Crim Law and Philos (2015) 9:367378

DOI 10.1007/s11572-013-9250-9
BOOK REVIEW

Can/Should We Purge Evil Through Capital

Punishment?
Matthew H. Kramer: The Ethics of Capital Punishment: A
Philosophical Investigation of Evil and its Consequences. Oxford
University Press, New York, 2011
Carol S. Steiker

Published online: 25 September 2013

 Springer Science+Business Media Dordrecht 2013

Abstract Matthew Kramers The Ethics of Capital Punishment: A Philosophical Investigation of Evil and its Consequences explores the morality of capital punishment and
develops his own purgative rationale in support of the practice. I present my objections
to Kramers purgative rationale and trace our disagreement to differences over the nature
of evil, the autonomy of human character formation, and the concept of defilement.

Matthew Kramer opens his erudite and engaging book on the ethics of capital punishment
with a personal preface that motivates his account. Kramer explains that upon learning
about the Holocaust as an eight-year-old child, he began to develop in an inchoate way
what he now defends as a purgative rationale for the use of the death penaltythe claim
that capital punishment is a morally required response to certain extreme evils. He
remembers thinking that it would have been morally grotesque if the trials of some of the
major Nazi leaders had ended with sentences that would involve the devotion of resources
to sustaining the lives of those leaders (v).
I have a contrasting childhood-reaction-to-the-Holocaust story that similarly motivates
my own views on capital punishment and my response to Kramers argument in particular.
At a slightly older age (12 or 13?), I immersed myself in the standard American Jewish
teenage girl Holocaust canon of the early 1970sAnne Franks Diary of a Young Girl and
Elie Wiesels Night, among other workswhich made a strong impression on me. I
remember ruminating not so much about the proper treatment of the surviving Nazi
leaders, but more about how so many Germans (and others) could have come to see an
entire people (my people) as not really people at allas something less than human. When
I later chose to work as a public defender early in my legal career, I would often advert to
this childhood experience when friends, relatives, and even fellow lawyers questioned my
choice to take on a role that they saw as morally dubious at best. I would explain that I
viewed the representation of (allegedly) heinous criminals as a kind of extreme civil rights

C. S. Steiker (&)
Harvard Law School, Cambridge, MA, USA
e-mail: steiker@law.harvard.edu

123

368

Crim Law and Philos (2015) 9:367378

workthe championing of the rights and dignity of exactly those people that rightthinking folks are inclined to view as not really people at all. I argued that by zealously
representing those whom we are most motivated to hate, I was keeping the world safe for
whoever else might at some point fall into that categoryJews, blacks, gays, etc. I hoped
that by contextualizing my clients offenses and giving voice to their storiesespecially
when they were in fact guilty as chargedI was working against our collective tendency to
dehumanize others, by combating this propensity in its most compelling form.
Thus, while Kramer saw in the Holocaust a challenge to respond adequately to the evil
of the Nazis and their ilk, I saw in the Holocaust a challenge to resist the human tendency
toward the dehumanization of others, even (maybe especially) wrongdoers. These differing
responses are not logically inconsistent with one another, but unearthing them may help
explain or at least motivate the concerns I raise at each step in Kramers argument, just as
they help explain and motivate his own account. In what follows, I will focus my attention
on responding to Kramers purgative rationale for capital punishment, as this is the most
novel and provocative part of his wide-ranging book. My response raises four different
doubts about Kramers argument: (1) epistemic doubts about the search for what Kramer
calls extravagant evil; (2) moral doubts about the search for extravagant evil; (3) doubts
about the concept of defilement on which Kramers purgative argument depends; and (4)
doubts about the necessity and permissibility of death as a punishment.
At the outset, in order to do justice to Kramers achievements in this book, I note that he
does not turn his full attention to the purgative rationale for capital punishment until the
sixth of his seven chapters. The first five chapters consist of an introduction, followed by a
chapter each on deterrence, retribution, incapacitation, and denunciation as possible justificatory rationales for capital punishment. For each rationale, Kramer explores a range of
possible arguments in support of the morality of capital punishment and responds to each
in turn, ultimately concluding that no version of any of these familiar rationales offers a
satisfying moral justification for the practice. Thus, most of Kramers book consists of
arguments against what most peoplescholars and laypersons alikeconsider to be the
moral foundations of the death penalty. There is a tremendous amount of valuable insight
in these detailed discussions, which offer a pithy and lively roadmap of each rationale for
the punishment-theory novice as well as in-depth engagement with the arguments of a wide
variety of scholars on the specific issue of capital punishment for the specialist. There is
also a great deal to debate in these chapters, though I will limit myself to engaging
Kramers arguments about the rationales that he rejects for capital punishment only to the
extent that such engagement will illuminate my disagreement with the purgative rationale
that he supports.
Kramer begins his account of the purgative rationale for capital punishment with an
anchoring illustrationthe biblical story of the stoning and burning of Achan and his
family, at Gods command, in response to Achans personal appropriation of some of the
Israelites spoils of war that were meant for a devotional sacrifice to God. In Kramers
view, the story of Achan illustrates the phenomenon of defilement, as distinct from collective responsibility. The gist of Kramers purgative rationale is that some crimes are so
iniquitous that the continued existence of the thugs responsible for them is a blot on the
moral order of the community in which those thugs are kept alive (186). Kramer explains,
Just as the Israelites were all tainted by the continuing existence of Achan, a community
is taintedin other words, its moral integrity is lessenedby the continuing existence of
anyone who has perpetrated some especially hideous crimes and who is within the jurisdiction of the community or otherwise specially connected to it (18687).

