Capital Punishment as a Response to Evil

Capital Punishment as a Response to Evil Peter Brian Barry Associate Professor of Philosophy Saginaw Valley State University pbbarry@svsu.edu Forthcoming in Criminal Law and Philosophy (2013). The final publication is available at link.springer.com: http://link.springer.com/article/10.1007%2Fs11572-013-9254-5 DOI 10.1007/s11572-013-9254-5 Some jurisdictions acknowledge, as a matter of positive law, the relevance of evil to capital punishment. At one point, the state of Florida counted that the fact that a murderer’s crime was "especially wicked, evil, atrocious or cruel” as an aggravating factor for purposes of capital sentencing. Fla. Stat. § 921.141(5)(h)(1991) allowed that a jury is entitled to find as an aggravating factor that the capital crime was “especially wicked, evil, atrocious or cruel.” However, in Espinosa v. Florida, the Supreme Court of the United States (SCOTUS) determined that ‘evil crime’ is so vague that no jury could make a principled determination that some crime is evil and thus struck down the Florida statute as unconstitutional. Espinosa v. Florida, 505 U.S. 1079 (1992). SCOTUS’ reasoning is puzzling here for various reasons, not the least of which is that there are equally vague standards relevant to capital sentencing that are widely utilized. As I write, seventeen of the thirty-three states in the United States in which capital punishment is legal regard the fact that some crime is “especially heinous, atrocious or cruel” as aggravation for purposes of capital sentencing. Alabama: Code of Alabama 13A-5-49; Arizona: A.R.S. 13-703 F; Arkansas: A.C.A. 5-4-604; California: Cal. Penal Code 190.2; Colorado: C.R.S.A. 16-11-103; Florida: F.S.A. 921.141(5); Idaho: I.C. 19-2515(3)(b); Kansas: K.S.A. 21-4625; Louisiana: La.C.Cr.P. Ar. 905.4.1; Missouri: V.A.M.S. 565.032.2; Nebraska: Neb. Rev. Stats. 29-2523; New Hampshire: N.H. Rev. Stat. 630:5 VII; North Carolina: N.C.G.S.A. 15A-2000(e); Oklahoma: Okl. St. ann. 21-701.12; Tennessee: T.C.A. 39-13-204(I); Utah: U.C.A. 76-5202; Wyoming: W.S. 6-2-102(h). It is difficult to discern why juries should be able to come to principled decisions about, say, a crime’s heinousness or atrociousness but not its evilness. Richard Rosen, “The ‘Especially Heinous’ Aggravating Circumstance in Capital Cases—The Standardless Standard,” North Carolina Law Review 64 (1986), pp. 941-92. SCOTUS’ concerns not withstanding, I contend that Florida may well be onto something. In this paper, I consider a thesis about capital punishment that strikes me as plausible on its face: if capital punishment is ever morally permissible, it is permissible as a response to evil. Call this the Punishment as a Response to Evil thesis, or simply PRE. PRE differs from other familiar claims about capital punishment and evil—that capital punishment is a lesser evil, a necessary evil, and so forth. The truth of PRE would not settle questions about the moral permissibility of capital punishment. Indeed, PRE might be part of an argument that capital punishment is never morally permissible: for if capital punishment is not morally permissible as a response to evil, then, according to PRE, it is not morally permissible, period. But that seems right. As I note below, there is a sense of ‘evil’ on which the term functions as a superlative: it “has the quality of ne plus ultra: Where do you go from there?” Lance Morrow, Evil: An Investigation (Cambridge: Basic Books, 2003), p. 53. If capital punishment cannot be justified as a response to evil in that sense, when can it be justified? Thus, both proponents and opponents of capital punishment should be interested in PRE. I am hardly the first to consider the possibility that capital punishment might be justified as a response to evil. The head of a crime victims’ rights group contends that “We need the death penalty because evil exists.” See Vince Beiser, “Vengeance is Mom’s,” Mother Jones (March/April 2006). Available online at http://www.motherjones.com/politics/2006/03/vengeance-moms (last accessed December 21, 2012). No less than Immanual Kant draws a connection between evil and punishment quite generally when he suggests that a criminal’s “own evil deed draws the punishment upon himself.” Immanual Kant, The Philosophy of Law: An Exposition of the Fundamental Principles of Jurisprudence as the Science of Right (Edinburgh: T. & T. Clark, 1887), p. 244. More recently, the esteemed Cambridge philosopher Matthew Kramer has defended “the purgative rationale” which implies that capital punishment is morally permissible—indeed, morally obligatory—“when the application of it brings to an end the defilingly evil existence of someone who has committed particularly flagitious crimes.” Matthew Kramer, The Ethics of Capital Punishment: A Philosophical Investigation of Evil and Its Consequences (Oxford: Oxford University Press, 2011), p. 8. Still, PRE is difficult to assess for various reasons: one problem is that PRE utilizes terminally ambiguous language and PRE itself can be disambiguated in importantly different ways. Initially, I distinguish between two different readings of PRE and recapitulate a theory of evil that I have articulated and defended at greater length elsewhere. See my Evil and Moral Psychology (New York: Routledge, 2013). I then argue for my favored version of PRE and consider the implications of my argument for the actual practice of capital punishment. Two Theses About Capital Punishment Again, PRE admits of at least two distinct readings. As evidence of this, note that we are told both that capital punishment ought to be reserved for the most wicked criminals Robert Blecker, “Among Killers, Searching for the Worst of the Worst,” Washington Post, December 3rd, 2000. and that since capital punishment is an extreme sanction it ought to be reserved for the most extreme crimes. Gregg v. Georgia, 428 U.S. 153 (1976). John Stuart Mill seems to endorse both assertions in his defense of capital punishment: When there has been brought home to any one, by conclusive evidence, the greatest crime known to the law; and when the attendant circumstances suggest no palliation of the guilt, no hope that the culprit may even yet not be unworthy to live among mankind, nothing to make it probable that the crime was an exception to his general character rather than a consequence of it, then I confess it appears to me that to deprive the criminal of the life of which he has proved himself to be unworthy—solemnly to blot him out from the fellowship of mankind and from the catalogue of the living—is the most appropriate as it is certainly the most impressive, mode in which society can attach to so great a crime the penal consequences which for the security of life it is indispensable to annex to it. John Stuart Mill, “Speech in Favor of Capital Punishment,” from Utilitarianism, 2nd ed., ed. George Sher (Indianapolis: Hackett, 2001), p. 65. Mill notes approvingly that capital punishment is an appropriate response to murder—“the greatest crime known to the law”—suggesting that it is the extreme nature of an agent’s action that warrants his execution. But Mill also thinks it crucial to determine whether there is any “palliation of the guilt” and whether the perpetrator’s crime is “an exception to his general character.” If not, then it seems to be the extreme nature of the agent that warrants execution. In the spirit of Mill’s distinction, we can separate two different theses concerning evil and capital punishment: (CP1): If capital punishment is ever morally justified it is justified as a punishment for the morally worst offenses—that is, evil crimes. (CP2): If capital punishment is ever morally justified it is justified as punishment for the morally worst offenders—that is, evil criminals. Here too, neither CP1 nor CP2 implies that capital punishment is ever morally permissible nor that the death penalty can be practiced within familiar moral and legal restrictions. CP1 only implies that if it morally wrong to execute perpetrators of evil crimes then capital punishment is never morally justified. Similarly, CP2 only implies that if it is morally wrong to execute evil offenders then capital punishment is never morally justified. Clearly these are two different theses, but both are supposed to capture PRE’s conditional assertion that if capital punishment is ever morally permissible, it is permissible as a response to evil. While Mill seems sympathetic with CP1, I contend that there is good reason to