiff, and for that reason, has motivated efforts to siphon off those gains to the public rather than plaintiffs themselves. See, e.g., Ciraolo, 216 F.3d at 245 (arguing that extracompensatory damages seek to make society whole); Sharkey, supra note 11, at 375–80 (detailing the development of split-recovery statutes as a means to divert punitive damages to the public).
  84. .Zipursky, supra note 67, at 136–41 (critiquing broadening compensatory theories of punitive damages).
  85. .See Peters, 79 F.3d at 34 (“Because courts insist that an award of compensatory damages have an objective basis in evidence, such awards are likely to fall short in some cases, especially when the injury is of an elusive or intangible character.”). Compare Redish & Mathews, supra note 37, at 15 (arguing that while traditional exemplary damages both punished defendants and compensated plaintiffs for intangible or immeasurable harms, subsequent iterations of punitive damages “made no pretense of compensating the victim”), with Bruce Chapman & Michael Trebilcock, Punitive Damages: Divergence in Search of a Rationale, 40 Ala. L. Rev. 741, 764–68 (1989) (conflating aggravated and punitive damages and arguing that they may serve to compensate for dignitary losses). This Article does not discuss weaker versions. One is that punitive damages cure shortfalls that come with litigation financing through, say, contingency fee arrangements. See, e.g., Schlueter, supra note 26, § 2.2(B)(1), at 33 (“Another justification for punitive damages is that it assists a successful plaintiff in recovering some of the costs involved in ‘prosecuting’ the defendant for malicious acts.”). But this version does not explain why shortfalls should be cured only when the underlying tort involved malice.
  86. .See Tenhopen v. Walker, 55 N.W. 657, 658 (Mich. 1893) (noting also that “the line between ‘actual’ and . . . ‘exemplary’ damages cannot be drawn with much nicety”).
  87. .295 N.W.2d 50 (1980).
  88. .Peisner v. Detroit Free Press, Inc., 304 N.W.2d 814, 817 (Mich. Ct. App. 1981) (using “exemplary” and “punitive” interchangeably), aff’d as modified, 364 N.W.2d 600 (Mich. 1984).
  89. .Kewin, 295 N.W.2d at 55 (Mich. 1980) (citing Tenhopen, 55 N.W. at 658; McChesney v. Wilson, 93 N.W. 627, 629 (Mich. 1903)).
  90. .Id. at 55 (internal quotation marks omitted) (quoting McFadden v. Tate, 85 N.W.2d 181, 184 (Mich. 1957)).
  91. .See, e.g., Hall v. Motorists Ins. Corp., 509 P.2d 604, 607 (Ariz. 1973) (“Punitive damages are not allowed for mere negligence.”); Lewis v. Horace Mann Ins. Co., 410 F. Supp. 2d 640, 664 (N.D. Ohio 2005) (same under Ohio law); Grogan v. Gamber, 19 Misc. 3d 798, 808 (N.Y. Sup. Ct. 2008) (same under New York law).
  92. .304 N.W.2d 814 (Mich. Ct. App. 1981), aff’d as modified, 364 N.W.2d 600 (Mich. 1984).
  93. .Id. at 817 (citing Kewin, 295 N.W.2d at 55).
  94. .See id. at 819 (Brennan, J., concurring) (“Since Michigan does not permit punitive damages, the term should be eliminated from our legal parlance. Jury instructions . . . which use the term ‘punitive damages’, while at the same time disclaiming that damages can be given to ‘punish’ the defendant, will inevitably lead to jury confusion . . . .”).
  95. .See id. (noting that the Michigan Supreme Court had reaffirmed that “only exemplary damages which are compensatory in nature are allowed”).
  96. .The Michigan Supreme Court’s jurisprudence in this area sometimes conflates aggravated damages with punitive damages. See Tenhopen v. Walker, 55 N.W. 657, 658 (Mich. 1893) (“They are properly based upon all the circumstances of the aggravation attending it. The real purpose is to compensate the plaintiff for the injuries he has suffered.” (emphasis added)).
  97. .See generally Sharkey, supra note 11, at 351–52 (proposing a new category of damages, “compensatory societal damages,” to compensate widespread harms).
  98. .See id. at 400–01 (theorizing that under the societal damages theory, the problem of a plaintiff’s windfall damages would be reduced or eliminated because the court or legislature would look to “either directly or indirectly, [] redress [the] societal harms” caused by the defendant (emphasis omitted)).
  99. .See Farnsworth & Grady, supra note 21, at 552 (distinguishing between compensatory damages and punitive damages and saying the latter are “usually sought in cases where the defendant has committed gross misconduct”).
  100. .Goldberg & Zipursky, supra note 73, at 961 (“The standard ‘under-deterrence’ explanation provided by deterrence theorists fails entirely to explain the rules for when punitive damages [are] awarded, as well as the amounts in which they are awarded. The same goes for accounts of punitive damages that cast them as compensatory of losses suffered by [third] persons . . . .”).
  101. .See Sharkey, supra note 11, at 375–77 (explaining the rationales behind, and implementations of, split-recovery statutes in several states).
  102. .For additional reasons to doubt compensatory approaches, see Zipursky, supra note 67, at 136–37 (arguing that this “broadening compensatories” approach assumes without warrant that tort remedies must be exclusively concerned with repairing harm).
  103. .There is a final problem with compensatory views that won’t become clear until this Article elaborates on its positive theory in Part II. As argued below, the idea that punitive damages are extracompensatory—the very idea that they aspire to award damages beyond that which is necessary to make a plaintiff whole—is itself important because it manifests, expresses, or realizes the ideal that a victim can make herself better in some material way than before she was victimized. Tossing punitive damages into the compensatory bin undermines this message. Perhaps this is a price that we should pay, all things considered. But we should at least recognize that construing punitive damages as just another form of compensation is not costless.