123

Crim Law and Philos (2015) 9:367378

369

Kramer acknowledges that his purgative rationale depends crucially on an account of

evil and its capacity to taint a community. He thus spends considerable time developing a
definition of evil, concluding that evil conduct is underlain by sadistic malice or heartlessness or extreme recklessness that is connected to severe harm in the absence of any
significant extenuating circumstances (223). But Kramer is quick to note that the purgative rationale does not cover every instance of evil conduct. Rather, it extends only to
instances of evil conduct that are so heinous as to render defilingly abominable the continued existence of their perpetrators (224). Kramer then develops an account of defilement, using two semi-fictional cases (based loosely on actual crimes) as paradigmatic
cases of extravagantly evil conduct (226). The cases involve repeated perpetration of
some combination of sadistic and remorseless rape, torture, mutilation, and murder,
including of infants and young children, producing horrible physical and psychological
pain. Such extravagantly evil conduct, argues Kramer, requires the execution of its perpetrators, in order to restore the moral standing of the communities in which they abide.
The need for purgation, in Kramers view, arises not from the relationship between the
communities and an external being (such as God in the story of Achan), but rather from the
normative relationship between each community and the rest of humanity (230).
Kramer concludes that only by executing perpetrators of extravagant evil (though it must
be a humane rather than a torturous execution) can a community purge the taint of such
evil, because otherwise the community would be allocating resources to the prolongation
of the perpetrators lifea life that constitutes a grotesque rebuff to humankind (237).
Kramer offers a number of nuances and caveats to his purgative rationale, including an
entire chapter (Chapter 7, The Death Penalty in Operation), on whether and in what
circumstances problems in the administration of capital punishment (such as arbitrariness,
discrimination, and risk of error) undermine the affirmative case that he makes in support
of the use of the death penalty. While Kramer acknowledges that some such problems, if
sufficiently severe, could undermine the legitimacy and thus the moral permissibility of a
system of capital punishment, he concludes that it is realistically possible to reduce any
problems in the administration of the death penalty to levels at which those problems do
not undermine the legitimacy of the whole system (327). He offers no definitive
assessment about whether or not the administration of capital punishment in the United
States (or anywhere else) is legitimate under his criteria, but he urges liberal-democratic
societies to impose a moratorium on capital punishment until the practice is reoriented
toward the sole objective of purgation, with sufficient procedural protections. If this
reorientation is successful achieved, however, Kramer concludes that the use of capital
punishment by such a society would be not merely morally permissible, but morally
obligatory (16).
My disagreement with Kramers purgative rationale stems from doubts that arise around
the three central working parts of his argument: the concept of extravagant evil, the
concept of defilement, and the significance of death as a punishment. My doubts are of
different types and will likely strike others (as indeed they strike me) as of different
weights as well. Together, however, they are of sufficient collective magnitude to
undermine Kramers purgative foundation for a morally permissible or required practice of
capital punishment.
Epistemic doubts about extravagant evil Kramer spends a great deal of time distinguishing extravagantly or defilingly evil conduct from ordinarily evil conduct, which
he acknowledges cannot morally be punished by death. Kramer develops two paradigmatic
cases of extravagant evil, which involve repeated and remorseless infliction of sickening,
sadistic torment on innocent victims that include infants, children, and the elderly