suppose that it is false and that CP2 is correct. Accordingly, sentencing entities in capital cases should be concerned first and foremost with determining whether a defendant is evil and not necessarily whether his crimes are. Eventually, I shall argue for CP2 and against CP1, although CP1 has able philosophers as defenders—Kramer, most notably. But CP1 it has its advocates outside of philosophy as well. Having reviewed dozens of appellate court decisions, forensic psychiatrist Michael Welner is convinced we currently lack the resources to properly instruct juries tasked with determining, for example, whether some crime is especially heinous, atrocious or cruel. Michael Welner, “Response to Simon: Legal Relevance Demands that Evil Be Defined and Standardized,” The Journal of the American Academy of Psychiatry and the Law 31 (2003), p. 417. To assist, Welner has developed the “Depravity Scale,” a mechanism designed to: …measure the crime itself, and not the perpetrator—that is, it will judge the "what" of a crime, as opposed to the "who" or the "why.” This instrument will provide an objective, evidence-based standard by which to assess the intents, actions and attitudes associated with a given crime that distinguish it as worse than other crimes. See https://depravityscale.org/depscale/faq.php#q5. Welner’s stated purpose is to illuminate what legal classifications like ‘heinous’ and ‘atrocious’ and ‘evil’ mean for purposes of sentencing because “with no guidance… distinguishing the worst of crimes is arbitrary.” Michael Welner, “Classifying Crimes by Severity: From Aggravators to Depravity,” in Crime Classification Manual, 2nd ed., John E. Douglas, Ann W. Burgess, Allen G. Burgess, and Robert K. Ressler, eds. (San Francisco: Jossey-Bass, 2006), p. 60. So, insofar as I understand his project, Welnar’s Depravity Scale is primarily intended to assess the depravity of a defendant’s crime rather than the depravity of the perpetrator, and that suggests that Welnar endorses CP1 and not CP2. The Depravity Scale is very much a work in progress. However, Welner’s current project has isolated twenty-six items reproduced in Table 1 “distilled and organized under headings shaped by psychiatric diagnoses” including antisocial personality disorder, narcissistic personality disorder, psychopathy, sadism, and malignant narcissism to isolate “the most pernicious behavior.” Welner, “Classifying Crimes by Severity,” p. 62. TABLE 1: Welner’s twenty-six criteria for depravity In 2006, Welner neither 6) nor 12) were listed as items under study. See Welner, “Classifying Crimes by Severity,” p. 69. Depravity Items Under Consideration During Phase B: Intent to traumatize the victim emotionally, maximizing terror, through humiliation, or to create an indelible emotional memory of the event Intent to maximize damage or destruction, by numbers or amount if more than one person is victimized, or by suffering and degree if only one person is victimized Intent to cause permanent physical disfigurement Intent to carry out a crime for the excitement of the criminal act Committing a crime to gain social acceptance or attention, or to show off Choices for carrying out the illegal act were available that did not involve depravity Carrying out a crime to terrorize others Intentionally targeting victims based on prejudice Targeting victims who are not merely physically vulnerable, but helpless Exploiting a close and trusting relationship with the victim Excessive response to trivial irritant; actions clearly disproportionate to the perceived provocation Escalating the depravity; inspiration for more Carrying out an attack in unnecessarily close proximity to the victim Indulgence of actions inconsistent with the social context Unusual quality of suffering of the victim; victim demonstrated panic, terror, and helplessness Prolonging the duration of a victim’s physical suffering Unrelenting physical and emotional attack; amount of attacking Exceptional degree of physical harm; amount of damage Influencing criminality in others to avoid prosecution or penalty Influencing depravity in others to destroy more Falsely implicating others, knowingly exposing them to wrongful penalty and the stress of prosecution Disregarding the known consequences to the victim Experiencing pleasure in response to the criminal actions and their impact Projecting responsibility onto the victim; feeling entitlement to carry out the action Disrespect for the victim after the fact Indifference to the actions and their impact Apparently, there is evidence that sixteen of the twenty-six items are at least somewhat representative of depravity and one item has been removed from further consideration Michael Welner, “The Justice and Therapeutic Promise of Science-Based Research on Criminal Evil,” The Journal of the American Academy of Psychiatry and the Law 37 (2009), p. 445. (although Welner also declares that there is “strong support” for all twenty-six items Welner, “Response to Simon,” p. 420. and that “almost all of the studied items… have drawn an overwhelming endorsement for being especially or somewhat representative of depravity” Welner, “Classifying Crimes by Severity,” p. 64.). And apparently, there is evidence of broad consensus among clinical professionals that as many as twenty-three items are representative of depravity, although the final list will probably include fewer. Welner, “Response to Simon,” p. 420. The final product of Welner’s research will be the Welner Depravity Scale—the WDS—a mechanism that promises to enable juries to classify crimes as depraved in an informed and principled manner. Welner, “The Justice and Therapeutic Promise of Science-Based Research on Criminal Evil,” p. 446. Again, the Depravity Scale is a work in progress so it would be premature to produce any final assessment. Still, if CP1 is mistaken then Welner’s project is deeply flawed whatever criteria the WDS winds up tracking. Again, if CP1 is false and CP2 is correct, then the primary concern of capital sentencing entities should not be the actions of the accused but his character. But if CP2 is correct, then what needs to be determined is not whether the defendant’s actions are evil but whether he, the person, is. That is not to say that the outcome of the Depravity Scale would be of no interest; it may well be. But if my arguments for CP2 and against CP1 are correct, then the WDS will have to serve a rather different sort of function. Again, one problem with PRE is that it is consistent with two different theses: CP1 and CP2. Having distinguished those two theses, I now turn to another problem plaguing PRE: the ambiguity of ‘evil’. Two Senses of ‘Evil’ Famously, in On the Genealogy of Morals, Nietzsche traces the origins of judgments that others are evil to a kind of “slave revolt in morality” that begins with ressentiment. Friedrich Nietzsche, On the Genealogy of Morals and Ecce Homo, trans. Walter Kaufmann (New York: Vintage, 1989), p. 36. Whereas noble elites first judged themselves to be good and then judged the common masses to be bad by comparison, Nietzsche supposes that the masses came to regard those elites as evil—not merely bad, but evil—and then judged themselves to be good in comparison. What interests me at present is Nietzsche’s willingness to carve out conceptual space between merely being bad on the one hand and being evil on the other. Nietzsche references one of his own earlier works to make the distinction explicit: Whoever begins at this point, like my readers, to reflect and pursue his train of thought will not soon come to the end of it—reason enough for me to come to an end, assuming it has long since been abundantly clear what my aim is, what the aim of that dangerous slogan is that is inscribed at the head of my last book Beyond Good and Evil—At least this does not mean “Beyond Good and Bad.” Ibid, pp. 54-5. In ordinary moral discourse, ‘evil’ is sometimes synonymous with paler moral language like ‘bad’. The Oxford English Dictionary offers “The antithesis of good in all its principal senses” as the primary definition of ‘evil’. Equating evil with mere badness suggests a comparatively thin conception of evil, one consistent with using ‘evil’ to describe just about anything that one disapproves of. Yet Nietzsche’s observation is, for all that, surely apt. Consider Dan Haybron’s observation that if you: …call Hitler or the Holocaust evil… you are unlikely to arouse much disagreement. On the contrary: you will have better luck generating dissent