German Idealism and the Concept of Punishment (Modern European Philosophy)

This page intentionally left blank

GERMAN IDEALISM AND THE CONCEPT OF PUNISHMENT Against the background of early modernism – a period that justified punishment by general deterrence – Kant is usually thought to represent a radical turn toward retributivism. For Kant, and later for Fichte and Hegel, a just punishment respects the humanity inherent in the criminal, and serves no external ends: it is instituted only because the criminal deserves it. In this original study, Jean-Christophe Merle uses close analysis of texts to show that these philosophers did not in fact hold a retributivist position, or even a mixed position; instead he traces in their work the gradual emergence of views in favor of deterrence and resocialization. He also examines Nietzsche’s view that morality rests on the rejection of retribution. His final chapter offers a challenge to the retributivist position, and a defense of resocialization, in the context of current legal theory and practice concerning the punishment of crimes against humanity. j e a n - ch r i s t o p h e m e r l e is a senior researcher in philosophy at the University of Tours, an Honorary Professor at the University of Saarland, and a lecturer at the University of Tu¨bingen.

MODERN EUROPEAN PHILOSOPHY General Editor ROBERT B. PIPPIN, University of Chicago Advisory Board GARY GUTTING, University of Notre Dame ROLF-PETER HORSTMANN, Humboldt University, Berlin

Some recent titles Daniel W. Conway: Nietzsche’s Dangerous Game John P. McCormick: Carl Schmitt’s Critique of Liberalism Frederick A. Olafson: Heidegger and the Ground of Ethics Gu¨nter Zo¨ller: Fichte’s Transcendental Philosophy Warren Breckman: Marx, the Young Hegelians, and the Origins of Radical Social Theory William Blattner: Heidegger’s Temporal Idealism Charles Griswold: Adam Smith and the Virtues of Enlightenment Gary Gutting: Pragmatic Liberalism and the Critique of Modernity Allen Wood: Kant’s Ethical Thought Karl Ameriks: Kant and the Fate of Autonomy Alfredo Ferrarin: Hegel and Aristotle Cristina Lafont: Heidegger, Language, and World-Disclosure Nicholas Wolsterstorff: Thomas Reid and the Story of Epistemology Daniel Dahlstrom: Heidegger’s Concept of Truth Michelle Grier: Kant’s Doctrine of Transcendental Illusion Henry Allison: Kant’s Theory of Taste Allen Speight: Hegel, Literature, and the Problem of Agency J. M. Bernstein: Adorno Will Dudley: Hegel, Nietzsche, and Philosophy Taylor Carman: Heidegger’s Analytic Douglas Moggach: The Philosophy and Politics of Bruno Bauer Ru¨diger Bubner: The Innovations of Idealism Jon Stewart: Kierkegaard’s Relations to Hegel Reconsidered Michael Quante: Hegel’s Concept of Action Wolfgang Detel: Foucault and Classical Antiquity

Robert M. Wallace: Hegel’s Philosophy of Reality, Freedom, and God Johanna Oksala: Foucault on Freedom Be´atrice Longuenesse: Kant on the Human Standpoint Wayne Martin: Theories of Judgment Heinrich Meier: Leo Strauss and the Theologico-Political Problem Otfried Ho¨ffe: Kant’s Cosmopolitan Theory of the Law and Peace Be´atrice Longuenesse: Hegel’s Critique of Metaphysics Rachel Zuckert: Kant on Beauty and Biology Andrew Bowie: Music, Philosophy and Modernity Paul Redding: Analytic Philosophy and the Return of Hegelian Thought Kristin Gjesdal: Gadamer and the Legacy of German Idealism

GERMAN IDEALISM AND THE CONCEPT OF PUNISHMENT

JEAN-CHRISTOPHE MERLE translated from the German by

JOSEPH J. KOMINKIEWICZ with

JEAN-CHRISTOPHE MERLE and

FRANCES BROWN

CAMBRIDGE UNIVERSITY PRESS

Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo, Delhi, Dubai, Tokyo Cambridge University Press The Edinburgh Building, Cambridge CB2 8RU, UK Published in the United States of America by Cambridge University Press, New York www.cambridge.org Information on this title: www.cambridge.org/9780521886840 © Jean-Christophe Merle 2009 This publication is in copyright. Subject to statutory exception and to the provision of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published in print format 2009

ISBN-13

978-0-521-88684-0

Hardback

Cambridge University Press has no responsibility for the persistence or accuracy of urls for external or third-party internet websites referred to in this publication, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

. . . we all know today’s executioners are humanists. Albert Camus

CONTENTS

page xi

Preface

xiii

List of abbreviations Introduction

1

part i desert as the sole justification for punishment 1 The two Kantian concepts of right

17

2 Kant’s legal justification of punishment

44

3 Kant’s moral justification of punishment

72

part ii punishment as a means of rehabilitation 4 Fichte’s “expiation contract”

87

5 Hegel’s “negation of crime”

107

part iii retributivist inhumanity 6 Nietzsche and punishment without remorse

149

7 What is the purpose of punishing crimes against humanity?

171

Conclusion

187

Bibliography

197

Index

204

ix

PREFACE

“Nemo prudens punit quia peccatum est sed ne peccetur,” says Seneca in De ira, and many philosophers who have come after him recommend such a justification of punishment by deterrence. Since Immanuel Kant, a completely different concept has spread among philosophers, considerably more so than among legal scholars and lawyers. According to Kant, the question of justification of punishment should not read: For what purpose punish? Rather, according to Kant’s absolutist or categorical imperative regarding punishment, punishment can only be carried out because the malefactor is deserving of the punishment. Everything else is allegedly unjust, and is detrimental to the malefactor’s human dignity as a moral subject. Such a theory of retributive justice, which draws not only from Kant but also from G. W. F. Hegel, inspires a great deal of fascination in many philosophers, but that notwithstanding it still stands on shaky ground. A precise analysis of Kant’s and Hegel’s philosophy of law and morality leads rather to a special form of deterrence theory. I will attempt to conduct this analysis within the confines of this book. The analysis begins with Kant, continues with J. G. Fichte and Hegel, leads to Friedrich Nietzsche, and then concludes with a discussion of the justification of punishment for crimes against humanity. This closing discussion should be seen as the touchstone. Should my position be able to explain this difficult case, then it should be even more able to explain cases of lesser difficulty. I would like to thank Manfred Frank and Anton Schindling for their comments on the manuscript as well as two anonymous referees from Cambridge University Press. Special thanks go to Sharon Byrd, Philippe Coppens, Roman Eisele, George Fletcher, Thomas Grundmann, Jan C. Joerden, Matthias Kaufmann, John Kleinig, Eugeˆnio Pacelli de xi

xii

preface

Oliveira, Herve´ Pourtois, Alexandre Travessoni Gomes, Luiz Moreira and Jean-Claude Wolf. I have also benefited from questions and remarks from: European Network “Applied Global Justice,” KantGesellschaft, Fichte-Gesellschaft, CAPPE (ANU Canberra), Chaire Hoover and Centre de Philosophie du Droit (Louvain), University of Grenoble, PUCRS Porto Alegre, UFMG Belo Horizonte, UFSC Florianopolis, Ankara Bar Association, University of Fribourg, Graduiertenkolleg Globale Herausforderung (Tu¨bingen), Charles Sturt University at Wagga Wagga, and a conference of the European Network “Global Justice” in Otzenhausen, as well as from my seminars on these topics in Aachen, Saarbru¨cken, Tours and Tu¨bingen. Last, but not least, I am very grateful to Hilary Gaskin of Cambridge University Press, as well as to Frances Brown.

ABBREVIATIONS

Kant GMS

Idee

KpV

KrV

Pa¨d

Rel

Groundwork of the metaphysics of morals (Grundlegung zur Metaphysik der Sitten) (1785, Ak iv:385–464) Immanuel Kant, Practical philosophy, ed. Mary Gregor (Cambridge: Cambridge University Press, 1996), pp. 37–108 Idea for a universal history with a cosmopolitan purpose (Idee zu einer allgemeinen Geschichte in weltbu¨rgerlicher Absicht) (1784, Ak viii:15–32) Immanuel Kant, Political writings, ed. Hans Reiss, trans. H. B. Nisbet, second edition, (Cambridge: Cambridge University Press), pp. 41–53 Critique of practical reason (Kritik der praktischen Vernunft) (1788, Ak v:1–164) Immanuel Kant, Practical philosophy, ed. Mary Gregor (Cambridge: Cambridge University Press, 1996), pp. 133–272 Critique of pure reason (Kritik der reinen Vernunft) (1st edn 1781, 2nd edn 1787, Ak iii:1–552) Page numbers are from the second edition Immanuel Kant, Critique of pure reason, ed. and trans. Paul Guyer and Alan W. Wood (Cambridge: Cambridge University Press, 1997) Lecture On pedagogy (Pa¨dagogik) (1803, Ak ix:437–99) (no translation) Religion within the boundaries of mere reason xiii

xiv

RL

TL

VE

ZeF

Fichte GNR

Hegel GPhR

NRSW

list of abbreviations

(Religion innerhalb der Grenzen der bloben Vernunft) (1793, Ak vi:1–202) Immanuel Kant, Religion within the boundaries of mere reason, in Kant, Religion within the boundaries of mere reason and other writings, ed. Allen Wood and George di Giovanni (Cambridge: Cambridge University Press, 1998), pp. 31–192 The doctrine of right (Part 1 of The metaphysics of morals) (1st edn 1797, 2nd edn 1798, Ak vi:203–372) Immanuel Kant, Practical philosophy, ed. Mary Gregor (Cambridge: Cambridge University Press, 1996), pp. 363–506 The doctrine of virtue (Part 2 of The metaphysics of morals) (1st edn 1797, 2nd edn 1798, Ak vi:373–493) Immanuel Kant, Practical philosophy, ed. Mary Gregor (Cambridge: Cambridge University Press, 1996), pp. 507–615 Immanuel Kant, Lectures on ethics, ed. Peter Heath and J. B. Schneewind, trans. Peter Heath (Cambridge: Cambridge University Press, 1997) (Eine Vorlesung Kants u¨ber Ethik)(c. 1875–80, Ak xxvii:286) Toward perpetual peace (Zum ewigen Frieden) (1795, Ak vii:341–86) Immanuel Kant, Practical philosophy, ed. Mary Gregor (Cambridge: Cambridge University Press, 1996), pp. 311–52 Foundations of natural right, according to the principles of the Wissenschaftslehre (Grundlage des Naturrechts nach Principien der Wissenschaftslehre) Johann Gottlieb Fichte, Foundations of natural right: Grundlage des Naturrechts nach Principien der Wissenschaftslehre, ed. Frederick Neuhouser, trans. Michael Baur (Cambridge: Cambridge University Press, 2000) Elements of the philosophy of right (Grundlinien der Philosophie des Rechts) G. W. F. Hegel, Elements of the philosophy of right, trans. H. B. Nisbet (Cambridge: Cambridge University Press, 1991) Lecture on Natural law and the science of state (Vorlesung u¨ber Naturrecht und Staatswissenschaft) (1818–19) (No translation)

list of abbreviations

PhR

xv

Lecture on The philosophy of right (Vorlesung u¨ber Philosophie des Rechts (1824–5)) (No translation)

Nietzsche GdM On the genealogy of morality (Zur Genealogie der Moral) Friedrich Nietzsche, On the genealogy of morality, ed. Keith Ansell-Pearson, trans. Carol Diethe (Cambridge: Cambridge University Press, 2007), pp. 1–128 WuL On truth and lies in a nonmoral sense ¨ ber Wahrheit und Lu¨ge) (U Friedrich Nietzsche, Writings from the early notebooks, ed. Ladislaus Lo¨b, Raymond Geuss and Alexander Nehamas (Cambridge: Cambridge University Press, forthcoming)

Note on translations of primary and secondary literature Every effort has been made to find published English translations of all foreign-language texts. Where there is no published translation, German passages have been translated for the purposes of this book. The reader should assume that if a quotation is from a German work for which no English-language citation is given, then the text has been newly translated. In the interests of simplifying the footnote citations, this will not always be noted unless there is a specific need for clarification.

INTRODUCTION

A nearly total unanimity prevails with regard to the fundamental necessity of a public penal system.1 Even among those few who advocate the abolition of all punishments, a large majority advocates instituting alternatives to the usual prison sentence, rather than calling for the abolition of punishment without anything to replace it. When seen in this light, the existence of public penal law can be regarded as being completely justified. The manner in which punishment might actually be justified, however, remains just as controversial a subject as determining the appropriate amount of punishment. This is because these issues are closely related to one another. Every theory of punishment currently advocated shares the rejection of the system of punishment which was prevalent in the early modern age. This rejected system, illustrated by such penal provisions as the Constitutio criminalis Carolina, enacted in 1532, was placed in opposition to the modern system of punishment by Michel Foucault in Discipline and punish. The early modern system differs from the modern system in the sense that the latter prefers either prison sentences or (if any) the most painless and most decent death sentences possible.2 It is worth noting that well into the eighteenth century more than one hundred crimes were capital offenses. Torture,

1. For an example of the few exceptions, see Herman Bianchi, “Abolition: assensus and sanctuary,” in Alexander R. Duff and David Garland (eds.), A reader on punishment (Oxford: Oxford University Press, 1994), pp. 336–51. 2. See the beginning of Michel Foucault, Discipline and punish: the birth of the prison, trans. Alan Sheridan, second edition (New York: Random House, 1995); see also the Constitutio criminalis Carolina in Friedrich-Christian Schroeder (ed.), Die Carolina: die Peinliche Gerichtsordnung Kaiser Karls V. von 1532 (Darmstadt: Wissenschaftliche Buchgesellschaft, 1986).

1

2

german idealism and the concept of punishment

which was already systematically employed as an interrogation method, was often a component of the punishment, as well as constituting an intensification of the death sentence. Even though torture as a means for investigation and security is currently being propagated again3 and even though the death penalty is still supported,4 there are no theorists to be found who would come out in support of a return to early modern practice. All contemporary theorists show themselves to be guided by the humanitarian spirit of the Eighth Amendment to the United States Constitution: “Excessive bail shall not be required, nor excessive fines imposed, nor cruel and unusual punishments inflicted.” What exactly this humanism consists of, and where the limits of this humanism might lie, are still controversial issues. However, the fundamental discrepancy is generally seen to be somewhere between the theories of retributive justice and general deterrence. Retributivist theories justify the punishment of a criminal on the grounds that retribution is demanded by justice to compensate for the inequities created by the crime. On this basis, one may make a subsidiary distinction and ask whether what needs to be compensated for is the gravity of the offense itself or the malevolence of the criminal that was illustrated by the said offense. Theories of general deterrence, on the other hand, justify the punishment on the grounds that all of the citizens will be deterred from carrying out an offense before it occurs, either through the threat of a certain punishment or through the enforcement of the said punishment; the latter option relies on the example it displays. On the one hand, the modern advocates of general deterrence (examples include Thomas Hobbes, Samuel von Pufendorf, Christian Wolff, Cesare Beccaria, Anselm Feuerbach and Arthur Schopenhauer) consider that the uselessness inherent in punishments that do not serve to deter others from committing crimes is inhumane. The current system of positive criminal law also requires that each punishment contains elements of general prevention as stated in the initial paragraphs of the German Penal Code: 3. See, for example, Winfried Brugger, “Darf der Staat ausnahmsweise foltern?,” Der Staat, 35 (1996), 67–97; and “Vom unbedingten Verbot der Folter zum bedingten Recht auf Folter?” Juristenzeitung, 35, no. 4 (18 February 2000), 165–73. For a detailed repudiation of torture, see, for example, Henry Shue, “Torture,” Philosophy and Public Affairs, 7, no. 2 (1977–88), 124–43. 4. See Ernest van den Haag, “Why capital punishment?,” Albany Law Review, 54, nos. 3–4 (1990), 501–14.

introduction

3

By serving a prison term, a prisoner should eventually become able to lead a life in which he or she commits no crimes (which is the goal of the execution of punishment). The completion of a term of imprisonment also serves to protect the general public from further crimes.5

On the other hand, theories of general deterrence often draw the criticism that they treat criminals inhumanely, because the intended aim of punishment is conceived of solely to serve the interests of the other citizens without taking into consideration the dignity of the criminal. This criticism is leveled in its most intense form when it is postulated that general deterrence allows the punishment of an innocent person.6 This objection can be understood in at least two different ways. General deterrence can be objected to on the one hand, for the reason that the aim of punishment ignores the interests of the convicted, or on the other hand, in accordance with retributivism, for the reason that any kind of interest – whether of the criminal or of the fellow citizens – should be disregarded by the penal sentence, because a punishment justified in a retributivist way is merely about inflicting on the criminal what he or she intrinsically deserves because of the deed. A punishment situated in the retributive model should be concerned with inflicting a punishment that is in line with what the prisoner merited because a certain crime was perpetrated. The latter objection is raised by retributivism. The former objection is raised by positions that hold that the rehabilitation of the perpetrator should be the punishment’s goal. Admittedly, the latter position recognizes that for the purpose of reaching rehabilitation a certain period of time of specific deterrence may become necessary, in which society is protected from further crimes through incapacitation of the criminal. 5. From the German National Code of Enforcement of Sentences, the “Strafvollzugsgesetz” (StVollzG), published by the German Federal Ministry of Justice, } 2. The above passage is the precept to the later statutes in the codex. Compare as well with the judgment of the German Federal Court of Justice (Bundesgerichtshof) on December 8, 1970, 1 StR 353/70. (Translation mine.) 6. Cf. Peter Koller, “Probleme der utilitaristischen Strafrechtfertigung,” Zeitschrift fu¨r die Gesamte Strafrechtswissenschaft, 91 (1979), 45–95; Kristian Ku¨hl, Die Bedeutung der Rechtsphilosophie fu¨r das Strafrecht (Baden-Baden: Nomos, 2001), p. 29; Peter Landau, “Karl Christian Friedrich Krauses Rechtsphilosophie,” in Klaus-Michael Kodalle (ed.), Karl Christian Friedrich Krause (1781–1832): Studien zu seiner Philosophie und zum Krausismo (Hamburg: F. Meiner, 1985), pp. 80–92 (p. 29). For a refutation of this objection, compare Fred Rosen, “Utilitarianism and the punishment of the innocent: the origins of a false doctrine,” Utilitas, 9, no. 1 (March 1997), 23–37.

4

german idealism and the concept of punishment

Retributivism criticizes theories of general deterrence as well as theories of rehabilitation considered as the aim of punishment for much the same reason, that is, because these theories treat punishment as a mere means to an end. Retributivists themselves hold that punishment ought to be justified without any reference to further goals, invoking the rationale that the criminal deserves it because he or she knowingly violated the law. It is for this reason that legal theorists term retributivism an absolute theory, because, according to retributivism, punishment represents a good that does not depend on any goal. On the contrary, in the “relative” theories the justification of punishment always depends on its relation to a goal. According to the proponents of retributivism,7 it derives its superiority from the fact that it alone – as the only theory of criminal justice that views the punishment solely as a goal in itself – treats the malefactor not as a simple means to an end, but as a subject possessing human dignity. In this work, I will attempt to refute these theses. I hope to show that it is not retributivism but rather rehabilitation that meets this requirement. Proponents of rehabilitation obviously consider the aim of punishment to be to grant the criminal the best “possible” status, by which it is understood that this is the way of treating the criminal that is both the most benevolent and still compatible with the protection of society against further crimes. Thus, with rehabilitation as an aim in punishment, clear limits are set for specific deterrence. Without specific deterrence, rehabilitation would be unthinkable, for if there were no public enforcement of the law, there could in turn be no reintegration back into society, for there would be no rule of law into which a criminal could be reintegrated after the rehabilitation has been completed. Unlike rehabilitative and specific deterrent punishments, retributivism does not concern itself with the future of the malefactor beyond the duration of the punishment. In this respect, the theory of rehabilitation is the only one that can categorically exclude those sorts of punishment that – as mentioned at the beginning of this introduction – the proponents of all the theories of punishment reject resolutely: the “cruel and unusual punishments.” The vice president of the Federal Constitutional Court of Germany Winfried Hassemer rightly 7. See, for example, Otfried Ho¨ffe, Gerechtigkeit: eine philosophische Einfu¨hrung (Munich: C. H. Beck, 2004), p. 83.

introduction

5

observes: “The goal of rehabilitation is necessarily contained in opting for prison sentences. Corporal and capital punishments do not need any further justification than retribution.”8 Even when one assumes that the punishment is merited and therefore justified as being compensation either for the crime or for the criminal’s own wickedness, and asserts the claim that one should treat the criminal humanely, this allegedly humane view of the criminal should also concern itself with the situation in which the criminal will be after the complete term of imprisonment has been served. The absence of appropriate rehabilitative measures leads not only directly to the creation of a durable criminal environment with an ensuing reduction in public safety,9 but also to a state in which the criminal is punished twice for a crime, by being continually stigmatized, instead of enabling him or her to express remorse and to reach a reconciliation with society.10 Therefore, concern for the convict’s future after the sentence necessitates appropriate treatment while the sentence is being served. Retributivism may attempt to fulfill this requirement of humaneness, along the lines of what Paul Ricur attempted.11 In doing so, retributivism stumbles upon what Hassemer terms the “antinomy of punitive goals,” which refers to the fact that in many cases the various existing theories of punishment do not allow for the same amount of punishment.12 Hassemer observes that A period of punishment limited by the proportionality principle and as required by the goal of retribution normally does not suffice for a treatment, so that the goal of rehabilitation itself will fail. A period of punishment can also be too long for a reasonable treatment of the prisoner.13

In view of this antinomy of the punitive goals, a priority rule must be set. Either retributivism should be deemed to be the primary aim and rehabilitation the secondary aim – meaning that the rehabilitation will be carried out only as far as it does not interfere with the retribution – or the rehabilitation should be given priority over the retribution. 8. Winfried Hassemer, Einfu¨hrung in die Grundlagen des Strafrechts, second edition (Munich: C. H. Beck, 1990), p. 286. 9. Cf. John Braithwaite, Crime, shame and reintegration (Cambridge: Cambridge University Press, 1989), p. 102. 10. Cf. Braithwaite, Crime, shame and reintegration, p. 101; Hassemer, Grundlagen des Strafrechts, p. 289. 11. Paul Ricur, Le Juste (Paris: Editions Esprit, 1995), p. 203. 12. Hassemer, Grundlagen des Strafrechts, p. 291. 13. Hassemer, Grundlagen des Strafrechts, p. 291.

6

german idealism and the concept of punishment

The form of retributivism that is given priority regards its moral superiority as stemming from its being the only genuinely humane justification for punishment. It sees itself as better than rehabilitation for the reason that it bases itself not upon concern for the criminal, but instead solely upon his or her merit – or for that matter, upon his or her guilt, responsibility or malevolence. However, in doing so, retributivism ignores the following points. First, the following differentiation is necessary. Retribution can be understood, on one hand, in its minimalist sense, as meaning that the guilt of the punished, without any exception, should be the prerequisite for any given punishment.14 According to this understanding, all theories of punishment, that is, theories both of general deterrence and of rehabilitation, are retributivist theories.15 On the other hand, retribution can also be understood to mean (1) a response to the offense that strives to provide equal compensation for the criminal’s merit or guilt and (2) that this equal compensation is the sole just punishment. Thus, it excludes any goal in punishment (for instance, general deterrence, rehabilitation and specific deterrence). This is what one usually understands under the term retributivism. Retributivism consists in the acceptance of the latter (disputable) thesis. When I speak of retribution in this book, I will be making reference to the latter understanding of retribution. Secondly, in the justification of the punishment as retaliation, it is not the criminal’s future, but rather his or her past that is taken exclusively into account. As we have seen, this occurs with appeal either to the criminal’s merit or to his or her guilt. The concept that lies at the root of this guilt can be from a modern, humane perspective only the concept of responsibility. In fact, in every contemporary retributive theory – as opposed to the cases of the deterrent and rehabilitative theories – there is the imperative to respect the responsibility of the criminal as constituting his or her own dignity and to respond to this dignity with equal compensation. Throughout this book, I will proceed as would a retributivist, assuming that a human being’s responsibility for his or her actions, as opposed to other living beings, is actually what constitutes the special status of human beings (their dignitas). But then, the following 14. Cf. Otfried Ho¨ffe, Gibt es ein interkulturelles Strafrecht? Ein philosophischer Versuch (Frankfurt a.M.: Suhrkamp, 1999), p. 72. 15. Cf. Ulfried Neumann and Ulrich Schroth, Neuere Theorien von Kriminalita¨t und Strafe (Darmstadt: Wissenschaftliche Buchgesellschaft, 1980), p. 6.

introduction

7

differentiation must be taken into account. There are always two dimensions to the assertion that a human being bears responsibility for his or her actions. On the one hand, this responsibility means that different actions will lead to different consequences, and especially that all actions, which from the perspective of morality are valued differently, will also lead in some cases to different consequences. Among other things, it means that actions violating the law necessarily could or should lead to a worse situation. On the other hand, the human’s status as a being capable both of reason and of assuming responsibility is an inalienable status that cannot be taken away from that human or any other human. The assertion that the human being’s past illegal actions should carry consequences should not lead us to stop treating this human being as a being capable of reason, except when this prevents his or her fellow human beings from exercising that same status. Otherwise, the perpetrator would be treated as a person capable of assuming responsibility only up to the crime; after the conviction, however, the criminal would lose this status, that is, his or her worth. In this way, retributivism would lack exactly the sort of behavior toward the criminal which it regards as its moral superiority over the other theories of punishment. In short, retributivism does not sufficiently differentiate between the actor and the action. In this respect, the viewpoint that would trace a retributivist influence back to Christianity proves not to be truly convincing.16 Rehabilitation is equally rooted in the Christian tradition, as the rehabilitation theorist John Braithwaite suggests in his plea for “reintegrative shaming”: It is shaming which labels the act as evil while striving to preserve the identity of the offender as essentially good. It is directed at signifying evil deeds rather than evil persons in the Christian tradition of “hate the sin and love the sinner.”17

Thirdly, retributivism relies on the assumption that the consequences of the crime, for which the criminal should take responsibility, should consist of providing equal compensation, because that is what justice demands. While retributivism focuses in this manner on demanding an equivalent punishment, it misses an indisputable consequence of the crime : at least temporarily, the existence of a commonwealth between the criminal and other citizens is made impossible by the crime. 16. For this viewpoint, cf. Claus Roxin, Strafrechtliche Grundlagenprobleme (Berlin: De Gruyter, 1973), p. 3; and Neumann and Schroth, Neuere Theorien, p. 13. 17. Braithwaite, Crime, shame and reintegration, p. 101.

8

german idealism and the concept of punishment

A short thought experiment may illuminate this point. Let us assume the retributivist position as the starting point for this experiment, which is a position that holds that the severity of punishment should be equivalent to the gravity of the crime and that this alone should represent – without any specific aim of punishment – the only justified solution. Then, let us assume that in this fashion we reach the conclusion that a criminal has merited a twenty-year prison sentence. It is usually the case that serving the prison term begins, at the very latest, after the judgment has taken effect and after all recourse has been exhausted (in reality, however, the criminal – still considered officially innocent – has already been in custody either since the investigation commenced or since the issue of the arrest warrant). In this case, retributivism should raise no objections if the enforcement of the sentence were to be postponed. If the criminal were, for example, twenty years old, this would allow him or her to be in prison from age thirty to age fifty, but to remain a free person from twenty to thirty.18 It is not only for pragmatic reasons (for instance, because of administrative concerns regarding prison capacity) that our society would reject such a reform of the prison system; our society would find this fundamentally and wholly unacceptable because it would severely endanger public safety. Above all, such a reform would be rejected because of a specific deterrent rationale. Now, the danger to society is clearly consequent to the crime. Following the specific deterrence model, the criminal is therefore liable for this consequence caused by his or her criminal offense. Whereas it is controversial whether the moral demand of retribution for the guilt that is addressed by retribution can be seen as a consequence of the crime, the danger to society is an unquestionable consequence arising from the actions of the criminal. Retributivism, however, does not actually take into account the degree to which society is endangered by these crimes, which is at odds with retributivism’s own self-portrayal of itself as being the only theory of criminal justice that requires the criminal to shoulder the burden of the consequences of his or her actions. Fourthly, the disregard shown for the consequences of the action attests to a disregard for the legal dimension of crime as well as for the punishment of a crime. Retributivism focuses on the guilt of the perpetrator. Admittedly, it also emphasizes that the punishment of 18. Such a postponement is actually possible under German law, though admittedly only in a limited number of cases with short sentences.

introduction

9

the criminal represents justice for the victim.19 Apart from that, concern for the commonwealth does not play a role in retributive theory’s justification of punishment. Consequently, if in the retributivist justification of punishment only the criminal is taken into account, this begs the question of why the punishment falls under the jurisdiction of the judicial powers, that is, why it belongs to the commonwealth – which, in other cases, maintains distance from the private sphere of the individual. If, in the retributivist justification of punishment, only the criminal and the victim are taken into account, then it is inevitable to ask why the punishment could not just be decided in a civil trial. Under these circumstances, we should not be surprised to observe that retributivism is hardly ever supported by legal theorists, even though it enjoys a wide esteem among philosophers for its supposed morality, as well as majority support.20 This discrepancy between the view of philosophers and the view of legal theorists does not, unfortunately, receive much attention from philosophers. Unlike the theories of deterrence and rehabilitation, the justified implementation of retributivism is not accountable to performance criteria, or to its “output” (put differently: “According to retributivism . . . the significance of punishment lies outside the realm of social reality”).21 If especially high recidivism figures were noticed in relation to one sort of crime that was being handled with measures involving rehabilitation of the criminal, there would then be questions as to the legitimacy of this sort of penal mechanism. However, the implementation of retributivism is, by its very essence, not dependent on its effectiveness. The justification for a system of retributive justice is not empirically verifiable through criminological studies, for example. Rather, the criticism directed toward retributivism must be at the conceptual level. For this 19. This perspective is very questionable, because, in a modern constitutional state, the criminal proceedings are differentiated from civil proceedings, among other things, by the former being carried out by the state’s attorney as the representative of the interests of the commonwealth, while the latter transpires between two private parties. At the most, the victims appear alongside as joint plaintiffs and the punishment is in no way seen to be compensation for the infringement on their rights. Since the notion that punishing the malefactor represents justice for the victim seems not to be central to the core of retributivism I will disregard this aspect of it, at least in this work, and concentrate only on the main argument of retributivism. Were this main argument omitted, the (alleged) justice on the part of the victim for a proponent of retributivism would also not be enough justification for the ills that the criminal would be forced to undergo as a component of his or her punishment. 20. Cf. Roxin, Strafrechtliche Grundlagenprobleme, p. 182. 21. Neumann and Schroth, Neuere Theorien, p. 11.

10

german idealism and the concept of punishment

reason, the criticism of retributivism that will be made in this book will present hardly any empirical and interdisciplinary aspects. Instead, it will concentrate on the conceptual arguments, that is, on arguments dealing with legal ethics and moral philosophy. In this respect, this book will proceed no differently than do the depictions of those legal theorists who still treat “absolute theory” in relation to Kant’s and Hegel’s argumentation – or more exactly, to its theoretical foundation. But my conceptual critique will aim to facilitate the convergence of the philosophical with the legal debate over the justification of punishment. Last but not least, the importance of empirical studies for the justification of punishment should not be overestimated. Hassemer calls to our attention that reliable knowledge about the successes of rehabilitation can hardly be obtained. The favorite argument based on the recidivism figures, which because they fluctuate between 30 and 40 percent are said to discredit the concept of rehabilitation, is untenable upon closer examination. First, from such statistics one knows only the manifest, determined and judged criminality . . . Secondly, experiments and empirical lines of argument face a fundamental problem in penal law: one cannot isolate the intervening variables; one cannot try out how it would be if one were to seek to achieve the reform of the offender with another form of penal consequence.22

Even if an especially high recidivism rate were to be observed in regard to rehabilitative punishments of certain crimes, it would still be impossible to draw a reliable conclusion from that observation. Is the limited efficiency of the punishment still better than nothing? Or, on the contrary, should the punishment of these crimes be completely abolished? Should a retributivist degree of punishment be substituted in place of a rehabilitative punishment? Or should an effective general deterrence punishment be introduced, such as the death penalty? The decision remains to be made, and it requires overall legal and moral guidance regarding the justification of punishment. The consequences for the commonwealth should also be brought into consideration. I differ in this point somewhat from the following view of Kristian Ku¨hl, for example: “All the theories of punishment that are geared to certain future goals must assert the suitability of the 22. Hassemer, Einfu¨hrung, p. 288. Also, cf. George P. Fletcher, Basic concepts of criminal law (Oxford: Oxford University Press, 1998), p. 31.

introduction

11

punishment for reaching these goals, and this assertion really must be proven with empirical data.”23 In my view, the debate about justification of punishment is first and foremost a legal and a moral discussion. In this work I will conduct the discussion accordingly. The four aforementioned points urge skepticism in relation to the claim of moral superiority that is made by retributivism, because retributivism, unlike rehabilitation, does not pay heed to the criminal’s interests, but instead only to his or her merits or responsibility for the consequences of every act. First, equal compensation is neither necessarily nor self-evidently the consequence of a criminal offense. Secondly, retributivism considers only one dimension of responsibility, but not the inalienable status of a human being as a being capable of reason. Thirdly, retributivism would allow important consequences of the criminal offense – such as those for the commonwealth – to be disregarded. Fourthly, an explanation is missing in retributivism for why a punishment intended retributively, whose justification only concerns the criminal – and possibly the criminal’s victim as well – does not, however, concern the commonwealth, but yet is still imposed by the commonwealth. All of these four objections form the basis of my critical debate with retributivism. Many exponents of it are already cognizant that these four points are at least problematic. They respond to skepticism by dispensing with the exclusivity of retributivism in justifying punishment and in determining the degree of punishment. The opinion is often expressed that retributivism only represents one particular ethical framework for criminal law in which goals of punishment are also allowed to be fully pursued. When this point is made in order to limit the application of the other theories of punishment through a kind of limitation of power, there is a presupposition that retributivism enjoys legal and moral superiority, for it allegedly respects the rights of the individual human being more than the other theories of punishment. Claus Roxin, for example, states: If guilt gives the state a right to retribute, or if it is the means by which the needs of the many as opposed to individual freedom can be reined in, this seems to me to be a more important question for criminal law than the question about the existence of guilt in general. The answer must be compatible with the second option.24 23. Ku¨hl, Die Bedeutung der Rechtsphilosophie, p. 30. 24. Roxin, Strafrechtliche Grundlagenprobleme, p. 21.