  104. .See, e.g., Morris, supra note 20, at 1174 (“So, in the liability with fault cases [including case involving punitive damages] there is an admonitory function as well as a reparative function: and the linkage of these two functions supplies a reason for taking money from the defendant as well as one for giving it to the plaintiff.”).
  105. .Georgia utilizes such a split-recovery scheme. See Ga. Code Ann. § 51-12-5.1(e)(2) (West 2010) (“Seventy-five percent of any amounts awarded under this subsection as punitive damages, less a proportionate part of the costs of litigation, including reasonable attorney’s fees, all as determined by the trial judge, shall be paid into the treasury of the state through the Office of the State Treasurer.”).
  106. .The spirit of this argument should seem familiar: it reflects versions of corrective justice arguments against instrumentalist understandings of tort law more generally. Leading formulations rejecting economic analysis and functionalism as insufficient to explain developments in tort law appear in Ernest J. Weinrib, The Idea of Private Law 46–48 (1995); Jules L. Coleman, The Practice of Principle 13–24 (2001); and Martin Stone, The Significance of Doing and Suffering, in Philosophy and the Law of Torts 131, 141–52 (Gerald J. Postema ed., 2001).
  107. .O.W. Holmes, Jr., The Common Law 2–3 (1881).
  108. .For an argument that civil recourse theory is just revenge by another name, see Emily Sherwin, Interpreting Tort Law, 39 Fla. St. U. L. Rev. 227, 232–33 (2011). For a reply, see Gabriel Seltzer Mendlow, Is Tort Law a Form of Institutionalized Revenge?, 39 Fla. St. U. L. Rev. 129 (2011).
  109. .Benjamin C. Zipursky, Civil Recourse, Not Corrective Justice, 91 Geo. L.J. 695, 736 (2003); John C.P. Goldberg, Two Conceptions of Tort Damages: Fair v. Full Compensation, 55 DePaul L. Rev. 435, 436 (2006); Goldberg & Zipursky, supra note 73, at 946–47.
  110. .Zipursky, supra note 109, at 734; John C.P. Goldberg, The Constitutional Status of Tort Law: Due Process and the Right to a Law for the Redress of Wrongs, 115 Yale L.J. 524, 606 (2005); Goldberg & Zipursky, supra note 24 , at 111–18; see Linda Radzik, Making Amends: Atonement in Morality, Law, and Politics 168–69 (2009) (“According to this Lockean model, the state becomes a stakeholder only when citizens forfeit their rights to retaliation and reparation to the state in return for increased security.”).
  111. .Goldberg & Zipursky, supra note 24, at 171.
  112. .See, e.g., id. (explaining how tort law allows plaintiffs to be punitive or vindictive towards defendants who have wronged them but noting constraints on this exercise in keeping with the “civil” nature of tort redress).
  113. .See, e.g., Hershovitz, supra note 78, at 96–97 (arguing that tort damages do “justice through the message that they send about the victim’s standing and the wrongdoer’s responsibility”); see generally Scott Hershovitz, Treating Wrongs as Wrongs: An Expressive Argument for Tort Law, 10 J. Tort L. 405 (2017) [hereinafter Hershovitz, Treating Wrongs as Wrongs] (explaining the kinds of messages that are conveyed through torts and tort litigation). Cf. Scott Hershovitz, Corrective Justice for Civil Recourse Theorists, 39 Fla. St. U. L. Rev. 107, 120–21, 125–27 (2011) [hereinafter Hershovitz, Corrective Justice for Civil Recourse Theorists] (discussing civil recourse as a method of corrective justice).
  114. .See Hershovitz, Treating Wrongs as Wrongs, supra note 113, at 408–09 (explaining the expressive function of tort law and arguing that “tort is an expressive institution—not just incidentally, but primarily”).
  115. .Hershovitz, Corrective Justice for Civil Recourse Theorists, supra note 113, at 126–27.
  116. .See, e.g., Nathan B. Oman, The Honor of Private Law, 80 Fordham L. Rev. 31, 70 (2011) (arguing that, in the context of tort law, punitive damages protect a plaintiff’s right to re-establish their honor).
  117. .See, e.g., John Gardner, Torts and Other Wrongs, 39 Fla. St. L. Rev. 43, 45–46 (2011) (arguing that civil recourse theory fails to provide an account of tort law, narrowly, suggesting instead that it offers an account of civil litigation of private law claims more generally).
  118. .Cf. Erik Encarnacion, Corrective Justice as Making Amends, 62 Buff. L. Rev. 451, 489 (2014) (“[I]f the goal of tort law, revenge, and the lex talionis are all the same, and if that aim is punishment of wrongdoers, then tort law’s default remedy should be punitive rather than compensatory damages.”).
  119. .Holmes, supra note 107, at 3.
  120. .Id. (emphasis added).
  121. .Id. For a view that negligence is a moral wrong and liability for negligence should take into account this feature, see generally Seana Valentine Shiffrin, The Moral Neglect of Negligence, in 3 Oxford Studies in Political Philosophy 197 (David Sobel et al., eds., 2017).
  122. .See Dobbs, supra note 29, § 34.4, at 862–63.
  123. .Zipursky, supra note 67, at 106.
  124. .Anthony J. Sebok, Punitive Damages: From Myth to Theory, 92 Iowa L. Rev. 957, 1031 (2007).