if you refer to Hitler or the Holocaust merely as bad or wrong: “Hitler was a bad person, and what he did was wrong.” …such tepid language seems terribly inadequate to the moral gravity of this subject matter. Prefix your adjectives with as many “verys” as you like; you still fall short. Only ‘evil’, it seems, will do. Daniel Haybron, “Moral Monsters and Saints,” The Monist 85:2 (2002), p. 260. Similarly, in her penetrating study of evil, Claudia Card commends a heuristic that calls for focusing on, not just any morally dubious events, but atrocities—paradigmatically evil events like genocide, slavery, torture, and rape—partly because “the core features of evils tend to be writ large in the case of atrocities” and because atrocities are “uncontroversially evil.” Claudia Card, The Atrocity Paradigm: A Theory of Evil (Oxford: Oxford University Press, 2002), pp. 8-9. And plausibly enough, Card holds that “in defining evil, it is necessary… to distinguish evils from other horrors.” Ibid., p. 12. Sometimes only ‘evil’ will do. A theory of evil should acknowledge this ambiguity in ordinary moral discourse. In his discussion of the traditional problem of evil, Peter van Inwagen allows that ‘evil’ connotes something like “the extreme reaches of moral depravity” but insists that the ordinary meaning of ‘evil’ equates it with ‘bad’ and nothing more. Peter van Inwagen, The Problem of Evil (Oxford: Oxford University Press, 2008), p. 12. Following van Inwagen, distinguish the ordinary sense of ‘evil’ from its extreme sense. The ordinary sense is consistent with exercising only mild opprobrium, while on the extreme sense ‘evil’ is “the worst possible term of opprobrium imaginable.” Marcus Singer, “The Concept of Evil,” Philosophy 79 (2004), p. 185. See also John Kekes, The Roots of Evil (Ithaca, NY: Cornell University Press, 2005), p. 1. Given the ordinary sense, it is perfectly legitimate to describe two people as evil but insist that one is more evil than the other, just as it is perfectly legitimate to suppose that two things can be bad and that one is worse. Given the extreme sense, it makes little sense to speak of “lesser varieties” of evil and evil people. Peter Dewes, The Idea of Evil (Malden, MA: Blackwell Publishing, 2008), p. 10. ‘Evil’, so understood, is a superlative and the difference between evil and the merely bad is a qualitative difference. Cf. Luke Russell, “Is Evil Action Qualitatively Distinct from Ordinary Wrongdoing?,” Australasian Journal of Philosophy 85:4 (2007), pp. 659-77. All this matters for assessing PRE and its variants. Since the idea under consideration is that capital punishment is permissible as a response to evil if it is ever morally permissible, then surely it is the extreme sense of ‘evil’ that is in play here. If ‘evil’ is understood in the ordinary sense on either CP1 or CP2 then there are implausibly many candidates for capital punishment: there are any number of merely bad acts and any number of ways to be a merely bad person that are transparently insufficient to justify the use of the death penalty. Thus, it simply has to be the case that any plausible version of PRE is implicitly utilizing ‘evil’ in the extreme sense, not the ordinary sense. A failure to keep these two senses of ‘evil’ separate generates a problem for one of the most well-developed defenses of CP1. The Purgative Rationale for Capital Punishment In a systematic work, the distinguished Cambridge philosopher Matthew Kramer argues that none of the familiar rationales succeed in justifying capital punishment save for the purgative rationale, a sui generis defense of capital punishment. Kramer’s discussion of the purgative rationale is laden with metaphor appealing to defilement and purgation, but the nerve of his position is clear: [The purgative rationale] maintains that some misdeeds are so monstrously evil as to render morally noxious the lives of the people who have committed them. The continued existence of the perpetrator of any such misdeed(s) will have defiled the moral character of the community in whose control he or she abides. Ibid., p. 15. Note that Kramer implicitly endorses CP1 insofar as he supposes that the moral legitimacy of capital punishment turns on it being a response to evil actions and not to evil people. So, while Kramer sometimes speaks of a perpetrator’s life being “imbued with defilingness” he is clear that it is the perpetrator’s evil misconduct that imbues his life as such. Ibid., p. 187. As Kramer recognizes, philosophical respectability demands explaining what it is for misdeeds to be evil—indeed, monstrously evil—and providing an account of defilement. Ibid. Evil actions, says Kramer, are morally wrong acts for which an agent is culpable, but he denies that just any state of mind sufficient for culpability can produce evil action: negligence, for example, can lead to terribly grave harm, but the “moral gravity” of a negligent action is apparently never enough to make an action evil. Ibid., p. 188. Rather, evil actions must be the product of either sadistic malice (which involves taking pleasure in the suffering of others) or heartlessness (which involves a lack of inhibitions about causing or witnessing the adversities of another) or extreme recklessness. Ibid., p. 203. Kramer suggests that extreme recklessness “involves a frame of mind utterly at odds with a proper level of respect and concern for others” but the same could probably be said of the other two evil-producing states of mind: someone genuinely possessed of sadistic malice is probably willing to use others for their own pleasure in a manner inconsistent with a proper level or respect and someone possessed of heartlessness lacks any real concern for others. Yet an evil action is not just an act that results from one of these egregious mental states since those states might produce only slight harm. Evil actions are also connected in some repugnant manner to seriously injurious consequences (that actually ensue or are desired or knowingly risked). Ibid., p. 204. Thus, on Kramer’s two-pronged account, evil conduct is underlain by one of these three especially egregious mental states, themselves connected with severe harm in the absence of significant extenuating circumstances. Ibid., p. 223. Kramer’s mental state requirement and serious harm requirement helps to make sense of his account of defilement and its relevance to the purgative rationale. Again, the mental states uniquely suited to produce evil actions involve a lack of proper respect and concern for others. But a community that houses a perpetrator of evil crimes would similarly display an inadequate level of respect and concern for others if it permits a defiled perpetrator to live. Of course, merely bad (but not quite evil) actions might similarly display an inadequate level of respect and concern for others. Thanks to an anonymous reviewer for noting this. Still, only evil actions are, well, evil and arguably only they display such inadequate respect and concern that they uniquely demand that a community purge the offender. The problem is not that the community who fails to execute a perpetrator of evil crimes somehow becomes collectively responsible for his crimes after the fact. The problem is that such a community fails to live up to its moral obligations to other communities and to humankind as a whole. Here is Kramer: …although his crimes initially tarnish only his relationship with the rest of humankind—and his more specific relationships with his own community and with the individual victims of his outrages—the relationship between his community and the rest of humankind is now at stake. …the continuation of his life—beyond the time required for a legally prescribed execution—is something that defiles the relationship between his community and the whole of humanity. His continued existence thenceforward is an affront to human kind. Ibid., p. 233. If a community executes a murderer but in a less-than-timely manner, it fails to show sufficient respect and concern to those wronged by his actions, insofar as it lacks the urgency we would expect of someone seriously motivated to correct the gravest of injustices. And finally getting around to executing him will not change the fact that the community did lack the appropriate sense of urgency. Timely purgation thus does not remove the moral taint of a community; it prevents taint in the first place. That is why simply banishing a perpetrator is not sufficient on the purgative rationale: a