12

german idealism and the concept of punishment

The restriction, which Roxin admits partly contradicts retributivism, proceeds, according to Roxin, by prohibiting a punishment from exceeding the retributivist degree of punishment, while allowing it to fall short of the retributivist degree. The punishment is not allowed to exceed the degree of guilt . . . On the contrary, it is allowed that the punishment befitting the guilt is undercut. Actually, such a practice shows itself to be unacceptable for a consistent retributivist theory, because it means a partial relinquishment of the suffering in punishment that provides equal compensation.25

The justification for this mixed theory is found by Roxin, as by so many other mixed theorists who have dominated the debate since the 1980s, in a form of division of labor between each of the theories of punishment. This division of labor proceeds by assuming the following observation: Each theory of punishment focuses unilaterally on certain aspects of criminal law – the specific deterrent theory focuses on the enforcement, the retributive notion on the judgment and the general deterrent conception on the goal of threatening punishment – and neglects the remaining guises of the penal power, even though, however, each one of them implies specific interferences with the freedom of the individual.26

According to the division of labor in the mixed theories, every theory will allegedly confine itself to its main focus. However, such a construct obviously succumbs to the aforementioned “antinomy of punitive goals.” My investigation will therefore select the search for a single primary justification of punishment as the guideline. In doing so, it is not out of the question that the demands of the other theories of punishment may be partly met; however, if they are met at all, they will only be partly met. This is because the other theories of punishment are incompatible with one another. They are not, however, contrary theories of punishment. In what follows, I will attempt not to lose sight of the conceptual argument’s concrete meaning and its concrete consequences. However, arguing from the point of view of legal ethics, my critique will, whenever necessary, allow itself to diverge from positive law, as retributivism also does. In this respect, I will not, for example, accept 25. Roxin, Strafrechtliche Grundlagenprobleme, pp. 23–4. 26. Roxin, Strafrechtliche Grundlagenprobleme, p. 12.

introduction

13

out of hand the criticism against rehabilitation theory that it allows neither a minimum nor a maximum time-limit for the degree of punishment. Roxin regards it as an objection against rehabilitation that rehabilitation should “consequently” aim for a “treatment until an ultimate reform . . . even when it is of unforeseeable duration,” and “that every punishment has to be dispensed with, even with the gravest crimes, if there is no danger of recurrence.”27 It may be true that many theorists of rehabilitation28 do not want to draw these conclusions, and are, on account of that, inconsistent. This does not mean, however, that rehabilitation as an aim of punishment should be abandoned, but instead that one should defend it – if one is to defend it at all – in a consistent manner, and for this reason accept and assert these consequences. I want to do exactly this. Therefore, in this text, I would like to criticize and refute retributivism in that very aspect which it sees as being the core of its legal and moral superiority over the other theories of punishment: I will attempt to show that it is not retributivism, but instead solely rehabilitation, that takes the responsibility truly seriously that a human being – and especially a criminal – has for his or her actions. For the aforementioned reasons, I will relate my critique first and foremost to Kant’s and Hegel’s arguments. My critique’s foundation will be the Kantian concept of right that builds directly upon the very human dignity to which retributivism appeals: “Right is therefore the sum of the conditions under which the choice (Willku¨r) of one can be united with the choice of another in accordance with a universal law of freedom.”29 Kant’s concept of right can be interpreted in two ways, which I present and critically discuss in the first chapter, that is, either as a liberal concept of right, or as a moral concept of right, which is moral in the narrowest sense of the term. According to the liberal interpretation, the Kantian concept of right (Rechtsbegriff ) means the coexistence of the freedoms of action according to the principle of equal rights, without any consideration of the content of the “choice” (Willku¨r), that is, not dependent on how humans want to employ this freedom. However, according to the moral interpretation in the narrow sense, the task falls to the legal system to implement the 27. Roxin, Strafrechtliche Grundlagenprobleme, p. 7. Also, cf. Fletcher, Basic concepts, p. 38. 28. For example, John Braithwaite and Philip Pettit, Not just deserts: a republican theory of criminal justice (Oxford: Clarendon Press, 1990), p. 101. 29. RL Ak vi:230. Immanuel Kant, Practical philosophy, ed. Mary Gregor (Cambridge: Cambridge University Press, 1996), p. 387.

14

german idealism and the concept of punishment

complete categorical imperative to such an extent in the legal system as the will of a single human being will allow itself to be moved to the morally right through coerced implementation. In the rest of Part i, the attempt will be made to show that Kant’s own retributivist theory is inconsistent with both the first, liberal interpretation of his concept of right (see Chapter 2) and the second, moral (in the narrower sense) interpretation of it (see Chapter 3). Rather, the theory of rehabilitation is compatible with both interpretations. In Part ii, I shall proceed from the statement that the Kantian concept of right comes to be adopted by Fichte as his concept of right, and by Hegel as the “abstract right.” Out of Kant’s concept of right, Fichte and Hegel develop similar consequences to those that Kant, in my view, should have logically drawn. Whereas in Fichte the fate of the criminal being punished is central (see Chapter 4), Hegel, on the contrary, follows restoration of rights as a guideline and through this offers a more systematic presentation of a position that, in its results, is very similar to the Fichtean theory of punishment (see Chapter 5). The retributivist way of interpreting Hegel, which is frequently deemed to be axiomatic, will prove to be unfounded and wrong. In Part iii, I shall attempt to show that, in opposition to Kant’s claims to it, retributivism respects the human dignity in the criminal’s person less than the alternatives of deterrence. I will deal in Chapter 6 with Nietzsche’s radical critique of retributivism, according to which the original motive for the institution of a retributivist punishment is to be found by no means in a respect for human dignity, but instead in a cruelty to the criminal free of morality, which rather hinders than encourages the onset of bad conscience and remorse. In contrast to that, I advocate, in Chapter 5, out of the perspective of a combination of specific deterrence and rehabilitation, a treatment of criminals that respects the dignity in human beings. In order to comply with the promise that has been made in this introduction to accept all of the consequences of the alternative to retributivism that I support – and even, as the case may be, the most radical, the most unusual and the most unpopular consequences – I shall even use, in the seventh chapter, the example of the gravest and most inhuman crimes, that is, of crimes against humanity.

PART I DESERT AS THE SOLE JUSTIFICATION FOR PUNISHMENT

1 THE TWO KANTIAN CONCEPTS OF RIGHT

1.1. The retributivist turn The lex talionis is a rule of distribution for the degree of punishment that is at least as ancient as the biblical verse “Breach for breach, eye for eye, tooth for tooth: as he hath caused a blemish in a man, so shall it be done to him again.”1 It never really was a self-evident, absolute principle for all punishments that needed no further justification. Plato, along with several major schools of ancient philosophy, adapted that premise formulated by Seneca into the famous sentence “Nemo prudens punit quia peccatum est sed ne peccetur.”2 It contains two decisive ideas. First, it implies that punishment is an evil.3 If punishment is to be allowed at all, this evil must be outweighed by the good it brings with it. Secondly, punishment is, as any institution, a means toward the realization of the bonum commune, which is the good that makes all other goods realizable in a sustainable way. Early modern thought in penal law drew important consequences from these premises. I would like quickly to highlight the commonalities of the early modern theorists of penal law, however significant their divergences from other aspects of law and penal law may be. First, punishment is a means to obtain compliance with the law from individuals. This is independent of the question whether the purpose of the law is primarily self-conservation, or perhaps the 1. Lev. 24:18–20; see also Exod. 21:23–25; Deut. 19:21 (KJV). 2. “A sensible person does not punish a man because he has sinned, but in order to keep him from sin.” Seneca, De ira, i.19.7 in Moral essays, trans. John W. Basore (3 vols., London: Heinemann/Cambridge, Mass.: Harvard University Press, 1928), vol. 1, pp. 106–355 (pp. 158–9). 3. See, for instance, Henry Sidgwick, The elements of politics, reprint of 1891 edition (New York: Cosimo Classics, 2005), p. 109.

17

18

desert as the sole justification for punishment

pursuit of happiness (Hobbes), or happiness through the least possible limitation of individual freedoms with an emphasis on the less privileged (Beccaria), or happiness as the greatest happiness of the greatest number (Bentham), or any other form of bonum commune. Indeed, selfconservation – and, in general, protection of citizens – is a condition for pursuing happiness, however one may conceive of the latter. Classical utilitarian thinkers certainly do belong to this tradition of thought in penal law: for example Jeremy Bentham, as well as Cesare Beccaria, who some see as a forerunner to utilitarianism (according to him, the “cool observer of human nature” recommends the following maxim of public ethics: “the greatest happiness shared among the greatest number”).4 However, this tradition is not limited to utilitarian authors, but also encompasses most of the authors of the natural law tradition. Thus, I shall sketch the most prevalent elements of early modern thought without addressing the many differences between natural law and utilitarianism that exist in the realm of public ethics, public law and the foundations thereof. Punishment is considered to be a sanction, meaning that it attaches a consequence to certain behaviors and certain deeds by means of a legal institution in order negatively to influence the motivations that lead to these deeds. As a sanction, its purpose is to deter a citizen from breaking the law; thus, punishment must be felt by its addressees to be an evil, and to that end the punishment must be impressive. In this case, the addressees are both the convicted and all other citizens whose passions may incline them to violate the law5 and to disregard reasons that demonstrate that obedience to the law benefits the entire commonwealth. While Grotius,6 Pufendorf7 and Bentham explicitly mention both goals, the latter states that “example is the most important end of all, in proportion as the number of the persons under temptation to offend is to one.”8 As a means of general deterrence, Bentham 4. Cesare Beccaria, On crimes and punishments, in Beccaria, On crimes and punishments and other writings, ed. Richard Bellamy, trans. Richard Davies (Cambridge: Cambridge University Press, 1995), pp. 1–113 (p. 7). 5. Book ii, Chapter xx, Section xxix(1) in Hugo Grotius, The rights of war and peace, ed. Richard Tuck (3 vols., Indianapolis: Liberty Fund, Inc., 2005), vol. 2, p. 1003. 6. Book ii, Chapter xx, Section vi(1f) in Grotius, The rights of war and peace, vol. 2, pp. 961ff. 7. Book viii, Chapter iii(9) in Samuel von Pufendorf, De jure naturae et gentium, ed. James Brown Scott, trans. C. H. and W. A. Oldfather, Classics of International Law (2 vols., Oxford: Clarendon Press / London: Humphrey Milford, 1934), vol. 2, pp. 1175–6. 8. Footnote in Chapter 13, } 1 of Jeremy Bentham, An introduction to the principles of morals and legislation, reprint of 1823 edition (Oxford: Clarendon Press, 1907), p. 171.

the two kantian concepts of right

19

mentions making an example of a criminal; as a means to specific deterrence he mentions “reformation” of the criminal’s will as well as “disablement” of his or her physical power.9 Secondly, innocents are not to be punished, for the following two reasons. To begin with, “there can arrive no good for the CommonWealth, by Punishing the Innocent.”10 In fact, the punishment, no longer being linked to the infringement of the law, would not be a sanction – meaning a disincentive – any longer. Thus, the punishment of innocents would contradict the very goal of punishment: deterrence. Furthermore, “for seeing all Soveraign Power, is originally given by the consent of every one of the Subjects, to the end they should as long as they are obedient, be protected thereby; the Punishment of the Innocent, is a rendring of Evill for Good.”11 Thus, the punishment of innocents would damage the very reason for establishing the commonwealth, the preservation of which is the proper aim of punishment. Thirdly, as for the degree of punishment, on the one hand, it is not (according to Pufendorf) “necessary that a man suffer exactly what he has done to another, that is, that crimes be always punished by talion.”12 Hugo Grotius stresses the point that most often penal law did not in practice enforce the law of talion, and that this is even the case with Moses, who most definitely never instituted it in reality. In fact, according to Grotius, the practice of penal law inflicted, rather, a degree of punishment proportionate to the crime, rather than equal to it: most of the time, the true degree of punishment exceeds the severity of the crime by many multiples.13 On the other hand, the early modern thinkers make a plea for introducing more differentiation into the degree of punishment that is in accordance with the severity of the crime committed. There is a sort of ordinal proportionality to punishment of crimes, wherein the severity of each crime is pegged to an ordinal scale of punishment. The purpose of this scale, upon which a less severe crime is always punished more mildly than a more severe crime, is deterrence of potential criminals from committing 9. Bentham, Principles of morals and legislation, p. 170. 10. Part 2, Chapter 28 in Thomas Hobbes, Leviathan, ed. A. R. Walter (Cambridge: Cambridge University Press, 1904), p. 229. 11. Part 2, Chapter 28 in Hobbes, Leviathan, p. 229. 12. Book viii, Chapter iii (27) in Pufendorf, De jure naturae et gentium, vol. 2, pp. 1213–18. 13. Cf. Book ii, Chapter xx, Section xxxii(f) in Grotius, The rights of war and peace, vol. 2, pp. 1010–13.

20

desert as the sole justification for punishment

more severe crimes rather than lesser crimes, as well as to deter criminals who are committing a crime from committing a further crime, for instance to deter robbers from murdering the victims who could identify them.14 Yet this rule for determining the degree of punishment is not to be confused either with talion law – which would imply an absolute equality between crime and punishment – or even with an arithmetic equality. (An arithmetic equality would imply proportionality between the difference in severity among the crimes, on the one hand, and the difference of severity between the degrees of punishment to which criminals are sentenced, on the other hand.) Since early modern penal theorists argue against punishments harsher than those required for deterrence, and since they assume that the practice of talion law exceeds the alleged equality of talion law, they see themselves, in comparison with talion law, as promoting clemency toward the convicts. The characteristic of avoiding “superfluous and needless pain” from being inflicted on the convict is termed “frugality” by Jeremy Bentham.15 Fourthly, the inner logic of the requirement for punishment to be no harsher than deterrence requires leads Cesare Beccaria to reject punitive torture and to prefer the punishments that simultaneously make the most efficient and durable impression on the people and are the least painful for the convict. The goal is not the real infliction of an evil or of pain, but the impression made first of all on other citizens, secondarily on the criminal himself or herself. For example, as the utilitarian Henry Sidgwick puts it, “from a utilitarian point of view . . . punishment should be, so far as possible, what Bentham calls ‘exemplarity,’ that is greater in appearance than in reality, since it is chiefly appearance that deters.”16 Mill defends the same view when he rejects the abolition of the death penalty and its replacement by a life sentence because he reproaches the abolitionists with being “driven to inflictions less severe in appearance, and therefore less efficacious, but far more cruel in reality.”17

14. Cf. Chapter 27 of Beccaria, On crimes and punishments, p. 63: “The harsher the punishment and the worse the evil he faces, the more anxious the criminal is to avoid it, and it makes him commit other crimes to escape the punishment of the first.” 15. Chapter 15, } 11 in Bentham, Principles of morals and legislation, p. 194. 16. Sidgwick, The elements of politics, p. 120. 17. John Stuart Mill, “April 1868 speech on capital punishment,” in Mill, Utilitarianism, ed. George Sher (Indianapolis: Hackett, 2001), pp. 65–70 (p. 65).

the two kantian concepts of right

21

Fifthly, punishment is not considered as following any of the traditional principles of justice such as commutative justice, distributive justice or corrective justice. Grotius and Pufendorf present the following arguments for rejecting the classification of punishment in any of the usual categories of justice. Since the aforementioned proportionality, which would correspond to a geometric equality – in an Aristotelian sense – has been rejected, punishment does not follow any principle of distributive justice. Does punishment follow a principle of commutative justice? Punishment is not owed to the criminal as being either the payment of a debt or the fulfillment of any other agreement.18 Nor is punishment to be thought of as compensation for damage. Also, there exists a duty that requires one to give back what each person merits only in the case of a positive merit and not in the case of demerit.19 Instead of only following a principle of justice, punishment is subject, first and foremost, to prudence (prudentia) and to public utility.20 Admittedly, punishment can be said to be “just,” but rather in a negative sense, by which no injustice arises by punishing someone who deliberately caused an evil to the other members of the commonwealth. Yet it is not justice but rather prudence that evaluates the degree of punishment necessary for obtaining a deterrent effect. Furthermore, although criminals ought to understand that being punished is not unjust, justice does not require them either to inflict punishment upon themselves or to cooperate in the infliction of punishment upon their own person. It is solely the other citizens who must cooperate in the latter case. Kant is the first theorist of penal law who derives from talion law not only a rule for determining the degree of punishment, but also a justification for punishment. In so doing, he introduces an alternative theory that is external to the idea of deterrence, to which all competing theories hitherto belonged. Therefore, Kant is not only a central figure in retributivism, but also its founding father. The Kantian turn concerns all five elements mentioned. Concerning the first point, Kant excludes neither all the deterrent effects nor all the deterrent goals of punishment. On the contrary, in the Doctrine of virtue, for example, he even rejects any “renunciation of rigorous 18. Cf. Grotius, Book ii, Chapter xx(2), in The rights of war and peace, vol. 2, p. 995; and Book viii, Chapter iii(4f) in Pufendorf, De jure naturae et gentium, vol. 2, pp. 1152–6. 19. Cf. Book viii, Chapter iii(15) in Pufendorf, De jure naturae et gentium, vol. 2, pp. 1186–7. 20. Cf. Pufendorf, Book viii, Chapter iii(24), in Pufendorf, De jure naturae et gentium, vol. 2, p. 1210.

22

desert as the sole justification for punishment

means (rigorosa) for preventing the recurrence of wrongs by others; for then a human being would be throwing away his rights and letting others trample on them.”21 What Kant actually rejects is the consideration of punishment as merely being a means toward the fulfillment of an end, for instance toward the fulfillment of an end of preventing crime before it occurs. According to Kant, first, punishment is justified only by the fact that the criminal violated the law, and, secondly, so that the only such punishment that complies with this justification is the one that corresponds to “justice” as defined by ius talionis. In all cases in which punishment cannot be justified in this way, it is unjust and therefore prohibited by Kant, however strong the deterrent effect may be. The consequence of this is that, concerning the fifth point, Kant regards the justification of punishment as concerning only justice – and not prudence. According to Kant, not only is it not unjust that criminals are punished, but justice even requires their punishment according to talion law. However, it is unclear whether for Kant this requirement of justice belongs to the realm of legal justice or of moral justice, which will become apparent below (see Section 1.2). As for the fourth point, the impression made by punishment on the citizens and the criminal does not matter to Kant. What matters to him is, instead, solely the evil inflicted upon the criminal who is deserving of the punishment. Concerning the third point, Kant unrestrictedly adopts talion law, although with some adaptations. He considers that talion law inflicts more severe degrees of punishment than deterrence theory, in particular more severe than Beccaria does – for Beccaria is allegedly “moved by overly compassionate feelings of an affected humanity.”22 One must note that today’s defenders of Kant’s theory consider that talion law imposes an upper limit on the degree of punishment, which, they allege, is exceeded by deterrence theories. Interestingly enough, in regard to the second point, it was not Kant himself but today’s Kantian retributivists who reproach the deterrence theories for tolerating the punishment of innocents under certain circumstances. Owing to the contrast in the first point, that is, between the deterrence theories and Kant, who comes to link the punishment and its 21. TL Ak vi:461. Immanuel Kant, Practical philosophy, ed. Mary Gregor (Cambridge: Cambridge University Press, 1996), p. 578. 22. RL Ak vi:335. Practical philosophy, ed. Gregor, p. 478.

the two kantian concepts of right

23

degree only to the moral demerit of the criminal (something that only a human being is capable of bearing), Kantian retributivists also see the core of their theory as presenting the only theory that respects the humanity in the person of the criminal. In this respect, Kantian retributivism radically diverges from the ancient function of talion law, as described by Henry Sidgwick: “at an earlier stage of social and intellectual development, th[e] distinction [between retribution and reparation] is obscure, or but faintly perceptible; the penal loss of an ‘eye for an eye’ or a ‘tooth for a tooth,’ was commonly regarded as a kind of reparation to the person originally maimed.”23 Thereby, respect for humanity is not to be understood in the utilitarian or natural law sense, that is, as keeping the evil or the pain inflicted upon the criminal as low as possible. Rather, in Kant’s view, respect for the humanity in human beings considered as beings capable of reason consists in two dimensions. The first dimension is treating the humanity in the criminal’s person as an end in itself. In this book, I shall not enter into Kant’s complex theory of autonomy and free will. Nor do I need to deal with this theory. I limit myself to observing that there cannot be any respect for the free will of rational beings without first allowing them external freedom, provided that it conforms to the limits imposed by the freedoms of the others. The second dimension in which respect for humanity consists is the infliction of as much evil on the criminal as is merited; it is respectful of the humanity in the criminal’s person because this demerit is the choice of the criminal, and therefore everybody ought to have the foreseeable consequences of their actions expressed, that is, of their merits and demerits. In } 49e of the Doctrine of right, which Kant devotes to penal law, he assumes that the latter requirement – meaning that everybody ought to accept the consequences of their actions – belongs to the former one. Yet he does not provide any argument supporting this point, and I think that none is possible in the Kantian system. Thus I shall assess the questions of whether retributivism can be grounded in the first dimension and of whether it can be grounded in the second dimension as two separate issues. The importance of proceeding through these two separate approaches is enhanced by the link between the two dimensions of respect for humanity, on the one hand, and the controversial issue of the relationship between law and morality, on the other hand. There 23. Sidgwick, The elements of politics, p. 107.

24

desert as the sole justification for punishment

are two theses that pertain to this relationship. Either the legal system is largely independent from morality and its moral premise does not go beyond preserving the first dimension that involves respect for human dignity by mutually limiting external individual freedom of each member of the legal system, or law is subordinated to morality, which it intends to enforce to some extent – which may be very wide. Whereas the first dimension involves respect for human dignity and belongs, incontrovertibly, to the competence of the law, the answer to the question whether the second dimension can belong to the competence of the law depends on the answer to the question about the relationship between law and morality in general. Now, if it appears that retributivism cannot be justified by respect for human dignity as taken in its first dimension, one should still inquire into the possibility of justifying it by respect for human dignity in the second dimension, for the case of an extensive subordination of the legal system under morality. In what follows, this twofold question of whether retributivism can be supported by either of the two dimensions of respect for human dignity mentioned above will eventually be answered in the negative in both dimensions. Thus, unlike the case in which only one of the two questions is answered in a negative way, the other being answered positively, I will not be obliged to choose either between the two dimensions or between the two theses mentioned above. I shall thus content myself with exploring the consequences of both dimensions of respect for human dignity for the justification of punishment. It will ultimately appear that they both support the same justification of punishment and the same degree of punishment.

1.2. Two Kantian concepts of right Some like to advance the notion that Kant created a revolution in legal philosophy just as much as he did in moral and theoretical philosophy, and that he made a significant contribution to today’s discussion. According to this notion, we have Kant’s Doctrine of right24 to thank for a liberal concept of right: right (Recht) as the 24. Throughout the rest of this work, an italicized and capitalized Doctrine of right will refer to the title of the work, whereas the doctrine itself, that is, Kant’s conception of right, will be lower-case roman. According to the common usage in Kant’s age, Kant employs the term doctrine of right in the sense of ius (RL Ak vi:229), referring therefore to law (Recht).

the two kantian concepts of right

25

coexistence of empirical freedoms according to the universal law of equal rights for all legal persons. This concept of right offers two advantages. On the one hand, it claims to comply not with a demanding moral justification, but only with the mere reciprocity in the coexistence of empirical freedoms. On the other hand, it appears not to contradict morality, because it demands, just as the categorical imperative does, a universal law. In this sense Wolfgang Kersting, for example, writes: The rational principle of right requires from everyone the very curtailment of freedom to which everyone who influences each other through actions in their freedom would agree to under fair conditions in a situation free of threats, namely, a strictly universalizable curtailment of freedom that limits everyone in the same way.25

Likewise, Otfried Ho¨ffe writes about the Doctrine of right that, on the one hand, “the political and legal theory that derives from it [from the categorical legal imperative] consists of a political . . . liberalism.”26 On the other hand, according to Ho¨ffe, the Metaphysics of morals orients “the principles of moral philosophy in the (metaphysical) doctrine of right toward an external, and in the (metaphysical) virtue theory, toward an internal lawgiving.”27 I believe, though, that these aspects are not compatible with one another. In believing so, I am not in any way disputing that Kant advanced them. He just did not advance them in the same work. The first aspect, that is, right as coexistence of empirical freedoms with equal rights, is advanced by Kant in, for example, Toward perpetual peace (1795) as “a republican constitution,” while requiring that the “civil constitution in every state” should “be republican,”28 and defining the republic in the following manner: A constitution established, first on principles of the freedom of the members of a society (as individuals), second on principles of the dependence of all

25. Wolfgang Kersting, Wohlgeordnete Freiheit: Immanuel Kants Rechts- und Staatsphilosophie, second edition (Berlin: De Gruyter, 1993), p. 27. 26. Otfried Ho¨ffe (ed.), Immanuel Kant: metaphysische Anfangsgru¨nde der Rechtslehre (Berlin: Akademie Verlag, 1999), p. 8. 27. Otfried Ho¨ffe, “Der kategorische Rechtsimperativ: Einleitung in die Rechtslehre,” in O. Ho¨ffe (ed.), Immanuel Kant: metaphysische Anfangsgru¨nde der Rechtslehre (Berlin: Akademie Verlag, 1999), pp. 41–62 (p. 48). 28. ZeF Ak viii:349. Immanuel Kant, Practical philosophy, ed. Mary Gregor (Cambridge: Cambridge University Press, 1996), p. 322.

26

desert as the sole justification for punishment upon a single common legislation (as subjects), and third on the law of their equality (as citizens of a state) . . . is a republican constitution.29

Let it be noted here that equality before the law is by no means justified by the categorical imperative. The second aspect is advanced by Kant in his Doctrine of right, which is why the Doctrine does not offer a liberal concept of right. With regard to the relation of right to morality in Kant, one typically distinguishes three main options: (1) the thesis of complete independence (initially proposed by Julius Ebbinghaus); (2) the dependence thesis of the necessary derivation of right from the principle of morality (initially advanced by Kantian legal theorists in the years in between the Groundwork of the metaphysics of morals and the Doctrine of right30); and (3) the thesis of limited dependence, arguing dependence on morality for the validity (Geltung) or for the adjudication of right, and, in contrast to that, arguing independence from morality for the realization or for the execution of right. Independence in the execution of right has the essential feature that, whereas morality envisions the execution – that is, the carrying out of actions – out of duty (Pflicht), right relies on coercion (Zwang) as the incentive. This third thesis appears to me to be philologically correct. The systematic evaluation connected to it overlooks, however, the scope of right’s dependence on morality in the adjudication. Interpreters of Kant concentrate, though, on what differentiates Kant from the Kantian legal theorists, who, even before the publication of the Doctrine of right (1797), had tried to develop a theory of right on the basis of the Groundwork of the metaphysics of morals (1785). Most of the Kantian legal theorists, such as Gottlieb Hufeland and Theodor Schmalz, derived right from the principle of morality, and, in fact, from the principle of morality as an adjudicative principle. In doing so, they understand right, just as Kant does, to mean that which regulates the external relations between humans. External actions and relations are already objects of moral duty. The difference between moral duty and right is placed by these Kantian legal theorists in the differentiation between, on the one hand, what is morally commanded or forbidden and, on the other, what is morally allowed. 29. ZeF Ak viii:349f. Practical philosophy, ed. Gregor, pp. 322–3. 30. Cf. W. Kersting, “Sittengesetz und Rechtsgesetz: die Begrundung des Rechts bei Kant und den fru¨hen Kantianern,” in Reinhard Brandt (ed.), Rechtsphilosophie der Aufkla¨rung: Symposium Wolfenbu¨ttel 1981 (Berlin: De Gruyter, 1982), pp. 147–77.

the two kantian concepts of right

27

The derivation from the principle of morality is admittedly twofold. On the one hand, a part of right consists of the morally allowed or is in the realm of the moral adiaphoron. On the other hand, the Kantian legal theorists appeal to the implication that whatever is morally commanded is therefore morally allowed. Thus, moral duty, next to the moral adiaphoron, belongs to right: I have the right to do the morally allowed as well as the morally commanded. Depending on the author, right is therefore classified into absolute or conditional, or into alienable and inalienable, or even into complete and incomplete rights. One could possibly distinguish a first step to the contemporary differentiation between human rights and the other rights, in so far as human rights count as being absolute rights, whereas the other rights differ from commonwealth to commonwealth and from situation to situation. Such a twofold derivation of right from the principle of morality leads, however, to three major problems. The first of these is the derivation that assembles right out of two heterogeneous parts relying on two very different normative justifications, which calls into doubt whether right as such is derived, that is, if right really forms a whole and can be differentiated from morality, or whether it is, rather, a weakened version of morality. Secondly, the authority to coerce is lacking. Thirdly, the authority to coerce in the case of the merely morally allowed appears more difficult to justify, if at all, than in the case of the morally commanded.

1.3. Kant’s concept of right derived from the principle of morality in the Doctrine of right The division of Kant’s Metaphysics of morals into the Doctrine of right and the Doctrine of virtue obviously attempts to propose a doctrine of right that is not subject to the Kantian legal theorists’ aforementioned problems. The core relation between the doctrine of right and the doctrine of virtue consists of two aspects. First, the doctrine of right and the doctrine of virtue can be derived from the same principle, that is, from the categorical imperative; this derivation, in the case of the law, implies no weakening of morality: the obligation remains the same. Secondly, the difference between the doctrine of right and the doctrine of virtue rests on the different motives for their respective execution. The execution of the doctrine of right only has to deal with coercion as its motivation, whereas the doctrine of virtue is responsible for the execution out of duty.

28

desert as the sole justification for punishment

It arises from the second aspect that “the substantive boundaries of the duties of right are defined . . . through the range of the external motivation that is related to them in juridical lawgiving: wherever coercion fails to motivate, juridical lawgiving does not rule.”31 Like most of the other interpreters, Kersting highlights the second point: Therefore, every duty of right is always also an indirectly ethical duty; the ethical mode of assigning obligation a priori inherent in every duty is not rescinded for the category of duties of right just because they can be issued in a juridical manner.32

I would, however, like to accentuate the first point: the area of right encompasses every ethical duty that can also be fulfilled through coercion. In other words, right should provide for the fulfillment of every ethical duty that could be fulfilled through coercion by coercion itself (that is, through both the threat of coercion and the application of coercion). But which duties can be fulfilled through coercion? Here too Kant’s answer appears to be clear: not maxims, but actions. Bernd Ludwig formulates this in the following manner: Now, what is the fundamental substantive difference between the principle of right and the categorical imperative? . . . When the “disposition” (Gesinnung) [Ak vi:393] of the actor – which is not accessible to external lawgiving – both cannot and is not allowed to be included, it is also not possible to address demands to it. However, that his or her actions at least can coexist with a universal lawgiving is indeed externally enforceable – and without having to exert influence for it on the maxims of the person concerned.33

It appears to me that there is an inaccuracy in Ludwig’s formulation. The third formula of the categorical imperative, to which Ludwig obviously refers, reads: “act in accordance with a maxim that can at the same time make itself a universal law.”34 If right is derivable, on the one hand, from the categorical imperative, and it is, on the other hand, responsible for that part of this imperative that allows it to be imposed through coercion, then the procedure for determining the content of right should look something like the following. As a first step, maxims 31. 32. 33. 34.