  125. .Campbell v. Gov’t Emps Ins. Co., 306 So. 2d 525, 531 (Fla. 1974); see also Barr v. Interbay Citizens Bank, 635 P.2d 441, 444 (Wash. 1981) (asserting that “[t]he arguments for punitive damages have generally been delineated as these four: compensation, punishment and deterrence, revenge, and promotion of justice” (emphasis added)).
  126. .One federal court writing in the 1960s acknowledged, almost sheepishly, that “[a] few isolated cases recognize the vindictive nature of punitive damages, a combination of personal and public revenge that may be justified as a means of discouraging self-help.” Nw. Nat’l. Cas. Co. v. McNulty, 307 F.2d 432, 434 n.3 (5th Cir. 1962). If we construe this comment as suggesting that outright revenge is not often identified as a justification for punitive damages, the Fifth Circuit seems correct; if we interpret the claim as suggesting that revenge is never even mentioned in the same breadth of other more common justifications for punitive damages, the Fifth Circuit’s observation has not aged well.
  127. .Harless v. First Nat’l. Bank in Fairmont, 289 S.E.2d 692, 702–03 n.17 (W. Va. 1982); see also Perry v. Melton, 299 S.E.2d 8, 13 (W. Va. 1982) (“We recognized that punitive damages also may provide a substitute for personal revenge by the wronged party.”); Davis v. Celotex Corp., 420 S.E.2d 557, 560 (W. Va. 1992) (“[T]he possibility of recovering punitive damages can . . . also serve as ‘a substitute for personal revenge by the wronged party.’” (citing Hensley v. Erie Insurance Co., 168 W. Va. 172, 183 (1981))).
  128. .Woodard v. City Stores Co., 334 A.2d 189, 191 (D.C. 1975) (“As the purpose of punitive damages is to punish the wrongdoer, deter him and others from similar conduct, and act as a substitute for personal revenge by the wronged party, they are only available against the wrongdoer himself or against one to whom his motives can be imputed.”); Winkler v. Hartford Accident and Indem. Co., 168 A.2d 418, 422 (N.J. Super. Ct. App. Div. 1961) (“Exemplary damages are allowed to punish the wrongdoer for a willful act and to vindicate the rights of a party in substitution for personal revenge, thus safeguarding the public peace.”); Sec. Aluminum Window Mfg. Corp. v. Lehman Assocs., Inc., 260 A.2d 248, 251 (N.J. Super. Ct. App. Div. 1970) (same (quoting Winkler, 168 A.2d at 422)); Roginsky v. Richardson-Merrell, Inc., 378 F.2d 832, 838 (2d Cir. 1967) (discussing “a variety of rationales: redressing affronts to personal feelings not susceptible of measurement, . . . financing the cost of deserving litigation where only small compensatory damages can be expected, diverting the plaintiff’s desire for revenge into peaceful channels, and serving as punishment for and deterrence from socially disapproved conduct.” (citation omitted)).
  129. .Murphy, supra note 78, at 15.
  130. .Mark A. Geistfeld, Hidden in Plain Sight: The Normative Source of Modern Tort Law, 91 N.Y.U. L. Rev. 1517, 1531 (2016) (describing but not endorsing this “widespread agreement”).
  131. .Cf. John C.P. Goldberg & Benjamin C. Zipursky, Civil Recourse Revisited, 39 Fla. St. U. L. Rev. 341, 345 (2011) (speaking of tort law in general rather than punitive damages specifically).
  132. .Winkler, 168 A.2d at 422.
  133. .Roginsky, 378 F.2d at 838.
  134. .Holmes, supra note 107, at 2.
  135. .See Hershovitz, supra note 78, at 87–89 (arguing that tort law can, at times, act as a substitute for revenge, albeit an inadequate one).
  136. .Cf. Sherwin, supra note 108, at 232–33 (claiming that if “the state provides injury victims with a controlled form of revenge . . . . [A] defense of tort law as a law of recourse for wrongs needs to confront head-on this vengeful aspect of victims’ demands for recourse”).
  137. .See Hershovitz, supra note 78, at 94–95 (arguing that properly calibrated revenge can be “a kind of justice, and indeed corrective justice”).
  138. .See Encarnacion, supra note 118, at 489 (noting that Hershovitz’s account of tort law, like revenge, aims to afford victims the opportunity to “get even”).
  139. .See, e.g., Geistfeld, supra note 130, at 1531 (referencing Holmes to suggest that because early forms of tort liability were based on revenge, they were inadequate to address “the fundamentally different problem of determining responsibility for unintentional, noncriminal harms”).
  140. .See Jeffrie G. Murphy, Getting Even: Forgiveness and Its Limits 17 (2003) (“Vengeance is the infliction of suffering on a person in order to satisfy vindictive emotions,” which are in turn “harsh negative passions—anger, resentment, even hatred—often felt by victims toward those who have wronged them.”).
  141. .See Hershovitz, supra note 78, at 93 (“But our question is not whether [plaintiff] would have actually taken revenge had he not been able to file a tort suit. It is what he would have gotten from revenge had he taken it.”).
  142. .See supra subpart I(A).
  143. .A possible vengeance-based view holds that windfalls serve an expressive role subordinate to the aim of securing revenge. Revenge is, in a sense, “sweeter” if the wrongdoer’s suffering counts as the victim’s gain. Structurally, this view points in the right direction, as I will argue below, because it answers the bipolarity problem that I’ve used as a cudgel against the dominant theories. But I don’t think that this view works for the revenge-substitute version of the unorthodox accounts; it must embrace the view that punitive damage is a form of revenge, which seems for that reason morally unattractive. I thank Jonathan Gingerich for bringing this possibility to my attention.
  144. .See supra subpart II(A).
  145. .See, e.g., Zipursky, supra note 109, at 699–704 (describing Weinrib and Coleman’s “bipolarity” argument).