defiled perpetrator banished to the most remote of deserts is not thereby banished from the human community at large. On the purgative rationale, capital punishment is not simply permissible for a narrow range of offenders but mandatory since it is the only way to prevent a community’s defilement. Ibid., pp. 234-5. The proponent of CP1 need not follow Kramer in holding that execution is mandatory for perpetrators of evil crimes, but the purgative rationale does imply that capital punishment is permissible as a response to the morally worst offenses and therefore it entails CP1. What, then, to say about the purgative rationale? Strictly, it is not evil actions that demand purgation since the death penalty “is appropriate for only a small proportion” of evil conduct. Ibid., pp. 223-4. Instead, Kramer refers variously to “extravagantly” and “monstrously evil” actions, to the “most heinous offences” and “particularly flagitious” ones, even to “extravagantly heinous” acts to narrow the scope of the purgative rationale. This suggests that Kramer is not using ‘evil’ in the extreme sense; if he were, then calling a crime “extravagantly” or “monstrously” evil would be redundant since ‘evil crime’ already picks out the worst sort of crime. Still, I take it Kramer is interested in marking out some class of crimes as the very worst—crimes that are not merely evil, but extravagantly so—and it is those crimes for which capital punishment is supposed to be an apt response. But there is still a problem. Kramer’s account of evil action is clear enough, but what is the difference between merely evil actions and extravagantly evil ones? They are not distinguished by our reactions to them Ibid., p. 225. and while extravagantly evil acts usually involve multiple murders committed in especially vile ways they need not. Ibid., p. 230. At one point, Kramer states that it is “instances of evil conduct that are so heinous as to render defilingly abominable the continued existence of their perpetrators” which require purgation. Ibid., p. 224. But if (extravagantly) evil actions just are those that defile their agents, then we seem to have a circular account of defilement. (Kramer asks rhetorically “When is conduct so evil as to imbue the life of its perpetrator with defilingness?” Ibid., p. 187. Well, when it is extravagantly evil, and not merely evil per se. But what is it for conduct to be extravagantly evil? Well, when it renders defilingly abominable the continued existence of its perpetrator.) He also says that only extravagantly evil acts bespeak “utter contempt for particular individuals and for the moral order of the community in which the iniquities occur” as well as utter contempt “for basic human capacities and thus for humankind as a whole.” Ibid., p. 230. But I confess I do not see why. Why can’t merely evil actions bespeak utter contempt for particular individuals, for basic human capacities, or for humankind as a whole, especially if spawned by sadistic malice or thoroughgoing heartlessness or extreme recklessness? The mental states required to produce evil actions per se are surely going to preclude anything that an evildoer might point to in order to demonstrate that she did not show such contempt. Things are further complicated by Kramer’s discussion of redemption. While responding to a series of objections, Kramer considers an objection that the purgative rationale leaves no room for repentance, especially life-changing repentance expressed by genuine and lasting contrition. Kramer is rightly skeptical that someone guilty of multiple extravagantly evil actions over an extended period of time could become genuinely and lastingly contrite. After all, the agent of such actions probably has a deeply fixed and settled character marked by extreme vices that permitted him to act as such and the deep fixity and settled nature of his character is going to make genuine redemption realistically unimaginable. But Kramer allows that someone guilty of a single extravagantly evil act “might be able to redeem himself through genuine and perdurable repentance” Ibid., pp. 239-40. such that we can separate that single extravagantly evil event from the rest of his life. If so, then executing him is impermissible despite his act. Ibid., p. 240. If it matters, what Kramer says about evildoing and redemption strikes me as independently plausible and consistent with the common intuition that it is possible, if extremely difficult, for an evil person to become morally decent. Claudia Card, The Atrocity Paradigm: A Theory of Evil (Oxford: Oxford University Press, 2002), p. 21; Daniel Haybron, “Moral Monsters and Saints,” The Monist 85:2 (2002), pp. 277-8; Luke Russell, “Dispositional Accounts of Evil,” Philosophical Studies 149 (2010), p. 232. But this result does force the proponent of the purgative rationale onto the horns of a dilemma. Either the advocate of the purgative rationale can make room for repentance or she cannot. If she cannot make room for repentance, then the purgative rationale is consistent with executing both the virtuous and the repentant, an odd result if one was hoping to give content to the thought that capital punishment is justified as a response to evil—more on this problem below. So, unless it makes room for repentance, the purgative rationale is consistent with executing genuinely morally redeemable individuals. But suppose that the proponent of the purgative rationale can make room for repentance and allows that such an accomplishment makes capital punishment inappropriate. If so, then it appears that the purgative rationale is not really an expression of CP1 after all. If the significant achievement of life-changing repentance makes capital punishment inappropriate, then concerns about character must matter to the advocate of the purgative rationale. But that means that the advocate of the purgative rationale is interested in more than just the moral status of evil action; the moral status of the action’s agent matters too. Interestingly, while Kramer’s official position is that the purgative rationale authorizes capital punishment as a response to an offender’s actions and not to the offender herself, he sometimes acknowledges a link between action and agent. Kramer allows that the executions of many putative evil persons—including “vile murderers” like Charles Manson and Ted Bundy and Dennis Rader, among others—are morally justifiable in principle. Ibid., p. 16. And he allows that capital punishment is appropriate when it puts an end to the “defilingly evil existence” of the executed. Ibid., p. 8. It is entirely possible, as far as I can tell, to embrace the purgative rationale that Kramer advocates yet endorse CP2. On this view, the performance of evil actions—or extravagantly evil ones, or whatever—is evidence of being an evil person. Even if evil acts must be the product of especially egregious mental states, someone who suffers from these states to the degree that they lead her to act and to do terrible, terrible things certainly counts as an evil person on the face of it. Thus, I conclude that Kramer’s discussion of the purgative rationale fails as an implicit defense of CP1. However, the problems that plague Kramer’s discussion of purgation and capital punishment run deep and likely doom any particular version of CP1. I explain why in the following section. Evildoing and Evil People Michael Stone, Professor of Psychology at Columbia University, is author of the tellingly titled The Anatomy of Evil Michael Stone, The Anatomy of Evil (Amherst, NY: Prometheus Books, 2009). and actively participates in Most Evil, a television show that documents and profiles serial killers. The centerpiece of Stone’s project is a 22-level hierarchical scale. Officially, this scale concerns evil acts Ibid., p. 11. but Stone fairly easily slips into talk of evil persons. For example, Stone commends thinking of a spectrum that includes persons “so devoid of human sentiments and so driven to commit evil acts repeatedly” that we call them ‘evil’. Ibid., p. 343. This spectrum is manifest in his 22-level “Gradations of Evil Scale” (GES) on display in Table 2. Ibid., p. 46. TABLE 2: Stone’s “Gradations of Evil Scale” (GES) Killing in Self-Defense or Justified Homicide Category 1: Justified homicide Impulsive Murders in Persons without Psychopathic Features Category 2: Jealous lovers, egocentric, immature people, committing crimes of passion Category 3: Willing companions of killers, impulse-ridden; some antisocial