Kersting, Wohlgeordnete Freiheit, p. 176. Kersting, Wohlgeordnete Freiheit, p. 176. Bernd Ludwig, Kants Rechtslehre (Hamburg: F. Meiner, 1988), p. 95. GMS Ak iv:436f. Immanuel Kant, Practical philosophy, ed. Mary Gregor (Cambridge: Cambridge University Press, 1996) p. 86.

the two kantian concepts of right

29

that survive the test should be determined. As a second step, the actions (by which I mean both active actions and abstaining from realizing actions) that these maxims lead to should be investigated. As a third step, an investigation must really seek to determine which of those actions can be realized by means of coercion; right must then see to it that these actions are met with a threat of coercion or an execution of coercion. Therefore, Ludwig’s formulation should be modified thus: “However, that his or her actions at least can coexist with a universal lawgiving of the maxims, that is, with the categorical imperative, is indeed partly externally enforceable – and without having to exert influence as a counterpart on the maxims of the person concerned”; and it should also be enforced by means of coercion in so far as it is enforceable by coercion. In contrast, Ludwig’s formulation (“that his or her actions at least can coexist with a universal lawgiving”) leaves open the possibility that law is nothing other than the equality of all legal persons with regard to actionable rights. Such a conception of right is not compatible with Kant’s Doctrine of right, as I will show in what follows. The premises stating that the actions, at the very least, can coexist with the universal lawgiving of the maxims, that is, with the categorical imperative, and that this is partly externally enforceable by means of coercion, and also should be enforced by this, in so far as it is enforceable, imply the conclusion not only (1) that rights are actionable by the persons concerned, but also (2) that action must be taken, and (3) that the enforcement should be compelled by coercion from those who would like to abstain from it. I shall content myself with only one example of this here; Kant writes: Thus ethics commands that I still fulfill a contract I have entered into, even though the other party could not coerce me to do so; but it takes the law (pacta sunt servanda) and the duty corresponding to it from the doctrine of right, as already given there. Accordingly the giving of the law that promises agreed to must be kept lies not in ethics but in Ius. All that ethics teaches is that if the incentive which juridical lawgiving connects with that duty, namely external constraint, were absent, the idea of duty itself would be sufficient as an incentive.35

It is certainly the case that no legal system would ever want to enforce fulfillment of every promise. Not even the relevant specifications from Kant’s Doctrine of right fulfill this challenging requirement (this is often 35. RL Ak vi:219–20. Practical philosophy, ed. Gregor, pp. 383ff.

30

desert as the sole justification for punishment

overlooked by most of the interpreters of Kant).36 The law – the Kantian right as well – deals with promises in very different ways. It is possible to differentiate between the following conceivable cases: 1 Always, and without action by the person concerned, the fulfillment of promises is enforced by means of coercion, or, alternatively, compensations are compelled from a promise breaker. 2 Only by the suit brought by the person concerned – but in this case always – is either the fulfillment of the promise enforced, or, alternatively, the delivery of compensations compelled from a promise breaker. 3 By the concerned person’s suit, only is the fulfillment of some promises (for example, only the fulfillment of contracts that fulfill certain conditions) enforced, or are remedies compelled for nonfulfillment of some promises (for example, for the non-fulfillment of contracts that fulfill certain conditions). 4 The fulfillment of some contracts that fulfill certain states of laws is, under certain circumstances and despite the action brought, not enforced, nor are compensations compelled. When we take Kant’s formulation cited above seriously, then only the first, or at most also the second, option should be possible in his doctrine of right. But Kant advances different options, which is exactly what occurs in every legal system. Here is one example for each option that appears in Kant’s writings. 1 The preliminary articles of Toward perpetual peace are in no sense at the disposal of the contractual partners, even if all partners were to be in agreement to abrogate an article.37 2 Kant claims that “if one of the partners in a marriage has left or given itself into someone else’s possession, the other partner is justified, always and without question, in bringing its partner back under its control.”38 3 The adherence to contracts with “servants” is only compelled in limited variations: “The contract of the head of a household with servants can therefore not . . . [be concluded] for life, but at most only for an unspecified period of time, within which one party may give the other notice.”39 36. 37. 38. 39.

Cf. Ku¨hl, Die Bedeutung der Rechtsphilosophie, p. 41. ZeF Ak viii:343–7. Practical philosophy, ed. Gregor, pp. 317–21. RL Ak vi:278. Practical philosophy, ed. Gregor, p. 427. RL Ak vi:283. Practical philosophy, ed. Gregor, p. 432.

the two kantian concepts of right

31

4 The sentence “sale breaks a lease”40 provides for the termination of rental contracts without compelling compensation for loss (but with adherence to the cancellation period). In Kant’s view, as with the current legal system, only in very few cases (for example, in the case of lying in court or in the case of breach of trust, etc.) are lies legally proscribed; still less is the truth compelled, even though with Kant, as is generally known, the moral prohibition against telling lies applies without exception. Even in penal law, in which Kant counts as being the strict supporter of a categorical retributive imperative, he does not accept all the consequences of his concept of right. On the one hand (to name a well-known case) if all the inhabitants of an island leave their homeland, he nevertheless requires that all convicts, including those who have been condemned to death, should actually serve their sentences out, but then, on the other hand, he concedes a power of pardon to the sovereign. In this vein, he writes, on the one hand: Punishment by a court . . . can never be inflicted merely as a means to promote some other good for the criminal himself or for civil society. It must always be inflicted upon him only because he has committed a crime . . . The law of punishment [that is, of criminal law] is a categorical imperative.41

On the other hand, however, he claims: With regard to crimes of subjects against one another it is absolutely not for him [the sovereign] to exercise it [the power of pardon]; for here failure to punish (impunitas criminis) is the greatest wrong against his subjects. He can make use of it, therefore, only in case of a wrong done to himself (crimen laesae maiestatis).42

Here the enforcement of the legal–ethical consequences of a crime stands at the disposal of other persons. At this point, the categorical imperative turns into – in opposition to Kant’s concept of right – a mere hypothetical imperative. Such isolated inconsistencies muddle not only the effective range of Kant’s concept of right, but also the systematic differentiation that ultimately proves to be untenable and leads to the situation that with Kant some actions, which, according to his Doctrine of right, should be 40. RL Ak vi:361. Practical philosophy, ed. Gregor, p. 496. 41. RL Ak vi:331. Practical philosophy, ed. Gregor, p. 473. 42. RL Ak vi:337. Practical philosophy, ed. Gregor, pp. 477–8.

32

desert as the sole justification for punishment

the concern of a legal command or proscription, turn out not to be so, because Kant separates many duties toward external actions from their corresponding actions in an inconsistent way. As is generally known, Kant differentiates between duties of right and duties of virtue. He claims that duties of virtue cannot be subject to “external lawgiving,” because they have to do with an end which (or the having of which) is also a duty. No external lawgiving can bring about someone’s setting an end for himself (because this is an internal act of the mind), although it may prescribe external actions that lead to an end without the subject making it his end.43

But if external actions are actually “commanded” by the corresponding categorical imperative or by the corresponding duty, then these actions – because they are external, and provided that they are enforceable by means of coercion as such – are legally dictated, that is, legally enforceable by means of coercion. In this respect, we should not allow ourselves to be led astray either by the asymmetry in the presentation of duties in the Groundwork of the metaphysics of morals – for instance, in the commentary on the categorical imperative’s formula of humanity as an end in itself – or by qualifying as duties of right the duties toward the preservation of humanity as an end in itself. The classification of duties follows two dichotomic criteria: “dut [ies] to oneself” are opposed to “dut[ies] to others,” just as the “preservation of humanity as an end in itself” is opposed to a “furtherance” of humanity as an end in itself. The asymmetry arises out of Kant referring to the proscription of false promises and the “assaults on freedom and property” as duties for the preservation of humanity as an end in itself, whereas no duties toward its furtherance are named. At least three arguments speak against combining the asymmetry in the presentation offered by Kant with a conceptual asymmetry. First, it would be absurd to believe that the proscription of false promises and the attack against freedom and property apply unconditionally. If property or freedom were to be abused as the means of a criminal action, both could then be vastly curtailed or suspended for longer periods of time. Secondly, the proscription of false promises is to be understood as a duty of right. A false promise is a promise that the 43. RL Ak vi:239. Practical philosophy, ed. Gregor, p. 395.

the two kantian concepts of right

33

promiser does not intend to keep. This intention is, however, not an external action. An external action is present if, instead of observing a false promise, one observes the non-fulfillment of a promise; in this way, however, the moral prohibition of false promises becomes a legal proscription of non-fulfillment of promises. Last but not least, thirdly, the duties toward the furtherance of humanity as an end in itself can be formulated as duties toward external action, for instance the duties toward establishing basic schooling or basic medical care for all. Being duties toward external – and enforceable – actions, these duties should definitely belong to the duties of right, whereas in Kant they, being duties toward the furtherance of humanity, count as being duties of virtue. For these reasons, the identification of the differentiation between duties toward the preservation of humanity and duties toward the furtherance of humanity, on the one hand, with a differentiation between duties of right, in the legal sense of enforceable rights, and duties of virtue, in the sense of non-legally enforceable duties, on the other hand, is generally untenable. But how far does the range of Kantian right actually extend? The categorical imperative cannot surely determine, a priori, either the single proscribed, dictated and allowed maxims or the single proscribed, dictated and allowed actions, because the moral appraisal of maxims and actions presupposes knowledge of their empirical attributes and of their context. In contrast, however, it is always possible to determine with the categorical imperative whether a given maxim or action is contrary to duty or in conformity with it, and, in the latter case, whether it is morally dictated or just morally allowed. Apart from that, every maxim, being as general and indirect as it can be, is focused on external actions. If we have a general maxim for guiding our lives, we can really determine for all cases that this or that external action is compatible or incompatible with our maxim, and if we necessarily have to carry out the action when following our maxim. That is why one can say about all external actions – taking into consideration their entire context – whether, in each case, they are morally proscribed, dictated or merely allowed. All external actions can potentially be influenced by coercion to some extent. Therefore, all external actions are not only objects of moral judgments, but also objects of right in the sense of Kant’s concept of right. Apart from that, the moral and the legal judgments of external actions should be decided identically every time. The only thing in the moral judgment that does not have any influence on the

34

desert as the sole justification for punishment

legal judgment is the answer to the question whether the morally and legally dictated action was executed out of duty or merely in conformity with duty.44 In this respect, morality is certainly more demanding than law. The range of right presented here, nevertheless, greatly surpasses whatever belonged to the area of law. The proximity of law and morality is likewise here much closer than it ever was in a legal system. Especially if one considers that the categorical imperative issues, amongst other things, dictates and proscriptions about aspects of personal lifestyle pertaining neither to the rights nor to the legitimate interests of other individuals, then Kant’s conception of right appears to be hardly liberal and, in fact, very odd.

1.4. The liberal interpretation of the Doctrine of right In view of the fact that Kant, like Locke and Mill, belongs to that group of authors one tends to refer to for the definition of what the liberal state is, these implications may astound. Yet, the liberal interpretation of Kant is based on two elements. Regarding the first element, the derivation of the concept of right from the categorical imperative is usually underestimated. Marcus Willaschek, for example, goes so far as to claim: Kant nowhere really says that the principle of right can be derived from, or is based on, the categorical imperative. The moral law and the categorical imperative are not even mentioned in }}A–E of the “Introduction to the doctrine of right,” where Kant introduced the principle of right.45

And Thomas Pogge begins his enquiry into the concept of right directly with the last lines from } B: “Kant defines Recht as ‘the whole of the conditions under which the choice of one can coexist . . . with the choice of the other according to a universal law of freedom.’”46 44. Otfried Ho¨ffe, in “Ko¨nigliche Vo¨lker”: zu Kants kosmopolitischer Rechts- und Friedenstheorie (Frankfurt a.M.: Suhrkamp, 2001), p. 11, correctly remembers that the differentiations between morality and legality and between the duties of law and virtue are two different differentiations. 45. Marcus Willascheck, “Why The doctrine of right does not belong in The metaphysics of morals,” Annual Review of Law and Ethics, 5 (1997), 205–27 (p. 230). 46. Thomas W. Pogge, “Is Kant’s Rechtslehre a ‘comprehensive liberalism’?” in Mark Timmons (ed.), Kant’s Metaphysics of morals (Oxford: Oxford University Press, 2002), pp. 133–58 (p. 137).

the two kantian concepts of right

35

In the “Introduction to the metaphysics of morals,” the division of the Metaphysics of morals is undertaken according to the motive of the fulfillment of the dictates of the categorical imperative. The decisive definition of right is found in the following formulation: “The doctrine of right and the doctrine of virtue are therefore distinguished not so much by their different duties as by the difference in their lawgiving, which connects one incentive or the other with the law.”47 This definition consists of (1) the duties that are common to morality and to right, which result from the categorical imperative,48 and (2) the difference in the respective lawgiving and incentive. In } b of the “Introduction to the doctrine of right” the second and third paragraphs give a definition of right. Right at the beginning, the duties that are common to morality and to right are recalled: “The concept of right, insofar as it is related to an obligation corresponding to it (i.e., the moral concept of right).”49 The three following points do indeed draw on the difference between right and morality (right pertains only to, first, “the external and indeed practical relation” of people; secondly, choice [Willku¨r] in contrast to mere wishes; and, thirdly, the form of choice in contrast to its matter). One usually overlooks, though, that at the end of this very same paragraph duties that are common to right and to morality are again mentioned: Kant speaks of the form of choice “insofar as choice is regarded merely as free, and whether the action of one can be united with the freedom of the other in accordance with a universal law.”50 Hence, it is about choice as form, that is, about choice “insofar as choice is regarded merely as free.” This clearly means that the freedom we are talking about here is not the freedom of action in choice, but instead the pure freedom required in Kantian morality, that is, a freedom of will in autonomy. Therefore, the “universal law” concerned here cannot just be equality of rights, but instead must be a law suited to the actor’s freedom of 47. RL Ak vi:220. Practical philosophy, ed. Gregor, p. 384. 48. Nothing changes with these duties if, like Willaschek, one presupposes an additional level of the imperative (“Act according to duty from duty,” p. 214), which is why Willaschek’s “alternative thesis” is much closer to the official thesis than he himself believes, and why it is not really relevant for the treatment of our problem. Willaschek himself believes his thesis to be a thesis of limited independence of the law from morality; in his own words, “rather an independent, basic law of practical rationality” (p. 223). 49. RL Ak vi:230. Practical philosophy, ed. Gregor, p. 387. 50. RL Ak vi:230. Practical philosophy, ed. Gregor, p. 387.

36

desert as the sole justification for punishment

will: it is the same “universal law” as the one in the third formula of the categorical imperative (“act in accordance with a maxim that can at the same time make itself a universal law”).51 Because the individual formulas of the categorical imperative are versions of the same imperative, and therefore equivalent, it also says: “So act that you use humanity, whether in your own person or in the person of any other, always at the same time as an end, never merely as a means.”52 The “reciprocal relation of choice” is, therefore, more demanding in relation to Kant than the mere coexistence of the powers of choice or the freedoms of action under the principle of equality as in the liberal understanding of a just legal system. The reference to a demanding moral freedom is indeed present in the concluding formulation of } B, in the “Introduction to the doctrine of right,” though not emphasized: “Right is therefore the sum of the conditions under which the choice of one can be united with the choice of another in accordance with a universal law of freedom.”53 One should, at this point, take note that Kant uses the expression “in accordance with a universal law of freedom,” which is neither the same as just saying “in accordance with a universal law” nor the same as “according to a universal law of choice.” In the former case, “freedom” (Freiheit) would be redundant, and should have been “choice” (Willku¨r), as in the middle of the sentence. In the latter case, “of freedom” would be superfluous. At the beginning of } C, we find a variant of the very same definition of right in which it is stated more 51. GMS Ak iv:436f. Practical philosophy, ed. Gregor, p. 86. Allen W. Wood rejects this point with the following argument: “as to the ‘universal principle of law’ itself, it is hard to sustain the view that it can be derived from the moral imperative. This principle is: ‘any action is right if it can coexist with everyone’s freedom in accordance with a universal law, or if on its maxim the freedom of choice of each can coexist with everyone’s freedom in accordance with a universal law’. . . This principle may bear a superficial resemblance to the Formula of Universal Law: ‘Act only in accordance with that maxim which you can at the same time will that it become a universal law . . .’ Like that formula, the principle of right provides us with a test only of the permissibility . . . But the principle of right says nothing about willing maxims as universal laws.” A. W. Wood, “Kant’s Doctrine of right,” Introduction to Otfried Ho¨ffe (ed.), Immanuel Kant: Metaphysische Anfangsgru¨nde der Rechtslehre (Berlin: Akademie Verlag, 1999), pp. 19–39 (p. 35). Wood’s argument against the derivation of law from the categorical imperative is, however, only valid against a derivation of his “execution” from the categorical imperative; it is not convincing against an adjudication of actions from the categorical imperative. 52. GMS Ak iv:429. Practical philosophy, ed. Gregor, p. 80. For the meaning of the term “humanity” see below p. 64. 53. RL Ak vi:230. Practical philosophy, ed. Gregor, p. 387.

the two kantian concepts of right

37

precisely which aspect of choice is meant (“the freedom of choice”54 [Freiheit der Willku¨r], that is, the form of the freedom, or the freedom of will in choice), and this extraneous element “according to a universal law of freedom” (emphasis mine) becomes dispensable and disappears: “Any action is right if it can coexist with everyone’s freedom in accordance with a universal law.”55 In the remaining part of } C, as well as throughout the entire Doctrine of right, the “freedom” of “choice” and “universal law” is mentioned at every repetition of the definition of right. Explicit references to the categorical imperative and to duty, however, appear more seldom, so that the framework, and hence the exact meaning of “freedom” and “universal law,” can easily be overlooked. Regarding the second element: the liberal interpretation of Kant also rests on a minimalist understanding of the second formula of the categorical imperative. Kant in fact applies (as shown above) only one part of the categorical imperative to law, because he is of the opinion that the duties toward the furtherance of humanity are allowed to relate only to an “end” (Zweck) and not to an external action so that they cannot constitute any “external lawgiving.” In contrast, for Kant, the duties toward the preservation of humanity as a goal in itself can constitute an external lawgiving, and indeed all of these duties as a whole can constitute such an external lawgiving. Now, the minimalist understanding leads one to believe that, even if one neither overlooks nor forgets the derivation of the Kantian concept of right from the categorical imperative, the requirement addressed by the categorical imperative to the legal system is strictly limited to the protection of the fellow citizens or to the prohibition of homicide, slavery, withdrawal of citizenship and the like. If law means as much as the coexistence of the freedom of action of all humans according to the law of equality of right, then what is being dealt with here is a liberal definition of law, which is derived from the harm principle without any reliance on Mill’s utilitarian background or on any comprehensive moral justification. In such a context, the requirements of just such a formula as the second formula of the categorical imperative, which has been conceived of minimally, are not especially difficult to fulfill (one of the exceptions consists of the case, for example, of suicide, which is forbidden by the categorical imperative). 54. RL Ak vi:230. Practical philosophy, ed. Gregor, p. 387. 55. RL Ak vi:230. Practical philosophy, ed. Gregor, p. 387.

38

desert as the sole justification for punishment

A liberal concept of right and a genuine independence of law from morality can be won not through the adoption of coercion as the motivation for the execution, but instead, first through an independent adjudication, that is, first when the content of law is no longer derived from the categorical imperative. It should be obvious that such a conception of right amounts by no means to legal positivism. The liberal concept of right comprising the set of conditions for the coexistence of external freedoms, according to the law of equality, contains completely normative requirements (of protection of the existence of all individuals, of equality before the law, etc.). Many legally positive and valid legal systems do not, however, conform to these criteria; rather, they infringe them.

1.5. The liberal concept of right In the introduction I argued that both of the following aspects can be found in Kant’s writings: (1) a liberal concept of right and (2) a derivation of right from the categorical imperative that leads to another concept of right, which is incompatible with the liberal concept of right. After I have developed the second aspect, I would like to return to the first. Here I shall content myself with presenting two passages that document Kant’s liberal concept of right: 1. On the concept of right as found in Plato’s Republic, Kant writes: A constitution providing for the greatest human freedom according to laws that permit the freedom of each to exist together with that of others (not one providing for the greatest happiness, since that would follow of itself) is at least a necessary idea, which one must make the ground not merely of the primary plan of a state’s constitution but of all the laws too.56

2. Kant designates as “the highest task which nature has set for mankind” the “establish[ment of] a society in which freedom under external laws would be combined to the greatest possible extent with irresistible force, in other words [the establishment of] a perfectly just civil constitution.”57 Kant states more precisely that it concerns that sort of society that “has not only the greatest freedom, and therefore a continual antagonism among its members, but also the most precise 56. KrV B 373. Immanuel Kant, Critique of pure reason, ed. and trans. Paul Guyer and Alan W. Wood (Cambridge: Cambridge University Press, 1997), p. 397. 57. Idee, Proposition 5, Ak viii:22. Immanuel Kant, Political writings, ed. Hans Reiss, trans. H. B. Nisbet, second edition (Cambridge: Cambridge University Press, 1991), pp. 45–6.

the two kantian concepts of right

39

specification and preservation of the limits of this freedom in order that it can co-exist with the freedom of others.”58 The context of both of these quotes, as well as further passages in Kant’s works where this concept of right is mentioned, is a teleology of history, what Kersting terms “a conception of right belonging to moral teleology.”59 Kersting rejects, however, that selfsame teleological perspective as being normatively irrelevant. He raises the following objection against the teleological perspective on right, which he diagnoses as having too strong an “affinity” with Wolff’s idea of human perfection: What kind of sense, though, does that sort of teleological, moral– pragmatic justification of right make if the right of coercion toward the freedom to fulfill duty relates neither to striving for perfection nor to self-preservation, but instead must relate itself to a duty that demands the formally perfect treatment of others, and if the moral duty were not to be grounded in a lex naturalis but in an internal lawgiving that would not require the realization of natural designs, but instead require an act out of respect for the law, and for that reason cannot depend on external assistance?60

Kersting’s rhetorical question is astonishing for three reasons. First, Kersting overlooks that the natural design in Kant’s moral philosophy is precisely none other than that the human “natural capacities” are “developed” “completely” and “in conformity with their end” “directed towards the use of his reason.”61 Since a reasonable being who follows reason – and not, for instance, his or her impulses – is really obeying the categorical imperative, the natural design in humans, according to Kant in his Idea for a universal history from a cosmopolitan point of view (1784), is clearly not only the cultivation of his or her relation to nature, the civilization of his or her behavior toward fellow humans and the formation of a civil society or state of law, but also the moralization of humans themselves, in short, exactly the “acting out of a respect for the law” required by Kersting, as well as the “moral duty” as an “internal lawgiving,” which is likewise required by him. Secondly, “the right of coercion toward the freedom to fulfill duty” is surely sufficiently present, as Kersting believes, because Kant speaks of the “greatest human freedom according to laws” without guaranteeing 58. 59. 60. 61.

Idee, Proposition 5, Ak viii:22. Political writings, ed. Reiss, p. 45. Kersting, Wohlgeordnete Freiheit, p. 142. Kersting, Wohlgeordnete Freiheit, p. 151. Idee, Propositions 1 and 2, Ak viii:18. Political writings, ed. Reiss, pp. 42–3.

40

desert as the sole justification for punishment

this freedom to the individual only under the condition that he or she fulfills any duties: individuals only need to heed the laws of coexistence in order to enjoy these freedoms guaranteed by coercion. That means that they enjoy them equally, no matter if they act morally or immorally, just not contrary to right. However, freedom is enabling for an individual striving for the fulfillment of moral duty to fulfill in an appropriate way the duties, which Kant derives from the categorical imperative, toward the preservation and the furtherance of humanity as an end in itself. The individual receives explicitly through freedom more and better means to preserve and to advance humankind as an end in itself. In this respect, the freedom to fulfill duty contains neither a duty to achieve perfection nor a duty toward self-preservation (otherwise we would not be dealing with a liberal concept of right); rather the freedom to fulfill duty is contained in the duty toward self-preservation and the duty to achieve perfection (in one of the passages mentioned above, Kant maintains, for example, that “happiness would follow of itself”).62 For this reason, right actually belongs to the “moral– pragmatic” without thereby betraying Kant’s system of reason. Thirdly, Kersting assumes that “right of coercion toward the freedom to fulfill duty . . . must relate itself to a duty that demands the formally perfect treatment of others . . . grounded . . . in an internal lawgiving . . . requiring an act out of respect for the law, and for that reason cannot depend on external assistance.” Kersting shows here that he locates the motivation for compliance to law in coercion, yet sees the source of law as an institution entitled to coerce in the moral duty to establish “the right of coercion toward the freedom to fulfill duty.” In no case do I deny that, according to Kant, it is a part of duty to enter into a state of laws wherever the state of nature still exists and to contribute to the establishment of a just legal system to the best of one’s ability. Kant does not say, however, that the legal system should originate from a human’s conscious fulfillment of duty. On the contrary, he mentions misery in the state of nature, that is, nature’s coercion being the drive for the establishment of a legal system, which is opposed to the inclination of humans. Immediately after one of the citations introduced above, we find the following remark: Man, who is otherwise so enamoured with unrestrained freedom, is forced to enter this state of restriction by sheer necessity. And this is

62. KrV B 373. Critique of pure reason, ed. Guyer and Wood, p. 397.

the two kantian concepts of right

41

indeed the most stringent of all forms of necessity, for it is imposed by men upon themselves, in that their inclinations make it impossible for them to exist side by side for long in a state of wild freedom. But once enclosed within a precinct like that of civil union, the same inclinations have the most beneficial effect.63

Kant repeats this conception in his Critique of the power of judgment and in the text Toward perpetual peace: For the problem is not the moral improvement of human beings but only the mechanism of nature, and what the task requires one to know is how this can be put to use in human beings in order so to arrange the conflict of their unpeaceable dispositions within a people that they themselves have to constrain one another to submit to coercive law and so bring about a condition of peace in which laws have force.64

1.6. A legal and a moral task If there are really two entirely different concepts of law in Kant, as I believe, which one should count as being the more “Kantian,” that is, which of the two is more consistent with Kant’s system, and which the more convincing concept? The answer to the second question appears clearly to be: the liberal conception of right accords not only with the reality of the legal systems we rate as just and possibly even the reality of legal systems generally, but also with our normative intuitions. The answer to the first question is more difficult. We can, however, begin with observing, in Kant’s Doctrine of right, the numerous deviations from those actions that, from the perspective of the categorical imperative, should be held to be necessary, for instance those concerning the proscription of lying, the dictate of the fulfillment of promises, etc. In Kant’s Doctrine of right, for instance, the right to property entails the right to allow fruit to rot without regard for other people actually needing the fruit or being able to make better use of it. The categorical imperative does forbid such waste, whereas Kant’s right does not. Furthermore, the liberal concept of right is, in contrast to the other concept of right in Kant, not found only in one work: standing alone opposed to the Critique of pure reason (1781), the Idea for a universal history from a cosmopolitan point of view (1784), the Critique of the power of judgment (1790), On the common saying (1793) and the 63. Idee, Proposition 5, Ak viii:22. Political writings, ed. Reiss, p. 46. 64. ZeF Ak viii:366. Practical philosophy, ed. Gregor, p. 335.

42

desert as the sole justification for punishment

text Toward perpetual peace (1795) is the Metaphysics of morals (1797); all the works listed belong to the critical section of Kant’s oeuvre. That the concept of right is more comprehensively developed in the Doctrine of right than in the other works should count less than the repetition of the other concept in systematic contexts. Seen systematically, the liberal concept of right is also more convincing than the others, because it reinforces the independence of right from morality, expands the range of coercion from compliance with the law to the establishment of a legal system, and saves the Metaphysics of morals from dichotomization into a Doctrine of right and a Doctrine of virtue. Just to name some external evidence: it is striking that many strange – no, even embarrassing – legal regulations found in Kant’s Doctrine of right, and derivable from a direct application of the categorical imperative, are shamefully, or out of respect for such an otherwise impressive philosophy, left unconsidered. In this context, Kant’s conception of marriage can be named, for instance, or his rigorous proscription of “public prostitution (venus volgivaga),” the island example, the remark on bestiality, etc.65 Fortunately, such embarrassments cannot justifiably be found in the passages of the aforementioned works. Not even Vittorio Ho¨sle, who, on the basis of Kant, Fichte and Hegel, also wants to punish “those crimes whose impunity . . . dissolves the metaphysical dignity of the person,”66 and in doing so advances a non-liberal concept of right, goes so far as to draw such conclusions. Last but not least, many Kant interpreters appear to me usually to misunderstand the Doctrine of right as a depiction of a liberal concept of right and thereby they choose more or less consciously the interpretation that they best hold to fit into Kant’s system. In this respect, I am more sympathetic to this – in my opinion, false – interpretation. Should I not be mistaken and were it to be correct, I would furthermore regret that Kant in the Doctrine of right did not derive the corresponding legal regulations from the liberal concept of right. Owing to this, because the interpretation of Kant is still open, I will, in what follows, inquire into the respective consequences that arise from both of the interpretations. A further reason compels me to such an investigation. The argument from } 49e in the Doctrine of right in 65. RL Ak vi:277–80; RL Ak vi:325; RL Ak vi:333; RL Ak vi:363. Practical philosophy, ed. Gregor, pp. 426–30; p. 467; p. 474; p. 498. 66. Vittorio Ho¨sle, “Was darf und was soll der Staat bestrafen?” in Ho¨sle, Rechtsphilosophie des deutschen Idealismus (Hamburg: F. Meiner, 1989), pp. 54–5.

the two kantian concepts of right

43

favor of a theory of retributivism cannot be classified, at first glance, into a liberal legal framework. Just as often, we find in the Doctrine of right elements67 with which the liberal concept of right is incompatible. If Kant speaks, for example, of a proportionality of punishment “to his [the criminal’s] inner wickedness” he obviously means an internal wickedness as opposed to an external wickedness, that is, to a mere contrariness to right of external freedom. It is completely possible that such elements actually do not belong to the liberal concept of right, though yet being compatible with it, and that we are dealing with two heterogeneous yet consistent theories of penal law. In this case we would have to observe that Kant does not keep to the division into the Doctrine of right and the Doctrine of virtue; a Kantian retributivist penal theory would still, however, remain tenable, although in this case it would have to be based on a moral argument. It would then simply reinforce the non-liberal interpretation of Kant’s concept of right. Therefore, after we have first investigated the arguments that are neutral to the moral theory as well as those arguments that are influenced by moral theory in support of retributivism, we can count Kant’s retributivist penal theory as being refuted. In what follows, I will accordingly test initially the arguments independent of the moral theory in favor of retributivism (the second chapter of this book) and then the arguments influenced by moral theory (the third chapter). Both the liberal concept of right and Kant’s moral theory will prove to be incompatible with his retributivist penal theory. 67. Cf. Jean-Christophe Merle, “Il punto di vista educativo e religioso dei Contributi dentinati a rettificare il giudizio del pubblico sulla Rivoluzione francese: la dimensione politica del Saggio di una critica di ogni rivelazione,” in Aldo Masullo and Marco Ivaldo (eds.), Filosofia trascendentale e destinazione etica (Milan: Guerini, 1995), pp. 303–25.

2 KANT’S LEGAL JUSTIFICATION OF PUNISHMENT

Utilitarianism and deontological ethics have traditionally dominated the debate about the justification of punishments and have thereby been associated respectively with the theories of deterrent and retributivist punishment. Nowadays, the debate about the justification of punishment is increasingly dominated by mixed theories.1 Most of these mixed theories represent the attempt of deontological philosophers, that is, mainly those influenced by Kant, to break with the traditional conception of the deontological, that is, the rather Kantian justification of punishment, as a purely retributive theory. Such theories were actually – with good reason – suspected of resting more on private morality than on legal principles. In what follows, I will assess the success of these attempts to strengthen the Kantian retributivist theory through the integration into the justification of punishment of an element of some form of deterrence or prevention. First, I hope to show that these attempts are only superficially mixed, whereas in reality they are completely based on a retributive foundation. Secondly, I will inquire into the arguments that Kant employs to justify the principle of retribution and the type of retribution he adopts, and attempt to determine the weaknesses of this argument. Finally, I will propose a principle of punishment that, in my opinion, fits in better with Kant’s principle 1. Cf. Don E. Scheid, “Kant’s retributivism,” Ethics, 93 (1983), 262–82; Sharon Byrd, “Kant’s theory of punishment: deterrence in its threat, retribution in its execution,” Law and Philosophy, 8, no. 2 (1989), 151–200; Thomas E. Hill, “Kant on punishment: a coherent mix of deterrence and retribution,” Annual Review of Law and Ethics, 5 (1997), 291–314; Sarah Holtman, “Toward social reform: Kant’s penal theory reinterpreted,” Utilitas, 9 (1997), 3–21; Otfried Ho¨ffe, “Vom Straf- und Begnadigungsrecht,” in Ho¨ffe (ed.), Metaphysische Anfangsgru¨nde der Rechtslehre, pp. 213–33.