123

370

Crim Law and Philos (2015) 9:367378

(Richards Case and Josephs Case, 22728). And of course the Nazis always stand in
the wings as baleful exemplars of the extremities of evil (231). Despite the power of these
examples, I remain, perhaps naively, agnostic about the very existence of extravagant evil,
as sufficiently distinct from the ordinary kind that Kramer agrees cannot justify the
imposition of capital punishment. Yes, evil exists on a continuum, and there are metrics
(intention, harm, remorselessness, sadistic pleasure, etc.) that can help us place conduct
along this continuum. But I do not share Kramers confidence that there exists a category
of extravagant evil that is distinct enough from lesser kinds of evil so as to have such
significant implicationsto move from precluding the imposition of death to requiring it.
The metrics by which evil should be measured are contestable both in the number and their
weight. For example, should intentions count more than harms? Or vice versa? Should
remorse count at all? Of what significance is the existence of cognitive impairment, mental
illness, or trauma on the part of the evildoer? Kramer addresses these questions, but the
source of his answers remains obscure, and even armed with his answers, it remains
difficult to demarcate the line between extravagant evil and something less. While Kramer
acknowledges the concern that the range of borderline cases associated with each of the
chief concepts in the purgative rationale is disquietingly ample (253), he assumes that
there in fact IS a border that can be demarcated. I am not so sure.
Moreover, and quite a bit more emphatically, I doubt our capacity to tell who belongs in
such a category if indeed it does exist. This is the borderline cases problem to which
Kramer adverts. Kramer assumes that we have access to the quality of moral depravity that
lies beneath extravagantly evil conductpresumably through the actions, statements,
personal histories, and perhaps psychological evaluations of the perpetrators of such
conduct. But the essential nature of other people is more obdurately opaque than Kramer is
willing to admitespecially the nature of heinous evildoers, who by definition are
exceptional and thus presumably not very much like the rest of us. The opacity of evil is a
subject that literature often explores, by imagining the back-story of some opaque fictional
evillike John Gardners story of Beowolfs monster in the novel Grendel, or Jean Rhyss
perspective on Charlotte Brontes madwoman in the attic in Wide Sargasso Sea, or perhaps
most pertinently, the exploration of the crimes of the Wicked Witch of the West in Gregory
Maguires novel and the hit musical derived from it, Wicked.1 All of these fictional
explorations attempt to demonstrate through art that what we think we know about
unquestioned evil is, in fact, up for questioning and may be quite different from what it
appears so blatantly to be.
In particular, I worry about our ability to distinguish the quality of moral depravity from
the symptoms of defects or diseases that might not only remove an actor from the realm of
the extravagantly evil, but even place him or her outside the legitimate reach of the
criminal sanction. Kramer seems convinced that the line between evil on the one hand,
and sick or broken on the other, is sufficiently clear. For example, it seems evident to
Kramer that a mentally impaired woman who kills her children in order to save the
children themselves from Satan is not extravagantly evil, but a mentally impaired woman
who does so to prevent herself from going to Hell bears full responsibility for her
repellently evil conduct (246). However, if the latter womans belief was accuratethat
is, if she was truly threatened with eternal torture unless she killed her childrenthen she
might well have a defense of duress for doing so, or at the very least a strong mitigating
case in terms of deserved punishment. What makes Kramer confident that he, or we, can
tell the difference between terrible harms perpetrated by people with no moral compass and
1

See, Gardner (1989), Rhys (1966), and Maguire (2004).