traits Category 4: Killing in self-defense, but extremely provocative towards the victim Category 5: Traumatized, desperate persons who kill relatives or others, yet have remorse Category 6: Impetuous, hotheaded murderers, yet without marked psychopathic traits Persons with a Few or No Psychopathic Traits; Murders of a More Severe Type Category 7: Highly narcissistic person, some with a psychotic core, who murder loved ones Category 8: Murders sparked by smoldering rage—resulting sometimes in mass murder Psychopathic Features Marked; Murders Show Malice Aforethought Category 9: Jealous lovers with strong psychopathic traits or full-blown psychopathy Category 10: Killers of people “in the way” (including witnesses); extreme egocentricity Category 11: Fully psychopathic killers of people “in the way” Category 12: Power-hungry psychopaths who murder when “cornered” Category 13: Inadequate, rageful psychopaths; some committing multiple murders Category 14: Ruthlessly self-centered psychopathic schemers Spree or Multiple Murders; Psychopathy is Apparent Category 15: Psychopathic, cold-blooded, spree or multiple murderers Category 16: Psychopathic persons committing multiple vicious acts (including murder) Serial Killers, Torturers, Sadists Category 17: Sexually perverse serial killers; killing is to hide evidence; no torture Category 18: Torture-murderers, though the torture element is not prolonged Category 19: Psychopaths driven to terrorism, subjugation, rape, etc. short of murder Category 20: Torture-murderers but in persons with distinct psychosis (such as schizophrenia) Category 21: Psychopaths committing extreme torture but not known to have killed Category 22: Psychopathic torture-murderers with torture as their primary motive; need not be sexual. Note that almost all the items in the GES describe agents: the GES distinguishes jealous lovers, hot-headed killers, power-hungry psychopaths, sexually perverse serial killers, torture-murderers, and so forth. There is an implicit moral ordering of the categories: the higher the number, the more likely the person will be described as ‘evil’. Ibid., p. 53. Torture appears to be a super-aggravator for Stone: the worst grouping of categories in the GES includes serial killers, torturers, and sadists; three of the five worst categories make a specific reference to torture; the very worst offenders in the GES are psychopathic torture-murderers with torture as their primary motive. Thus, Stone’s GES initially seems to be of interest to the advocate of PRE and CP2 in particular: it codifies a whole range of nasty individuals, some of whom are plausibly regarded as evil people, and allows us to determine who is the worst of the bunch. The SEP thus functions as a kind of heuristic for sentencing entities in capital cases for generating informed and principled judgments about the appropriateness of the death penalty. Does the defendant fall into some sufficiently high-ranked category? If so, then capital punishment is appropriate. If not, then not. But there are at least two complications here. First, where is the line to be drawn? How do we determine which category marks the difference between being an evil person and not being an evil person? Clearly, not just any category in the GES picks out an evil person; Category 1—justified homicide—is “not evil at all”; it functions as an “anchor,” as the “absolute lowest point, where the notion of evil has disappeared.” Ibid., p. 30. Sympathetic and remorseful killers fall into Category 5 of the GES, a comparatively modest ranking that suggests they are not evil. The most horrific cases that people would “just about unanimously” call ‘evil’ includes torture-murderers with torture as their primary goal in Category 22. Ibid., p. 33. A psychopath guilty of extreme torture but not known to have killed would fall into Category 21 on the GES, the second worst of the categories on that scale, but not the most damnable category. Are only members of Category 22 and not Category 21 eligible for execution? Does that mean that psychopathic serial torturers are not evil? But why are only torture-murderers legitimate candidates for capital punishment? Why not sexually perverse serial killers? Why not torture-murderers who didn’t torture for prolonged periods of time? Certainly our actual practice does not suggest that we think that only serial torture-murderers are apt candidates for the death penalty. There are certainly individuals who occupy fairly modest rankings on the GES—say, Categories 7 and 8—who would likely be executed in any number of jurisdictions that practice capital punishment because their crimes suggest familiar kinds of aggravators: for example, if their crime resulted in the death of a vulnerable victim or multiple persons. If the GES does not pick out twenty-two different categories of evil criminals deserving of the death penalty, then we need some further means for determining which categories house legitimate candidates for capital punishment and which do not. Relatedly, and this is the second problem, why think that the GES really does accurately rank order the twenty-two different kinds of personalities under consideration? Why think that killers of people “in the way” suffering from extreme egocentricity (Category 10) are worse sorts of people than jealous lovers with strong psychopathic traits (Category 9)? Why are power-hungry psychopaths who murder when “cornered” (Category 12) not worse sorts of people than inadequate and rageful psychopaths (Category 13)? Stone’s answer might follow from his apparent sympathy with ordinary language philosophy; he frequently appeals to our actual use of the term ‘evil’ when discussing the meaning of that term and he endorses Wittgenstein’s declaration that “The meaning of a word is its usage” Ibid., pp. 14-5. At one point Stone declares that “if a judge, journalist, and the public concur that a particular crime was evil—well, then, it was evil. For that establishes the meaning of evil—down here on Earth” (15). But are our common intuitions about evil people this fine-grained? Is there a consensus that psychopaths driven to terrorism but stop short of murder are morally worse persons than torture-murderers who spare their victims prolonged torture? If not, then what justifies putting the former in Category 19 and the latter in Category 18? Is there any consensus that extreme torturers who are not known to have killed are morally worse people than serial sexual killers? If not, then what justifies putting the former in Category 21 and the latter in Category 17? Absent some answer to these questions, the rankings utilized in the GES are arbitrary, a concern that Stone himself notes. Ibid., p. 34. What is needed at this point is a theory of evil, one that will illuminate the difference between actions and persons that are evil and those that are not. It is beyond the scope of this paper to develop such a theory, but I can gesture at one. For a fuller discussion and defense of such a theory, see my Evil and Moral Psychology. I take it that being an evil person is a matter of having a certain sort of character; it would be odd to regard someone as evil while denying that she suffers from any moral vices, for example. Insofar as the virtues (and vices) are plausibly understood as multi-track dispositional traits, including dispositions concerning motivation and cognition and affect, it should be no surprise that evil character is composed by a series of such dispositions. Now, it is a mistake to suppose that if someone is evil in virtue of being extremely cruel, for example, that she must tend to perform many cruel acts. Cf. Dan Haybron, “Consistency of Character and the Character of Evil,” in Haybron (ed.), Earth’s Abominations: Philosophical Studies of Evil (Amsterdam: Rodopi, 2002), p. 63. Such an evil person will certainly be disposed to perform cruel actions, but familiarly dispositions can be masked such that they never actually manifest: poisons are disposed to kill even if antidotes abound, for example. And that matters. For if being an evil person involves being disposed to act wrongly but being disposed to act wrongly does not require acting wrongly, then it is a mistake to suppose that an evil person must be an evildoer. There is similarly good reason to think that the converse does not hold either—that is, there is good reason to doubt that evildoers must be evil people. Genocidal actions are commonly regarded as examples of evil actions. Card, The Atrocity