44

kant’s legal justification of punishment

45

of right than his own conception of punishment. In so doing, I will consciously and consistently employ Kant’s principle of right as a criterion for judging the correctness of that principle of punishment. The “Introduction to the doctrine of right” formulates the principle of right in the following manner in } B: “Right is therefore the sum of the conditions under which the choice of one can be united with the choice of another in accordance with a universal law of freedom.”2 Upon this foundation, I will criticize Kant’s theory of punishment, and attempt to construct one anew.

2.1. The interpretation of Kant’s penal theory as a mixed theory For a long time the philosophical debate about penal theory was dominated by the dichotomy between retribution, exemplarily illustrated by the Kantian school, and general deterrence. By the latter term, I mean deterring the commission of future crimes by deterring not only the criminal but also other citizens. Recently, however, the situation has radically changed: the discussion is now dominated by mixed theories, which conceive of the theories of retributivism and of deterrence as reciprocally restricting and complementing each other. On the one hand, general deterrence theorists clearly profess culpability as being a prerequisite to every punishment and endorse some kind of proportionality between a crime and its punishment. On the other hand, retributivists generally agree with the deterrence theorists on the observation that although the principle of retribution in itself indeed forms a moral principle it does not qualify as a principle of right in the cases in which punishment demanded by it obviously would not deter from the commission of any crime. Each of these positions complements the other, because the element that is stronger in one is weaker in the other, and vice versa. The general deterrence theories remedy a weakness in the principle of retribution by providing a rationale for punishment that clearly connects it not with a private morality, but instead with a system of public law. Retributivism improves the theory of deterrence by furnishing a principle that gives a simple answer to the question of proportionality of punishment and seems to be unobjectionable with relation to justice toward individuals. Sharon Byrd’s important article about 2. RL Ak vi:230. Practical philosophy, ed. Gregor, p. 387.

46

desert as the sole justification for punishment

Kant’s conception of penal law3 brilliantly contrasted this combination with the traditional view of Kant’s penal theory according to which it is through and through a retributivist penal theory. Byrd argues that legal systems must be secured against the inclination of citizens toward breach of the law and that they obtain this guarantee through threatening punishment. Hence, according to Byrd, deterrence constitutes the fulfillment of the public right to coerce. But once the aim of punishment is determined in that way, the execution – that is, the type and the degree – of the punishment no longer follows the principle of deterrence, but instead the principle of retribution. The main reason Byrd proposes for this shift from deterrence to retribution is that only the latter treats human beings not merely as means but also as ends in themselves. Thus, Byrd points out, in Kant’s Doctrine of right deterrence is found in the threat of executing the punishment, whereas the retributive principle is found in the actual execution of the punishment, so that deterrence and retribution reciprocally “limit” each other. Even though Byrd is of the opinion that threatening and retribution reciprocally limit each other in this relation, there is, however, an obvious asymmetry between them, which, in my opinion, privileges retribution and explains the appeal of Kant’s retributivism in recent years. Indeed, even though in Byrd’s reconstruction general deterrence constitutes the aim of punishment, it is retribution that determines the “amount” of punishment. I would like to express this in the following manner. Citizens should be punished for their crimes if and only if: 1 the threat of punishment could deter them (deterrence condition); and 2 the punishment punishes the crime (retributivist condition); and 3 the amount of punishment is determined by the retributive principle (I will define this principle more precisely below). The first two conditions are purely negative prerequisites: if they are not satisfied then both the punishment and the threat of it are prohibited. They do not, however, positively determine the type and amount of punishment. For clarification of this point I would like to look at the strongest example upon which Byrd grounds her new interpretation of Kant, that is, Kant’s expounding on “right of 3. Byrd, “Kant’s theory of punishment.”

kant’s legal justification of punishment

47

necessity (Ius necessitatis)” by way of what Kant terms “the plank of Carneades”: There can be no penal law that would assign the death penalty to someone in a shipwreck who, in order to save his own life, shoves another, whose life is equally in danger, off a plank on which he had saved himself. For the punishment threatened by the law could not be greater than the loss of his own life. A penal law of this sort could not have the effect intended, since a threat of an ill that is still uncertain (death by a judicial verdict) cannot outweigh the fear of an ill that is certain (drowning), but only unpunishable (impunibile).4

Kant adopts two premises. The first premise states that the penal law “would assign the death penalty” (retributive determination of the degree of punishment, cf. the third point above). The second premise is formulated indirectly, that is to say, negatively: “A penal law of this sort could not have the effect intended.” The logical opposite to this is our requirement that penal law prevent some crimes, which means that the threat cannot be proven to have no effect. The logical opposite does not mean that the effect of deterrence must be either proven or certain or maximal. This prerequisite is so weak that I have to ask myself if there is any imaginable punishment that could ever fail to satisfy it. Is there truly no one who would prefer a certain drowning to a possible death by a judicial verdict? Provided that the risk of being caught (for example because there were witnesses to the event or for some other reason) is not negligible: is there really no person who prefers drowning and being mourned because of this tragic accident over the public dishonor of a judicial sentence possibly or probably leading to death? To stress this point, let us assume that the court will probably accept mitigating circumstances and sentence the criminal to a prison sentence of twenty years, so that the person who has the choice between murder and drowning could be sure that the death sentence would not be imposed. Can we exclude the possibility that some persons might prefer to die tragically rather than face a life in prison without honor and career chances? If that is so, then the threat of the death penalty (even of a prison sentence) really does prevent some murders even in such “a case of necessity.” I admit that a “not uncertain” death by judicial verdict would increase the deterrent effect, that is, it would prevent some 4. RL, “Appendix to the Introduction of The doctrine of right,” at Ak vi:235f. Practical philosophy, ed. Gregor, pp. 391–2.

48

desert as the sole justification for punishment

murders upon which the threat of an uncertain death or prison sentence by a judicial verdict would have no effect. At least in some cases, “a threat of an ill that is still uncertain” truly can and will “outweigh the fear of an ill that is certain.” The logical opposite to the expression “could not have the effect intended” requires only some deterrent effect, but by no means the maximal deterrent effect. That is why there can be a penal law that also imposes a death sentence (or a lesser sentence) for murder in such a case of necessity. In more general terms, I do not see any punishment, any imposition of pain on guilty persons, of which one can claim that it does not have in one way or another a deterrent effect (the punishment of innocent people, on the contrary, would probably completely fail, because such a punishment would not be related to a crime). Even if death was sanctioned with something as ridiculous as either a one-month period in jail or a 10,000 dollar fine, these punishments would still deter some potential murderers; therefore, they would still have a deterrent effect. Because the principle of retribution prescribes a punishment for all crimes, it seems that the first condition listed above (that the punishment must have a deterrent effect) will always be satisfied. Thus, the condition of deterrence does not in any way limit the principle of retribution. Nor does the principle of retribution limit the principle of deterrence, because the latter only requires some deterrent effect. However if we (in opposition to Byrd’s interpretation of Kant) were to understand the deterrence principle in such a way that it also determined the degree of punishment, then it would aim to maximize the deterrence effect. In this case, the principle of retribution would actually limit the principle of deterrence. In our case, of necessity, the maximization of the deterrence effect would require not only the death penalty, but also death after a lengthy and varied torture. One might object by contending that the prohibition of torture only means that a punishment should not surpass a certain threshold, which, for instance, would be established by the prohibition of a cruel or inhumane treatment, though below this threshold the principle of deterrence can still require the adoption of the punishment that is the most deterrent. This could be an alternative approach, but it is certainly not Kant’s or Byrd’s, because the principle of retribution does not allow any latitude to apply any sort of leximin. Rather, the principle of retribution allows only one solution, for instance the death penalty for murder as in the example given above.

kant’s legal justification of punishment

49

If, in Byrd’s essay, no incompatibility is present between the retributive and the deterrent elements, then it is due only to the weakness of the criterion that is adopted for the latter: the punishment must at least deter some future crimes. The retributive principle satisfies this criterion and is thus not really limited by it. Hence, all that this mixed theory adds to the classical theory of retribution is the condition that classical retributivism would not be justified if the prescribed threat of punishment would not actually prevent at least one crime. The claim that the aim of punishment for Kant rests in deterrence shows just one way in which the case for retributivism could be made stronger. For the rest, the mixed theory prescribes the same degree of punishment as the classical theory of retribution. In order to express more clearly my point that mixed theory displays no actual relationship between retributivism and deterrence, I would like to differentiate the possible meanings of a retributive theory. Retributivism can correspond to at least the four following theses: 1 All criminals, and only criminals, should be punished. 2 The punishment of criminals serves as retribution for the crimes committed. 3 The degree of punishment should be proportional (ordinally, not cardinally) to the crime, which means that the relation of the crimes amongst themselves should correspond to the relation of the punishments amongst themselves. By that I mean that a serious crime should be punished more harshly than a minor crime, and that two equally serious crimes should be punished with equivalent harshness. 4 The degree of punishment must be equivalent to the crime. The third thesis compares two classes of relations, that is, the relation between different crimes and the relation between different punishments. In contrast to that, the fourth thesis directly links a crime to a punishment without consideration of proportionality. The third thesis obviously does not imply the fourth thesis, it merely prohibits punishing a shoplifter more harshly than a murderer. Now consider the following possibilities. The first possibility is that the thief is sentenced to a week of community service, and the murderer to twenty years behind bars. The second possibility is that the thief is sentenced to one year in jail, and the murderer is sentenced to death. The third possibility is that the thief is sentenced to one week of community service, and the murderer is sentenced to death. All three examples comply with the third thesis. However, they are not just

50

desert as the sole justification for punishment

representing very different approaches to penal legislation. The first possibility definitely does not comply with the fourth thesis, because the murderer is not sentenced to death. The criterion for the fourth thesis appears to me to be better fulfilled by the third possibility than by the second, but I do admit that this might be controversial. I do not wish to discuss here the question whether a different penal law could be imagined that complies with the fourth thesis without complying with the third thesis, even though this appears to me not to be impossible. The only reason for my differentiation between the third and the fourth thesis is to reject the fourth thesis as being incompatible with Kant’s concept of right, even though Kant defends this thesis. Yet, I admit that the third thesis is statistically true (I will explain later what I mean by “statistically”).5 I observe that theses one through four remain untouched by Byrd’s mixed theory. I would like to distinguish now the possible theses concerning the theories of deterrence. 1 Future crimes are prevented by the punishment of actual criminals (in contrast to the following two theses, this descriptive thesis belongs to no normative theory of deterrence). 2 Future crimes should be prevented by the punishment of actual criminals. 3 Citizens should be punished in such a manner as to provide the most effective deterrent to future crimes. 4 Criminals and only criminals should be punished, and only in such manner that provides the most effective deterrent to future crimes. I observe that the third and fourth theses are not supported by mixed theory. Because the first thesis is a purely descriptive thesis, the only general deterrent thesis within the mixed theory that remains is the second thesis. Therefore, I will now concentrate on the principle of retribution. As a first step, I will inquire into Kant’s justification and attempt to show that his justification is clearly insufficient. As the second step, I will show that Kant’s principle of retribution contradicts his principle of right. 5. I take “statistically” to mean that the seriousness of the offense correlates with the degree of punishment actually carried out, even though there will also be cases where more serious crimes are punished more mildly than less grave ones. This will be explained at the end of this chapter as well as being illustrated in Chapter 7, using the example of punishment of crimes against humanity.

kant’s legal justification of punishment

51

2.2. Retributivism as a justification of punishment Before we inquire into Kant’s justification of the principle of retribution we will have to distinguish between two meanings of retribution, a stronger one and a weaker one: these are Vergeltung (the second retributivism thesis, or mere retribution) and Wiedervergeltung (the fourth retributivism thesis, or ius talionis).6 Most of the current Kantian retributivists argue for Vergeltung and reject Wiedervergeltung.7 In } 49e of the Doctrine of right, Kant argues in a two-step process: first, he attempts to justify the right to punish in general; secondly, he attempts to justify Wiedervergeltung, or ius talionis, as a principle for deciding the degree of punishment. In the first step, Kant rejects the utilitarian conception of penal law: Punishment by a court . . . can never be inflicted merely as a means to promote some other good for the criminal himself or for civil society . . . The law of punishment is a categorical imperative, and woe to him who crawls through the windings of eudaimonism.8

Most interpreters implicitly derive the principle of retribution (Vergeltung) from this rebuttal. Yet neither the word retribution (Vergeltung) nor its concept is present before the second step. Even then, one cannot find mere retribution (Vergeltung) but only equality in retribution (Wiedervergeltung). Therefore, I suspect that the interpretation of the first step as a justification of mere retribution comes from the interpretation of the second step, that is, from the justification of retaliation. Illustratively, the second step contains yet again the same rejection of the utilitarian theory as the first step: according to Kant, all principles except ius talionis “are fluctuating and unsuited for a sentence of pure and strict justice because extraneous considerations are mixed into them.”9 In part five of the Appendix, Kant makes a plea for rejecting the utilitarian concept of punishment in favor of ius talionis.10 6. The German terms Vergeltung (that is, requiring comparable recompense for a wrong) and Wiedervergeltung (that is, demanding wholly equal recompense) correspond roughly to the English terms retribution and retaliation (etymologically: ius talionis), respectively. 7. Cf. Schied, “Kant’s retributivism”; Jeffrie G. Murphy, “Does Kant have a theory of punishment?” Columbia Law Review, 87 (1987), 509–32; and Ho¨ffe, “Vom Straf- und Begnadigungsrecht.” 8. RL Ak vi:331. Practical philosophy, ed. Gregor, p. 473. 9. RL Ak vi:332. Practical philosophy, ed. Gregor, p. 473. 10. RL Ak vi:363. Practical philosophy, ed. Gregor, pp. 497f.

52

desert as the sole justification for punishment

Kant’s formulation of retaliation, that is, ius talionis, confirms my suspicion: Whatever undeserved evil you inflict upon another within the people, that you inflict upon yourself. If you insult him, you insult yourself; if you steal from him, you steal from yourself; if you strike him, you strike yourself; if you kill him, you kill yourself.11

The first sentence cites an “undeserved evil” and therefore does not base itself on degree of punishment; thus it cannot concern retaliation, that is, equality in retribution. Furthermore, whoever wishes to understand the passage as a plea for mere retribution will have to derive this concept from the second sentence. In fact, there are no examples to be found in the second sentence for the “like-for-like” principle of punishment. For instance, the following interpretation of the first sentence would be completely sufficient: if you commit a crime, you throw society back to the state of nature in which you are not protected against undeserved evils that are committed against you. In this condition, there is no retribution, not even “mere” retribution. Such an interpretation obviously falls short of explaining the second sentence. Therefore, the interpretation of the first step as the justification of retribution is based merely on retaliation. This interpretation, though, bases itself singularly and alone on the rejection of the utilitarian position in which it is tacitly assumed that there cannot be a third alternative to retributivism and the utilitarian theory of deterrence. This assumption appears wrong to me. In order to demonstrate this, I will suggest another solution. First, however, I would like to make clear that Kant himself allows exceptions with regard to retaliation. The first exception rests in the alleged right of necessity in the example listed above in Section 2.1: There can be no penal law that would assign the death penalty to someone in a shipwreck who, in order to save his own life, shoves another, whose life is equally in danger, off a plank on which he had saved himself. For the punishment threatened by the law could not be greater than the loss of his own life . . . Hence the deed of saving one’s life by violence is not judged inculpable (inculpabile), but only unpunishable (inpunibile).12

11. RL Ak vi:332. Practical philosophy, ed. Gregor, p. 473. 12. RL Ak vi:236. Practical philosophy, ed. Gregor, p. 392.

kant’s legal justification of punishment

53

This case is supported by neither the second, the third or the fourth thesis of retributivism. There is a further exception, if the number of a murderer’s accomplices is so great that the state, in order to have no such criminals in it, could soon find itself without subjects . . . then the sovereign must also have it in his power . . . [to] pronounce a judgment that decrees for the criminals a sentence other than capital punishment, such as deportation.13

The first retributivism thesis must therefore be modified to: “All criminals and only criminals should be punished by the state, unless either the crime occurred in the state of nature [as in the first exception] or the punishment would lead to reversion back to that state.” Or, formulated differently: “All criminals, and only criminals, should be punished, presupposing that the state can ubiquitously enforce the law from the time before the deed till after the execution of the punishment.” This means: “In a stable state, all criminals and only criminals should be punished.” From this we can conclude that for Kant even the weakest retribution thesis, that is, the first thesis – and therefore also the strongest thesis, that is, the fourth retribution thesis (the principle of retaliation, or Wiedervergeltung) – possesses no absolute validity, but instead is subject to the realization of his principle of right.

2.3. The ambiguity of the concepts of retribution and of right Now I would like to explain why the principle of retribution has nothing to do with Kant’s principle of right, and furthermore why even its application in the Doctrine of right can stand in the way of his principle of right. It has already been demonstrated how the principle of retaliation suddenly arises in } 49e, without Kant attempting to deduce it from his principle of right. If we search for an argument for this principle in the Doctrine of right, then we find the following in the “Preliminary concepts of the metaphysics of morals (philosophia practica universalis).” If someone does the duty “he can be constrained by law to do”

13. RL Ak vi:334. Practical philosophy, ed. Gregor, p. 475.

54

desert as the sole justification for punishment he does what is owed (debitum) . . . if what he does is less than the law requires, it is morally culpable (demeritum). The rightful effect of what is culpable is punishment (poena) . . . conduct in keeping with what is owed has no rightful effect at all.14

The debt (debitum) results from the criminal not fulfilling what the law requires. It is here that the ambiguity of the German term Schuld (“guilt”) comes into play. It can mean either debt (debitum) or culpability (culpa). Accordingly, retributivism can be schematically described in the following manner: 1 Whoever makes him- or herself culpable of a crime has not paid his or her debt to society. 2 Unpaid debts must be paid off. 3 Punishment is the paying off of debts. To take the two meanings of the word Schuld as being equivalent is misleading and leads to fallacies in the case of crimes that cause damage for which no compensation is possible, that is, for crimes for which there is no reparation. All infringements that are not irreparable are private crimes not falling under the jurisdiction of criminal courts, but rather under that of civil courts. Kant gives the example of “embezzlement, that is misappropriation of money or goods entrusted for commerce, and fraud in buying and selling when committed in such a way that the other could detect it.”15 Such crimes endanger the existence not of the commonwealth, but only of the individuals who trusted and freely entered into a contract with the criminal. On the other hand, public crimes do not damage private legal contracts, but instead the commonwealth itself. For such crimes, as in the case of the exercise of possibly deadly force upon a victim, there is no possible compensation. How can, for instance, a prison or death sentence ever provide compensation for the victim of such a crime? What could ever provide compensation for the insecurity of all citizens caused by a murder? The debts (debita) to the commonwealth can only be paid off when there is abstention from committing crimes in the future: as soon as a crime occurs, though, the criminal is no longer able to pay off his or her debts. Once the criminal has lost civil personality, then this person has lost even the possibility of paying off the debt to the community in the future, that is, to abide by the law in the future. 14. RL Ak vi:227f. Practical philosophy, ed. Gregor, p. 382. 15. RL Ak vi:331. Practical philosophy, ed. Gregor, p. 472.

kant’s legal justification of punishment

55

The objection could be raised that there is no real possible compensation between the commonwealth and the criminal, but at least an internal compensation is possible. An internal compensation of this sort stands possibly connected with the problem of the “highest good” – that is, with the “exact correspondence of happiness with morality.”16 Kant seems to support himself on such an idea when he writes: This fitting of punishment to the crime, which can occur only by a judge imposing the death sentence in accordance with the strict law of retribution, is shown by the fact that only by this is a sentence of death pronounced on every criminal in proportion to his inner wickedness.17

There are two kinds of objections that can be raised against such a justification of retributivism.18 The first and most obvious objection is that, according to Kant, the state exercises control in a state of laws, that is to say, it regulates the relation between the external freedom of the individuals to one another. The highest good is, on the contrary, part of an internal and therefore individual relation.19 Secondly, even if procurement of the highest good falls under the purview of the state, then the requirements for the highest good would never justify a retributivist penal theory, but would rather refute it. The highest good would require that the relative relation between immorality and unhappiness, or pain (or remorse), should be the same. Actually, Kant defines punishment as “the right a ruler has against a subject to inflict pain upon him because of his having committed a crime.”20 What should the pain consist of so that it could be comparable with the amount of immorality? Let us rely on Kant’s definition from the Critique of practical reason: Happiness is the state of a rational being in the world in the whole of whose existence everything goes according to his wish and will, and rests, therefore, on the harmony of nature with his whole end as well as with the essential determining ground of his will.21 16. KpV Ak v:125. Practical philosophy, ed. Gregor, p. 240. 17. RL Ak vi:333. Practical philosophy, ed. Gregor, p. 474. 18. For a more detailed refutation of theories of punishment as restoration of harmony, cf. Jean-Claude Wolf, “Strafe als Widerherstellung eines Gleichgewichts,” Jahrbuch fu¨r Recht und Ethik, 11 (2003), 199–216. 19. Cf. Thomas E. Hill, “Kant on wrongdoing, desert, and punishment,” Law and Philosophy, 18 (1999), 407–41 (p. 429); see also below, Chapter 3. 20. RL Ak vi:331. Practical philosophy, ed. Gregor, p. 472. 21. KpV Ak v:124. Practical philosophy, ed. Gregor, p. 240.

56

desert as the sole justification for punishment

If pain is the opposite of happiness, then pain must be the state of a rational being to whom, in the whole of existence, nothing occurs according to his or her “wish and will.” Therefore, pain should consist in the whole absence of harmony between the person’s nature and his or her end, as well as between the end and the “essential determining ground of his will.” Owing to the punishment, the criminal, who is forced to undergo stringent discipline in order to make the “wish and will” spotless again, does not profit from the criminal deed. Therefore, one must grant that punishment for the purpose of reform brings with it much pain. Ius talionis also does not make the criminal happy. It does, however, allow that something in the world occurs according to the “wish and will” of the criminal. The principle of punishment would really be decided according to the criminal’s maxim: if the criminal really wanted to kill, then the state will kill the criminal; if the property of another is taken by a thief, then the property of the thief will be taken, etc. Paradoxically, the punishment reaches something like a universalization of the criminal’s maxim. Now, for Kant, the universalization of a maxim poses something of a test for morality. Even the merely partial universalization of an evil maxim as a maxim for action is invariably forbidden. This should be understood to be a warning signal against ius talionis. It is also, coincidentally, the maxim upon which a vendetta, that is, a personal revenge or blood-feud, is based. My arguments are indifferent to the debate as to whether ius talionis is to be taken literally regarding Kant. Everyone knows Hegel’s ironic commentary against the literal application of ius talionis.22 However, Kant declines punishing sex offenders with rape – he rather demands that they be castrated.23 He sentences the bandit to slave labor, but surely not in order to allow the bandit to fall victim to the same violence felt by the victims of banditry.24 Because enough Kant interpreters have emphasized this point,25 it is not necessary to explain any further why my argument does not have to distinguish between the third thesis and the fourth thesis, but instead allows me to reject these theses along with the second thesis. The only retribution taking place in the punishment is purely a negative type. Kant writes: “Whoever steals makes the property of 22. 23. 24. 25.

Cf. GPhR } 101 Anm. RL Ak vi:363. Practical philosophy, ed. Gregor, p. 498. RL Ak vi:333. Practical philosophy, ed. Gregor, pp. 474–5. For example, Ho¨ffe, “Von Straf- und Begnadigungsrecht,” p. 227.

kant’s legal justification of punishment

57

everyone insecure and therefore deprives himself (by the principle of retribution [Wiedervergeltung]) of security in any possible property.”26 This means nothing other than the criminal being shut out of the commonwealth and sent back into the state of nature. Therefore, only criminals can and should be punished, under the condition given by the first modified retributivism thesis27 that a stable state exists. Retributivism is correct then only if we understand it in that way. The use of the word “retribution” in this case is indeed misleading. In fact, both the reciprocal limiting of rights in the commonwealth and reciprocal protection of the citizens of a state could be called reciprocal retribution. Recognition of the rights of another and the protection of them are actually a type of reciprocal contribution or “tribute.” In contrast to that, a crime is merely negative and in no respect a contribution. For this reason, the terms “retribution” and “retaliation” appear to be ill suited to describe the modified first thesis. The simplicity and strictness of ius talionis may be fascinating for some people. The strictness of a punishment ought not to be mistaken, however, for strictness regarding a concept. Ius talionis cannot be deduced from Kant’s concept of right, but rather contradicts that very concept; it does not even originate from common sense for it is disputed in equal measure between experts on penal law and between citizens. Above all else, it is possible to reconstruct Kant’s theory of punishment directly from his concept of right without ius talionis.

2.4. Exclusion from the commonwealth In fact, besides the dichotomy between the utilitarian theory of deterrence and the so-called Kantian theory of retribution, there is a third, authentically Kantian and legal approach. Neither the prevention of repetition of a crime nor the criminal’s reform, which could be called specific deterrence, is given much attention in today’s debate about punishment, because both ideas are alleged to replace the punishment’s true meaning with social considerations that have nothing to do with the crime. I regard such an objection as being completely baseless, and believe that this type of specific deterrence could provide rationale for the punishment itself. 26. RL Ak vi:333. Practical philosophy, ed. Gregor, p. 474. 27. The modified thesis: “All criminals and only criminals should be punished by the state, unless either the crime occurred in the state of nature [as in the first exception] or the punishment would lead to reversion back to that state.”

58

desert as the sole justification for punishment

In the following, I will describe the theory of punishment that Kant should have built upon his principle of right, in order then to criticize the theory he actually constructed upon it, which, in my opinion, contradicts this principle. I will refer as far as possible, though, to individual passages in Kant’s Doctrine of right so long as they appear to me to be compatible with his principle of right. I will argue for the following theses: 1 Criminals ought to be punished in order to prevent them from carrying out further crimes. This ought to be accomplished by measures (2) and (3). 2 Punishment ought to deprive the criminal of the status of a citizen until (3) is accomplished. 3 The punishment ought to reeducate the criminal so that the criminal’s status as a citizen can be recognized again. In the following, I will connect the principle of right with the categorical imperative in order to reconstruct a Kantian theory of punishment. Besides the rejection of utilitarianism, Kant names two other arguments in his first step (see Section 2.2), two affirmative arguments: first, the criminal must be punished “because of his having committed a crime”; secondly, the penal law is a “categorical imperative.”28 Both these points contain a descriptive and a normative element: on account of the crime committed by the criminal, the said criminal must be punished for the sake of the categorical imperative. There is, however, in the Doctrine of right no mention of an imperative that specially relates to the penal law. In the proper meaning of the term, only one single categorical imperative can exist, for which the Groundwork of the metaphysics of morals offers different formulations. The Doctrine of right develops an imperative of right as a particularization of the categorical imperative by factoring in the circumstance that a large number of people share the same limited world. I propose to conceive of penal law’s “categorical imperative” as being a particularization of the categorical imperative of right, under the premise that a crime has been committed. By definition, because of the crime, a state of nature prevails over the relationship between the criminal and the rest of the commonwealth. Kant regards the state of law and the state of nature as a dichotomy that excludes any third possibility. In fact, Kant terms the 28. RL Ak vi:331. Practical philosophy, ed. Gregor, p. 472.

kant’s legal justification of punishment

59

situation after the crime’s occurrence thrice as “state of nature”: in the case of a murder with many accomplices, in the case of a duel and in the case of infanticide.29 In an original state of nature before the establishment of a commonwealth, as well as during periods of regression back to the state of nature, the categorical imperative demands the creation of a legal system. Should the establishment of the institutions constitutive of a state of law not be immediately possible, then permissive law30 demands achievement of it for everyone – therefore, for the criminal as well – in the quickest way possible. I am arguing that nothing other than punishment can rehabilitate and make the criminal a member of the commonwealth again, because one must consider that the criminal is not immediately able to reenter the commonwealth. By committing a crime the criminal has obviously accepted the regression back to a state of nature. In a state of nature, nobody has any rights and external freedom is continually threatened, whereas, however, what the categorical imperative demands is that the latter be respected. The criminal accepts the disappearance of his or her external freedom as well. This means that the will of the criminal commits a type of suicide, which makes this person “unfit to be a citizen.”31 The criminal cannot be treated any longer by the state as a free person, as a rational being. The type and the degree of punishment that Kant requires demonstrate this in a clear way. The most famous example is surely the death sentence for murder. Hermann Cohen and some modern interpreters32 have raised a convincing objection against it, saying that this punishment contradicts the Kantian principle of morality because the punishment irrevocably destroys a rational being. For this reason they have proposed alternative punishments. Their argument has been so successful that nowadays there is hardly a 29. RL Ak vi:334, 336. Practical philosophy, ed. Gregor, p. 475, pp. 476–7. 30. ZeF Ak viii:347f. “Toward perpetual peace,” in Practical philosophy, ed. Gregor, p. 321. Cf. Reinhard Brandt, “Das Erlaubnisgesetz, oder: Vernunft und Geschichte in Kants Rechtslehre,” in R. Brandt (ed.), Rechtsphilosophie der Aufkla¨rung: Symposium Wolfenbu¨ttel 1981 (Berlin: De Gruyter, 1982), pp. 233–85. 31. RL Ak vi:331. Practical philosophy, ed. Gregor, p. 472. 32. Hermann Cohen, Schriften zur Philosophie und Zeitgeschichte, ed. Albert Go¨rland and Ernst Cassirer (Berlin: Akademie Verlag, 1928), p. 341. Cf. Robert A. Pugsley, “A retributivist argument against capital punishment,” Hofstra Law Review, 9 (1981), 1501–23 (p. 1516); Steven S. Schwarzschild, “Kantianism and the death penalty,” Archiv fu¨r Rechts- und Sozialphilosophie, 70 (1985), 343–77; Attila Ataner, “Kant on capital punishment and suicide,” Kant-Studien, 97, no. 4 (2006), 452–82.

60

desert as the sole justification for punishment

philosopher who dares to endorse this part of Kant’s penal law. (An interesting exception would be Tom Sorell, who argues in favor of both Kantian retributivism and the death penalty.)33 If one accepts the Kantian theory of penal law, as Cohen does, then such objections are, however, not convincing. They become so only when one reconstructs an alternative theory of penal law from Kant’s concept of right, which is explicitly counter to Kant’s theory of penal law.34 In fact, Kant would deal with these objections in the following way. First, a rational being ceases to be such as soon as a crime has been committed, and not only after he or she has been punished. The punishment only derives the consequence from the factor that through the crime a rational being denies its rational character. Secondly, the degree of punishment that Kant imposes for all other crimes he mentions shows clearly that he no longer sees the criminal as a rational being. In this vein, I wish to provide the following example: But what does it mean to say, “If you steal from someone, you steal from yourself”? Whoever steals makes the property of everyone else insecure and therefore deprives himself (by the principle of retribution) of security in any possible property. He has nothing and can also acquire nothing; but he still wants to live, and this is now possible only if others provide for him. But since the state will not provide for him free of charge, he must let it have his powers for any kind of work it pleases (in convict or prison labor) and is reduced to the status of a slave for a certain time, or permanently if the state sees fit.35

At this point, Kant fails to differentiate between two issues: (a) that someone without property has to work for his or her livelihood and (b) that the work for one’s own livelihood means enslavement, instead of, for example, working as a day laborer. Actually, there is a commonality between this enslavement and the death penalty, just as between every other punishment mentioned in the Doctrine of right, for example deportation,36 “permanent expulsion from civil society,” or castration.37 To Kant, castration is a “partial” murder, as is self-castration: “To deprive oneself of an integral part or organ (to maim oneself) . . .

33. 34. 35. 36. 37.

Tom Sorell, Moral theory and capital punishment (Oxford: Blackwell, 1987), p. 162. Cf. Section 2.5. RL Ak vi:333. Practical philosophy, ed. Gregor, p. 474. RL Ak vi:334. Practical philosophy, ed. Gregor, p. 475. RL Ak vi:363, Appendix 5. Practical philosophy, ed. Gregor, p. 498.

kant’s legal justification of punishment

61

are ways of partially murdering oneself.”38 Even if we leave aside the question whether suicide is punishable or not, then castration by another person still remains a type of partial death. With all of these kinds of punishment, the criminal is neither seen as being a constituent of the commonwealth nor treated as being such. Kant himself speaks against slavery, just as he does against voluntary slavery as well as against selling one’s own children or against inherited slavery. Yet he makes a single exception expressly against a person who has “forfeited his personality by a crime.”39 The exact consequence of this exclusion appears to have been expressed most clearly by Fichte: either the criminal is deported like an “outlaw” into the desert, where an even more horrible death awaits than the legal execution of a death penalty, or the criminal becomes someone who anyone may kill, as dangerous animals are killed. This extreme conclusion resulting from exclusion from civil society has only a heuristic function, however, since just as it is nowhere near Fichte’s last word on the subject, so little shall it be the last word in our reconstruction of Kant’s theory of punishment.