123

Crim Law and Philos (2015) 9:367378

371

those perpetrated by people whose mental processes (rather than moral faculties) are
impaired? The line between depravity and disorder (or in clinical terms, the differential
diagnosis between psychopathology and other conditions) is not so clear. Even psychological experts, let alone judges and lay decision makers like juries, have difficulty
assessing the effects of mental illness, mental retardation, and childhood abuse (or other
forms of mental abnormality and trauma) on cognition, judgment, intention, and volitional
controlall of which are among the metrics that distinguish gradations of evil conduct.
The two contrasting jury verdicts in the Texas prosecution of Andrea Yates (a mentally ill
woman who drowned her five children to save them from Satan)one that rejected her
insanity defense and another that accepted it after her conviction was appealed and
remandedillustrate well this difficulty.2
Consider the recent case of Anders Behring Breivik, the Norwegian mass murderer who
killed 77 people and wounded hundreds of others, most of them children, in a horrific
bombing and shooting attack in July, 2011.3 Breivik calmly hunted down his victims at a
summer camp, publicly declared that he would have liked to kill more people, and smiled
and raised his fist in a fascist salute in court.4 Breiviks premeditated atrocities, the horrible
harms he inflicted, his evident enjoyment of the suffering of others, and his complete lack
of remorse make him a perfect fit for Kramers paradigm of extravagant evil. Yet the
prosecutors in the case sought to have the court declare Breivik insane, so as to have him
committed to a mental hospital for treatment rather than punished criminally.5 Despite the
prosecutions arguments, a five-judge panel declared Breivik sane, and he was sentenced to
the maximum penalty of 21 years, with a possible extension at the end, should he still be
considered a threat after serving his sentence.6 Even if Kramer might deem Breiviks
conduct a clear case of extravagant evil, the question of whether Breivik was evil or sick is
clearly one that others, including the government officials entrusted with investigating and
prosecuting the crime, found much more difficult to answer.
The difference between Kramer and the Norwegian prosecutors is not a difference over
ethics; rather, it is a difference over the wellsprings of human action. Kramers account of
extravagant evil depends upon a very strong view of self-authorshipstronger even than
that of many committed retributivists. Kramer acknowledges that offenders genetic predispositions, impairments, and environments may create constraints and pressures in
their lives (241), but he insists that evil offenders are reflective agents rather than automatons and that they form[] their profoundly evil characters within the limits of their
genes and their environments and are thus rightly held fully responsible for their heinous
crimes against humanity (243). This view stands in sharp contrast with the work of many
medical and social scientists that locates the roots of violent anti-social behavior in strong
genetic predisposition, social deprivation, and situational constraints, rather than in
autonomous character formation.7 In Kramers view, only when offenders have not
played any roles in forming their own monstrous characters should they be held less than
fully accountable for their heinous crimes (243, emphasis added). Kramer gives as
2

See Kadish et al. (2012, pp. 95960).

See Lewis and Lyall (2012).

See id.

See id.

See id.

See, e.g., Raine (2013) (exploring biological roots of crime); Goldberg (1996, p. 259) (maintaining that
malevolence is caused by the lack of fulfillment of some profound human need); Zimbardo (2007)
(emphasizing the importance of situational factors in producing evil conduct).

123

372

Crim Law and Philos (2015) 9:367378

examples a person whose character is profoundly altered by falling into a coma or by being
slipped a personality-altering drug. As for mental illness, Kramer would absolve from
criminal responsibility only those who are completely disengaged from reality or whose
insanity leads them to incorrectly assess the damagingness of their conduct (247).
Kramer contrasts mental aberrations that produce egregious misapprehension of moral
principles, which should not preclude full responsibility for evil under his purgative
rationale (246).
In Kramers account of responsibility, there are no shades of grey. Errors of moral
judgment are sharply delineated from cognitive impairments, mental illnesses, or traumarelated responses. Responsibility for egregious moral errors is either total or non-existent
(in rare and extreme circumstances). Only because he holds such an improbably sharply
delineated account of moral responsibility for evil can Kramer be satisfied that the problem
of vagueness in his scheme of purgative justice is manageable. Restricting capital
punishment to clear-cut cases of monstrously evil conduct and relying on the prosecutions burden of proof can work to manage the problem of vagueness only if we share
Kramers confidence in the clarity of his clear-cut cases. But Kramers clarity on the
issue of extravagant evil is founded on an implausibly rigid view of human character
formation and agency.
Moral doubts about extravagant evil The epistemic doubts described above give rise to
an obvious concern about false positives. If we cannot be as sure as Kramer that such a
thing as extravagant evil exists or that we can tell who belongs in such a category if it
does exist, we should worry more than he does about wrongly condemning offenders to
this categoryand by implication, wrongly condemning them to death. Moreover, the
ubiquitous problem of over-inclusion looms larger here than in many other legal contexts
because the eligible and the ineligible may be radically discontinuous rather than arrayed
along a spectrum. That is, those who may wrongly be deemed extravagantly evil and
thus eligible for capital punishment are likely to be those who are insane (or cognitively
impaired or victims of trauma) and thus may not be criminally responsible for their conduct
at all. Consequently, the false positives are not likely to be offenders who are almost
extravagantly evil, but rather those who are not correctly deemed evil at all in the first
place. Moreover, if I am correct that the line between evil and sick is much more fuzzy and
difficult to discern than Kramer alleges, then we should also worry that false positives will
be influenced by conscious or unconscious biases about who is likely disposed to be evil
that is, by racial (and other) biases. Thus, epistemic doubts about determinations of
extravagant evil necessarily give rise to moral doubts about the consequences that flow
from such determinations.
Moreover, Kramers purgative rationale for capital punishment generates a broader and
deeper moral qualm than the concern about false positives. The project of seeking out
extravagantly evil offenders with the purpose of purging them from a society so as to
restore that societys moral purity is a project that fails to acknowledge collective contributions to the conduct of extravagantly evil offenders. Such offenders are made, not
born. The communities in which heinous offenders commit their offenses doubtless have
contributed in some waysoften in crucially important waysto their characters and their
crimes. The collectives contributions may lie in failing to offer treatment for mental
illness or other defects, in failing to protect an offender from trauma, abuse or neglect, or
more affirmatively in creating institutional or cultural conditions that promote anti-social
violence. The purgative rationale depends on Kramers underlying account of complete
self-authorshipan account that is not only implausible as a descriptive matter but also