Paradigm, p. 9; Joel Feinberg, Problems at the Roots of Law: Essays in Legal and Political Theory (Oxford: Oxford University Press, 2003), p. 160. However, the perpetrators of various genocidal actions are not that different from you and I save for the circumstances they found themselves in; most of the perpetrators of mass killing are extraordinary “only by what they did, not by who they were.” James Waller, Becoming Evil (New York: Oxford University Press, 2002), p. 8. And all sorts of research from social psychology suggests that many otherwise morally decent people will do terrible, terrible things when placed in the right circumstances. See, for example, Philip Zimbardo, The Lucifer Effect: Understanding How Good People Turn Evil (New York: Random House, 2007). Further, it is commonly thought that evil people are rare. Haybron, “Moral Monsters and Saints,” p. 278; John Kekes, Facing Evil (Princeton, N.J.: Princeton University Press, 1990), p. 5; Russell, “Dispositional Accounts of Evil Personhood,” Philosophical Studies 149 (2010), p. 232. But acts of genocide are not that rare, not when it is estimated that some 60 million men, women, and children have been the victims of mass killings and genocide in the 20th century alone. James Waller, Becoming Evil: How Ordinary People Commit Genocide and Mass Killing (New York: Oxford University Press, 2002), p. xii. Evil people become implausibly common if we suppose that evildoers must be evil people. Thus we have reason to think that the Identity Thesis—the thesis that evil people just are evildoers—is simply false. We have reason to reject the Identity Thesis left-to-right because we have reason to doubt that evil people must actually engage in evildoing and we have reason to reject the Identity Thesis right-to-left because we have reason to doubt that the agents of evil actions must themselves be evil people. Actually, I doubt that even a disposition to engage in evildoing would be enough to ensure that an evildoer is an evil person. One problem is that a disposition to do evil could make it the case that an agent lacks other properties necessary for moral responsibility; Godzilla (in some movies) is disposed to perform paradigmatically evil acts but is weirdly regarded as a moral responsible agent. A graver problem is that someone could be disposed to do evil but feel great regret or remorse or shame in response; a repentant child molester who struggles in vain against his urges and hates himself for failing is deeply flawed but, I think, wrongly regarded as evil. I do discuss these matters at greater length in my Evil and Moral Psychology, especially Chapter 4. No surprise that the Identity Thesis is so commonly rejected. Card, The Atrocity Paradigm, p. 22; Paul Formosa, “A Conception of Evil,” The Journal of Value Inquiry 42 (2008), p. 233; Eve Garrard, “Evil as an Explanatory Concept,” The Monist 85, p. 321; Haybron, “Moral Monsters and Moral Saints,” p. 279; Adam Morton, On Evil (New York: Routledge, 2004), p. 65; Luke Russell, “Dispositional Accounts of Evil,” p. 232; Singer, “The Concept of Evil,” p. 190. Even without a fully developed theory of evil in place, I submit that rejecting the Identity Thesis is sufficient to establish that CP1 is mistaken. I argue in the next section that unless the Identity Thesis is true, evildoing does not necessarily justify capital punishment as a response. Since the Identity Thesis is false, CP1 ought to be rejected too. Evildoing and Capital Punishment One especially powerful reason for rejecting the Identity Thesis is that a fair bit of evildoing, perhaps very much of it, is the upshot of anomalous circumstances that powerfully effect the behavior of the agents who find themselves in those circumstances and lead them to act in ways that are not really reflective of their actual character. Russell, “Dispositional Accounts of Evil,” p. 237. Again: most of the perpetrators of mass killing are extraordinary only in virtue of what they do, not in virtue of who they are. Philosophical dissatisfaction with the Identity Thesis is reflected in the institutions of our criminal justice system given that “juries want more than evidence to convict; they want to be certain that the person they are sending to prison or to another world is an evil creature as well as an evildoer.” Janet Malcom, “Iphigenia in Forest Hills: Anatomy of a Murder Trial,” The New Yorker (May 3, 2010), p. 57, emphasis added. Such dissatisfaction is evident, I submit, in our actual practices of administering the death penalty in the United States. Familiarly, SCOTUS struck down capital punishment (as it was practiced in the United States in 1972) as unconstitutional in Furman v. Georgia. Furman v. Georgia, 408 US 238 (1972). The Court’s rationale is somewhat elusive, but the 1976 decision, Gregg v. Georgia Gregg v. Georgia, 428 US 153 (1976). that effectively resurrected capital punishment is illuminating. In response to Furman, Georgia amended its capital sentencing statutes to require, among other things, a hearing to determine whether at least one of ten specified aggravating circumstances exists beyond a reasonable doubt and whether there are any mitigating circumstances; absent aggravation or given the presence of mitigation, a capital sentence may not be handed down. In Gregg, SCOTUS held that Georgia’s revised capital sentencing statutes are constitutional and at present every state that practices capital punishment enumerates a list of aggravating circumstances to guide sentencing entities. It is not my hope to enter into complicated debates about the constitutionality of the death penalty in the United States, but SCOTUS’ interest in aggravation and mitigation is relevant for present purposes. Consider some familiar examples of aggravation for purposes of capital sentencing: just about every state practicing capital punishment counts the fact that a murder was committed for “pecuniary gain” as an aggravator and the most common aggravator concerns the murderer’s status as a previously convicted violent felon. And multiple states—twenty-eight of the thirty-three that practice capital punishment—allow that either 1) being especially heinous or atrocious or cruel The Model Penal Code includes 1) but not 2) or 3). See the Model Penal Code §210.6(3)., 2) being outrageous or wantonly vile, or 3) being torture or unnecessarily torturous suffices as aggravation. But importantly, the mere existence of aggravating circumstances is not sufficient to render a defendant deserving of the death penalty. Again: it must also be the case that there are not sufficient mitigating circumstances. In the criminal law, any factor tending to show that a criminal defendant, while guilty, is less culpable than his act alone would indicate constitutes mitigation. And quite generally, mitigation includes any evidence relevant to the defendant’s character, propensities, or circumstances of the offense, including facts about her situation and upbringing—that is, facts about her and not just her crime. And all this should seem right once we have abandoned the Identity Thesis: for any putative evil action, it will be an open question whether or not there are any mitigating circumstances—any facts about her or her situation—that show that a defendant is less culpable then she would be in absence of those circumstances. The fact that a heinous crime was committed in anomalous and deeply stressful circumstances might count as mitigation, as might genuine contrition for that crime: either suggests that even if the defendant’s action is an uncontroversial instance of evildoing, it does not follow that the defendant is herself an evil person. But consider the consequences of this observation for Welnar’s project: as long as it seems plausible to suppose that mitigation should have something to do with capital sentencing, no version of the WDS can settle whether some perpetrator should be sentenced to death. For the Depravity Scale is concerned with actions and not with agents, and facts that are sufficient to show that an act is evil do not suffice to show that its agent is evil. This problem must be faced by any proponent of CP1. Again: CP1 implies that if capital punishment is ever morally permissible, it is permissible as a response to evil actions. But the arguments of this section suggest that evildoing is possible even in the face of serious mitigation and the arguments of the previous section suggest that evildoing need not reflect the actual character of the