2.5. Rehabilitation by discipline Even though the criminal may have lost his or her “personality,” it is forbidden from the Kantian standpoint to treat the criminal as one would a thing or a (possibly dangerous) animal, for the following reasons. The most important provided by Kant in } 49e of the Doctrine of right seems to rely on the following formulation of the categorical imperative: “So act that you use humanity, whether in your own person or in the person of any other, always at the same time as an end, never merely as a means.”40 At first glance, the Doctrine of right seems to apply this formulation to the instance of the criminal: For a human being can never be treated merely as a means to the purposes of the other or be put among the objects of rights to things: his innate personality protects him from this, even though he can be condemned to lose his civil personality.41 38. 39. 40. 41.

TL Ak vi:423. Practical philosophy, ed. Gregor, p. 547. RL Ak vi:283. Practical philosophy, ed. Gregor, p. 431. GMS Ak iv:429. Practical philosophy, ed. Gregor, p. 80. RL Ak vi:332. Practical philosophy, ed. Gregor, p. 473.

62

desert as the sole justification for punishment

Although these sentences resemble one another, they do not appear to use exactly the same dichotomy: whereas the Groundwork of the metaphysics of morals speaks of “humanity” and “personality,” the Doctrine of right uses the words “innate personality” and “civil personality.” However, if we inquire into the meaning of the latter dichotomy, the formulation of the Groundwork would then appear relevant for explaining the sentence from the Doctrine of right. Kant defines personality in relation to the concept of imputability. In the “Preliminary concepts of the metaphysics of morals (philosophia practica universalis)” he gives the following definition: A person is a subject whose actions can be imputed to him. Moral personality is therefore nothing other than the freedom of a rational being under moral laws (whereas psychological personality is merely the ability to be conscious of one’s identity in different conditions of one’s existence). From this it follows that a person is subject to no other laws than those he gives to himself (either alone or at least along with others). A thing is that to which nothing can be imputed. Any object of free choice which itself lacks freedom is therefore called a thing (res corporalis).42

Thomas Pogge reads this passage – in my opinion, correctly – as consisting of two definitions: By italicizing “moral,” Kant flags that “moral personality” is more specific than “personhood.” The most plausible specification, which would also vindicate the “therefore,” is this: Having moral personality means being a subject whose inner actions are capable of imputation, a subject with (transcendental) freedom of the will . . . Persons in the wider, weaker sense are then subjects whose external actions can be imputed to them as expressive of their will, choice, or intentions.43

I shall term the latter a subject with freedom of action, in contrast to the former, which, in addition, possesses freedom of the will. According to Pogge’s view, a moral personality presupposes the other, weaker personality and, hence, freedom of action as well, which can be guaranteed only by a legal system. This means that moral personality presupposes civil personality. On the other hand, one

42. RL Ak vi:224. Practical philosophy, ed. Gregor, p. 378. 43. Thomas Pogge, “Is Kant’s Rechtslehre comprehensive?” Southern Journal of Philosophy, 36 supplement (1997), 161–87 (p. 163).

kant’s legal justification of punishment

63

can construct a legal system that simultaneously ensures freedom of action and neither presupposes nor demands that one dispose of freedom of the will. Of what does the “innate personality” consist to which Kant refers? Since he separates it from the civil personality, it can concern either the moral personality or a third type of personality. But since the moral personality presupposes the civil personality, but the “innate personality” is opposed to the civil personality, the “innate personality” cannot then be the moral personality. From the term “innate personality” we can derive that this personality is original (that is, not acquired) and inalienable. From the term “personality” we can draw that it is somehow dependent on or relates to a type of imputability. Since Kant, while distinguishing persons from things, identifies “imputation” with “free choice,” that is, identifies it as being freedom of action, one could conclude that the innate personality, as well as the civil and indirectly the moral personality, should be guaranteed by a legal system. Actually, the innate personality receives the protection of penal law. However, the innate personality is not the same as the civil personality, and one can be the bearer of the former without being the bearer of the latter. I now turn to the reason why the “innate personality” according to } 49e of the Doctrine of right cannot simply be treated as a mere means. According to the Groundwork of the metaphysics of morals,44 the reason consists in the humanity in either my person or the person of someone else. In this context, the concept of “humanity” can now be understood in a misleading way, that is to say, by linking it to a dichotomy that only plays a role where the differentiation between morality and right is at stake. Now the human being as a natural being that has reason (homo phaenomenon) can be determined by his reason, as a cause, to actions in the sensible world, and so far the concept of obligation does not come into consideration. But the same human being thought [of] in terms of his personality, that is, as a being endowed with inner freedom (homo noumenon), is regarded as a being that can be put under obligation and, indeed, under obligation to himself (to the humanity in his own person).45 In the doctrine of duties a human being can and should be represented in terms of his capacity for freedom, which is wholly supersensible, and

44. GMS Ak iv:429. Practical philosophy, ed. Gregor, p. 80. 45. TL Ak vi:418. Practical philosophy, ed. Gregor, p. 544.

64

desert as the sole justification for punishment so too merely in terms of his humanity, his personality independent of physical attributes (homo noumenon), as distinguished from the same subject represented as affected by physical attributes, a human being (homo phaenomenon).46

In both of these citations, “humanity” clearly means moral personality or freedom of the will. Understanding “humanity” in the formulation of the categorical imperative in this way would then make its statement redundant: “So act that you use [moral personality], whether in your own person or in the person of any other, always at the same time as an end, never merely as a means.”47 The Groundwork confirms this reading: A human being, however, is not a thing and hence not something that can be used merely as a means, but must in all his actions always be regarded as an end in itself. I cannot, therefore, dispose of a human being in my own person by maiming, damaging or killing him.48

In the specific context of the Doctrine of right, the following reading prevails: Every human being has a legitimate claim to respect from his fellow human beings and is in turn bound to respect every other. Humanity itself is a dignity; for a human being cannot be used merely as a means by any human being (either by others or even by himself) but must always be used at the same time as an end. It is just in this that his dignity (personality) consists, by which he raises himself above all other beings in the world that are not human beings and yet can be used, and so over all things.49

Forbidding that human beings be “used” can be read in two different ways. Seen from the standpoint of the categorical imperative, it is surely justified, since absolutely every human is able to develop autonomy, that is, free will. From the standpoint of the legal system, it means that this system in no way protects only those citizens who abide by it and limit their freedom of action in such a way as to be compatible with the freedom of action of every other citizen. Rather, the legal system protects all humans, even those who are actually unable to abide by it, for example children and criminals. By analogy 46. 47. 48. 49.

RL Ak vi:239. Practical philosophy, ed. Gregor, p. 395. GMS Ak iv:429. Practical philosophy, ed. Gregor, p. 80. GMS Ak iv:429. Practical philosophy, ed. Gregor, p. 80. TL Ak iv:462. Practical philosophy, ed. Gregor, p. 579.

kant’s legal justification of punishment

65

to the manner in which Kant deals with children, I propose that criminals are differentiated from citizens by the use of the term “citizens of the world.” In fact, Kant writes about parents: They cannot destroy their child as if he were something they had made (since a being endowed with freedom cannot be a product of this kind) or as if he were their property, nor can they even just abandon him to chance, since they have brought not merely a worldly being but a citizen of the world into a condition which cannot now be indifferent to them even just according to concepts of right.50

Neither children nor criminals are in a position to abide by the law, that is, to limit their freedom of action by the freedom of action of others. Yet both of them are capable of acting; therefore, they are endowed with freedom, though at the same time lacking the ability to develop, in a way compatible with the freedom of all the other persons (borrowing Pogge’s phrasing), “their will, choice, or intentions.” Therefore, the legal system cannot allow them freedom of action, though it must at the same time provide the possibility of obtaining this freedom, as soon as and for as long as they are in the position to respect the freedom of others. In my opinion, that is the meaning of the term “citizen of the world” in the aforementioned passage. At the very least, I see two important confirmations of my interpretation. First, since Idea for a universal history with a cosmopolitan purpose (1784), Kant has assumed, in human beings, a “natural disposition” (Naturanlage) for developing reason within a historical process till a “perfectly just civil constitution” is established.51 In this context, “reason” is not to be conflated with the transcendental meaning of autonomy or freedom of the will, but instead to be seen as being “a faculty which enables the creature to extend far beyond the limits of natural instinct the rules and intentions it follows in using its various powers, and the range of its projects is unbounded.”52 Thus, according to Kant, the innate character belonging to all agents requires their own development toward the establishment of and membership in a commonwealth, which is a notion that finds further development in the Doctrine of right. Other writings – first and foremost Toward perpetual peace – contain this perspective.

50. RL Ak vi:281. Practical philosophy, ed. Gregor, p. 430. 51. Idee, Proposition 5, Ak viii:22. Political writings, ed. Reiss, p. 46. 52. Idee, Proposition 2, Ak viii:18. Political writings, ed. Reiss, p. 42.

66

desert as the sole justification for punishment

Secondly, even criminals sentenced to slavery are expressly not subjected to regulation by the “right of things” (or the law of property). For Kant, the difference between a slave and any animal consists of the slave being able to fulfill duty, even when he or she possesses no rights whatsoever. In the “Division of the metaphysics of morals as a whole,” Kant defines the legal status of things and animals as being “the relation in terms of rights of human beings toward beings that have neither rights nor duties.” Kant says, with regard to “the relation in terms of rights of human beings toward beings that have only duties but no rights,” that it is missing (“vacat”) in the doctrine of right “for these things would be human beings without personality (serfs, slaves).”53 Though these human beings lack status as citizens, they cannot, however, be either used for “shameful purposes” or “depose [d] of . . . life and members.”54 One could object to such an interpretation on the grounds that it is supported by metaphysical assumptions. Indeed, Kant obviously makes metaphysical assumptions in the Doctrine of right, for example relating to child-rearing: “the offspring is a person, and it is impossible to form a concept of the production of a being endowed with freedom through a physical operation.”55 Yet one need not accept such assumptions to recognize agents as potential citizens or as actual citizens, as soon as and for as long as they are able to respect the rights of others for whatever reasons they choose, be it due to the observance of the categorical imperative or be it simply due to selfinterest for having their own rights recognized by the commonwealth. Every “citizen of the world” has a claim on the state maintaining the possibility for him or her to become an actual citizen of the said state. Moreover, the establishment of a commonwealth amongst all human beings is a postulate or a prerequisite. It is a prerequisite not only from the standpoint of the categorical imperative, but also when we accept with Pogge’s “one-way thesis” that as soon as we accept Kant’s categorical imperative we also have to accept his system of right, though not the other way around. Actually, Kant’s doctrine of right is not only descriptive, but also normative – it clearly contains the criterion of universalizability and applies it to all subjects of external freedom, for example agents. Therefore, Kant’s system of right does not create 53. RL Ak vi:241. Practical philosophy, ed. Gregor, p. 396. 54. RL Ak vi:330. Practical philosophy, ed. Gregor, p. 471. 55. RL Ak vi:280. Practical philosophy, ed. Gregor, p. 429.

kant’s legal justification of punishment

67

agency, but instead presupposes it, and ought to incorporate as much as possible the principle of the reciprocal restriction of freedom. In my opinion, this requires that the legal system promote and develop the ability to follow the law. In fact, Kant explicitly deals with the development of this ability in children and in “savages.” Now, there is no difference between criminals and children in respect to their legal status. Therefore, I will view the treatment of criminals as being analogous to Kant’s treatment of children. In Pedagogy, Kant conceives of child-rearing as a twofold task: “discipline” and “culture.”56 This goal is to be followed regarding children as well as “savages.” Kant’s concept of right consists in the reciprocal coercion of citizens guaranteed through public coercion exercised by the state. In contrast thereto, the criminal is like a child and a “savage” and an object of unilateral coercion, which is termed by Kant in Pedagogy “discipline” or “breeding.” “Discipline submits the human being to the laws of humanity and begins to let him feel the coercion of the laws.” Kant conceives of discipline as being a prerequisite for culture, that is, for the positive part of education. Respect for the law must first be acquired through an external force applied against human beings, before the ability is gained to obey the law voluntarily and – most importantly – to be one’s own legislator, which constitutes freedom under law, which is equivalent to the civil state. When we coerce a dog to obey a rule what we really want is the dog to obey our rule. With a human being, however, we ought to try to promote humanity as being the goal, that is, encourage external freedom under the law. Punishment ought to promote the criminal’s reform. Kant considers the possibility of a time-limit on the punishment. The thief “is reduced to the status of a slave for a certain time, or permanently if the state sees fit.”57 Unfortunately, however, this timelimit is found nowhere else in the Doctrine of right. The reason for this might be that Kant assumes that the education of “savages” is much more difficult than educating children. He explains: “But the human has from nature such a large tendency to freedom that if he has grown used to it for a while he sacrifices everything for it.”58 However 56. Pa¨d, Ak ix:449. (Translation mine) 57. RL Ak vi:333. Practical philosophy, ed. Gregor, p. 474. 58. Pa¨d, Ak ix:442. (Translation mine)

68

desert as the sole justification for punishment

nowhere does Kant explicitly exclude for a criminal the possibility of rehabilitation. In Kant’s Doctrine of right, a murderer does not always have to die. The cases listed above (see Section 2.2) are not the sole exceptions: not only in the case in which a murderer has too many accomplices, for example, but also in the case in which a sovereign exercises “his right of majesty” to grant “clemency . . . only in individual cases.”59 Therefore, the murderer can eventually be rehabilitated and released, unless his or her release would threaten the security of the other citizens, i.e. unless he or she is not yet “disciplined.” Now, the possibility of rehabilitation is completely incompatible with the death penalty. The alternative to retributivism that I am presenting here is in no way utilitarian. It absolutely does not view punishment as being “merely a means to promote some other good for the criminal himself or for civil society.”60 Admittedly, the murderer certainly profits more from his or her rehabilitation than from the death penalty. Yet the rehabilitation model that I have sketched out here supports itself solely on the goal of restoring the civil state destroyed by the crime in such a way that the criminal can be reintroduced into that very state. The only available means to this end is punishment with unilateral coercion. During the time when the “discipline” is being carried out, the commonwealth is protected from the risk of the criminal repeating the crime. In the model of punishment I have sketched out, punishment therefore carries out an intention that is not only inherent in the concept of right but also inherent in the goal of preserving the humanity in every person. I do concede that the theory of general deterrence also follows a goal inherent in the concept of right, in so far as the punishment motivates the other citizens to respect the law; however, a theory that requires deterring the rest of the population from committing a crime does not really take the goal of preserving the humanity in every person seriously. In order to maximize the deterrent effect, the theory of general deterrence could end up increasing the punishment in duration and in degree beyond what is necessary to rehabilitate the criminal back into a full-fledged citizen. One may object to specific deterrence, which I propose as being an alternative to the principle of retribution, on the grounds that it infringes the fairness principle, according to which a punishment is 59. RL Ak vi:334. Practical philosophy, ed. Gregor, p. 475. 60. RL Ak vi:331. Practical philosophy, ed. Gregor, p. 473.

kant’s legal justification of punishment

69

not allowed to be lengthened or formulated anew once the criminal has already been sentenced. Such an objection would, however, rely on the false assumption that the punishment’s unrevisability implies that it is set to an exact duration, that is, to a sentence of a certain amount of either months or years. But the duration is clearly defined in my proposal: the punishment ends with the criminal’s rehabilitation, that is, with the reestablishment of his or her ability to abide by the law like other citizens, so that the criminal poses no more of a threat to the law than they do. A weightier objection might lie in the difficulty in determining exactly when the criminal is able to reenter the commonwealth as a full-fledged citizen. In order to assess whether a criminal is rehabilitated one certainly needs, in most cases, a complex decision in which there always exists a danger of error. Yet this danger is also present in every court sentence. Moreover, today’s judicial system does actually apply to a certain extent specific deterrence: most prisoners are released on parole, which reduces the effective term of imprisonment, and recidivists are often punished more harshly than first-time offenders, and so on. Let us imagine for a moment how a mechanism might look with which one could eliminate the risk of arbitrariness in assessing the degree of rehabilitation. One could imagine an institutional system that would gradually remove the burden of proof from the criminal and place it on competent judges. After a while, the burden of proof would shift and it would become increasingly more difficult for the appropriate judges to prove that the punishment should continue. Aside from that, such measures as parole, community service and more severe sentences against recidivists could constitute a sort of second “test.” The second test would be found in “real” life and would constitute at the same time both a threat to and an incentive for the prisoner to reintegrate sincerely into society and respect the legal system. Finally, the punishment could also include specific measures of specific deterrence. The judge Marianna Pfaelzer not only sentenced the hacker Kevin Mitnick to exactly four years in jail, but also forbade him from using computers and mobile telephones without written approval of his probation officer. Certainly specific deterrence and rehabilitation – the latter meaning strict discipline – prevent numerous crimes from being committed by other citizens. Therefore, both theories always include a certain dimension of general deterrence, even though they do not target its maximization. Both specific deterrence and rehabilitation, in

70

desert as the sole justification for punishment

accordance with Kant’s definition of punishment, “inflict pain upon” criminals61 and threatening pain for a breach of the law certainly is an important component of deterring potential criminals from crimes. Yet general deterrence does not become the rationale for punishment since we would still punish criminals even when not a single other citizen would be deterred from committing a crime by the punishment. Nonetheless, the effect of general deterrence caused by specific deterrence certainly does constitute a strong and very fortunate side effect of the punishment. I would, however, like to make it clear that the concept of punishment as being rehabilitation is in no way expressive of “overly compassionate feelings of an affected humanity,” for which reason Kant reproaches Beccaria’s theory of punishment.62 Discipline, “breeding,” enslavement and unilateral coercion certainly appear to criminals as being neither attractive nor enjoyable ways to be punished. Specific deterrence and rehabilitation may also frequently coincide with the proportionality of punishment required by the retributivist thesis formulated above on p. 49. Although one might not be able to determine in principle that a thief probably might be rehabilitated more quickly than a hostage-taker or even a murderer, one can do it empirically. However, there are certainly some cases in which just the opposite may be true. Last but not least, retributivism and general deterrence do not exclude the death penalty. In fact, Kant even explicitly prescribes it even though many Kantians either deplore this or maintain a careful silence on the subject. For a long time, there was a debate about the usefulness of the death penalty as a means of general deterrence.63 Yet all the participants in this debate agreed that it would be justified if it could be proven that the death penalty was effective in preventing crimes by other citizens. In opposition to both retributivism and general deterrence, my alternative radically excludes the death penalty. It also excludes sentences condemning the criminal to life in prison without the opportunity for a reduction, although it does not forbid keeping a criminal behind bars for life if all attempts at rehabilitation should fail. The solution that I have proposed appears

61. RL Ak vi:331. Practical philosophy, ed. Gregor, p. 472. 62. RL Ak vi:334f. Practical philosophy, ed. Gregor, p. 475. 63. A classic – and skeptical – representation of the debate is provided in H. L. A. Hart, Punishment and responsibility (Oxford: Clarendon Press, 1968).

kant’s legal justification of punishment

71

to me to reach in the best possible way what the twofold requirement of all theories of penal law ought to be: ensuring the safety of all citizens and respecting the humanity of all criminals.

2.6. Summary The “mixed theories” of penal law fail in their attempt truly to link retributivism and general deterrence. They necessarily fail because both theories clearly differentiate themselves from each other as justification for punishment. I have therefore proposed we undertake a twofold modification of the terms in the debate between deontological ethics and deterrence theory. First, I have attempted to show that Kant’s retributivist theory of punishment does not draw on his concept of right. Secondly, I have consciously avoided attempting to mix the Kantian concept of right with a principle of general deterrence, but instead have argued that the Kantian concept of right is completely compatible with a theory of specific deterrence, which encompasses both incapacitation and the rehabilitation of criminals. This theory is not a mixed theory, and nor does it have to be one. Therefore, if Kant’s theory of punishment is justifiable on the basis of the Kantian system, then this justification could only occur while attempting a moral argument from which legal consequences arise. In the following chapter, I wish to show that Kant’s moral theory cannot afford this justification. Quite to the contrary, it will confirm the end result of this chapter, which had as its point of departure the theory of right.

3 KANT’S MORAL JUSTIFICATION OF PUNISHMENT

Now that I have rejected Kant’s theory of retribution on the basis of a liberal interpretation of his concept of right in favor of a combination of specific deterrence and rehabilitation, I would like to investigate whether the alternative, stronger or moral interpretation of his concept of right can substantiate Kant’s theory of retaliation. As was shown in the first chapter, this interpretation consists of an implementation of the categorical imperative in the legal system so far as the commands and prohibitions of the categorical imperative are able to be implemented by force.

3.1. Which type of moral proportionality is at issue? The current retributivist theories of punishment often radically differ from Kant’s own theory because they assume that punishment is justified by a goal, however modest this goal may be. The position held by expressivists, for instance, states that punishment should be seen retributively so that society’s moral judgment about the punishable action can be expressed.1 Another theory of punishment, such as the one held by Jean Hampton and Jeffrie Murphy, regards the punishable action as comprised of the criminal putting his or her value over the worth of the victim.2 The retributive punishment that is then stipulated is a punishment that restores the proper relation between the worth of the victim and the worth of the criminal. 1. Cf. Thomas E. Hill, “Kant on punishment: a coherent mix of deterrence and retribution,” Annual Review of Law and Ethics, 5 (1997), 291–314 (pp. 320–1). 2. Cf. Jean Hampton and Jeffrie Murphy, Forgiveness and mercy (Cambridge: Cambridge University Press, 1988), pp. 45–53.

72

kant’s moral justification of punishment

73

In opposition to such theories as well as similar ones, and also against the legal tradition following Seneca’s dictum (as I have mentioned before), “nemo prudens punit quia peccatum est sed ne peccetur,”3 is Kant’s view, according to which the punishment should not be allowed to have a goal (Zweck) but rather should be inflicted only “because of his having committed a crime.”4 Kant regards this requirement as a moral justification of punishment, by which he defines the criminal’s worthiness of punishment. In the following, I would like to conduct an interpretation of Kant’s moral retributivism that satisfies this requirement. I shall attempt to show that although one can derive from Kant’s moral theory a retributivist conception of a criminal deserving of punishment, one can, however, justify not a retributivist punishment in the legal system, but rather that combination of specific deterrence and rehabilitation that already ensues from the liberal concept of right (see Section 2.5). At this point, I would like to introduce a classification through an additional differentiation, in which I will enumerate several possible retributive theses. H. L. A. Hart divides the retributive position into three theses: First, that a person may be punished if, and only if, he has voluntarily done something morally wrong; secondly, that his punishment must in some way match, or be equivalent of, the wickedness of his offence; and thirdly, that the justification for punishing men under such conditions is that the return of suffering for moral evil voluntarily done, is itself just or morally good.5

3. “A sensible person does not punish a man because he has sinned, but in order to keep him from sin.” Seneca, De ira, i.19.7. Seneca uses as a basis Plato’s Laws 11.933e–934b, where it is said that every thief should not only provide compensation for the item that was taken, but also suffer a more lenient or a severer punishment depending on his motives: “This additional penalty is to be inflicted not because of the crime (what’s done can’t be undone), but for the sake of the future: we hope that the offender himself and those that observe his punishment will either be brought to loathe injustice unreservedly or at any rate recover appreciably from this disastrous disease.” Plato, Laws, trans. Trevor J. Saunders, in John M. Cooper and D. S. Hutchinson (eds.), Plato: Complete works (Indianapolis: Hackett, 1997), pp. 1318–1616 (pp. 1585–6). Therefore, Plato’s – as well as Seneca’s – intention is clearly directed, not only toward the compensation of the victim, but toward the reform of the criminal as well as toward general deterrence, which is conceived of as a collective reform. Seneca, De ira, in Moral essays, ed. John W. Basore (3 vols., London: Heinemann/Cambridge, Mass.: Harvard University Press, 1928), vol. 1, pp. 106–355 (pp. 158–9). 4. RL Ak iv:331. Practical philosophy, ed. Gregor, p. 472. 5. H. L. A. Hart, Punishment and responsibility (Oxford: Clarendon Press, 1968), p. 231.

74

desert as the sole justification for punishment

The first thesis does not constitute a specific characteristic of retributivism, because, in my opinion, no alternative theory of criminal justice ever allowed the punishment of the innocent.6 The third thesis can correspond to as many as the four following subtheses: 1 All criminals – and only criminals – should be punished. 2 The punishment of criminals acts as retribution for a past criminal act. 3 The degree of punishment should be proportional (in an ordinal, not in a cardinal way) to the crime, that is, the proportion of punishments to one another should correspond to the proportion of the crimes to one another. 4 The degree of punishment must be equal to the crime. The third thesis connects a legal descriptive observation (the descriptive observation of a punishable action) to a legal consequence (the punishment of an action). In contrast, the second thesis connects a legal consequence (the punishment of an action) to a descriptive moral observation (the wickedness of the perpetrator). Thus, retributivism’s moral justification pertains to the second thesis; however, it can support or contradict the third thesis. In what follows, I will show that it contradicts the third thesis. Even the point of reference is lacking for proportionality between the moral wickedness and the punishment. Kant distinguishes, in fact, actions in conformity with duty from actions contrary to duty, and with the latter he distinguishes actions that are merely in conformity with duty from actions that are done out of it. However, neither a gradation of how much an action stands in conformity with duty nor a gradation of contrariness to duty can be found in Kant’s works. Kant merely states: If someone does more in the way of duty than he can be constrained by law to do, what he does is meritorious (meritum); if what he does is just exactly what the law requires, he does what is owed (debitum); finally if what he does is less than the law requires, it is morally culpable (demeritum).7 Virtue ¼ þ a is opposed to negative lack of virtue (moral weakness) ¼ 0 as its logical opposite (contradictorie oppositum); but it is opposed to vice ¼ – a as its real opposite (contrarie s. realiter oppositum).8 6. Cf. Fred Rosen, “Utilitarianism and the punishment of the innocent: the origins of a false doctrine,” Utilitas, 9, no. 1 (March 1997), 23–37. 7. RL Ak vi:227. Practical philosophy, ed. Gregor, p. 382. 8. TL Ak vi:384. Practical philosophy, ed. Gregor, p. 516. (There is a small error in the translation that I have corrected.)

kant’s moral justification of punishment

75

Indeed, the gravity of the punishable action should not be confused with the correspondence between the intention and the duty. Whereas a punishable action is more or less serious, any injury of duty counts as being contrary to duty, and Kant does not differentiate any further than that. Kant even emphasizes that an exception from the moral law or from the fulfillment of duty should never be allowed to be trivialized, but should be considered as really representing a complete misconduct.9 The individual instances of misconduct can be traced back to a supreme maxim, which does not obey the categorical imperative. Because the assertion of a gradation of evil is problematic with Kant, I will then, in what follows, confine Kant’s retributive moral theory to the assertion stating that the punishment is not imposed in order to fulfill an aim, but instead merely because the criminal did injury to the law.10

3.2. The highest good and the necessary relationship between guilt and punishment The requirement for a relationship between the moral evil of the individual who acts contrary to duty and his or her punishment is derived from the postulate of the highest good as it is formulated in the Critique of practical reason. It is generally known that this postulate requires a necessary connection between the two components of the idea of the highest good, that is, between virtue and happiness (that is, between Tugend and Glu¨ckseligkeit); however, Kant further refines this connection into a subordination of happiness beneath virtue, which serves as a prerequisite for happiness. Kant writes: Now, inasmuch as virtue and happiness together constitute possession of the highest good in a person, and happiness distributed in exact proportion to morality . . . constitutes the highest good of a possible world, the latter means the whole, the complete good, in which, however, virtue as the condition is always the highest good, since it has no further condition above it, whereas happiness is something that, though always pleasant to the possessor of it, is not of itself absolutely and in all respects good but always presupposes morally lawful conduct as its condition.11 9. Cf. TL Ak iv:424. Practical philosophy, ed. Gregor, p. 76. 10. For example, Hugo Bedau, in “Retribution and the theory of punishment,” Journal of Philosophy, 75, no. 11 (1978), 601–20, shows that it is also problematic in general. 11. KpV Ak v:110f. Practical philosophy, ed. Gregor, p. 229.

76

desert as the sole justification for punishment

The highest good means postulating the reign of the principle of right over the physical world, which usually abides by the laws of physics. The highest good can only be realized under the reign of the principle of right. Retaliation against evil, for the time being, plays no role in this; instead, it is only the reward to the moral disposition that plays a role. How can the relation between having wicked dispositions and the worthiness of punishment be derived using such a definition of the highest good? Next, we must refer to the status and the substance of the highest good, whose negative variant is the worthiness to be punished. Being a postulate of the practical reason, the highest good has the status of a noumenal, synthetic and necessary connection between virtue and happiness; in the sensible world this connection – wherever it might happen to be present – can only incidentally come into existence.12 Kant even finds that, it must seem strange, that philosophers both of the ancient and modern times could nevertheless have found happiness in precise proportion to virtue already in this life (in the sensible world), or [let themselves be] persuaded that they were conscious of it.13

If the relation between guilt and punishment were to show the inverse of the highest good, and therefore have the same status as the highest good, then the relation must belong exclusively to the noumenal world. In the Doctrine of virtue, we find an acknowledgment of this very assumption: Punishment is not an act that the injured party can undertake on his own private authority but rather an act of a court distinct from him, which gives effect to the law of a supreme authority over all those subject to it; and when (as we must in ethics) we regard human beings as in a rightful condition but in accordance only with the laws of reason (not civil laws), then no one is authorized to inflict punishment and to avenge the wrongs sustained by them except him who is also the supreme principle of right giver; and he alone (namely God) can say “Vengeance is mine; I will repay.”14

As already outlined in the Lecture on ethics (written sometime between 1775 and 1780), deterrent punishment inheres in the worldly ruler and retributive punishment inheres in the moral ruler: 12. Cf. KpV Ak v:114f. Practical philosophy, ed. Gregor, pp. 231ff. 13. KpV Ak v:115. Practical philosophy, ed. Gregor, p. 232. 14. TL Ak vi:460. Practical philosophy, ed. Gregor, p. 578.

kant’s moral justification of punishment

77

All punishments are either deterrent or retributive . . . All punishments by authority are deterrent, either to deter the transgressor himself, or to warn others by his example. But the punishments of a being who chastises actions in accordance with morality are retributive.15

There are two reasons for the highest good’s noumenal status. First, for us human beings, actions done out of duty and actions that are merely in conformity with duty can never be discerned from one another.16 Secondly, we will never find a worldly ruler of whom we can be completely sure that he or she will always abide by the categorical imperative.17 One could argue against assigning this exclusively noumenal status to the inverse of the highest good, that is, to the relation between guilt and punishment, by saying that with an action that is contrary to the legal law one is able to suggest – wholly and with absolute certitude – the existence of wicked dispositions; and with that, one dispenses with the first reason against a retributive punishment imposed by the earthly authorities. The second reason still persists, however, because there is no individual to be found, who acts out of duty, to be the highest judge. In what follows, I would like to show that, even when this second reason is disregarded, the earthly authority is still not authorized to impose retributive punishments, principally for reasons that result from the substance of the highest good. The highest good consists of the necessary connection between virtue and happiness, where virtue is the prerequisite for happiness. Therefore, the converse of happiness should be derivable from the converse of virtue, that is, from wicked dispositions. This appears prima facie to suggest that the converse of happiness is unhappiness, in the sense of an unhappy condition. However, this would be a fallacy, because the logical converse of happiness is the lack of happiness as a simple privatio. Now, Kant defines happiness in the following fashion: Happiness is the state of a rational being in the world in the whole of whose existence everything goes according to his wish and will, and rests, therefore, on the harmony of nature with his whole end as well as with the essential determining ground of his will.18

15. VE Ak xxvii:286. Immanuel Kant, Lectures on ethics, ed. Peter Heath and J. B. Schneewind, trans. Peter Heath (Cambridge: Cambridge University Press, 1997). 16. GMS Ak iv:407. Practical philosophy, ed. Gregor, p. 62, and VE 43. 17. Cf. Idee, Proposition 6, Ak viii:23. Political writings, ed. Reiss, pp. 46–7. 18. KpV Ak v:124. Practical philosophy, ed. Gregor, p. 240.