123

Crim Law and Philos (2015) 9:367378

373

objectionable as a moral matter in its radical separation of evil and purity, of guilt and
innocence.
Albert Camus famous and eloquent argument against the death penalty in Reflections
on the Guillotine elaborated powerfully on precisely this point, questioning whether
society could claim a degree of innocence necessary to justify the imposition of capital
punishment: Certainly the victim is innocent. But can society, which is supposed to
represent the victim, claim a comparable innocence? Is it not responsible, at least in part,
for the crime which it represses with such severity?8 Camus recognized that he was not
the first to insist on societal responsibility for crime, noting I need not continue a line of
argument which the most varied minds have elaborated since the eighteenth century.9 But
he did offer his own summing up of the familiar line of argument with the trenchant and
quotable observation that every society has the criminals it deserves.10 In light of the
inevitable existence of some degree of collective responsibility for individual acts of evil,
Camus rejected the implied moral message of purgation: To assert that a man must be
absolutely cut off from society because he is absolutely wicked is the same as saying that
society is absolutely good, which no sensible person will believe today.11
The persuasiveness of Camus rejection of the faulty moral accounting and the objectionable moral message implicit in a purgative rationale for capital punishment can be
assessed by a thought experiment. Imagine that at the end of World War II Germany itself
(rather than the Allied forces at Nuremberg) had criminally tried and executed a group of
the highest ranking Nazi officials pursuant to the purgative rationale that Kramer favors
and declared its duty to humanity to purge itself of extravagant evil done (even if it
continued to deal with lesser war criminals in a less absolute way). Surely, such an action
would have spawned compelling criticism of the German government for avoiding selfexamination by a form of convenient scape-goating. By placing the full weight of
responsibility for the most extravagant crimes of the Nazi regime on a small group of its
leaders, the government could fairly be seen as avoiding or denying the need for the
difficult task of investigating and acknowledging the complicity of many others further
down the chain of command and throughout German society in the worst excesses of the
Nazis. The purgative rationale for capital punishment invites this critique more than other
justificatory accounts because it suggests that what defiles a society is the continuing
presence of a heinous wrongdoer instead ofand in contrast tothe societys own contributions to the wrongdoers heinous crimes. It is not surprising that many years after the
judgments at Nuremberg, debate continues about how much of the weight of responsibility
for the terrible crimes of the Nazi regime should be placed on leaders in comparison to
ordinary citizens.12 Indeed, the purging by execution of Nazi officials as a result of the
Nuremberg trials may have been desirable to the Allied powers precisely because the
executions could symbolically (even if falsely) cleanse Germany as a society of the most
extravagant aspects of its war guilt so that it could be rebuilt and reintegrated into
European society.
In short, purgation of heinous criminals not only neglects to examine and acknowledge
collective responsibility for extravagant evil, it actively works against any such
8

Camus (1959, p. 30).

Id. at 31.

10

Id.

11

Id. at 47.

12

See, e.g., Goldhagen (1996).