  146. .Labelling the following views “retributivist” may be misleading according to some ways of carving up the taxonomic space. Although I’m not inclined to agree (I take certain expressivist accounts to explain, even if only partly, the intrinsic value of the good that punishment seeks to secure), I will not elaborate here. Those worried about taxonomy can feel free to construe the accounts as “expressivist” instead. For a recent, subtle discussion of these taxonomic questions, see Kleinfeld, supra note 70, at 1525–34.
  147. .Galanter & Luban, supra note 74, at 1432.
  148. .This is true in virtue of a more general idea. See Hieronymi, supra note 78, at 546 (“An action carries meaning by revealing the evaluations of its author.”).
  149. .Murphy, supra note 78, at 25. For another account of the “moral injury” that derives from wrongdoings, see Jean Hampton, Correcting Harms Versus Righting Wrongs: The Goal of Retribution, 39 UCLA L. Rev. 1659 (1992).
  150. .Hampton, supra note 78, at 44.
  151. .Jean Hampton, The Retributive Idea, in Forgiveness and Mercy, supra note 78,
    at 111, 125.
  152. .Haryanto v. Saeed, 860 S.W.2d 913 (Tex. App. —Houston [14th Dist.] 1993, writ denied).
  153. .Id. at 917.
  154. .Id.
  155. .Id.
  156. .Id.
  157. .Id.
  158. .See also Hieronymi, supra note 78, at 546 (observing that “without apology, atonement, retribution, punishment, restitution, condemnation” or some kind of other response that “recognize[s]” the wrong as such, a “past wrong . . . says, in effect, that you can be treated this way, and that such treatment is acceptable.”).
  159. .Hampton, supra note 151, at 125.
  160. .Id.
  161. .Id.
  162. .As with any expressive account of punishment, there is room to argue that punishment is unnecessary because we can communicate condemnation or equality through other means. But actions speak louder and more decisively than words.
  163. .Galanter & Luban, supra note 74, at 1433.
  164. .Id.
  165. .Juries have broad discretion to award pain and suffering damages in many jurisdictions, allowing these difficult-to-calculate awards to serve as proxies for punitive damages. This bolstered the Washington Supreme Court’s complaint that punitive damages seemed unnecessary given that compensatory damages included claims for hurt feelings that “frequently border on the imaginary,” and therefore leave no further need for plaintiffs to “exact[] a pound of flesh” through punitive damages. Spokane Truck & Dray Co. v. Hoefer, 25 P. 1072, 1074 (Wash. 1891).
  166. .Kleinfeld, supra note 70, at 1513 (citation and internal quotations omitted).
  167. .For an analogous argument that the unavailability of punitive damages in cases involving intentional breaches of contract risks communicating the moral triviality of broken promises, see Seana Valentine Shiffrin, The Divergence of Contract and Promise, 120 Harv. L. Rev. 708, 710, 726 (2007).
  168. .For similar “fair labelling” concerns, see generally James Chalmers & Fiona Leverick, Fair Labelling in Criminal Law, 71 Mod. L. Rev. 217 (2008).
  169. .See Galanter & Luban, supra note 74, at 1432–33.
  170. .Jeremy Waldron, Moments of Carelessness and Massive Loss, in The Philosophical Foundations of Tort Law 387 (David G. Owen ed., 1995).
  171. .But see generally Shiffrin, supra note 121 (arguing that negligent behavior is much more seriously wrongful than is typically recognized). Much of Shiffrin’s discussion pertains, however, to conduct that would qualify as “ill will,” insofar as it seems more aptly characterized as reckless or grossly negligent behavior.
  172. .In principle, behavior that expresses a judgment about the victim’s inferiority need not involve a subjective mental state that matches or contains that same judgment. One might “innocently” express that judgment without truly harboring any ill will towards a person and vice versa. We can imagine cases of serious moral wrongdoing that are motivated by ill will as indicated above but which fail to objectively manifest or express that same negative value judgment about inferiority. But for most practical purposes—and for most legal purposes—subjective ill will and objective expressions of inferiority walk hand in hand. After all, it is very difficult to identify a person’s mental states without access to externally manifested conduct or behavior. So, without any external expressions of ill will—will that somehow conveys the message of the victim’s inferiority—it will, in practice, be difficult to identify the relevant subjective ill will. Likewise, externally manifested expressions of the victim’s inferiority will provide very good evidence of subjectively held ill will absent some excuse or justification. For purposes of discussion, this Article assumes that subjective ill will and external expressions of victim inferiority will both be present whenever serious wrongdoings occur.
  173. .Beyond Hampton, Murphy, Galanter, and Luban, see Hieronymi, supra note 78, at 546 n.27 (quoting Murphy approvingly) and Hershovitz, supra note 78, at 93 (same). Joshua Kleinfeld folds a slightly modified version of Hampton and Murphy’s expressive account into a more comprehensive “reconstructionist” theory of criminal law. See Kleinfeld, supra note 70,
    at 1507–09.
  174. .See supra Part I.
  175. .Consider a sampling of recent “self-help” books published about the topic in the past few years alone. See generally Ross Edgley, The Art of Resilience: Strategies for an Unbreakable Mind and Body (2020) (describing the importance of resilience in mentally and physically accomplishing a swim around Great Britain); Tod Bolsinger, Tempered Resilience: How Leaders Are Formed in the Crucible of Change (2020) (describing the leadership transformation of a tempered resilient leader); Gail Gazelle, Everyday Resilience: A Practical Guide to Build Inner Strength and Weather Life’s Challenges (2020) (describing strategies to connect with and grow personal resilience); Rick Hanson & Forrest Hanson, Resilient: How to Grow an Unshakable Core of Calm, Strength, and Happiness (2018) (describing techniques to become a resilient individual). For another recent discussion linking resilience to stoicism, see William B. Irvine, The Stoic Challenge: A Philosopher’s Guide to Becoming Tougher, Calmer, and More Resilient 41–66 (2019) (describing examples of challenges that resilient people have overcome and describing the resilience continuum).