78

desert as the sole justification for punishment

According to this, the converse to happiness is a rational being’s condition in the world for whom nothing in his entire existence “goes according to his wish and will,” and therefore rests on the complete lack of correspondence between nature and his aim, that is, “the essential determining ground of his will.” In short, the will of the wicked human being should never be able to realize itself. The complete lack of effect of the evil will is something other than inflicting some sort of harm on the evil individual. In the clearest possible fashion, Kant rejects the latter as immoral and stringently distinguishes it from actual punishment: It is, therefore, a duty of virtue not only to refrain from repaying another’s enmity with hatred out of mere revenge but also not even to call upon the judge of the world for vengeance, partly because a human being has enough guilt of his own to be greatly in need of pardon and partly, and indeed especially, because no punishment, no matter from whom it comes, may be inflicted out of hatred.19

From the Kantian perspective, the punishment should be derived from the crime itself according to a moral law that reigns over the world. This means a necessity similar to that of the physical laws of nature. Kant himself uses the analogy: Before reason awoke, there were no commandments or prohibitions, so that violations of these were also impossible . . . From the moral point of view, therefore, the first step beyond this state was a fall; and from the physical point of view, this fall was a punishment, for it led to a host of hitherto unknown evils.20

Regarding crimes, this means that the moral legislator should prevent their perpetration, and do so, in fact, before the implementation of the criminal intention. Kant also did not ignore the fact that it does not usually happen like that in the real world. However, one is allowed to postulate that the necessary connection between guilt and punishment might be compensated in the noumenal world. When one has finally determined that, despite this postulate, the intention that is contrary to duty led to a deed that is contrary to duty, the question arises as to why the evil intention, whose realization could not be prevented, is allowed to be punished ex post facto. At this point, it is 19. TL Ak vi:460f. Practical philosophy, ed. Gregor, p. 578. 20. Conjectures on the beginning of human history, in Immanuel Kant, Political writings, ed. Hans Reiss, trans. H. B. Nisbet (Cambridge: Cambridge University Press, 1970), pp. 221–34.

kant’s moral justification of punishment

79

appropriate to differentiate between two cases: either the perpetrator remains an evil person after the punishable action or he or she converts to the side of the good. If the perpetrator were to remain an evil person, then he or she would want further to realize intentions that are contrary to duty. The necessary connection must lead to the perpetrator being prevented from realizing those intentions. This prevention occurs not for the sake of punishing the specific crime that he or she committed, but instead because of the evil manifested along with the crime’s perpetration. According to the Kantian perspective, in the case of the perpetrator converting to the side of the good since that time, one would have to look at this case from two different angles. In Kant’s work Religion within the boundaries of mere reason, he analyzes the relation of moral conversion with punishment as follows. In the case of conversion, the punishment takes place within a situation in which “he now leads a new life and has become a ‘new man.’”21 Kant locates punishment “in the situation of conversion itself”: the “conversion is an exit from evil and an entry into goodness,”22 and he sees that “by means of the very concept of moral conversion, we can think that situation as entailing such ills as the new human being, whose disposition is good, can regard as having been incurred by himself . . . and, [therefore], as punishment.”23 Kant describes the conversion as a simultaneous existence between the old and the new human: As an intellectual determination, however, this conversion is not two moral acts separated by a temporal interval but is rather a single act, since the abandonment of evil is possible only through the good disposition that effects the entrance into goodness, and vice-versa. The good principle is present, therefore, just as much in the abandonment of the evil as in the adoption of the good disposition, and the pain that by rights accompanies the first derives entirely from the second.24

The punishment of an evil human being consists wholly of the good human being’s birth pains; in short, it is comprised of that through which the person is made into a good human being. This displays, 21. Rel Ak vi:73. Immanuel Kant, Religion within the boundaries of mere reason, in Kant, Religion within the boundaries of mere reason and other writings, ed. Allen Wood and George di Giovanni (Cambridge: Cambridge University Press, 1998), pp. 31–192, (p. 89). 22. Rel Ak vi:73f. Religion, p. 90. 23. Rel Ak vi:73. Religion, p. 114. 24. Rel Ak vi:74. Religion, p. 90.

80

desert as the sole justification for punishment

par excellence, the evil human being’s lack of happiness, because in the conversion he or she breaks with the evil intentions and new, good intentions lead to happiness, that is, to their full effect. Kant evidently describes punishment in a clear form with similarities to the Pauline Epistles in the New Testament:25 The emergence from the corrupted disposition into the good is in itself already sacrifice (as “the death of the old man,” “the crucifying of the flesh”) and entrance into a long train of life’s ills which the new human being undertakes in the disposition of the Son of God, that is, simply for the sake of good, yet are still fitting punishment for someone else, namely the old human being (who, morally, is another human being).26

What appears to be “moral happiness” from the new point of view is perceived from the old point of view as ill-being and punishment.27 Here, the appeal of the usual interpretation of retributive punishment to Kant proves to be false. Thomas Hill sees a motive for future moral behavior in the moral human being’s pangs of conscience about his or her earlier delinquency.28 Exactly the converse is the case with Kant.

3.3. Worthiness of punishment and rightful punishment What consequence are we to draw, though, for rightful punishment in the phenomenal world from this moral punishment in the noumenal world? It is appropriate, owing to the case of the human being who continues to remain an evil human being, to use incapacitation29 as a means so that his or her evil intentions will not come to fruition in the future. In the case of the human being who has converted, there is no longer any reason for punishment provided that one can discern that there has truly been actual reform on the part of the human being. Now, a conversion in the phenomenal world – a conversion which was completed in the noumenal world – does not lend itself to being substantiated with absolute certitude. Admittedly, one could increasingly conjecture such a conversion in the case of a continued behavior in conformity with duty, so that, in the course of time, and in view of Rom. 6:2, 6; Gal. 5:24; Eph. 4:17–24; Col. 3:1–17. Rel Ak vi:74. Religion, p. 90. Cf. footnote in Kant, Rel Ak vi:75. Religion, p. 91. Cf. Hill, “Kant on punishment,” 358–60; similar to John Deith, “On the right to be punished: some doubts,” Ethics, 94 (1984), 191–211 (pp. 210–11). 29. Cf. Arthur Ripstein, Equality, responsibility and the law (Cambridge: Cambridge University Press, 1999), p. 144. 25. 26. 27. 28.

kant’s moral justification of punishment

81

such behavior, reintegration into society becomes more and more appropriate. In general, specific deterrence – in the form of incapacitation – with subsequent rehabilitation is most appropriate, because (at least in a first period of time) one cannot discern if one is dealing with the first or the second case when it comes to a convicted criminal. Neither specific deterrence nor rehabilitation is at the disposal of the public authorities, but they are their duties. On the one hand, the duty to provide assistance to those in need in itself also encompasses the duty to provide assistance to any person who may be a future victim of crime. On the other hand, however, we are not allowed to do harm or to deny freedom to any human beings – who became good in the meantime – owing to their previous transgressions, if either this harm or this denial of freedom is not necessary to public safety. At this point, we have reached the same point that our Kantian critique of Kant’s theory of criminal justice derived from the Kantian liberal concept of right reached. Different from the conclusion that Kant reached in the same year (1797) in his Doctrine of right, in the Doctrine of virtue Kant came to draw the same conclusion that we just have: It is, therefore, a duty of human beings to be forgiving (placabilitas). But this must not be confused with meek toleration of wrongs (mitis iniuriarum patientia), renunciation of rigorous means (rigorosa) for preventing the recurrence of wrongs by others; for then a human being would be throwing away his rights and letting others trample on them, and so would violate his duty to himself.30

The retributivist might want to celebrate half a victory because of this result, because – while it may not be the legal punishment – it is however the case that the moral punishment is conceived of retributively. As early as in the Lecture on ethics, Kant defines “Retributive punishments” as “those pronounced because the evil has occurred.”31 Unlike deterrent punishments, “moral punishments are imposed because a sin has been committed; they are consectaria of moral transgression.”32 By celebrating half a victory, however, the retributivist overlooks the fact that Kant, even though he does not say it explicitly, understands “moral punishments” to be the worthiness of punishment, whereas he understands the other punishments to be legal 30. TL Ak vi:461. Practical philosophy, ed. Gregor, p. 578. 31. VE 79. 32. VE 79.

82

desert as the sole justification for punishment

punishments, that is, punishments in the proper sense. Kant comments on the legal punishment in the Lectures on ethics, just as he does, at a later point, in the Doctrine of virtue : Authority punishes, not because a crime has been committed, but so that it shall not be committed. But every crime, in addition to this punishment, has a property of deserving to be punished, because it has taken place. Such punishments, which must therefore necessarily follow upon the actions, are moral in character, and are poenae vindictivae; just as a reward follows upon a good action, not so that further good actions should be done, but because there has been a good action done.33

Because the worthiness of punishment – which is to say, the moral punishment – belongs to the noumenal world, the other theories of punishment do not compete with this moral retributivism for a simple reason, that is, because the other theories are merely legal theories of punishment, which make no claim on the noumenal world. Therefore, the alternative theories do not express themselves against the second thesis in the manner in which it was formulated by H. L. A. Hart (see Section 3.1). They simply ignore the subject of retribution against moral evil and merely controvert the third thesis, which retributivism intends to substantiate with help from the second thesis. The retributivists want to derive the third thesis from the retributive moral thesis, which we have just examined, as a result of which a necessary proportionality between the punishable action and the punishment will come to exist. We have, however, just shown with our reconstruction of Kant’s moral argumentation that the third thesis by no means follows from the second thesis, but rather that, from it, a combination of specific deterrence and rehabilitation arises. Consequently, the third thesis can only ground itself on itself, as has been argued in this chapter in order to refute it. As a provisional conclusion, we are allowed to assert that neither the liberal (see Chapter 2) nor the moral (see Chapter 3) concept of right, which one can derive from two different interpretations of Kant (see Chapter 1), can justify a retributive theory of punishment. And there is more: if both paths of argumentation were to pay attention to the consistency of their deduction of punishment from the Kantian concept of right they would then lead in our sketch of the reconstruction of the arguments to a combination consisting of specific deterrence and of rehabilitation – which strictly excludes 33. VE 79.

kant’s moral justification of punishment

83

retributivism – without arguing for general deterrence. At this point, it finally comes into full view that the retaliative theory is not the only alternative to general deterrence, which is a view that is contrary to both Kant and many of the authors inspired by him. Fichte and Hegel then reconstructed what should have been Kant’s theory of punishment, if it had been consistent with Kant’s practical philosophy. Although this is a controversial issue, I believe that none of these authors was a retributivist.

PART II PUNISHMENT AS A MEANS OF REHABILITATION

I will proceed with my reconstruction of what should have been Kant’s theory of penal law, had it been derived from Kant’s concept of right, through an interpretation of the Fichtean and the Hegelian theory of punishment. Since the degree of punishment that both Fichte and Hegel support (incapacitation and resocialization, as we shall see) is nearly the same, interpreting both authors may seem redundant. Thus one may ask oneself whether treating the more famous of the two, who is at the same time the one often considered as being a retributivist, would not suffice. However, the ways in which they come to this shared conclusion could hardly be more divergent. As we shall see, whereas Fichte begins with the unilateral breaking of the social contract by the criminal and with its consequences (the exclusion from the commonwealth), and then introduces a normative obligation to realize the concept of right, Hegel starts with the validity and the enforcement of law that has to be manifested, without drawing on the figure of exclusion – on the contrary, the criminal does not cease to belong to the commonwealth and to enjoy by the same token respect for his or her human dignity. Whereas Fichte focuses on the fate of the criminal, Hegel takes the validity of the law as the basis for his approach. The fact that both authors eventually adopt the same degree of punishment on the basis of the same concept of law in spite of the very different ways of reaching it should make it clear that it is not the method but the adoption of this shared concept of right that matters for reconstructing a Kantian theory of punishment. Nor do the differences between Kant, Hegel and Fichte regarding the foundation of their respective systems and of their respective conceptions of autonomy matter for our investigation. Therefore, I shall deliberately put them aside in the following argumentation. 85

4 FICHTE’S “EXPIATION CONTRACT”

It is well known that the liberal interpreters of Kant’s concept of right (see Chapter 1) and Fichte share the same concept of right, even though they deduce it in fundamentally different ways:1 the reciprocal restriction of the external freedoms according to the universal law of equal rights. Nevertheless, Fichte develops a theory of penal law clearly incompatible with that of Kant. That is why I wish not simply to compare both of the theories in what follows, but rather – now that I have confronted Kant’s theory of penal law with his concept of right in Chapters 2 and 3 – also now to compare Fichte’s theory with the concept of right common to Kant and Fichte. In Chapter 2, I have already criticized Kant’s retributivism, and indeed both the classical interpretation of this retributivism as the sole justification of the existence of punishment and the newer interpretation of this retributivism as justification of the degree of punishment. In both interpretations, Kant’s retributivism cannot be derived, or deduced, from his concept of right: moreover, his retributivism even contradicts his concept of right. That is why I will now, as a lead in, sketch with what rationales Fichte rejects Kantian retributivism as a justification of the degree of punishment, even though he defines the degree of punishment according to the criterion of talion law. Fichte published the first part (}} 1–16) of his Foundations of natural right in 1796, half a year before the publication of Kant’s Doctrine of right. Half a year after the publication of Kant’s Doctrine of right, the second part (}} 17–24) of Fichte’s Foundations of natural right appeared, in 1797. In 1796, in the chapter in the Foundations of natural 1. Cf. Jean-Christophe Merle (ed.), “Eigentumsrecht,” in Merle (ed.), Johann Gottlieb Fichte: Grundlage des Naturrechts (Berlin: Akademie Verlag, 2001), pp. 159–72 (p. 161).

87

88

punishment as a means of rehabilitation

right about the right of coercion (}} 13ff.), Fichte developed this aspect of the Kantian mixed theory, which was then neglected, but since the 1980s it has again been receiving attention: general deterrence through threatening punishment. Fichte’s actual theory, as it was formulated in 1797 in } 20 of the Foundations of natural right, seems, however, to be a mixed theory: Fichte interpreters have shown that in the Foundations of natural right a combination of elements of general deterrence through threatening punishment, elements of specific deterrence and elements of reform and of rehabilitation can be found. In this respect, Kaufmann rightly sees a commonality between the Kantian penal law interpreted as mixed theory and Fichte’s actual theory of punishment: Here, in the principles of his theory of coercion, Fichte is very modern, in the respect that he propagates the unification theory overwhelmingly accepted nowadays of the connection between the theories of deterrence and retribution. The justification of punishment, on the whole, occurs through deterrence; in the determination of who should be punished, not only the damage plays a role, but also the extent to which a bad or an inadequate will played a role. The same is valid for determining the degree of punishment.2

Likewise, Lazzari observes “that both pillars of the Fichtean theory of punishment, the ideas of performing expiation and of deterrence, are widely in harmony with the diffuse requirement of a more humane system of punishment.” Lazzari praises the mixed character of the Fichtean theory of punishment, even though he is aware of the problems connected with it: A “mixed” attempt of that sort appears in many respects really illuminating and befitting the multiplicity of the aspects having to be taken into account. What is problematic in Fichte’s argument is, first and foremost, the incompatibility of the legal and theoretical assumptions, which substantiate the deterrent part of Fichte’s theory of punishment in the first part of the Foundations of natural right and the restricting framework in } 20.3

The objection that I raised earlier against the Kant interpretation as a mixed theory, however, holds ipso facto against Fichte’s mixed theory. Therefore, I will attempt to show in what follows that general 2. Matthias Kaufmann, Rechtsphilosophie (Freiburg i.Br.: K. Alber, 1996), p. 132. 3. Alessandro Lazzari, “Eine Fessel, die nicht schmerzt und nicht sehr hindert: Strafrecht,” in Merle (ed.), J. G. Fichte: Grundlage des Naturrechts, pp. 173–86 (pp. 183–4).

fichte’s “expiation contract”

89

deterrence through threatening punishment is not compatible with the Fichtean – or Kantian – concept of right. Thus, it is no coincidence that one year later, in the actual chapter in the Foundations of natural right about penal law (} 20), Fichte contradicts his chapter on the right of coercion and, in my opinion, justifies punishment primarily with specific deterrence, even though he never explicitly renounces his earlier mixed theory.

4.1. Retributivism’s lack of justification from the legal perspective I will now briefly remind the reader of Kant’s argumentation in favor of a retributivist grounding of penal law. In } 49e of the Doctrine of right, Kant advocates a retributivist position, more precisely talion law: “like for like.”4 Kant delivers no direct justification of talion law, but merely claims that there are no acceptable alternatives.5 An alternative can be acceptable only if it passes the test of the categorical imperative, especially its third formula: “So act that you use your humanity, whether in your own person or in the person of any other, always at the same time as an end, never merely as a means.”6 Kant applies this test to the criminal: For a human being can never be treated merely as a means to the purposes of another or be put among the objects of rights to things: his innate personality protects him from this, even though he can be condemned to lose his civil personality.7

Kant declares the retributivism of talion law as being the only theory of penal law that passes the test of this formula; all other theories of penal law treat the criminal merely as a means. Now, not only do the individual punishments determined by the retributivist talion law clearly infringe the aforementioned formula of the categorical imperative,8 but Kant’s negative argumentation in favor of retributivism also clearly applies to that theory, which regards general deterrence – whether by threatening punishment or by an example carried out – as being the primary justification of punishment, while leaving, in my opinion, the thesis of specific deterrence untouched.9 It is 4. 5. 6. 7. 8. 9.

RL Ak vi:332. See Section 2.2. GMS Ak iv:429. Practical philosophy, ed. Gregor, p. 80. RL Ak vi:331. See Section 2.4. See Section 2.5.

90

punishment as a means of rehabilitation

exactly this neglected option that Fichte reaches as the result of his deduction of penal law. Fichte proceeds under the assumption of the same concept of right: “Every relation of right is determined by this proposition: each person is to limit his freedom through the possibility of the other’s freedom.”10 He obviously regards Kant’s negative argumentation in favor of retributivism as a crude fallacy. In the second part of his Foundations of natural right, Fichte criticizes with sharp language Kant’s Doctrine of right, which had appeared nine months earlier: Punishment is not an absolute end. The claim that it is (whether stated explicitly or through propositions that implicitly presuppose such a premise, e.g. the unmodified, categorical proposition that “he who has killed, must die”) makes no sense.11

Some pages later, regarding the “claims” he is attacking, Fichte refers explicitly to Kant, “the great, though not infallible, man”12 who accuses Beccaria of alleged “overly compassionate (Empfindelei) feelings of an affected humanity.”13 Fichte turns against an absolute right of punishment, according to which judicial punishment is regarded not as a means, but as itself an end, which is said to be grounded on a categorical imperative that is itself not further examinable . . . By relying on what is supposedly unexaminable, this theory allows its proponents to exempt themselves from the need to prove their claims and so to charge those who think differently with sentimentality (Empfindelei) and an affected humanitarianism.14

Thereby, Fichte does not wish to contest, in any way, that retaliation is a principle of justice. In the second edition of his Attempt at a critique of all revelation (1793), Fichte definitely sees a motivation to moral action in the morally necessary proportionality between virtue and happiness, that is to say, in God as the infinite rational being whom Fichte declares as just.15 In the Foundations of natural right, Fichte claims furthermore: 10. GNR i/3 411. Johann Gottlieb Fichte, Foundations of natural right: Grundlagen des Naturrechts nach Principien der Wissenschaftslehre, ed. Frederick Neuhouser, trans. Michael Baur (Cambridge: Cambridge University Press, 2000), p. 109. 11. GNR i/4 60. Foundations of natural right, p. 228. 12. Footnote at GNR i/4 76. Foundations of natural right, p. 245. 13. RL Ak vi:334–5. Practical philosophy, ed. Gregor, p. 475. 14. GNR i/4 76. Foundations of natural right, p. 245. 15. Cf. Jean-Christophe Merle, “Il punto di vista educativo e religioso dei Contributi destinati a rettificare il giudizio del pubblico sulla Rivoluzione francese: la dimensione politica del Saggio di una critica di ogni rivelazione,” in Aldo Masullo

fichte’s “expiation contract”

91

“There is absolutely no dispute about whether a murderer has been treated unjustly, if he, too, should lose his own life in a violent manner.”16 However Fichte immediately stresses that the question of penal law is an “entirely different question,” that is, the question from where . . . [does] a mortal get the right of this moral world-order, the right to render the criminal his just deserts? and it was this purely juridical question that the noble Beccaria (who was certainly not unfamiliar with that kind of moral judgment) had in mind. Whoever ascribes this right to a worldly sovereign will surely be required . . . to regard . . . every government as a theocracy.17

4.2. Penal law as a specific type of the right of coercion In his chapters about the right of coercion (}} 13–16), Fichte identifies this with the right of coercion exercised by the state. Now, such an equation is by no means self-evident, as a comparison with Kant’s right of coercion and penal law shows. Numerous Kant interpreters proceed under the assumption, which they consider as obvious, that in Kant’s thought penal law naturally and directly belongs to the right of coercion, even though Kant neither claims nor suggests this interpretation. Next, let us investigate what exactly the right of coercion consists of for Kant. The right of coercion exercised by the state is derived and deduced from that very right of coercion inhering in every person as soon as his or her rights are infringed. This individual right of coercion entitles the battered person to curtail the freedom of a fellow human being, because, according to Kant, Whatever is wrong is a hindrance or resistance to freedom in accordance with universal laws . . . Therefore, if a certain use of freedom is itself a hindrance to freedom in accordance with universal laws (i.e., wrong), coercion that is opposed to this . . . is consistent with freedom in accordance with universal laws, that is, it is right.18

The consummate example of this individual right to coercion is the right to self-defense. The state, to which this right is assigned, employs and Marco Ivaldo (eds.), Filosofia trascendentale e destinazione etica (Milan: Guerini, 1995), pp. 303–25. 16. GNR i/4 77. Foundations of natural right, p. 246. 17. GNR i/4 77. Foundations of natural right, p. 246. 18. RL Ak vi:231. Practical philosophy, ed. Gregor, p. 388.

92

punishment as a means of rehabilitation

the equivalent of the individual right to self-defense when seeking to prevent the commission of an infringement of the law by means of police at the actual moment when the violation of rights is still just an attempt. The state is also authorized to exercise a right of coercion in the forms of nullification or annulment, seizures, and the revocations of licenses (withdrawing a concession or a privilege, in the Hohfeldian meaning, for example nowadays a driving license). Apart from that, the state is endowed with a right of coercion after the commission of a rights infringement (as with the forced restitution of stolen goods) and with the court-ordered delivery of compensation for damages, if this is possible (with some rights infringements – typically with murder – there is no fitting substitute restitution). These cases concern retribution in the narrow sense. Where it is possible (with some rights infringements – typically, once again, with murder – it is fully impossible)19 this retribution completely satisfies the legitimate individual claims of the victim of a rights infringement. It is in no case a punishment. Whereas Kant presents the right of coercion immediately in his “Introduction to the doctrine of right”20 and links each right with a right of coercion, he first treats penal law only much later and merely as one of the elements of “public law.”21 Fichte diverges from Kant not only in introducing penal law as early as in the deduction of the right of coercion in general, but also in his definition of the right of coercion as well as in the end he ascribes to it. Whereas the Kantian right of coercion merely resists an existing “hindrance to freedom in accordance with universal laws,” coercion, according to Fichte’s chapter about the right of coercion, already prevents the formation of such a hindrance even before the attempt: Now if things could be arranged so that the willing of any unrightful end would necessarily, and in accordance with an ever-operative law, result in the opposite of what was intended, then any will that is contrary to right would annihilate itself . . . It was necessary to present this proposition in its full, synthetic rigor, since all laws of coercion, or penal laws, (the entirety of penal legislation) are grounded on it.22

According to the respective definitions of the right of coercion delivered by Kant and Fichte, the right of coercion appears for Kant 19. 20. 21. 22.

See Section 2.3. RL Ak vi:231. Practical philosophy, ed. Gregor, p. 388. RL Ak vi:331–7. Practical philosophy, ed. Gregor, pp. 472–8. GNR i/3 426. Foundations of natural right, p. 126.

fichte’s “expiation contract”

93

initially in the perpetration of a rights violation, while for Fichte, on the other hand, it occurs already, in its very essence, in the intent of perpetration. Fichte states clearly: The security of the two parties is not supposed to depend on a contingency, but on a near-mechanical necessity that excludes every possible exception. There can be such security only if the law of right is the inviolable law of both parties’ wills.23

Incidentally, it is worth noticing that, in Kant’s justification of punishment, deterrence through threatening punishment plays an explicit role only in the passage about the right of necessity, and that it plays this role only in an indirect and negative way; therefore, with Fichte, deterrence receives a much larger importance.

4.3. General deterrence is not the true justification of penal law In } 20 of the Foundations of natural right – in the actual paragraph about penal law – Fichte speaks no longer about the threat of punishment, but about the punishment itself,24 which presupposes that deterrence by threatening punishment has failed – at least partly: the crime has already taken place. In view of the demanding, declared goal of the general deterrent threat of punishment – “The threat of punishment aims to suppress bad wills . . . in which case punishment will never be necessary”25 – the failure is obvious. Either the humbler task of general deterrence through example must inhere in the punishment of a criminal or it must have another end. In the first case, and in the absence of any other goal, the criminal would be treated as a mere means for general deterrence, which is, however, forbidden.26 Therefore, another primary justification of punishment must be searched for, which would in turn substantiate the normative 23. GNR i/3 424. Foundations of natural right, p. 124. 24. In Rainer Zaczyk, Das Strafrecht in der Rechtslehre J. G. Fichtes (Berlin: Duncker and Humblot, 1981), Zaczyk emphasizes that the chapter about the right of coercion “does not deal with penal law in the narrow sense, but instead only with a ‘coercion of will,’ in a meaning which remains to be defined more precisely.” The threat of punishment is nevertheless necessarily connected – though not in so wide a sense – with punishment. There is an important characteristic of the Fichtean theory here: the attempt to conceive of punishment as being a component of the right of coercion in general, instead of its being a right of coercion meant for special cases and being a special form of the right of coercion. 25. GNR i/4 60f. Foundations of natural right, p. 228. 26. See Section 4.1.

94

punishment as a means of rehabilitation

acceptability of deterrence by threatening punishment – indeed no longer being the primary aim of punishment, but instead, at best, being the secondary goal thereof. Now, whoever at the beginning of } 20 more closely examines the thesis and the antithesis of that deduction leading to the “expiation contract” (Abbu¨ßungsvertrag) will notice, first, that in the thesis there is no indication of the punishment to be found, and, secondly, that the antithesis no longer draws on the same concept of coercion as the earlier chapters about the right of coercion. The thesis depicts the crime as a breach in the civil contract on the part of the criminal, who excludes him- or herself ipso facto from society. Even though the Antithesis, in the strict sense of the term – by which I mean the paragraph that begins with “Antithesis,” not the following paragraph dedicated to the synthesis – refers to “these punishments” in contrast to “other punishments,”27 it obviously does not concern a punishment in the legal sense. This becomes clear as soon as one reads the passage about the destiny of the criminal, who can be considered to be incorrigible and in whose case the thesis of being excluded from the commonwealth remains valid. Fichte stresses that the criminal’s death is not a form of punishment, but only a means to ensure security. This gives us the entire theory of the death penalty. The state as such, as judge, does not kill the criminal; it simply cancels the contract with him . . . If, afterwards, the state also kills the criminal, then it does so not by virtue of its judicial authority, but through the police. This takes place, not in consequence of any positive right, but out of necessity.28

The substance of the synthesis begins with the sentence: “This can be arranged only through a contract of all with all,”29 even though there is, from the start of the Antithesis to the end of the chapter, no heading for “Synthesis.” The synthesis immediately and explicitly concerns a punishment in the legal sense. Yet, “public security” is not to be reached anymore by the right of coercion, in the sense of the chapter about the right of coercion, that is, by a 100 percent effective right of coercion in the form of general deterrence through threatening punishment.30 The synthesis rather deals, on the one hand, with specific deterrence – both by the criminal’s incapacitation 27. 28. 29. 30.

GNR i/4 59f. Foundations of natural right, p. 227. GNR i/4 74. Foundations of natural right, p. 243. GNR i/4 60. Foundations of natural right, p. 227. GNR i/4 426. Foundations of natural right, p. 228.

fichte’s “expiation contract”

95

(the convicts live on their own island) and by the threat of irrevocable exclusion from the commonwealth in the absence of any reform – and, on the other hand, with general deterrence by example: The [penal] law’s first aim was to prevent the criminal from committing a crime. Since this goal was not achieved, the state’s punishment of the criminal serves another purpose: to prevent other citizens, and to prevent the criminal in the future, from committing the same offense.31

If general deterrence – whether through threatening punishment or example – were the end of the punishment, then it would be better served by sentencing people to a death in the desert or summary execution than by a prison sentence or even by forced labor. Because Fichte provides for the possibility of reform and the rehabilitation of the criminal, criminals will possibly be released one day. In view of public security and general deterrence either by threatening punishment or by making an example of the criminal, every instance of recidivism means that the punishment recommended by Fichte appears to be less effective than exclusion from society, which, according to Fichte, equally leads either to the death of the criminal in the desert or to shooting the criminal like a wild animal. In the Antithesis, Fichte makes it clear that the prerequisite for legal punishment to replace the exclusion of the criminal from the commonwealth is that by legal punishment “the mutual security of the rights of all in relation to all others”32 is met as effectively as through exclusion from the commonwealth. By the acceptance of a possible incidence of recidivism, the conclusion can be drawn that public security – which in the chapter about the right of coercion still demanded the almost mechanical, full prevention of every crime even before the stage when it is attempted – is now understood only in a much more relative sense: public security may be an important good, but not an absolute good anymore. Whoever reads more closely the chapter about the right of coercion admittedly finds a similar inconsistency already there. Fichte determines a degree of punishment for general deterrence through the threat of punishment that is in no way optimal for that goal: ius talionis, which is supposed to countervail the criminal’s impetus for crime (“the opposite to A” – “A” being the criminal goal – so that the

31. GNR i/4 61. Foundations of natural right, p. 228. 32. GNR i/4 59. Foundations of natural right, p. 227.

96

punishment as a means of rehabilitation

opposite to “A” means the “counterpoise”).33 In } 20, Fichte explicitly refers back to } 14, in order to recommend that the “punishment must be equal to the offense: poena talionis.” According to him, it should be the “material principle of positive punishment within a state,” in short, a criterion. Fichte does not fortify his claim with any argument – other than the vague image of a counterpoise, that is, of the pans of a scale balancing – that both general deterrence, whether by the threat of punishment or by example, and specific deterrence are best achieved by way of talion law, that is, by a retributivist degree of punishment. Therefore, the conflict between the different theories of penal law pertains, allegedly, only to the grounds for punishment, not to its degree. Nevertheless, Fichte emphasizes that “exclusion from the state is the most terrible fate a human being can encounter”;34 he even rejects any form of torture before killing the criminal.35 The objection could be raised that killing the criminal without any court sentence, that is, as a mere police measure, provided that it is complemented by torture before the homicide and by making public both the torture and the killing, might deter even in cases in which talion law does not suffice to curb the criminal impetus. In his Antithesis, Fichte himself mentions that talion law is a much milder punishment than exclusion from the commonwealth.36 If one were to proceed under the assumption of Fichte’s psychological premise that the more severe the punishment is, the greater the criminal impetus drops off, then it follows that exclusion from the commonwealth is a more efficient deterrent than ius talionis. Nonetheless, Fichte ultimately rejects exclusion from the commonwealth. Moreover, not once does he attempt to show that this exclusion from the commonwealth would be ultimately linked to a loss of deterrence. Only because of Fichte’s unsubstantiated and implausible assumption that general deterrence is best reached by ius talionis do we have a paradox before us: legal punishment (retaliation as in talion law) meant to serve general deterrence by the threat of punishment is 33. 34. 35. 36.

GNR i/4 426f. Foundations of natural right, p. 229. GNR i/4 68. Foundations of natural right, p. 237. Cf. GNR i/4 74. Foundations of natural right, pp. 242–3. One could argue that exclusion from the commonwealth sets a limit for the punishment determined by general deterrence. It is indisputable that exclusion from the commonwealth is mostly less painful than torture before execution. Nevertheless, in most cases, exclusion from the commonwealth, which leads to a sure death “in the desert,” remains a worse fate than the degree of punishment recommended by Fichte’s talion law.

fichte’s “expiation contract”

97

milder than the lack of a legal punishment, that is to say, milder than extralegal punishment or exclusion from the commonwealth. Therefore, whoever compares the legal punishment to the lack of a legal punishment will come to see that the legal punishment does not imply punishment in the sense of the infliction of suffering due to the crime, but instead a mitigation of extralegal punishment. Under these circumstances, Fichte should have argued for the weaker antithesis according to which legal punishment, despite its lesser deterrent effect, does not lack every deterrent effect, as it would in the case in which the criminal were further to enjoy the status of a normal citizen, that is, impunity. With such an antithesis, the legal punishment recommended by Fichte must exert a not insignificant deterrent effect. However, because this legal punishment in no way optimizes general deterrence, it is impossible for general deterrence to be punishment’s primary end. If at all the case, general deterrence would rather be the primary end of an extralegal punishment (in my opinion, it is not even that, because exclusion from the commonwealth has a reason – the criminal is a potential danger for the commonwealth – rather than a general deterrent end). In this respect, Fichte is not able to make the claim anymore that, since exclusion from the commonwealth is not as effective a deterrent as legal punishment, there is “in such cases . . . no reason to exclude the offender; but admittedly . . . there would also be no reason not to exclude him. The decision would be a matter of free choice.”37 After Fichte has in this way wrongly assumed the parity of legal and extralegal punishment out of consideration for general deterrence, he reaches for a subsidiary criterion in order that the aforementioned “decision” should not be left to “free choice.” The subsidiary criterion could consist of the investigation into what directly arises from the combination of the concept of right and the concept of infringement of right. The result of this investigation should actually not deliver merely a subsidiary criterion for the appraisal of competing notions of degrees of punishment, but instead the primary grounding of punishment itself. But, at first glance, this option is not to be found. Now, the logical consequence of the reciprocal limitation of freedoms contained in the Fichtean concept of right combined with the infringement of these limits would be the complete exemption from legal punishment, that is, the unadorned 37. GNR i/4 60. Foundations of natural right, p. 227.