123

374

Crim Law and Philos (2015) 9:367378

examination or acknowledgement in its falsely tidy balancing of the figurative accounting

books. It is no coincidence that a widely used curriculum for teaching schoolchildren about
the Holocaust is entitled Facing History and Ourselves.13 Yet this is precisely what
purgative capital punishment, born in Kramers musings about post-Holocaust justice, fails
to attempt.
Doubts about defilement Even more central to Kramers account of the purgative
rationale for capital punishment than the concept of extravagant evil is the concept of
defilement. Defining and locating responsibility for extravagant evil is important to
Kramers exposition because of the effect that such evil has on a societynamely, the
effect of defilement, which in Kramers view is a kind of moral taint or contamination
that requires a society to purge the source from its midst. Yet this central working part of
Kramers account is the least theorized aspect of his argument, despite the heavy lifting
that he wants it to do.
Most essentially, where does the idea of defilement come from? As noted above,
Kramer starts (and pretty much ends) with the biblical story of Achan, who stole war booty
from God and was consequently stoned and burned, along with his family, for his transgression. Kramer briefly suggests that Kants much-quoted passage about an island society
on the brink of dispersal having a duty to execute its last murderer might be read as a protopurgative argument for the death penalty (18485). But Achan gets the lions share of the
exposition, and this biblical emphasis is telling, as it is extremely difficult to make sense of
the idea of defilement without reference to seemingly pre-modern notions of a community
being collectively accountable to a wrathful diety for the presence of sin within its midst.
The concept of defilement by wrongdoing and the need for cleansing inexorably brings to
mind the Old Testament world of the sacrifice, the fast, and the mikvah (ritual cleansing
bath), of banishment, flood, and fire. Or, more ecumenically, defilement conjures the
ancient Greek world of Oedipus and Antigone, in which terrible consequences naturally
flowed from the breaking of powerful, even sacred, taboos around sex and death. Kramer
maintains that, despite the metaphorical boost he derives from the story of Achan, he is
relying on nothing supernatural in his account of defilement (224). Nor, he claims, is he
relying solely on emotions, on the feelings of revulsion or of outrage and disgust that
decent people experience in response to heinous crimes (225). Rather, he asserts that
defilement is the reaction-independent ethical consequence of extravagant evil (225),
claiming in essence that his is a post-Enlightenment account of ethical duty derived from
reason. But without relying on supernatural displeasure or visceral emotion, Kramers
argument has a circular quality. Defilement is the consequence of extravagant evil, because
extravagant evil defiles.
Kramers account of defilement as the source of a duty to execute suffers in comparison
to retributive accounts of duties to execute, which have a firmer grounding in more
attractive moral facts or values. The classic Kantian strand of retributivisim derives the
duty to mirror as much as possible a wrongdoers offense in its punishment from the
assumed moral fact of human autonomy. The uniquely human capacity for moral agency
gives rise to a collective duty to respect a wrongdoers autonomy by treating his or her
actions as authorizing their universal application. If one starts with the attractive and
plausible assumption of human autonomy, one can follow Kants reasoning to the Categorical Imperative. Similarly, the Hegelian strand of retributivism emphasizes the duty to
punish wrongdoers commensurate with their wrongdoing as the necessary consequence of
a commitment to the equality of persons. Once again, one can start with an attractive and
13

See Facing History and Ourselves, available at http://www.facinghistory.org/.

123

Crim Law and Philos (2015) 9:367378

375

plausible moral value (equality) and reason to a duty to punish commensurate with
wrongdoing to uphold the equality of victims. The connection between extravagant evil
and defilement is both less attractive (because of its sticky ties to ancient superstitions,
rather than to post-Enlightenment accounts of human autonomy and equality) and less
susceptible to reason than the claims of retributivism. And in the absence of Kramers
crucial account of defilement, even the most extravagant evil fails to create a duty to purge
it (and thus the case for capital punishment in such instances must fall back on some other
source of collective duty to punish, like retributivism, deterrence, or the like).
Kramer claims that his purgative rationale is more convincing than retributivism
because retributivism relies on the vague concept of desert, which fails to establish that
death, rather than some less severe sanction, is necessary to punish even the most heinous
crimes. Therefore, Kramer concludes, the retributive case for capital punishment fails the
Minimal Invasion Principle, an overarching moral principle limiting government action to
the least restrictive means necessary to achieve substantial governmental ends that Kramer
adopts from philosopher and death penalty opponent Hugo Bedau (4 n.3). In contrast,
asserts Kramer, the purgative rationale requires death and only death, because there is no
other means that can sufficiently purge the taint of defiling evil (25758). Despite my own
serious doubts about the persuasiveness of any moral grounding for capital punishment, I
would put my money on retributivism over purgation in a battle of the rationales.14 Kramers argument against retributivism is the weakest of his knockouts of competing
rationales, both because he has to rely on the importation of a principle (the Minimal
Invasion Principle) from outside of retributivism itself in order to defeat it, and because it
is not clear that the Minimal Invasion Principle does in fact defeat retributivisms claims in
support of capital punishment, at least in some cases. Even if the central concept of desert
is vague and does not always specify a precise punishment, why doesnt retributivism
specify death as the only proportional punishment for crimes of the kind of extravagance
that Kramer describes as defiling? After all, Kramer acknowledges that retributivism has
some limited determinacy, in that multiple answers will be correct but at the same time
a myriad of intelligible answers will be incorrect (136). Kramer summarizes this point
deftly: Limited determinacy is limited determinacy as well as limited determinacy (136).
But he fails to convincingly establish that the limited determinacy of retributivism does
not require death in the small subset of the most heinous cases that are his concern, even if
retributivism lacks similar specificity in many others factual settings. Indeed, during the
difficult-to-read passages in which Kramer patiently paints a vivid portrait of the extravagantly evil actions of his paradigmatic cases (Richard and Joseph), I repeatedly wondered
why it wasnt simply the case that Richard and Joseph deserved to die for their crimes (as a
retributivist would have it), rather than the more complicated, less intuitive, and less
reason-based case that they should die because we are collectively defiled by their conduct,
even though we are not collectively responsible for it.
Doubts about death After explaining his concept of defilement, Kramer expounds on the
duty of a community to respond to defilement by purgation. He explains that extravagantly
evil conduct demonstrates a repudiation of humankind through monstrous behavior
(232) and thus constitutes a crime against the whole of humanity (235). Therefore,
Kramer argues, a community incurs an ethical obligation to the whole of humanity to purge
an extravagantly evil offender by executing him because the offenders life stands as an
affront to the dignity of humankind and as a blot on the relationship between humankind
14
I have explored in greater detail the difficulties of combating retributive arguments in favor of capital
punishment in Steiker (2011).