evildoer. Quite generally, the agent of some putative instance of evildoing might, but for that fact, be a clear and obvious case of a moderately virtuous and repentant person—in other words, a morally decent person. Thus, CP1 is consistent with executing otherwise morally decent people. That, I submit, is a fatal flaw of CP1. It is certainly an odd consequence of taking seriously the thought that if capital punishment is ever morally permissible, it is permissible as a response to evil. This argument against CP1 can be developed in other ways. I contend that CP1 is inconsistent with the most familiar justifications of capital punishment; at least the most familiar justifications of capital punishment do not support CP1. In Gregg, SCOTUS held that “Retribution and the possibility of deterrence of capital crimes by prospective offenders are not impermissible considerations for a legislature to weigh in determining whether the death penalty should be imposed” and considerations involving retribution and deterrence probably underlie the two most commonly offered justifications for capital punishment. Consider retribution and the theory of retributivism first. Familiar versions of retributivism hold that capital punishment is morally justified because the relevant offender deserve to be executed and that facts about what is deserved alone are relevant to determining the appropriateness of punishment. Thus, for most retributivists, the property of desert does not just function as a side-constraint protecting the innocent from just punishment; in the absence of special circumstances, it warrants the infliction of punishment on the guilty. Kramer, The Ethics of Capital Punishment, p. 72. But once we allow that the agent of an evil action need not herself be evil, we must be open to the possibility that even an evil action might be best explained in situational terms and not in terms of the acting agent’s vices or character more generally. But then it is an open question whether the agent of even the most heinous evildoing really deserves to be executed in response. Similarly, once we allow that contrition has some moral relevance to determining the appropriateness of punishment, it must be an open question whether an evildoer deserves to be executed supposing that she is genuinely contrite. The general problem for CP1 here is that facts about mitigation and desert, at least very often, are a function of facts about an agent herself and not simply her actions. CP1, with its exclusive concern with evil actions, is insensitive to facts about agents of evildoing and thus fails to make room for morally salient considerations concerning mitigation and desert that are central to any plausible formulation of retributivism. Accordingly, no appeal to the concerns that motivate retributivism is likely to justify CP1. Appeals to deterrence yield no better results. In debates about deterrence and capital punishment it is common to distinguish between specific and general deterrence: punishment that yields the former consequence affects only the individual being punished while punishment that yields the latter consequence also affects the public at large—say, by making it publically manifest that this is what happens to those who behave in the requisite manner. Focus on specific deterrence first. A concern for specific deterrence means that CP1 ought to be read as calling for executing the particular agent of the evil act under consideration. But since, as I suggest above, the overwhelming majority of evil actions are not the result of anything like evil character but of powerfully coercive circumstances that lead agents to act in ways that do not reflect their character, there is little chance that an appeal to specific deterrence will support CP1. Unless the circumstances that produced some evil act are themselves replicated, the non-evil agent of an evil act is not likely to perform it again: there is nothing to deter. Accordingly, any defense of CP1 that is motivated by a concern for deterrence is going to have to appeal to general, not specific, deterrence. But an appeal to general deterrence is of little help for similar reasons. Again, we have good reason to believe that the preponderance of evil actions are performed in greatly coercive circumstances and are best subject to situational explanations, not dispositional ones. But capital punishment is only going to generally deter if potential criminals “loosely calculate the costs and rewards of their behavior.” Richard Lempert, “Desert and Deterrence: An Assessment of the Moral Bases of the Case for Capital Punishment,” Michigan Law Review 79 (1981), p. 1193. Kramer (20) quotes Lempert approvingly. So it will be of little use to appeal to a potential murderer’s tendency to be risk-averse or instrumentally rational by promising to visit the harm of capital punishment on him if he performs some putatively evil act. The fact that most evil behavior is subject to situational explanation, not dispositional explanation suggests that the goal of preventing evil behavior is more likely to be realized if we focus on removing those circumstances that would otherwise lead to evil behavior rather than attempting to deter the public at large by threatening them with capital punishment. So neither an appeal to specific or general deterrence is likely to justify capital punishment as a response to evil action. So appealing to deterrence per se will not justify CP1. It is worth emphasizing that all of the problems that plague CP1 do not similarly plague CP2. One problem with CP1 is that it is consistent with executing the moderately virtuous and the repentant—individuals whose evil actions were prompted by powerful situational influences and not representative of their overall character—and thus permits executing otherwise morally decent people, an odd response to the thought that capital punishment is justified as a response to evil. By contrast, CP2 fares better. Again, I can only gesture at a full-blown theory of evil personhood, but I contend that evil character is partly constituted by the possession of extreme moral vices and the absence of morally redeeming emotions expressive of contrition. See my “Extremity of Vice and the Character of Evil, Journal of Philosophical Research 35 (2010), pp. 25-42. But more generally, insofar as CP2 is focused on evil people and not evil actions, CP2 is not consistent with executing morally decent people: whatever else can be said about evil people, they are not morally decent. CP2 is also consistent with at least one of the familiar justifications for capital punishment. Again, there are many different versions of retributivism but CP2 coheres especially well with any version of retributivism that links the degree of punishment to the culpability of the offender. Michael S. Moore, “The Moral Worth of Retributivism,” in Ferdinand Schoeman, Responsibility, Character, and the Emotions (Cambridge: Cambridge University Press, 1987), p. 179. Since evil people are especially morally culpable—indeed, it is commonly thought that evil people deserve our strongest moral condemnation Russell, “Dispositional Accounts of Evil,” p. 232; Singer, “The Concept of Evil,” p. 190. —then there is a strong case to be made that responding to evil people by executing them is warranted given that capital punishment is an especially dramatic response to culpable individuals. It is less clear whether or not concerns about deterrence really support CP2. On the one hand, as I suggest above, it is commonly thought that evil people are rare. And I have also suggested that the typical agent of an evil action is not an evil person, but a non-evil person in terribly coercive circumstances. And one might wonder whether anything at all will deter evil people, especially if one identifies evil people with individuals who are outside of our moral community, incapable of moral dialogue with us. See Gary Watson, Responsibility and the Limits of Evil: Variations on a Strawsonian Theme,” in his Agency and Answerability: Selected Essays (Oxford: Oxford University Press, 2004), pp. 219-59 and Michael McKenna, “The Limits of Evil and the Role of Moral Address: A Defense of Strawsonian Compatibilism,” The Journal of Ethics 2 (1998), pp. 123-42 for sensitive discussions of this thought. I concede the possibility that executing evil people because they are evil might have a fairly minimal deterrent effect, too minimal to actually justify institutionalizing capital punishment. But