98

punishment as a means of rehabilitation

exclusion from the commonwealth. It would primarily be concerned merely with the consequence of the infringement of rights, that is, with the elimination of an obstruction of rights, and not be concerned with goal-oriented measures, that is, with the prevention of obstruction of rights that has not yet occurred. Instead of this, Fichte adopts the reciprocal “usefulness” of the criminal and of the state to one another as a subsidiary criterion. It results from this criterion that the criminal, despite the crime, continues to be a member of the commonwealth. According to the subsidiary nature of this criterion, general deterrence by threatening punishment and by example remains, throughout the entire text of } 20 of the Foundations of natural right, the “primary end”;38 rehabilitation constitutes only a secondary justification of punishment. Now, Fichte introduced the subsidiary criterion only because of its false premise that exclusion from the commonwealth – extralegal punishment – and legal punishment out of consideration for deterrence are equally efficient. Because this premise is false, Fichte must have given up, de facto and tacitly, general deterrence as being the primary end of penal law that would require exclusion from the commonwealth. Admittedly, Fichte refers to the principle of general deterrence not only at the beginning, but also in the entire text of } 20. Nevertheless, he does not argue in favor of this principle, and does not derive any consequences from it regarding the degree of punishment. Therefore, we must draw the following conclusion: what Fichte introduces as being a subsidiary criterion constitutes in reality the single true, primary justification of legal punishment in } 20. It is now time to investigate this justification.

4.4. Specific deterrence and rehabilitation as justification of legal punishment Fichte justifies only very briefly why he opts for an alternative to the exclusion of the criminal from the commonwealth. His entire grounding can be summarized in two citations: It is just as much in the state’s interest to preserve its citizens . . . as it is in each individual’s interest not to suffer the loss of all rights for every single offense.39 38. GNR i/4 60. Foundations of natural right, p. 227. 39. GNR i/4 60. Foundations of natural right, p. 227.

fichte’s “expiation contract”

99

This expiation contract is useful for all (for the state as a whole) as well as for each individual citizen. Under it, the whole obtains both the prospect of preserving citizens whose usefulness outweighs their harmfulness, as well as the obligation to accept their expiation; the individual citizen obtains the perfect right to demand that some expiation be accepted in place of the more severe punishment that he deserves.40

Whereas the interest of the citizens of a state to be legally instead of extralegally punished is easily comprehensible, the usefulness of the criminal to society deserves at least some explanation. It would be a mistake to suspect that there is a utilitarian element in this usefulness. It would be in equal measure wrong to believe that this appeal to usefulness does not represent any legal rationale and that the legal concept of right remains unaffected in the exclusion of the criminal from the commonwealth only for the reason that the commonwealth persists amongst the other, innocent citizens of the state. Already in } 4, two instances of deduction of the concept of right attest to the fact that, for Fichte, exclusion from the commonwealth, owing to the denial of mutual recognition, cannot be a final and absolute exclusion. Exclusion should rather remain temporary for the duration of the denial of mutual recognition. This holds true for all finite, rational beings. Let us look at the aforementioned two instances. 1. Fichte writes: In each relation into which I enter with the individual C, I must refer to the recognition that has occurred and must judge him in accordance with it . . . Assuming that his action is indeed determined by the sensible predicates of his prior actions . . . but not determined by his having recognized me as a free being, i.e. assuming that, by means of his action, he robs me of the freedom that belongs to me and thus treats me as an object . . . Thus, in this case, I am able, with perfect consistency (which is my only law here), to treat him as a merely sensible being, until both sensibility and rationality are once again united in the concept of his action.41

Fichte explains his position in the following way: I . . . appeal to a law that is valid for us both, and apply that law to the present case . . . But, in so far as I appeal to that common law in my opposition to him, I invite him to be a judge along with me; and I demand that in this case he must find my action against him consistent and must 40. GNR i/4 60. Foundations of natural right, pp. 227–8. 41. GNR i/3 355–6. Foundations of natural right, pp. 45–6.

100

punishment as a means of rehabilitation

approve of it, compelled by the laws of thought . . . The source of this obligation is certainly not the moral law: rather it is the law of thought; and what emerges here is the syllogism’s practical validity.42

In the lines that follow that (heading “C”), Fichte universalizes this requirement for consistent behavior: it holds true for relations to every individual. 2. In } 4, Fichte in general attempts to show what his headings already clearly formulate: “(i) I can expect a particular rational being to recognize me as a rational being, only if I myself treat him as one.”43 “(ii) But in every possible case, I must expect that all rational beings outside me recognize me as a rational being.”44 “(III) . . . I must in all cases recognize the free being outside me as a free being, i.e. I must limit my freedom through the concept of the possibility of his freedom.”45 I will not address the reasons why, according to Fichte, the satisfaction of these requirements is a prerequisite of the possibility of self-consciousness in finite, rational beings; even less will I treat the question here as to whether Fichte was right about this issue. Here I only have to observe that, according to Fichte, the requirement that every finite, rational being should belong to the commonwealth and be treated as a member of it is a component of the concept of right, so that the decision about who belongs to the commonwealth is not even at the disposal of the commonwealth itself. The “laws of thought” demand that every finite, rational being is recognized as a member of the commonwealth, so long as this being recognizes other rational beings and treats them as such, or as soon as the being is again ready to do so after having denied this recognition for a time to others. Thus Fichte makes it clear in } 4 that exclusion from the commonwealth cannot be final. It is merely a suspension that lasts as long as it takes for the denier-of-recognition to recognize others. It becomes clear that Fichte’s use of the word “usefulness” is absolutely not applied in the utilitarian sense. Fichte’s background is neither the Scottish school nor the utilitarianism of the eighteenth century, but instead modern natural right, which is inspired more from Cicero than from Aristotle. According to Cicero, that same usefulness (utilitas), which is not a merely apparent usefulness, is defined as being 42. 43. 44. 45.

GNR i/3 356. Foundations of natural right, p. 47. GNR i/3 352. Foundations of natural right, p. 42. GNR i/3 353. Foundations of natural right, p. 43. GNR i/3 358. Foundations of natural right, p. 49.

fichte’s “expiation contract”

101

nothing other than that which contributes to right in the sense of a just ordering of the community.46 Fichte’s following explanation should be read with this definition of usefulness in mind: This contract is useful for all (for the state as a whole) as well as for each individual citizen. Under it, the whole obtains . . . the prospect of preserving citizens whose usefulness outweighs their harmfulness.47

Usefulness for state and citizens consists of the punishment contributing to what is just, that is, contributing to the establishment of a legal system. If the state of law and the state of nature amount to a dichotomy without a third option, which is the case with Kant and also with Fichte, then the crime poses a transition from a state of law to a state of nature between the criminal and other people. The concepts of right and usefulness require, however, that all humans are part of the commonwealth. The usefulness of legal punishment (in the twofold meaning of the word “usefulness”: in both the ancient and the modern senses of the term) therefore consists in reestablishing the state of law between the criminal and the community, and, indeed, as quickly as possible. The concept of right requires simultaneously the suspension and the fastest possible rehabilitation of the criminal as a member of the commonwealth, provided that he or she is again just as able as others to adhere to the laws. That is why Fichte quickly abandons ius talionis, which he had initially applied for the degree of punishment, in favor of specific deterrence and the theory of reform. At the beginning of } 20, Fichte still speaks of “punishment . . . equal to the offense: poena talionis.”48

46. Cf., for example, Cicero in De officiis, ii.ii.9f: “What I am going to next address is that which is labelled ‘beneficial.’ Custom has stumbled over this word and strayed from the path, gradually sinking to the point where she has severed honourableness from benefit, decreeing that something can be honourable which is not beneficial, and beneficial which is not honourable. Nothing more destructive than this custom could have been introduced into human life . . . For they [the philosophers] hold that whatever is just is also beneficial, and again, whatever is honourable is also just. Therefore it follows that whatever is honourable is also beneficial. Those who do not see this clearly often admire shrewd and crafty men and mistake wickedness for wisdom. Theirs is an error that must be uprooted; and their fancy must be wholly converted to that hope which consists of the understanding that they will achieve what they want by honourable policies and just deeds, and not by deceit and wickedness.” Marcus Tullius Cicero, On duties, ed. M. T. Griffin and E. M. Atkins (Cambridge: Cambridge University Press, 1991), p. 66. 47. GNR i/4 60. Foundations of natural right, p. 227. 48. GNR i/4 61. Foundations of natural right, p. 229.

102

punishment as a means of rehabilitation

Four pages later, Fichte says of the criminal that, “He must forfeit his freedom until it is clear that he has reformed; or else he must be excluded from the state without mercy.”49 Five more pages later, Fichte states more precisely: But these institutions for reform must also be prudently arranged. First, they must be actually separated from society and established according to the spirit of the law [specific deterrence through incapacitation]. The state has full responsibility for any damage caused by someone who, at the time, is being excluded from society. Therefore, these persons have lost all their freedom. However, if a person is to reform himself, and if his efforts at reform are to be subject to judgment, then he must be free. Therefore, a chief maxim is: such people must be free within necessary limits and must live in society among themselves.50

Therefore, what remains of this Thesis is the exclusion from the commonwealth, which is, however, still only a provisional exclusion. What remains of the Antithesis is the criminal’s affiliation to the commonwealth, which, however, is provisionally suspended and only fully enters into force after the criminal’s reform. The synthesis is the “expiation contract.” The expiation contract consists in offering a chance to the convicted criminal to escape exclusion from the commonwealth and from that status as a human being who, like a savage outlaw, can be freely shot like a harmful, wild animal. Instead of being excluded, the criminal can decide to be punished by spending time in a prison or upon an island with other convicts. Depending on the criminal’s behavior in the penal camp, after a specific period of time he or she will then be either readmitted into the commonwealth or permanently excluded from it.51 Fichte’s contract of expiation entails a double originality: first, it is not a contract between the ordinary citizens designed to legitimate the state’s authority to punish; and, secondly, it conceives of punishment not as it would of an evil, but, instead, as of an opportunity for the criminal to improve his or her fate. In these two regards, Fichte’s conception diverges, for example, from the “non-retributive Kantian approach to punishment” based on an appeal to a hypothetical social contract.52

49. 50. 51. 52.

GNR i/4 65. Foundations of natural right, p. 233. GNR i/4 70. Foundations of natural right, p. 239. GNR i/4 71. Foundations of natural right, p. 240. Michael Clark, “A non-retributive approach to punishment,” Ratio, 17, no. 1 (2004), 12–27.

fichte’s “expiation contract”

103

What soon proves to be Fichte’s justification of penal law (specific deterrence and rehabilitation) takes on two essential elements of Kant’s philosophy of right. First, the decision about who shall belong to the commonwealth is a decision that does not fall to the commonwealth, because every human should belong to it. Secondly, the state of law should be reached by the shortest possible path; detours are allowed – and dictated – only along this path (Kant’s “permissive law”).53

4.5. Is general deterrence really necessary for a justification of punishment? Naturally, the objection can be raised against my interpretation of } 20 that nowhere in this entire paragraph does Fichte explicitly abandon the justification of penal law with that sort of deterrence that consists of threatening punishment. Nevertheless, in the course of } 20, Fichte abandons general deterrence and talion law; in the end, the criminal’s reform – and not ius talionis – determines the date of release. This also means that sometimes the punishment should be harsher than what retaliation would require: It would be very prudent if the criminal himself were allowed to determine, in accordance with the degree of his depravity, the length of time of his reform – but with the proviso that he would later be free to extend it in accordance with a certain standard. But each criminal must be given a peremptory term for reform, in accordance with his particular crime.54

After this term expires, the criminal should either – in the case of reform – be reassumed into the commonwealth, or – in the case that reform did not occur – be permanently excluded from it.55 As a result of the occurrence of multiple justifications of penal law by Fichte, which exclude each other from being primary justifications 53. Cf. Section 2.4 above and Reinhard Brandt, “Das Erlaubnisgesetz, oder: Vernunft und Geschichte in Kants Rechtslehre,” in Brandt (ed.), Rechtsphilosophie der Aufkla¨rung (Berlin: De Gruyter, 1982), pp. 233–85. 54. GNR i/4 71. Foundations of natural right, p. 240. 55. Here it appears to me that Fichte contradicts his premise and antithesis, according to which every human should be offered legal punishment, provided that this person does not pose any danger – for instance, for the prison guards. Now, we cannot exclude the possibility that a criminal, after a certain period of time, still has not reformed, but without having posed an additional security problem ipso facto by continuing to remain in the “correctional facility.”

104

punishment as a means of rehabilitation

and also cannot be combined into a mixed theory because they lead to different notions of degree of punishment, one must either declare Fichte’s penal law as being hopelessly contradictory or investigate the individual arguments, as well as their role and relevance in Foundations of natural right, and then reach an assessment. In view of this situation, my argument has shown two things. First, if we do not accept the false premises, according to which legal and extralegal punishments would have an equally deterrent effect, then only Fichte’s argument for specific deterrence by incapacitation would remain as a primary justification for penal law. Secondly, that argument is the only one that is compatible with the degree of punishment, which Fichte conclusively adopts at the end of } 20, for this reason. That is why I am of the opinion that Fichte – admittedly only after lengthy comings and goings – substantiates penal law with specific deterrence and rehabilitation. Naturally, legal punishment exerts a deterrent effect by being openly disseminated (general deterrence by threatening punishment) and by the execution of punishment not being a secret (general deterrence by example). This effect certainly contributes to public security and, for that reason, is thoroughly welcomed by the commonwealth. Nevertheless, threatening punishment and public execution of punishment are only then allowed if the infliction of punishment is itself justified, because punishment exceeds the reciprocal limitation of freedoms contained in the concept of right. If general deterrence by threatening punishment were 100 percent effective and the punishment, ipso facto, never needed to be executed – as Fichte wishes in the chapter about the right of coercion – then the infliction of punishment would never be justified; then, however, the whole of } 20 of the Foundations would be superfluous. The chapter about the right of coercion clearly stipulated a 100 percent effective general deterrence by threatening punishment: “the security” of any two given persons is not supposed to depend on a contingency, but on a near-mechanical necessity that excludes every possible exception. There can be such security only if the law of right is the inviolable law of both parties’ wills.56

It is appropriate that the law of right should be normatively inviolable. In contrast, human freedom, in the sense of the power of choice, is potentially infinite; it is always able to transgress limits – and even 56. GNR i/3 424. Foundations of natural right, p. 124.

fichte’s “expiation contract”

105

normatively inviolable limits as well. Solely the laws of nature are unable to be infringed. Coercion is surely an effective yet not 100 percent effective means to obtain public security. As Hegel later wrote, “Only he who wills to be coerced can be coerced into anything.”57 Many a culprit accepts the risk of punishment, even if he or she later regrets it because of fear and may come to be aware of the fact that the punishment actually was too high a price to pay. Therefore, not only threatening punishment but also inflicting punishment needs to be justified. The threat is not first in line for justification. Hegel attains specifications of penal law, similar to Fichte’s, without needing to take a detour via general deterrence and via its shortcomings. Furthermore, Hegel proceeds under the assumption of the necessary validity of right and, therefore, does not need the thought experiment of exclusion from the commonwealth. In this respect, Hegel offers a systematically more stringent representation of the same view about penal law, as I would like to show in the following. An enlightening comparison to the conceptions of Fichte and Hegel is offered by Karl Christian Friedrich Krause (1781–1832) in his philosophy of law, which is based upon similar Kantian–idealist foundations and was published in the period between the publication of both of their works. Admittedly, only the first section (1803) and the posthumous second section about cosmopolitan law (1890) of Krause’s own Grundlage des Naturrechts (The foundation of natural right) have been published, whereas the planned section about public law and a subsection about penal law within this section remained stuck in the planning stages.58 But Krause’s Latin habilitation thesis of 1802, which he translated into German and upon which he made comments,59 provides an informative sketch of the philosophy of law. According to Krause, penal law should cause it to appear to the citizen that adherence to the law is more useful than adherence to impulses;60 in a footnote – probably from 1805 – he states this more precisely: punishment aims, according to Krause, “especially toward civicly 57. GPhR } 91. G. W. F. Hegel, Elements of the philosophy of right, trans. H. B. Nisbet (Cambridge: Cambridge University Press, 1991), p. 120. 58. Cf. Wolfgang Forster, Karl Christian Friedrich Krauses fru¨he Rechtsphilosophie und ihr geistesgeschichtlicher Hintergrund (Ebelsbach: Aktiv Druck und Verlag, 2000), pp. 210–16. 59. K. C. F. Krause, Abhandlung u¨ber die Idee und die Eintheilung der Philosophie und der Mathematik und den innigen Zusammenhang beider, in Paul Hohlfield and August Wu¨nsche (eds.), Philosophische Abhandlungen aus dem handschriftlichen Nachlasse (Leipzig: Schulze, 1889), pp. 5–40. 60. Krause, Abhandlung u¨ber die Idee, pp. 34–5.

106

punishment as a means of rehabilitation

reforming the criminal thoroughly and in a rational way, and toward also ensuring that punishment does not affect his other rights.”61 Krause remarkably proceeds in the determination of punishment from the assumption of the inalienable rights of the perpetrator, which, as a “formula,” indicates “at the same time the limits of punishment.”62 Krause clearly breaks away from retributivism. As the other published parts of Grundlage des Naturrechts63 as well as unpublished references in his letters show, Krause derives from that axiom, amongst others, the absolute inviolability of the body, and, along with that, a proscription of death as well as of all corporal punishment.64 As in other areas,65 the visionary Krause ironically comes somewhat closer in the consequences which he draws from his theory to the conclusions which one should expect from a liberal interpretation of the Kantian idealist concept of right than the more famous authors Fichte and Hegel.66 Krause, Abhandlung u¨ber die Idee, pp. 34–5 (footnote). Krause, Abhandlung u¨ber die Idee, p. 35. Part i in Krause, Abhandlung u¨ber die Idee, pp. 237–8. Cf. Forster, Krauses fru¨he Rechtsphilosophie, pp. 108–9, 303. For instance, in the theory of marriage, in K. C. F. Krause, Grundlage des Naturrechts oder philosophischer Grundriß des Ideals des Rechts (Jena: Gabler, 1803), pp. 169ff. 66. See the overview provided by Peter Landau, “Karl Christian Friedrich Krauses Rechtsphilosophie,” in Klaus-Michael Kodalle (ed.), Karl Christian Friedrich Krause (1781–1832): Studien zu seiner Philosophie und zum Krausismo (Hamburg: Meiner, 1985), pp. 80–92. 61. 62. 63. 64. 65.

5 HEGEL’S “NEGATION OF CRIME”

5.1. The controversial classification of the Hegelian theory of punishment Unlike in the secondary literature about Kant’s theory of punishment, which for a long time served as the model for all retributivist or absolutist theories of penal law, the divergence from the traditional – and often held as being self-evident1 – retributivist interpretation has quite a long history with regard to the interpreters of Hegel’s penal law. Whereas the interpretation of Kant’s penal law as being a mixed theory has only come about since the 1980s,2 Hegel’s penal law was very early on regarded by Christian Reinhard Ko¨stlin as a mixed theory,3 by British Neo-Hegelianism (for instance by Bernard Bosanquet)4 as being a deterrent theory and by John Ellis McTaggart5 1. Cf. Part i, } 62 of Arthur Schopenhauer, The world as will and idea, trans. R. B. Haldane and J. Kemp, seventh edition (4 vols., London: Kegan Paul, Trench, Tru¨bner, 1909?), vol. 1, pp. 430–52. Nowadays, cf., for instance, Ulrich Klug, “Abschied von Kant und Hegel,” in Ju¨rgen Baumann (ed.), Programm fu¨r ein neues Strafgesetzbuch: der AlternativEntwurf der Strafrechtslehrer (Frankfurt a.M.: Fischer, 1968), pp. 36–41; Claus Roxin, Strafrechtliche Grundlagenprobleme (Berlin: De Gruyter, 1973), p. 2; Winfried Hassemer, Einfu¨hrung in die Grundlagen des Strafrechts, second edition (Munich: C. H. Beck, 1990), pp. 283–4; Jean-Claude Wolf, Verhu¨tung oder Vergeltung? Einfu¨hrung in ethische Straftheorien (Freiburg i.Br.: Alber, 1992), p. 50; Otfried Ho¨ffe, Gerechtigkeit: eine politische Einfu¨hrung, second edition (Munich: C. H. Beck, 2004), p. 79. 2. See Section 2.1. 3. Christian R. Ko¨stlin, Neue Revision der Grundbegriffe des Kriminalrechts, reprint (Aalen: Scientia, 1970). Cf. Andrei A. Piontkowski, Hegels Lehre u¨ber Staat und Recht und seine Strafrechtstheorie, trans. Anna Neuland (Berlin: De Gruyter, 1960), p. 199. 4. Bernard Bosanquet, The philosophical theory of the state (London and New York: Macmillan, 1899). 5. John E. McTaggart, Studies in Hegelian cosmology (Cambridge: Cambridge University Press, 1918), chapter 5.

107

108

punishment as a means of rehabilitation

as being a reform theory. Admittedly, this last interpretation has not prevailed. Even today, there is controversy about whether Hegel is really a retributivist6 or whether he possibly also adopts general or specific deterrent elements,7 or whether at the opposite extreme, he is only a theorist of pure deterrence.8 Just as in Kantian studies, at this time mixed theories are also predominant in Hegelian studies. Some of the best interpreters – Allen Wood and Georg Mohr, for instance – combine the retributivist interpretation with that of deterrence. According to Mohr, Hegel’s theory of punishment delivers “a type of grounding, which was new for that time, for the theory of retribution, which was up to then rather poor in its arguments.” According to Mohr, Hegel is able to attain this through allotting “to the punishment as its inherent goal the task of a positive general deterrence.”9 Likewise, Wood initially contends that 6. Cf. especially Mark Tunick, Hegel’s political philosophy: interpreting the practice of legal punishment (Princeton: Princeton University Press, 1992); in addition, for instance, Ossip Kurt Flechtheim, “Die Funktion der Strafe in der Rechtstheorie Hegels,” in Flechtheim, Von Hegel zu Kelsen: rechtstheoretische Aufsa¨tze (Berlin: Duncker and Humblot, 1963), pp. 9–20 (p. 13); Flechtheim, “Zur Kritik der Hegelschen Straftrechtstheorie,” Archiv fu¨r Rechts- und Sozialphilosophie, 54 (1968), 539–48 (p. 542); Peter G. Stillman, “Hegel’s idea of punishment,” Journal of the History of Philosophy, 14 (1976), 169–82 (p. 173); Klaus Scala, “Hegels Begriff der Strafe und die moderne Strafvollzugsproblematik,” Hegel-Jahrbuch (1987), 164–70; Igor Primorac, “Punishment as the criminal right,” Hegel-Studien, 15 (1980), 187–98 (p. 193); and Igor Primoratz (these are two spellings of the same name), “Banquos Geist: Hegels Theorie der Strafe,” HegelStudien, supplement 29 (1986), 53; Kurt Seelmann, “Wechselseitige Anerkennung und Unrecht: Strafe als Postulat der Gerechtigkeit?”, Archiv fu¨r Rechts- und Sozialphilosophie, 79, no. 2 (1993), 228–36; Seelmann, “Versuch einer Legitimation von Strafe durch das Argument selbstwiederspru¨chlichen Verhaltens des Strafta¨ters,” Jahrbuch fu¨r Recht und Ethik, 1 (1993), 315–26; Matthias Kaufmann, Rechtsphilosophie (Freiburg i.Br.: K. Alber, 1996), p. 308; footnotes in G. W. F. Hegel, Principes de la philosophie du droit, trans. JeanFranc¸ois Kerve´gan (Paris: Presses Universitaires de France, 1998), pp. 180–1; Arthur Ripstein, Equality, responsibility and the law (Cambridge: Cambridge University Press, 1999), p. 93; Otfried Ho¨ffe,“Ko¨nigliche Vo¨lker”: zu Kants kosmopolitischer Rechts- und Friedenstheorie (Frankfurt a.M.: Suhrkamp, 2001), p. 79. 7. Cf. Section 1.6 of Ted Honderich, Punishment: the supposed justifications (New York: Harcourt, Brace and World, 1969); Wolfgang Schild, “Ende und Zukunft des Staatsrechts,” Archiv fu¨r Rechts- und Sozialphilosophie, 70 (1984), 71–112; Allen W. Wood, Hegel’s ethical thought (Cambridge: Cambridge University Press, 1990), p. 110; Matthias Kaufmann, “Zwangsrecht (}} 13–16),” in Jean-Christophe Merle (ed.), Johann Gottlieb Fichte : Grundlage des Naturrechts (Berlin: Akademie Verlag, 2001), pp. 125–37. 8. Cf., for instance, Stanley Benn, “An approach to the problems of punishment,” Philosophy, 33 (1958), 321–41; Anthony M. Quinton, “On punishment,” in Harry Burrows Acton (ed.), The philosophy of punishment (London: Macmillan, 1969), pp. 55–64. 9. Georg Mohr, “Unrecht und Strafe,” in Ludwig Siep (ed.), G. W. F. Hegel: Grundlinien der Philosophie des Rechts (Berlin: De Gruyter, 1997), pp. 95–124 (pp. 95, 105).

hegel’s “negation of crime”

109

“Hegel is a genuine retributivist. He rejects as ‘superficial’ all theories that try to justify punishment by the ‘good’ which is supposed to come of it.” On the next page, Wood observes: “but the state’s intention to reassert the validity of right in the face of wrong looks like an intention not to do justice as such, but to promote a good end, namely the public recognition of the validity of right.” Wood asks: Why is it important for the state to assert the validity of right, to express its disapproval of crime? Is there any reason for it to do this apart from its devotion to such consequentialist ends as preventing future crimes and reassuring people that their rights are being protected . . .?10

These very interpreters highlight in this way that Hegel argues more convincingly for retributivism than Kant. According to Mohr, “compared to Kantian retribution theory . . . Hegel’s theory of punishment is the argumentatively better executed foundation of this type of theory.”11 Wood says of Hegel’s theory of punishment: “This position agrees with Kant, whose commitment to retributivism is clear, but whose defense of it remains at best embryonic.”12 Just as for Kant’s penal law, for Hegel’s penal law as well it is fair to assume a mixed theory of punishment that ascribes to every argument a part of the task of justification. Thus, Mohr wishes to recognize Hegelian penal law’s merit of building up “a theoretic instrument” that “does profitable preliminary work underlying the current claims to a mixed theory.”13 Just like the currently prevailing interpretation of Kant’s theory, Mohr also sees a mixed theory of punishment in Hegel: Taking all the relevant passages from the Elements as the basis for the interpretation of Hegel’s theory of punishment yields the diversified picture of a modern mixed theory. It differentiates between justice in punishment as an institution rooted in retributive theory and justice in degree of punishment specified by deterrent theory.14

According to Mohr’s interpretation, the relationship between a certain element of retributive theory and a certain element of deterrent theory is formulated by Hegel the reverse way of that by Kant. 10. Allen W. Wood, Hegel’s ethical thought (Cambridge: Cambridge University Press, 1990), pp. 109–10. 11. Mohr, “Unrecht und Strafe,” p. 121. 12. Wood, Hegel’s ethical thought, p. 109. 13. Mohr, “Unrecht und Strafe,” p. 122. 14. Mohr, “Unrecht und Strafe,” p. 119.

110

punishment as a means of rehabilitation

According to the interpretation of the mixed theory, in Kant deterrent theory justifies punishment as an institution, and retributive theory justifies the degree of punishment. In contrast, Heiko H. Lesch sees in Hegel the same relation between both elements of the mixed theory as in Kant.15 The way in which Hegel’s penal law is interpreted as a mixed theory, however, only indirectly correlates to the process of argumentation in Hegel’s text. Indeed, the connection made by such a mixed theory between each of the arguments offered respectively by the theories of deterrence and retributivism does not concern the relationship of the argumentation based on “Abstract right” to the argumentation based on “Ethical life.”16 The internal relationship of the individual arguments based on “Abstract right” to one another is not inquired into by the mixed theory.

5.2. The twofold justifications of punishment It may astound one to see that the interpretations of Hegel’s theory of punishment can contradict one another so deeply that some interpreters seem to vacillate between multiple interpretations. These different interpretations, however, draw on similarly different passages in Hegel’s Elements of the philosophy of right (1821). Elements of the Hegelian theory of punishment can be found not only in the section “C. Coercion and crime” within “Abstract right,”17 but also in “Ethical life,” under the heading “B. The administration of justice,” specifically in “b. The existence (Dasein) of the law” and “c. The court of law.”18 The deterrence theory interpretations of Hegel often base themselves on “Ethical life,” whereas the retributivist interpretations either simply forget }} 218–20 or declare them to be unimportant. The latter interpretations categorize these paragraphs mainly as being a mere indication that a punishment must be tailored to the respective societal conditions without the rationale for punishment and the principle determining its degree being affected by }} 218–20.

15. Heiko Hartmut Lesch, Der Verbrechensbegriff: Grundlinien einer funktionalen Revision (Cologne: Heymanns, 1999), p. 97. 16. GPhR, Parts i and iii, respectively. 17. GPhR }} 90–103. 18. See GPhR }} 218 and 220, respectively.

hegel’s “negation of crime”

111

5.2.1. An objective and subjective justification of punishment. The reverse of those interpretations that wish to ignore or neglect }} 218–20 is an overestimation of the second part of the Elements of the philosophy of right (“Morality”). In “Morality,” there are indeed elements that clearly have to do with crime, especially in the section entitled “Purpose and responsibility.” Yet in the “Morality” section, the word “punishment” is found only three times. Moreover, it is not to be found either in the main body of the text or in the Remarks but instead in the Additions – and there it is only an aside. Punishment is only treated in the part of right as the “universal” or right “in itself” and in the part of the “subjective disposition, but of that right which has being in itself.”19 Punishment clearly belongs to the objective aspect of the idea of right. Yet in the passage out of “Abstract right,” many interpreters attempt to demonstrate the presence of two arguments that are both in favor of retributivism: an objective one and a subjective one. The clearest presentation of this twofold argumentation is provided by Ossip K. Flechtheim and by Igor Primorac, and in a divergent form by Kurt Seelmann.20 Flechtheim differentiates an “objective justification” from a “subjective justification,” since “the objective way to legitimate punishment appears to Hegel . . . as still being insufficient. He wishes to demonstrate that the criminal, by the criminal deed, also subjectively consents to the punishment.”21 In the objective justification, there is an argument that draws “on the nature of the crime and on the law that was injured by it.”22 Flechtheim emphasizes that: As early as in the theological writings of his youth, Hegel’s deduction has been based on the axiom that the relation between deed and punishment is inseparable; that the deed already contains the 19. GPhR } 141, Remarks. G. W. F. Hegel, Elements of the philosophy of right, ed. Allen W. Wood, trans. H. B. Nisbet (Cambridge: Cambridge University Press, 1991), p. 186. 20. Flechtheim, “Die Funktion der Strafe,” pp. 9–20; Flechtheim, “Zur Kritik der Hegelschen Straftrechtstheorie,” 539–48; and Flechtheim, Hegels Straftheorie (Berlin: Duncker und Humblot, 1975). Primorac, “Punishment as the criminal right,” 187–98; and Primoratz, “Banquos Geist”; Seelmann, “Wechselseitige Anerkennung,” 228–36; Seelmann, “Versuch einer Legitimation,” 315–26. Further authors assert the existence of a twofold argumentation in Hegel’s theory of punishment, for instance, Bertrand Guillarme, Penser la peine (Paris: Presses Universitaires de France, 2003), pp. 63–4. 21. Flechtheim, Von Hegel zu Kelsen, pp. 17–18. 22. Primoratz, “Banquos Geist,” 39.