123

376

Crim Law and Philos (2015) 9:367378

and his community, especially if his community is assigning large quantities of resources
to the sustainment of his existence (236). This is the last link in Kramers chain of
reasoning, and it, too, is subject to questionindeed, to questioning along the lines that
Kramer directs at other theories and practices of punishment. For example, just as Kramer
asks of retributivism, why isnt some less severe punishment sufficient to meet our collective obligation to respond to extravagant evil? After all, even the Old Testament from
which Achans story derives is full of alternative metaphorsstories of banishment as an
alternative to death, in response to transgressions that are arguably grosser than that of
Achan. Think of Adam and Eve expelled from the Garden of Eden; or Cain, marked and
banished to wander in the Land of Nod; or Jonah, swallowed but expelled alive by the
whale. Why isnt long-term or even life-long incarceration, our modern equivalent of
banishment, an adequate response even to extravagant evil? One can agree that our duties
to humankind urgently demand a response to crimes that demonstrate contempt for
humanity, but why that response must take the form of purgation through execution
remains mysterious.
The one argument that Kramer repeatedly offers for purgation through execution rather
than any other punishment is that only execution saves the community from expending its
resources to maintain the life of a heinous wrongdoer. But Kramer fails to explain why we
should not view paying for the imposition of a long-term or life-long sentence as an
expenditure of resources to punish the offender, rather than as an expenditure of resources
to sustain the offenders life. After all, Kramer recognizes that there are massive costs
involved in the special litigation process and post-conviction appeals that capital cases
requireexpenses that generally far exceed the cost of even lifetime incarceration.15 Yet
he does not object to these expenditures, recognizing that they are simply the necessary
costs of carrying out the sentence that he prefers. One could similarly view the costs of
maintaining the life of an offender as the costs of imposing a punishment of incarceration
that is otherwise just. It seems odd, and oddly formalistic, to invoke the costs of life
imprisonment as an argumentindeed, the sole argumentagainst its justice. Retributivisms answerthat life imprisonment is simply not commensurate with the gravity of
the offenders crimeonce again seems a more powerful and persuasive reason for deathand-nothing-but.
Moreover, the arguments that Kramer levels against any kind of torture as punishment
also can be aimed at the punishment of death. Kramer is clear that torture is never
permissible as punishment, because it is fundamentally at one with the worst type of evil
(63). What makes torture impermissible in Kramers view is its intentional exploitation of
the pain-experiencing capacity of human beings (63). There has been substantial litigation
in the United States about whether even execution by lethal injection carries too great a
risk of physical pain to be tolerated under the constitutional prohibition of Cruel and
Unusual Punishment.16 But even putting physical pain aside, surely psychological pain
should count as well in any prohibition of torture: for example, I doubt that Kramer (or
anyone else) would exclude from the definition of torture the rape or torture of a persons
child in his or her presence. Once we are on the ground of psychological pain, the lengthy
and indeterminate seclusion of condemned inmates on Death Row as they await their
executionsthe so-called Death Row Phenomenoncould plausibly be deemed a form of

15

See Steiker and Steiker (2010, p. 117).