  176. .Steven M. Southwick & Dennis S. Charney, Resilience: The Science of Mastering Life’s Greatest Challenges 8 (2d ed. 2018).
  177. .See, e.g., Building Your Resilience, Am. Psych. Ass’n (2012), https://‌‌/topics‌/resilience [https://‌‌/8HCW-K9VW] (pointing out that “resilience not only helps you get through difficult circumstances, it also empowers you to grow and even improve your life along the way” and emphasizing an ideal according to which a person “bounc[es] back” from challenges with “profound personal growth”).
  178. .Similar usages are not uncommon. See, e.g., Eric Greiteins, Resilience: Hard-Won Wisdom for Living a Better Life 3 (2015) (“Resilience is the virtue that enables people to move through hardship and become better.” (emphasis added)); Southwick & Charney, supra note 176, at 8 (“In some definitions, resilience also entails the ability to grow from adverse events and find meaning in them.”). Nassim Nicholas Taleb coined the term “antifragile,” which comes close to the idea. Nassim Nicholas Taleb, Antifragile: Things That Gain from Disorder 3 (2012) (“Some things benefit from shocks; they thrive and grow when exposed to volatility, randomness, disorder, and stressors and love adventure, risk, and uncertainty . . . Let us call [this property] antifragile.”). But, again, this Article avoids speaking in terms of “virtues,” “abilities,” or properties of things because these terms focus attention on whether a given subject presently has or lacks a stable trait. According to the present usage, a person can manifest resilience through her action or behavior on a particular occasion even if she has not fully developed the trait properly so called. Someone who lacks the stable disposition of being resilient may, in other words, act with resilience on particular occasions.
  179. .President Biden set forth his plans in a website named Statement by President-Elect Joe Biden on the November Jobs Report and Continuing Economic Crisis, Build Back Better (Dec. 4, 2020), https://‌‌/web‌/20201204183000‌/https://‌‌/press-releases‌/statement-by-president-elect-joe-biden-on-the-november-jobs-report-and-continuing-economic-crisis‌/ [https://‌‌/3AG7-5EEP].
  180. .Asma Khalid & Barbara Sprunt, Biden Counters Trump’s ‘America First’ With ‘Build Back Better’ Economic Plan, Nat’l Pub. Radio (July 9, 2020), https://‌‌/2020‌/07‌/09‌/
    889347429‌/biden-counters-trumps-america-first-with-build-back-better-economi c-plan [https://‌‌/7AUC-34XB].
  181. .Joseph R. Biden (@JoeBiden), Twitter (July 24, 2020, 7:15 PM), https://‌‌/JoeBiden‌/status‌/1286817188288970758?s=20 [https://‌‌/9RQP-3LAJ]; see also Joseph R. Biden (@JoeBiden), Twitter (July 14, 2020, 8:42 AM), https://‌‌/JoeBiden‌/status‌/1283034009895751680?s=20 [https://‌‌/Z6JX-KMQ4] (“[W]e’ll emerge stronger than before.”); Joseph R. Biden (@JoeBiden), Twitter (May 19, 2020, 8:00 PM), https://‌‌/JoeBiden‌/status‌/1262910915768315904?s=20 [https://‌‌/F2SG-NG63] (“The United States of America has always emerged stronger from every trial and tribulation—and we will again.”).
  182. .Stephane Hallegatte, Jun Rentschler & Brian Walsh, Glob. Facility for Disaster Reduction and Recovery, Building Back Better: Achieving Resilience Through Stronger, Faster, and More Inclusive Post-Disaster Reconstruction 1 (2018), https://‌‌/sites‌/default‌/files‌/publication‌/Building%20Back%20Better.pdf [https://‌‌/64ZY-EUL6]. The GFDRR is a global partnership administered by the World Bank. Glob. Facility for Disaster Reduction and Recovery, Partnership Charter 1 (2007), https://‌‌/sites‌/default‌/files‌/publication‌/partnership-charter.pdf [https://‌‌/HS55-DP3A].
  183. .Hallegatte, supra note 182, at ii.
  184. .Third U.N. World Conference, Sendai Framework for Disaster Risk Reduction 2015–2030, at 21 (Mar. 18, 2015) [hereinafter Sendai Framework], https://‌‌/files‌/43291‌_sendaiframeworkfordrren.pdf [https://‌‌/8L2W-5PXA].
  185. .How extensive Tim’s role must be to count as having participated will depend on the circumstances as well as Tim’s own interpretation of events. But nothing turns on this discussion for present purposes. Ultimately, initiating a lawsuit in which the plaintiff seeks punitive damages will count as sufficient participation in any event.
  186. .For more on this point, see infra section II(B)(2).
  187. .Thomas MacMillan, Officials Hail Return of Federal Workers to World Trade Center, Wall St. J. (Sept. 9, 2016, 7:04 PM), https://‌‌/articles‌/officials-hail-return-of-federal-workers-to-world-trade-center-1473462107 [https://‌‌/42YL-L69W].
  188. .Mayor Bill de Blasio, Transcript: Mayor de Blasio Delivers Remarks Commemorating The Federal Government’s Return to One World Trade Center, Off. Website N.Y.C (Sept. 9, 2016), https://‌‌/office-of-the-mayor‌/news‌/722-16‌/transcript-mayor-de-blasio-delivers-remarks-commemorating-federal-government-s-return-one [https://‌‌/NN65-AAHD].