I am inclined to think that this is a problem for deterrence-based justifications of capital punishment and not of CP2 itself. And again, CP2 is consistent with holding that capital punishment is not morally justified; it only holds that if capital punishment is ever morally permissible, it is permissible as a response to evil people. In summary, there are deep and fatal problems with CP1 while CP2 fares much, much better. So, if one is inclined to endorse PRE and think that capital punishment is a permissible as a response to evil if it is permissible at all, then CP2 should be the preferred formulation. However, I concede that CP2 may well suffer from its own sequence of problems. In the final section, I briefly consider some problems that arguably await the advocate of CP2. Not surprisingly, I think that these problems can be tended to. Conclusion: Putting CP2 Into Practice The main challenges to CP2 are practical ones: how exactly is CP2 going to be put into practice? My general response to these practical challenges is to stress that while putting CP2 into practice will require making some alterations to our current capital sentence mechanisms, what is going to be novel about putting CP2 into practice is not all that radical compared to our current practice. The primary challenge to putting CP2 into practice has already been suggested above. Aren’t SCOTUS’s concerns about vagueness still hovering? Can juries form non-arbitrary judgments about who is evil and who is not? There is bound to be some agreement about who counts as an evil person—just about everyone is willing to say that Hitler is evil, for example—but there is surely a fair bit of disagreement about the meaning and intension of ‘evil’ and its cognates. Juries and sentencing entities will surely require some guidance here. What sort of guidance could be supplied? Second, and relatedly, putting CP2 into practice seems to require that we ask sentencing entities to do something that they are neither accustomed to nor especially good at. Asking a jury, for example, to determine whether a defendant is evil goes well beyond asking them to determine familiar matters of legal fact. Putting CP2 into practice would require a jury to moral assess a defendant’s character, a rather different sort of task, and, absent the sort of guidance called for above, not the sort of task we have any reason to think that a typical jury will be up to. But note that we already ask sentencing entities to make assessments of a defendant’s character. For example, California makes a specific reference to the defendant’s character when enumerating mitigating circumstances. Cal. Penal Code 190.3. And the fact that a defendant acted under extreme mental or emotional disturbance, duress, or under the domination of another person—facts that suggest the defendant’s crime is the product of situational influences and is not reflective of her character—is widely taken to constitute mitigation. Admittedly, putting CP2 into practice requires drawing an overall assessment of a defendant’s character and not simply noting some consideration relevant to drawing an overall assessment, but it is simply not the case that juries and sentencing entities are never tasked with evaluating character. The first challenge requires a somewhat different response. If the concern is that juries, left to their own devices, will not be able to come to any principled judgment that the defendant is evil then the obvious solution is to avoid leaving them to their own devices. There is an analogy available here with the “especially heinous, atrocious or cruel” standard noted above. While SCOTUS has held that this standard is unconstitutionally vague absent any further elaboration Maynard v. Cartwright, 486 U.S. 356 (1988). it does allow the use of the “especially heinous, atrocious or cruel” standard if the sentencing entity has appropriate guidance or expertise. Walton v. Arizona, 497 U.S. 639 (1990). So long as juries, for example, are given some sort of mechanism to distinguish crimes that are especially heinous, atrocious, or cruel then they will have guided discretion, rather than unguided discretion, in determining the appropriateness of a capital sentence. By analogy, if juries are given some sort of standard to distinguish evil people from non-evil people, then they too will have guided discretion, rather than unguided discretion, in determining the appropriateness of a capital sentence. Admittedly, this response says nothing about what the legal standard needed to identify evil people would look like. But we can probably gain some insight, again, by looking at current practice. Any number of aggravators currently utilized in capital sentencing are suggestive of evil personhood. For example, the fact that a perpetrator “laid in wait” or proceeded in a “cold, calculated manner” arguably suggests something terrible about him, as does the fact that he showed “utter disregard for” or “extreme indifference to human life” even though only a handful of states utilize these as aggravators. Welner’s WDS too could be of some help here. As I have argued throughout, we cannot infer that a defendant is evil just because her crime is, but certainly a pattern of exceptionally depraved behavior is suggestive of evil character, and even if evidence of past depravity is not admissible at trial for the purposes of conviction, it would certainly be relevant to determining whether a defendant is evil. Other commonly recognized aggravators are less relevant to determining whether a defendant is evil. For example, a majority of states count the fact that a victim of murder was a public official—say, a peace officer or judge or prosecutor—as aggravation. But the murder of a police officer need not suggest that the killer was possessed of any dubious mental state aside from an intelligible desire to escape. Welner, “Classifying Crimes by Severity: From Aggravators to Depravity,” p. 61. The most problematic objection to CP2 is that it seems to license the execution of those who have committed no crime. After all, rejecting the Identity Thesis leaves open the possibility that an evil person has committed no crime, much less engaged in any evildoing. Perhaps an evil person has simply been thwarted prior to any attempt at engaging in criminal activity, or perhaps the opportunity just has not come up. CP2 is consistent with not simply punishing them but with executing them. This would not exactly violate a moral side-constraint prohibiting punishing the “innocent”: genuinely evil people are oddly described as such. But for those concerned about such things, a remedy is available: as a matter of policy, a community that endorses CP2 could both allow that being evil might make a person eligible for capital punishment and permit sentencing entities to determine if a defendant is evil only after the defendant has been convicted of some crime (or perhaps convicted of some limited class of sufficiently grave crimes). That would be enough to ensure that the “innocent” avoid punishment. Here, then, is how I imagine CP2 would be put into practice. The commission of some limited class of evil acts would be necessary to initiate a capital trial but what settles the appropriateness of a death sentence is the offender’s status as the morally worst sort of offender—as vile or monstrous or simply evil—and that determination would have to be made only after a conviction. Juries in capital cases currently have a dual role to play in a bifurcated trial: they must first determine whether, as a matter of legal fact, a defendant is guilty of some crime and then must determine sentence in a separate hearing. But a different sort of bifurcated trial could be had: we could ask juries only to determine a defendant’s guilt while tasking a separate sentencing entity with some expertise in these matters—say, a pool of select psychologists and psychiatrists, perhaps even philosophers!—to determine whether the defendant is evil. As a matter of positive law, the Sixth Amendment of the United States Constitution requires juries to determine the appropriateness of a death sentence, not judges. Ring v. Arizona, 536 U.S. 584 (2002). Whether that is a moral requirement for just punishment is an open question. But if my arguments above are cogent, there is nothing necessarily unjust about executing the morally worst sort of people. Consistent with CP2, one might insist: if being an evil person doesn’t justify capital punishment, nothing does. PAGE 35