16

See, e.g., Baze v. Rees, 553 US 35 (2008) (rejecting Eighth Amendment cruel and unusual punishment
challenge to lethal injection protocol).

123

Crim Law and Philos (2015) 9:367378

377

torture, as some courts have recognized.17 Indeed, not only death, but certain conditions of
confinementsuch as extended solitary confinementmight also plausibly be deemed a
form of torture. In sum, Kramer is too quick to conclude both that death and only death is
necessary to respond to extravagant evil, and (assuming that it is necessary) that death is
permissible within Kramers own account of permissibility.
But the strongest argument for the impermissibility of the death penalty in the purgative
form that Kramer advocates comes from outside of his own rubric of permissibility. The
most striking aspect of Kramers purgative rationale is how closely (though no doubt
unintentionally) it mirrors the Nazis own twisted rationale for their extermination campaign against Jews and other enemies of the Reichthe need to purge from Aryan society
a kind of defiling or contaminating taint. Kramer seeks to locate the moral foundation of
his purgative rationale for capital punishment in the duty that each community owes to
humankind as a whole to respond appropriately to heinous crimes against humanity.
However, in considering what duties we might plausibly owe to humankind as a consequence of our humanity and in response to atrocities like the Holocaust, I find more
plausible and foundational a duty not to arrogate the power to define for ourselves the
boundaries of the human. The central problem with the very concept of defilement is that it
implies a necessary separation between the body of humankind and a kind of cancer or
excrement that must be excised or voided. Punishment is one thing, as its justice can and
should be consistent with respect for the humanity of the punished; purgation is another
thing entirely. The project of purgation is at its very root a project of narrowing the
boundaries of the fully human. In an effort to cleanse evils stain from humanity, Kramers
purgative rationale shrinks the fabric of humanity in the wash. It would be ironic indeed if
this troubling proposal were to become one of the enduring lessons of the Holocaust.
Acknowledgments I thank Fionnuala N Aolain and Jordan Steiker for helpful conversation and
comments.

References
Baze v. Rees, 553 U.S. 35 (2008).
Camus, A. (1959). Reflections on the Guillotine. Trans. Richard Howard. Michigan City, Indiana: FridtjofKarla Publications.
Facing History and Ourselves, http://www.facinghistory.org/.
Gardner, J. (1989). Grendel. Vintage: New York.
Goldberg, C. (1996). Speaking with the Devil: Exploring Senseless Acts of Evil. New York: Penguin Books.
Goldhagen, D.J. (1996). Hitlers Willing Executioners: Ordinary Germans and the Holocaust. New York:
Alfred A. Knopf.
Kadish, S., Schulhofer, S., Steiker, C., and Barkow, R. (2012). Criminal Law and Its Processes, Ninth
Edition. New York: Wolters Kluwer.
Lewis, M. and Lyall, S. Norway Mass Killer Gets the Maximum: 21 years, New York Times, Aug. 24,
2012.
Maguire, G. (2004). Wicked: The Life and Times of the Wicked Witch of the West. New York: Harper
Collins.
Raine, A. (2013). The Anatomy of Violence: The Biological Roots of Crime. New York: Pantheon Books.
Rhys, J. (1966). Wide Sargasso Sea. New York: Popular Library.
Soering v. United Kingdom, 161 Eur. Ct. H.R. (ser. A) (1989).
17
See, e.g., Soering v. United Kingdom, 161 Eur. Ct. H.R. (ser. A) (1989) (holding that the extradition of a
young German national to face capital murder charges in the United States would violate Article 3 of the
European Convention on Human Rights, guaranteeing the right to be free from inhuman and degrading
treatment, because of the likelihood that Soering would experience the death row phenomenon).

123

378

Crim Law and Philos (2015) 9:367378

Steiker, C.S. (2011). The Death Penalty and Deontology, In J. Deigh and D. Dolinko, (Eds.), The Oxford
Handbook of Philosophy of Criminal Law. Oxford: Oxford University Press.
Steiker, C.S. and Steiker, J.M. (2010). Cost and Capital Punishment: A New Consideration Transforms an
Old Debate, University of Chicago Legal Forum.
Zimbardo, P. (2007). The Lucifer Effect: Understanding How Good People Turn Evil. New York: Random
House.

123