  189. .MacMillan, supra note 187.
  190. .September 11, 2016: Remembering and Looking Forward, U.S. Dep’t Homeland Sec. (Sept. 13, 2016, 5:15 PM), https://‌‌/blog‌/2016‌/09‌/13‌/september-11-2016-remembering-and-looking-forward [https://‌‌/B9ZS-K2U9] (emphasis added).
  191. .Politicians voiced similar sentiments after the Oklahoma City terrorist bombings. Here is Representative Markwayne Mullin: “It is because of the strength of our communities and the help from Americans across this great nation that our state rebounded stronger than before.” Mike Sanders, Delegation Remembers Bombing, Woodward News Op., Apr. 19, 2005, at 4A (emphasis added); see also Chris Casteel, House Resolution Observes 10th Anniversary of Attack, Daily Oklahoman, Apr. 21, 2005, at 8A (“Ten years after the bombing, we Oklahomans are stronger than ever.” (statement of Rep. Frank Lucas)); Sam Anderson, Boom Town: The Fantastical Saga of Oklahoma City, Its Chaotic Founding, Its Apocalyptic Weather, Its Purloined Basketball Team, and the Dream of Becoming a World-class Metropolis 359 (2018) (citing Hon. Steven Taylor as stating that after the bombing, the community could make “two choices. One was to go down and stay down. The other was to get up and prove that civics and citizenship and the rule of law will overcome and prevail. We are so strong and resilient that we are actually going to be better. That’s what we did.”).
  192. .Staff and Wire Reports, HUD Moves into New Building; Some Employees Like the Location Close to the Site of the Bombing, Daily Oklahoman, March 16, 2004, at 8A.
  193. .Susan Brison, Aftermath: Violence and the Remaking of a Self, at xii (2002).
  194. .Id. at 15.
  195. .Id. at 15–16.
  196. .Id. at 20 (emphasis added).
  197. .Id.
  198. .Id.
  199. .Id. at 115.
  200. .Id. at 20.
  201. .Id. at 14.
  202. .Michael Neenan, Developing Resilience: A Cognitive-Behavioural Approach 15 (2d ed. 2018).
  203. .Richard Posner, The Economic Analysis of Law 7 (Vicki Been et al. eds., 7th ed. 2007) (“‘Sunk’ (incurred) costs do not affect a rational actor’s decisions. . . . Rational people base their decisions on expectations of the future rather than on regrets about the past. They treat bygones as bygones.”). For a fascinating discussion of the economic conception of sunk costs, and a defense of honoring them, see Ryan Doody, The Sunk Cost “Fallacy” Is Not a Fallacy, 6 Ergo 1153 (2020).
  204. .Posner, supra note 203, at 7.
  205. .Sendai Framework, supra note 184, at 21.
  206. .See Sendai Framework, supra note 184 (stating that disasters create an opportunity to “Build Back Better”).
  207. .Jonathan Matusitz, Symbolism in Terrorism: Motivation, Communication, and Behavior 52 (2015); see also id. at 53 (collecting sources describing the 9‌/11 attacks in terms of their symbolic significance).
  208. .De Blasio, supra note 188; see also Matusitz, supra note 207, at 53 (“Osama bin Laden’s emphasis is not only on a triumph over Western imperialism or Zionism, but also on the symbolic obliteration of the institutional values that make Western society possible.”).
  209. .De Blasio, supra note 188.
  210. .More specific understandings of the symbolic significance of the attacks have been attributed, including implicit claims about Western hegemony. See, e.g., Matusitz, supra note 207, at 52–53 (collecting sources).
  211. .Id.
  212. .The discussion infra section II(B)(1) supports this point.
  213. .Supra subpart II(A).
  214. .De Blasio, supra note 188.
  215. .Id.
  216. .Id.
  217. .Hilde Lindemann Nelson, Damaged Identities, Narrative Repair 18 (2001).
  218. .Id.
  219. .See, e.g., De Blasio, supra note 188 (describing how Americans have reinterpreted 9‌/11).
  220. .Matusitz, supra note 207, at 52–53 (stating that “U.S. global hegemony” was the target of 9‌/11).
  221. .Similar thoughts may be couched in terms of dignity. Scott Hershovitz—who cites and accepts the above-referenced claims of Murphy and Hieronymi—argues that having access to legal redress is important for reasons of self-respect and dignity. Hershovitz, Treating Wrongs as Wrongs, supra note 113, at 411–12. At stake is nothing less than our dignity as rights holders, persons who can place demands on others and rightfully expect those demands to be deferred to. See id. at 417–18 (stating that human dignity “places a constraint on others, who must give us the space to move through the world”). Having the ability to mitigate the messages sent by wrongdoings is important primarily because it protects our dignity interests qua bearers of rights. See also Erik Encarnacion, Boilerplate Indignity, 94 Ind. L.J. 1305, 1325 (2019) (arguing that accountability waivers threaten dignity interests by denying individuals the legal power to sue in court). But this Article prioritizes a different audience than Hershovitz: the victim’s relation to herself. The account of self-respect and dignity that Hershovitz presupposes is one according to which dignity is a function of one’s social standing in a community, emphasizing a primarily social threat posed to dignity. Hershovitz, Treating Wrongs as Wrongs, supra note 113, at 417 n.43 (“To say that dignity does not demand social stratification is not to say that it does not depend on a notion of rank. I agree with Jeremy Waldron that it does.” (citing Jeremy Waldron, Dignity, Rank, and Rights (Meir Dan-Cohen ed., 2012))). This Article has no quarrels with this though it emphasizes the self as the primary audience of one’s concern. Frederick Douglass did not want to advertise his triumph over Edward Covey—at least not immediately—yet his victory proved to himself his own worthiness.
  222. .Lindemann Nelson, supra note 217, at 21.
  223. .Indeed, this is precisely the aim of “Clauswitzian terror,” defined as “terrorism used as a strategic weapon to wreak psychologically debilitating effects on mainstream citizens.” Matusitz, supra note 207, at 323.
  224. .Anne E. Ralph, Narrative-Erasing Procedure, 18 Nev. L.J. 573, 578 (2018) (citing Jerome Bruner, Life as Narrative, 54 Soc. Res. 11, 11 (1987)).
  225. .John Rawls, A Theory of Justice 348, 386 (Harv. Univ. Press rev. ed. 1999).
  226. .Lindemann Nelson, supra note 217, at 18. For another attempt to invoke the concept of narrative to interpret the harms giving rise to hedonic damages, see Sean Hannon Williams, Self-Altering Injury: The Hidden Harms of Hedonic Adaptation, 96 Cornell L. Rev. 535, 568–79 (2011).
  227. .At least if we take the costs of building One World Trade Center as a benchmark.
    Eliot Brown, Tower Rises, and so Does Its Price Tag, Wall St. J. (Jan. 30, 2012), https://‌‌/articles‌/SB10001424052970203920204577191371172049652 [https://‌‌/
  228. .U.S. Gen. Acct. Off., GAO-04-72, September 11: Overview of Federal Disaster Assistance to the New York City Area, at ii–iii (2003) (listing various budget line items for restoration and “enhancement”), https://‌‌/assets‌/gao-04-72.pdf [https://‌‌/
  229. .September 11th Victim Compensation Fund, https://‌‌/ [https://‌‌/845D-4DUV] .
  230. . A similar argument that tries to justify tort law’s remedial structure in terms of the question of who most fairly bears certain costs has been endorsed elsewhere. For a sophisticated version that invokes a conception of outcome responsibility for losses, see Stephen R. Perry, The Moral Foundations of Tort Law, 77 Iowa L. Rev. 449, 513 (1992). His account has been criticized for, among other things, failing to capture certain features of tort law. John C.P. Goldberg & Benjamin C. Zipursky, Tort Law and Responsibility, in Philosophical Foundations of the Law of Torts 17, 29–35 (John Oberdiek ed., 2014). But the idea that fairness matters in allocating costs as between wrongdoer and victim, as a normative resource, survives this criticism unless we further maintain that fairness-plus-responsibility suffices to explain and justify the moral foundations of tort law as a whole. That idea is not presupposed above.
  231. .See supra subpart I(B).
  232. .See supra subpart I(B).
  233. .Jason M. Solomon, Equal Accountability Through Tort Law, 103 Nw. U. L. Rev. 1765, 1801 (2009).
  234. .For a useful discussion of the distinction between broad and narrow conceptions of corrective justice, see Gregory C. Keating, The Priority of Respect over Repair, 18 Legal Theory 293, 316–17 (2012).
  235. .Frederick Douglass, Narrative of the Life of Frederick Douglass, an American Slave, Written by Himself 50 (William L. Andrews & William S. McFeely, eds., 1997).
  236. .Id.
  237. .Id.
  238. .This is an important point often overlooked in expressive accounts, which typically focus on what message is sent to a broader community. For the idea that one’s self is an important audience for one’s behavior, see generally Stephen Bero, The Audience in Shame, 177 Phil. Studies 1283 (2020).
  239. .Douglass, supra note 235, at 50.
  240. .Id.
  241. .See supra section II(A)(1).
  242. .Again, this is why Rawls considered self-respect to be perhaps the most important “primary good.” Rawls, supra note 225, at 348, 386.
  243. .A classic account of retributivism in the criminal law can be found in Michael Moore, Placing Blame (2010). For a compelling case against retributivist justifications of punishment, see Tadros, supra note 70. For a retributivist sur-reply, see Mitchell N. Berman, Rehabilitating Retributivism, 32 L. & Phil. 83 (2013).
  244. .See supra section II(B)(1).
  245. .See supra section II(B)(1).
  246. .Immanuel Kant, The Metaphysics of Morals 116 (Lara Denis ed., Mary Gregor trans., Cambridge Univ. Press rev. ed. 2017). It has been argued that punishing the guilty is good for the guilty insofar as it constitutes a form of recognition of a guilty person’s genuine moral agency. For a helpful discussion arguing that a moral accountability theory is more consistent with Kant than a reciprocity theory, see Margaret M. Falls, Retribution, Reciprocity, and Respect for Persons, 6 L. & Phil. 25 (1987). But at least in Kant’s case it seems a strange form of “good,” which results in an inmate’s death.
  247. .The idea, that the best goods are those that are not just good in themselves but also have good results, has an old pedigree. Whether justice is a good of this type has preoccupied philosophers for just as long. See Plato, Republic 33–35 (G.M.A. Grube tran., C.D.C. Reeve rev., 1992).
  248. .See subpart II(A).
  249. .Lindemann Nelson, supra note 217, at 18.
  250. .Id.; Mayra Schechtman, The Constitution of Selves 93–94 (1996) (arguing that personal identity is a function of one’s autobiographical narrative); Jerome Bruner, Making Stories: Law, Literature, and Life 63–64, 85–87 (2002) (same).
  251. .Douglass, supra note 235, at 50.
  252. .See Lindemann Nelson, supra note 217, at 18.
  253. .See supra subpart II(A)(2).
  254. .See Douglass, supra note 235, at 50 (recounting how a physical fight he had with Covey was a “glorious resurrection, from the tomb of slavery, to the heaven of freedom” that revived him with a sense of his own manhood).
  255. .See supra section II(B)(1).
  256. .See supra section II(B)(2).
  257. .See supra subparts II(A)–(B).
  258. .See supra subpart II(B).
  259. .See supra section II(B)(2).
  260. .Douglass, supra note 235, at 51.
  261. .Id.
  262. .Galanter & Luban, supra note 74, at 1432.
  263. .For one thing, some courts have fully embraced the punish-and-deter orthodoxy and shape their doctrines to reflect it. Indeed, the worry motivating this Article is that this same orthodoxy, taken to its logical conclusion, is consistent with eliminating the practice of transferring punitive damages awards to plaintiffs. If even the orthodoxy suffers from a problem of “fit,” this suggests that likely any attempt to bring a unifying theory of punitive damages will suffer too, to the extent existing doctrine reflects that orthodoxy.
  264. .For a similar methodological stance that criticizes the “rubber stamp” approach to theorizing areas of law, see Erik Encarnacion, Contract as Commodified Promise, 71 Vand. L. Rev. 61, 88 (2018).
  265. .Exxon Shipping Co. v. Baker, 554 U.S. 471, 492 (2008) (emphasis added).
  266. .Deters v. Equifax Credit Info. Servs., Inc., 202 F.3d 1262, 1272 (10th Cir. 2000); see also Tillett v. Lippert, 909 P.2d 1158, 1162 (Mont. 1996) (noting that punitive damages should serve to punish and set an example for the purpose of deterrence).
  267. . Chalmers & Leverick, supra note 168, at 219 (identifying “fair labelling” as a normative principle governing criminal liability).
  268. .The reform of “divorce” also coheres with attempts to respond to the problem of punitive redundancy, which arises when a defendant may be “punished” multiple times for the same transgression, given that several plaintiffs may prevail in successive rounds of litigation for the same underlying conduct. Each plaintiff, in my view, would be entitled to retributive damages to satisfy their resilience interests. This amount would not yield redundancy, given that each plaintiff would obtain what she is uniquely entitled to. Deterrence damages, however, should presumably be treated differently to avoid redundancies. For an illuminating discussion, see generally Bert I. Huang, Surprisingly Punitive Damages, 100 Va. L. Rev. 1027 (2014).
  269. .Dan Markel has similarly used the label “retributive damages,” but in so doing he refers to something different: an “intermediate” sanction, which places the plaintiff in the role of a private attorney general, and which essentially construes punitive damages as essentially a bounty for securing the public’s interests in meting out retribution. Markel, supra note 66, at 307 n.224.
  270. .See Dobbs, supra note 29, § 34.4, at 862 (identifying a defendant’s “malice, ill-will, intent to injure,” and his “oppressive, evil, [or] wicked” nature as factors considered in permitting punitive damages).
  271. .See, e.g., Ondrisek v. Hoffman, 698 F.3d 1020, 1028 (8th Cir. 2012) (citing BMW of N. Am., Inc. v. Gore, 517 U.S. 559, 568 (1996)); Berrier v. Thrift, 420 S.E.2d 206, 208 (N.C. Ct. App. 1992) (quoting the trial court’s jury instructions, which state: “[u]pon a showing of gross, willful or wanton negligence, whether to award punitive damages, and within reasonable limits, the amount to be awarded are matters within the sound discretion of the jury.”).
  272. .See, e.g., Smith v. Litten, 507 S.E.2d 77, 79 (Va. 1998) (quoting the trial court’s jury instructions, which state that if the jury finds the plaintiff is “entitled to be compensated for his damages” and that the defendant “acted under circumstances amounting to a willful and wanton disregard of [the plaintiff’s] rights,” the jury may “award punitive damages to punish [the defendant] and to serve as an example to prevent others from acting in a similar way.”).
  273. .See, e.g., Ill. Pattern Jury Instructions-Civil, 35.00, https://‌‌/Resources‌/51790bb6-f04a-4956-a47d-384c30af4b9e‌/35.00.pdf [https://‌‌/7HC6-D538] (describing factors including how reprehensible the conduct was, what actual and potential harm did the conduct cause the plaintiff, and what amount of money is necessary to punish and deter future wrongful conduct).
  274. .Murphy v. Waldrip, 692 S.W.2d 584, 591 (Tex. App.—Fort Worth 1985), writ ref’d n.r.e. (Nov. 13, 1985) (quoting the definition submitted by the trial court).
  275. .See, e.g., Horowitz v. Schneider Nat., Inc., 708 F. Supp. 1573, 1578 (D. Wyo. 1989) (“In considering the amount of punitive damages, the jury must consider the nature of the tort, the amount of the actual damages, and the wealth of the defendant.”).
  276. .Sometimes this point is made by reference to the goal of deterrence. Neal v. Farmers Ins. Exch., 582 P.2d 980, 990 (Cal. 1978) (“Also to be considered is the wealth of the particular defendant; obviously, the function of deterrence will not be served if the wealth of the defendant allows him to absorb the award with little or no discomfort.”). But the same point applies with respect to the retributive function of punitive damages, since a very small “punishment” for egregious conduct would qualify as a punishment in name only.
  277. . Mathias v. Accor Econ. Lodging, Inc., 347 F.3d 672, 677 (7th Cir. 2003).
  278. .Olson v. Walker, 781 P.2d 1015, 1018 (Ariz. Ct. App. 1989) (“Punitive damages are awarded primarily to punish the wrongdoer and deter others from similar conduct.” (emphasis added)).
  279. .St. Louis, Iron Mountain & S. Ry. Co. v. Williams, 251 U.S. 63, 66–67 (1919) (noting that punitive awards may not be “wholly disproportioned to the offense”).