112

punishment as a means of rehabilitation

punishment in itself; and that the punishment of the deed necessarily proceeds from the deed.23

I shall name this thesis the thesis of negation of the negation. According to Flechtheim, the thesis was formulated in } 93, where Hegel states: Because coercion24 destroys itself in its concept, it has its real expression (Darstellung) in the fact that coercion is cancelled (aufgehoben) by coercion; it is therefore not only conditionally right but necessary – namely, as a second coercion which cancels an initial coercion.25

In } 99, Hegel equates “the cancellation (Aufheben) of the crime, which would otherwise be regarded as valid” with “the restoration of right.”26 Flechtheim identifies the “restoration” (Wiederherstellung) explicitly with the classical thesis of retaliation (Wiedervergeltung), which is discussed in } 101.27 According to Flechtheim’s interpretation, the classical retributivist thesis forms the link between the objective and the subjective justification. That is because, unlike right in itself, the principle of retaliation draws not on the total legal system, but instead on the individual and on his or her individual deed. In the subjective justification, there is an argument that draws on the relation of punishment to the “empirical will of the criminal.” Flechtheim sees the subjective justification to be mainly located in } 100, which he interprets in the following manner: Allegedly, it is inherent in the crime, which is considered as being the action of a rational being, that the crime is something universal and that it lays down a law. The lawbreaker recognizes through his or her deed that this law applies to him- or herself, so that the lawbreaker is subsumable under that law, which is his or her own right. The criminal is honored as a rational being by his or her action being regarded in just such a way, that is, as if the action itself placed a claim on the punishment by virtue of it being its own right.28

Flechtheim comments that: From this conception of the legal offense, Hegel obtains the importance that he attributes to the guilty plea and to the jury court. If the wrongdoer 23. 24. 25. 26. 27. 28.

Flechtheim, Von Hegel zu Kelsen, p. 12. “Coercion” can also be regarded as being “force.” GPhR } 93. Elements, ed. Wood, p. 120. GPhR } 99. Elements, ed. Wood, p. 124. Flechtheim, Von Hegel zu Kelsen, p. 13. Flechtheim, Von Hegel zu Kelsen, p. 17.

hegel’s “negation of crime”

113

consents to his or her punishment, not only in so far as the wrongdoer is the personification of an objective rational will but also as he or she is the manifestation of his or her subjective particular will, hence every court and every legal proceeding becomes superfluous since the criminal can pass his own sentence him- or herself. But even Hegel does not dare to construe reality in terms of this “idea,” and contents himself with the requirement that “what the judges pronounce does not differ from what is in the consciousness” from the subjective self-consciousness of the wrongdoer.29

What Flechtheim quotes without providing a page reference is derived, however, from the Additions in } 227, which is expressly concerned with confessing to the elements of a crime – therefore, it is not, as Flechtheim assumes, concerned with the criminal’s consent to penal law. Then, according to Flechtheim’s conception, } 101 is seen as being a qualification of } 100. According to Flechtheim, the criminal should have recognized talion law. Meaning, he continues, that this criminal should recognize that he or she “deserves to be punished” and that “what the criminal has done should also happen to him.”30 This thesis is realized in the principle of “equality” between crime and punishment. In turn, this principle is specified not as being a “specific equality,” that is, as being in natura, but instead as being an equality “in terms of its value.”31 It remains unclear, however, why punishment should require such a twofold justification. The context in Flechtheim suggests that this twofold justification is meant to bring about the criminal’s reconciliation with society. An obvious objection arises, though, against this rationale for a twofold justification. Reconciliation was indeed Hegel’s goal in the writings of his youth, but punishment was seen by him at that time as something that is alien to the criminal as a human being. That is because it was not something subjective, but instead only something objective. At the outset of the section with the title “C. Coercion and crime,” the concern is about something objective: the cancellation (Aufhebung) of crime and the restoration (Wiederherstellung) of right. Admittedly, Hegel writes: “The injury . . . which is inflicted on the criminal is not only just in itself [objective aspect] . . . it is also a right for the criminal himself, that is, a right posited in his existent

29. Flechtheim, Von Hegel zu Kelsen, p. 17. Elements, ed. Wood, p. 257. 30. GPhR } 101, Remarks. Elements, ed. Wood, p. 127. 31. GPhR } 101, Remarks. Elements, ed. Wood, p. 128.

114

punishment as a means of rehabilitation

will [subjective aspect].”32 Yet this implies neither that this subjective aspect would be necessary for the justification of the punishment as a complement to the objective justification, nor that the criminal would have to consent to penal law in order that the punishment might be legitimate. Hegel’s formulation in } 100 completely allows that, after first being sentenced and then having served a punishment, the criminal gains the insight that his or her deed must imply the punishment, even though the deed implied the punishment in itself already while he or she was committing the deed. In such a case, the criminal’s consent would not be – unlike the case with the cancellation of the crime and the restoration of right – a (normative) prerequisite, but instead a result of the crime. Naturally, this does not mean that the subjective insight of the criminal into the justification of his or her punishment has no systematic locus in the Elements of the philosophy of right. Obviously, the process by which the criminal gains insight into the necessity of penal law makes possible the “Transition from right to morality.” In the part about morality that truly does constitute “a further advance in the inner conceptual determination of the will,” punishment is not at issue – the word “punishment” itself practically never appears.33 It is in the part about “Ethical life,” in which the opposition between the general will in itself and the general will for itself is cancelled, that punishment is discussed again for the first time. There it is handled from the perspective that dangerousness is a criterion for punishment in general as well as for legal punishment in particular, that is, for the opposite of private revenge. }} 218 and 220 clearly do not make the claim, however, to contributing to the justification of the existence of punishment. To this extent, the justification of punishment rests exclusively on the development of the general will in itself; therefore, 32. GPhR } 100. Elements, ed. Wood, p. 126. 33. Cf. GPhR } 104. Elements, ed. Wood, p. 131. The word Strafe (“punishment”) itself appears only in Hegel’s handwritten marginalia in GPhR } 118 as well as in the Remark on } 132, though Allen W. Wood, for instance in } 218, renders the German Ahndung as the English “punishment,” though it could also be rendered as “revenge for a wrong.” In these passages, the word Strafe is used only in the following contexts: “Suffering in general – including punishment – as a consequence in general of the action,” “punishment, infringement,” and “unhappiness merely as such has no ethical interest – just as injustice and punishment are not mere ills” (translation mine); “The sphere in which the above circumstances come into consideration as grounds for relaxing the punishment is not the sphere of right, but the sphere of clemency,” in Elements, ed. Wood, p. 161.

hegel’s “negation of crime”

115

it rests on the objective part. My conclusion is that the subjective aspect of punishment provides no justification of punishment as being the justification of a legal institution, even if there happen to be elements of the concept of right in that justification. In the subsection “Coercion and crime,” Hegel clearly formulates the affiliation of the justification of punishment to the objective aspect of the concept of right. In his Remark on } 99, Hegel speaks of “the objective consideration of justice, which is the primary and substantial point of view in relation to crime.”34 While the supposed subjective justification comes under the heading “the criminal’s own right,” its validity expressly remains conditional to the validity of the objective justification: “The injury . . . which is inflicted on the criminal is not only just in itself (and since it is just [emphasis mine], it is at the same time his will as it is in itself, an existence . . . of his freedom, his right).”35 And while the alleged subjective justification appears in the context of classical retaliation theory, then it is again expressly conditional to the validity of the objective justification: “The cancellation . . . of crime is retribution (Wiedervergeltung) in so far [emphasis mine] as the latter, by its concept, is an infringement of an infringement.”36 Hegel criticizes the classical theory of retaliation by opposing it with the argument of the nullity of crime: “Yet the concept itself must always contain the basic principle, even for the particular instance.” Hegel also terms this “the substance of the thing (Sache) itself.”37 Furthermore, no interpreter asks the question whether either the objective justification alone or the subjective justification alone is sufficient. If neither of the two would suffice then one would have to ask one more question to find to what extent they reciprocally complement each other and they together would be sufficient. What can be found regarding this point in the secondary literature tends not really to be convincing. Primoratz attempts to substantiate the twofold justification in the following way: This aspiration of Hegel to justify punishment both in the objective and in the subjective sense is in no way contingent. Nor is it an expression of his doubts about the rightness and convincingness of the one or the other justification and about the consequent need for a complementing argument . . . His fundamental objection that he brings forward against 34. 35. 36. 37.

GPhR } 99, Remark. Elements, ed. Wood, p. 125. GPhR } 100. Elements, ed. Wood, p. 126. GPhR } 101. Elements, ed. Wood, p. 127. GPhR } 101. Elements, ed. Wood, p. 128.

116

punishment as a means of rehabilitation

Plato consists in Plato not recognizing the principle of subjectivity, that is, there being no place in his republic for the individual.38

5.2.2. A legal argument and an argument based on recognition of the criminal’s humanity. It still remains for us to consider the alternative interpretation of the twofold justification of punishment given by Seelmann. He sees in Hegel “two different constructions for legitimating punishment and penal law”: the “argument from law” and the “argument from recognition.”39 Both link the objective with the subjective dimension. The argument from law is the thesis previously mentioned, that is, the thesis of “the criminal’s own right” or “subsumption” under the criminal’s own law.40 I will come back to this later. In what follows, I will address the question whether Hegel’s (objective) argument from law is linked to the (subjective) argument from recognition in a twofold justification. The argument from recognition relies on } 97 of the Elements of the philosophy of right: When an infringement of right as right occurs, it does have a positive external existence . . . but this existence within itself is null and void. The manifestation of its nullity is that the nullification of the infringement likewise comes into existence; this is the actuality of right, as its necessity which mediates itself with itself through the cancellation . . . of its infringement.41

According to Seelmann this argument can be understood “only if one takes as a point of departure Hegel’s understanding of right as being a universal relation of reciprocal recognition of free and equal persons . . . an understanding that Hegel borrowed from Fichte, while introducing some modifications.”42 Seelmann himself formulates the argument along the following lines: Since . . . recognition is a reciprocally conditioned relation [according to Hegel], the wrongdoer himself withdraws the recognition from himor herself. Therefore, punishment is nothing other than the mere 38. Primoratz, “Banquos Geist,” 40. 39. Seelmann, “Versuch einer Legitimation,” 319; cf. Seelmann, “Wechselseitige Anerkennung,” 228. 40. Cf. GPhR } 100. Elements, ed. Wood, pp. 126–7. 41. GPhR } 97. Elements, ed. Wood, p. 123. 42. Seelmann, “Wechselseitige Anerkennung,” 230.

hegel’s “negation of crime”

117

manifestation of this situation that the wrongdoer him- or herself has brought about, whereby punishment lowers the physically observable legal status of the wrongdoer in order to create a legal relationship based on equality.43

The argument from recognition is thus not to be understood as if punishment were justified by the crime being a refusal of recognition on the part of the criminal, because based on this denial of recognition one could just as well shoot the criminal as if he or she was a dangerous animal that anyone should be allowed to kill without any particular procedure – that is, without a legal sentence, as is the case in Fichte’s thought experiment, to which Hegel alludes in another passage.44 Rather, the argument from recognition signifies that the aim of punishment is situated in compelling the recognition of the victim as an equal by the wrongdoer: the aim of punishment is the creation of (in Seelmann’s words) “a legal relationship based on equality.”45 This means that the humiliation of the criminal, which goes along with the punishment, should only occur temporarily. Therefore, the punishment is justified not in itself, but merely by its aim. If retaliation is to play a role here, then it should only play a role as an instrument of rehabilitation. Before one adopts the argument from recognition that has been reconstructed in such a way, one must note that it lacks any basis in this paragraph’s text; the relation of the argument to } 97 is not substantiated by Seelmann.46 The argument from recognition sees disruption of “a reciprocally conditioned relation” in the nullity of the infringement of right.47 At least for right in the state as defined in the “Ethical life” section, this cannot be the case. In } 218 Hegel writes: Since property and personality have legal recognition and validity in civil society, crime is no longer an injury . . . merely to a subjective infinite, but to the universal cause . . . whose existence . . . is inherently . . . stable and strong.48 43. Seelmann, “Versuch einer Legitimation,” 320. 44. Fichte’s thought experiment is found in GNR i/4 59. Fichte, Foundations of natural right, p. 226. For more on Fichte’s thought experiment, see Lazzari, “Eine Fessel,” pp. 173–86. Hegel alludes to this in GPhR } 100. Elements, ed. Wood, p. 126. 45. Seelmann, “Versuch einer Legitimation,” 320. 46. It is substantiated neither in “Versuch einer Legitimation” nor in “Wechselseitige Anerkennung.” 47. Seelmann, “Versuch einer Legitimation,” 320. 48. GPhR } 218. Elements, ed. Wood, p. 250.

118

punishment as a means of rehabilitation

In the state, right, getting legal validity, is no longer dependent on recognition on the part of individuals. If the crime would not be cancelled by a “second coercion” then, indeed, it would constitute a precedent that might call into question the validity of right.49 In such a case, it would be not only the criminal’s legal status – as in Fichte – that would be at risk, but also – as in Fichte – the legal relationship of all persons with one another. Moreover, the crime does not necessarily imply the non-recognition of a legal person by the criminal. In his Remark on } 95, Hegel draws attention to the negation of the legal capacity of a person as only representing “shapes” of the crime. Hegel makes mention of “further development . . . and . . . further shapes,” for which he provides the examples of “perjury” and “counterfeiting, forgery, etc.,” and thus of crimes that are not directed against a single person but rather against the state. Furthermore: The substantial element within these forms is the universal, which remains the same in its further development and in the further shapes it assumes; thus its infringement, i.e. crime, also remains the same, in conformity with its concept.50

First and foremost, the crime’s “nullity” can be explained in another way. According to Hegel, “force or coercion immediately destroys itself in its concept, since it is the expression of a will which cancels . . . the expression or existence of a will.”51 The crime’s nullity relies on the will itself being destroyed in the crime. Thus, according to Hegel, the crime consists in a self-contradiction rather than in an infringement upon reciprocity. 5.2.3. A mixed theory of punishment. Another twofold objective and subjective justification of punishment is also presented on the model of Kant’s penal law: a mixed theory (see Section 2.1). Such a mixed theory of punishment consists of ascribing to retributivism and to the theory of general deterrence each a part of the task of justification. In this way, Mohr wishes to recognize “a theoretic instrument [that] does profitable preliminary work underlying the current claims to a mixed theory.”52 According to Mohr, the Elements combine a retributivist

49. 50. 51. 52.

GPhR } 93. Elements, ed. Wood, p. 120. GPhR } 95, Remark. Elements, ed. Wood, p. 122. GPhR } 92, Remark. Elements, ed. Wood, p. 120. Mohr, “Unrecht und Strafe,” p. 122.

hegel’s “negation of crime”

119

rationale of punishment as an institution with a general deterrent specification of the degree of punishment.53 Such an interpretation sees the element of deterrence theory in “Ethical life” (especially in } 218), whereas it locates the retributivist element in “Abstract right.” Therefore, it inverts that interpretation of Kant’s penal law as a mixed theory according to which deterrence theory justifies punishment as being an institution and retaliation theory justifies the degree of punishment. If Kant and Hegel are really mixed theorists, then Hegel is thus the only one of the two who justifies retributivistically the existence of the punishment itself. A mixed interpretation has to ask itself the question, however, about the compatibility of both of its components. Let us call to mind again the results of Chapter 2: with Kant’s conception, as it is generally known, one cannot avoid a collision between general deterrence in the execution of punishment and the categorical imperative’s application dictates that the humanity in the criminal should not be treated as a means, but always only as an end. General deterrence, which is mixed with retributivism, does not solely concern the threat of punishment. Kant’s (controversial) implicit assumption presupposes that the mere threat of punishment treats no one as a mere means, since, merely by threat, no harm is inflicted on anyone so long as this person does not commit a crime. Whoever commits a crime would then not be punished according to the degree dictated by general deterrence but rather according to the degree dictated by retributivism. Hence only the threat of punishment – not the punishment itself – is justified using general deterrence. Thus, both the existence of punishment and the degree of punishment remain grounded in a purely retributive way. That is due to Kant not allowing the degree of punishment to be defined other than retributively. In Kant, the threat of a non-retributivist degree of punishment is not allowed to lead to a non-retributivist degree of punishment during the execution of the punishment, which would contradict the categorical imperative. Mixed theory’s interpretation, hence, is only in appearance a mixed theory – in reality it remains a retributivist justification of punishment. Now how does all of this look in Hegel’s alleged mixed theory? To Mohr, } 218, with its character of general deterrence, is “a further relevant angle for the apportionment of punishment appended by 53. Mohr, “Unrecht und Strafe,” p. 122; see also Section 5.1.

120

punishment as a means of rehabilitation

Hegel to his conception of punishment first in the theory of civil society.”54 The possibility of a mixed theory presupposes that this relevant angle is compatible with the alleged retaliation from “Abstract right.” In regard to retaliation in the way it is meant by Kant, this would obviously not be the case. Hegel defines danger in relation to the “inherently . . . stable and strong” “existence” that the law should have, that is, in relation to compliance with the law.55 The more the crime endangers general compliance with the law, the more strictly the crime must be punished. If no state exists, the danger is, by definition, “infinite” and the revenge must then be in equal measure to the danger.56 If “the power of society has now become sure of itself” then the punishments could, and should, become much more lenient. Such a definition of punishment prescribes deterring further crimes as the aim of punishment. One cannot see, in Hegel’s view, any retributivism unless one were to conceive so broadly of the definition of retributivism that the proportionality, or the parity, between crime and punishment would end up amounting to the parity between danger and punishment. By so doing, though, the retributivist thesis actually would become indistinguishable from the deterrent thesis. Conclusion: a Hegelian mixed theory of punishment, in reality, is just as improbably a true mixed theory as can be the supposed Kantian mixed theory. A Kantian mixed theory remains fundamentally retributive; a Hegelian mixed theory ultimately remains a theory of deterrence.

5.3. Objections to the retributivist interpretations If one can find in Hegel’s theory of punishment neither a twofold justification nor a mixed theory, but instead only an objective interpretation, which draws either on retributivism or on a deterrent theory, then the retributivist reading seems likely. Therefore, it is the traditional and most supported reading. Now, however, compatibility of the classical principle of retaliation with the Hegelian concept of right proves itself to be problematic in at least two ways.

54. Mohr, “Unrecht und Strafe,” p. 118. 55. GPhR } 218. Elements, ed. Wood, p. 250. 56. GPhR } 218, Remarks. Elements, ed. Wood, pp. 250f.

hegel’s “negation of crime”

121

5.3.1. Hegel’s critique of talion law concerns more than its literal application. In the first passage in which Hegel refers to the theory of retaliation, this occurs solely with reference to the argument of the negation of the negation57 as well as in relation to taking into account the crime’s quantity and quality in the determination of the negation of the negation:58 The cancellation . . . of crime is retribution in so far as the latter, by its concept, is an infringement of an infringement, and in so far as crime, by its existence . . . has a determinate qualitative and quantitative magnitude, so that its negation, as existent, also has a determinate magnitude.

Immediately thereafter, Hegel expressly rejects classical retaliation as being “equality in the specific character of the infringement.” Hegel writes: “this identity [of crime and retribution], which is based on the concept, is not an equality in the specific character of the infringement, but in its character in itself – i.e. in terms of its value.”59 What “this identity . . . which is based on the concept” or this “equality in the specific character of the infringement” can possibly mean is elucidated by Hegel in the Remarks: It is only in terms of this specific shape [specific equality] that theft and robbery . . . and fines and imprisonment etc. . . . are completely unequal, whereas in terms of their value, i.e. their universal character as injuries . . . they are comparable.60

Therefore, the only “retaliation” (Wiedervergeltung) that Hegel announces himself as supporting is nothing other than the infringement of the infringement. Some interpreters are of the opinion that Hegel’s critique is directed only against a literal application of the classical theory of retaliation. According to these interpreters, Hegel’s critique can be reduced to these lines: “the sphere of externality, in which no absolute determination is in any case possible”; “in the realm of finite things, the absolute determination remains only a requirement, on which the understanding must impose increasing restrictions . . . [which admits] only an approximate fulfilment.” This aspect of Hegel’s critique surely explains the “insuperable difficulty” for the thesis of “specific equality” 57. 58. 59. 60.

GPhR }} 93 and 97. Elements, ed. Wood, pp. 120–1 and p. 123. GPhR } 96. Elements, ed. Wood, pp. 122–3. GPhR } 101. Elements, ed. Wood, p. 127. GPhR } 101, Remarks. Elements, ed. Wood, p. 129.

122

punishment as a means of rehabilitation

to determine punishments (especially if psychology also invokes the strength of sensuous motives . . . and, as a corollary, either the correspondingly greater strength of the evil will or – if we prefer – the correspondingly lesser strength and freedom of the will in general).61

If, however, one reads Hegel’s critique of classical retaliation in an unabridged form then one must particularly take into account the second half of Hegel’s argument: If we not only overlook this nature of the finite realm but [emphasis mine] also proceed no further than abstract and specific equality, an insuperable difficulty arises when we come to determine punishments . . . Furthermore [emphasis mine], it is very easy to portray the retributive aspect (Wiedervergeltung) of punishment as an absurdity (theft as retribution for theft, robbery for robbery, an eye for an eye, and a tooth for a tooth, so that one can even imagine the miscreant as one-eyed or toothless); but the concept has nothing to do with this absurdity, for which the introduction of that [idea of] specific equality is alone to blame.62

Thus the absurdity lies not in an erroneous, literal determination, but rather in the axiom that “what the criminal has done should also happen to him,” and therefore in the specific equality or in talion law in contrast to the Hegelian principle of the negation of the negation.63 It should be noticed that in this context Hegel, from the outset, does not ascribe to this axiom of classical retaliation the status of a philosophical determination of the concept of right. He introduces the remarks on } 101 as follows: It is usual in science for a determination – in this case, that of punishment – to be defined in terms of the universal representations . . . of conscious psychological experience. In the present case, this experience would indicate that the universal feeling of peoples and individuals towards crime is, and always has been, that it deserves to be punished, and that what the criminal has done should also happen to him.64

Primoratz maintains that this “attitude” of “reciprocity” is the very reciprocity of a “man towards punishment.” He annotates Hegel: “On this point, commonsense is correct, Hegel believes, and philosophy has

61. 62. 63. 64.

GPhR } 101, Remarks. Elements, ed. Wood, p. 128. GPhR } 101, Remarks. Elements, ed. Wood, p. 128. GPhR } 101, Remarks. Elements, ed. Wood, p. 127. GPhR } 101, Remarks. Elements, ed. Wood, p. 127.

hegel’s “negation of crime”

123

to abide by the judgment of common sense.”65 This is contradicted not only by the fact that Hegel has time and again either strictly differentiated between common sense philosophy and the mere science of understanding or expressly wanted to differentiate between the two. One of the many points of Hegel’s critique of common sense and of the science of understanding is that they become entangled in contradictions from which they attempt in vain to loosen themselves, for they cannot cancel these very contradictions philosophically. Again, in the passage about crime Hegel continues: It is incomprehensible how those sciences which derive their determinations from universal representations . . . should on other occasions accept propositions which contradict such so-called universal facts of consciousness.66

Hegel presumably means by this those “various theories of punishment as prevention, as a deterrent, a threat, a corrective, etc.” which presuppose, in the punishment’s rationale, the “superficial character of an evil.”67 Hegel sharply criticizes them: As a result of these superficial points of view, however, the objective consideration of justice, which is the primary and substantial point of view in relation to crime, is set aside; it automatically follows that the essential consideration is now the moral point of view, i.e. the subjective aspect of crime, intermixed with trivial psychological ideas.68

I proceed under the assumption that the contradictory “propositions” that Hegel criticizes in the Remark on } 101 are those very theories I have described above, because both these and retributivism were already then the two dominant competing theories of punishment: punishment was imposed either because the crime was committed (retributivism), or in order that something occur (rehabilitation, either general or specific deterrence; all following a consequentialist intent). Sometimes, these are referred to as “retrospectively” or “prospectively” directed theories. Hegel raises the same objection against retributivism as he raises against the various consequentialist theories of punishment, that is, that they rely on common sense, that they are subjective rationales and that they contradict one another. 65. 66. 67. 68.

Primoratz, “Banquos Geist,” 37. GPhR } 101, Remarks. Elements, ed. Wood, p. 127. GPhR } 99, Remarks. Elements, ed. Wood, p. 124–5. GPhR } 99, Remarks. Elements, ed. Wood, p. 125.

124

punishment as a means of rehabilitation

It is well known that, whenever the theories of understanding contradict one another and the dispute seems to be irresolvable, Hegel’s “science of reason” consists in canceling (aufheben) these contradictions.69 In this cancellation, the opposing theories prove themselves to be merely partial insights into the relevant concept. In the question of penal law as well, Hegel proceeds exactly according to this method in the Elements of the philosophy of right. Both retributivism and consequentialism are categorized expressly as partial insights into the concept of punishment by Hegel so that both, by their claim that they provide a rationale of punishment, distort that very concept of punishment. Hegel says of the consequentialist theories of punishment that the various considerations which are relevant to punishment as a phenomenon . . . and to its relation . . . to the particular consciousness, and which concern its effect on representational thought (as a deterrent, corrective, etc.), are of essential significance in their proper context, though primarily only in connection with the modality of punishment. But they take it for granted that punishment in and for itself is just. In the present discussion, we are solely concerned with the need to cancel (aufheben) crime – not as a source of evil, but as an infringement of right as right – and also with the kind of existence which crime possesses, which must also be cancelled.70

About retributivism, Hegel states: “It is this inner identity which, for the understanding, is reflected in external existence (Dasein) as equality.”71 By “inner identity” is meant the infringement of the infringement; by “equality” is meant classical retaliation. Before I move on to the Hegelian cancellation of the opposition between both of these theories in the concept of punishment, there is a point that contradicts the assumption of some interpreters, stating that the classical theory of retaliation and the Hegelian penal theory are compatible, to which I have yet to attend. 5.3.2. The negation of the negation is not retaliation. The aforementioned “negation of the negation” is not to be understood as being the negation of an opposing element, as in the case that A is the negation of :A, and therefore it is ::A. For Hegel, the negation of the negation is rather the cancellation of the negation, that is, the cancellation of 69. Cf., for example, Hegel’s Encyclopedia of the philosophical sciences, } 36, Addition. 70. GPhR } 99, Remarks. Elements, ed. Wood, p. 125. 71. GPhR } 101, Remarks. Elements, ed. Wood, p. 128.

hegel’s “negation of crime”

125

coercion. It is not ::A, but rather A. Yet it is not the absolute A, but instead an A having validity by the cancellation of the negation. Naturally, by the negation of the negation, :A becomes negated. That :A becomes negated does not mean, however, that the negation is on the whole negated. Now, right requires that the negation itself becomes negated, and therefore that no violence results from the punishment. The punishment is the special negation actually bringing to termination the very act of negating in general. Now, how does this relate to classical retaliation, according to which “what the criminal has done should also happen to him”?72 The criminal has negated the personality (A) of his or her victim: :A. Happening to the criminal what he or she him- or herself has done to the victim means the negation of the personality (B) of the criminal: :B. The relationship between :A and :B is naturally that B executed the action :A. Through :B, :A was indeed negated but B was also negated. Therefore, :B must be negated. In other words: whoever is responsible for making sure that what the criminal has done happens to the criminal him- or herself does exactly what the criminal has done. Therefore, what the criminal has done ought also to happen to him or her (whoever is responsible, that is). Retaliation reproduces the negation instead of canceling it. Hegel expresses this in the following way: In the sphere of the immediacy of right, the cancellation . . . of crime is primarily revenge, and its content is just so far as it constitutes retribution . . . Thus revenge, as the positive action of a particular will, becomes a new infringement; because of this contradiction, it becomes part of an infinite progression and is inherited indefinitely from generation to generation.73

Hegel clearly draws the consequence from this opposition between the negation of the negation and the classical theory of retaliation. To require that this contradiction [between the negation of the crime and its multiplication] . . . should be resolved . . . is to require a . . . punitive rather than an avenging justice. Primarily, this constitutes a requirement for a will which, as a particular and subjective will, also wills the universal as such.74 72. GPhR } 101, Remarks. Elements, ed. Wood, p. 127. 73. GPhR } 102. Elements, ed. Wood, p. 130. This progression of revenge is elucidated by the Addition to } 102 with the censure of “undying” “revenge” among “uncivilized peoples” that propagates itself as revenge and then revenge for that revenge from generation to generation. 74. GPhR } 103. Elements, ed. Wood, p. 131.

126

punishment as a means of rehabilitation

A common way of interpreting this passage involves reading it as merely being a requirement for impartiality of the judge against the risk of an interested – and thus partial – private justice. However, this implies an inadmissible abridgment. For even if this private justice committed itself to impartiality, the problem that Hegel addressed would still exist. The problem consists namely in that the victim “place[s] [his or her] infinity in any infringement . . . which occurs, and whose justice is therefore altogether contingent.” In this sentence, Hegel does not deny that the revenger acts justly. The “content” of revenge is, he writes, “just so far as it constitutes retribution,” and Hegel speaks of “[its] justice.”75 Therefore, this sentence is not concerned with the risk of the misuse of punishment. It is concerned with the right that is to be restored. Whereas in revenge two persons stand opposed to one another, who are both to be seen as being the existence (Dasein) of the will and therefore are allowed to make a claim on the right of personality, in the state every person is subject to the law, in which inheres the objective right of universality. Before the law, the person has only the subjective right of personality, which implies, for instance, that the state is not allowed to handle him or her arbitrarily. The consequence is that that which is to be restored as right is, in the case of revenge, the personality of the criminal but in the case of punishment the validity of the law.76 Now the victim’s personality does not directly contain the personality of the criminal, whereas the law, by definition, recognizes the personality of all members of the commonwealth. Hegel says of the punitive justice of the state: “Instead of the injured party, the injured universal now makes its appearance.” Hegel places the subjective and contingent revenge in contrast with the punishment: Objectively, this reconciliation applies to the law, which restores and thereby actualizes itself as valid through the cancellation (Aufheben) of the crime; and subjectively, it applies to the criminal in that his law, . . . which is known by him and is valid for him and for his protection, is enforced upon him in such a way that he himself finds in it the satisfaction of justice and merely the enactment of what is proper to him.77 75. GPhR } 103. Elements, ed. Wood, p. 130. 76. This differentiation is lacking, in my view, in Axel Honneth, The struggle for recognition: the moral grammar of social conflicts, trans. Joel Anderson (Cambridge, Mass.: MIT Press, 1996), p. 21. 77. GPhR } 220. Elements, ed. Wood, p. 252.

hegel’s “negation of crime”

127

What Hegel calls “retribution” (Wiedervergeltung), which is “an infringement of an infringement,” should not be confused, therefore, with the retaliation (Wiedervergeltung) of classical retributivism.78 In the degree of punishment as well, a difference now arises between the restoration of the personality of the individual – or of a group of individuals – and the restoration of the law. Restoring the law, that is, to reaffirm its validity, is completely feasible. Restoring the existence (Dasein) of the will in an individual person in which it was found before the crime is, on the contrary, only possible in some cases. These cases are treated in } 98 of the Elements of the philosophy of right: “The cancellation . . . of the infringement, where the latter has caused damage, is civil satisfaction in the form of compensation (in so far as any compensation is possible) [second emphasis mine].”79 In most of the cases, though, there is no compensation to be found. And in those cases as well, wherein compensation is possible, the “positive existence of the injury,” that is, “the particular will of the criminal,” remains unaffected: the criminal can provide the whole compensation while keeping his or her criminal intentions.80 Some authors attempt to justify the classical principle of retaliation not merely by appeal to reciprocity in general, but also in other ways. Jean Hampton mentions, for example, the explanation that talion law has the aim of destroying the criminal’s claim permanently to acquire supremacy over his or her victim. Poking out the criminal’s eye is, according to this conception, a way for the victim, who was made monocular by the criminal, to revoke that superiority unjustly gained by the criminal. The question of the rightness of this and of similar explanations remains to be addressed (Hampton herself rejects it). Bear in mind, though, that Hegel does not enter into such interpretations of classical retaliation theory.81 In its place, Hegel adopts a symbolic interpretation: “It is this inner identity which, for the understanding, is reflected in external existence . . . as equality.” In revenge, the restoration of right cannot really occur: therefore, it occurs symbolically – that is, as talion law, as a specific equality: “robbery for robbery, an eye for an eye, and a tooth for a tooth.”82 GPhR } 101. Elements, ed. Wood, p. 127. GPhR } 98. Elements, ed. Wood, p. 124. GPhR } 99. Elements, ed. Wood, p. 124. Cf. John Kleinig, “Punishment and moral seriousness,” Israel Law Review, 25, no. 3 (1991), 401–21 (p. 416). 82. GPhR } 101, Remarks. Elements, ed. Wood, p. 128. 78. 79. 80. 81.

128

punishment as a means of rehabilitation

The “absurdity” of such “punishments” consists of their having nothing to do with the restoration of right, with the termination of the coercion and with reconciliation – a reconciliation with the criminal as well. On the contrary, such punishments do not treat the criminal as an existence of freedom. Only in so far as the criminal is a member of a commonwealth is the preservation of the will’s existence in the criminal’s person necessary and required for the restoration of a commonwealth to which he or she belongs and by which his or her rights are protected as well. The criminal’s forced amputation is not necessary for such a restoration of the validity of law. Rather, amputations contradict the idea of a commonwealth of free persons. For that reason, amputations are “absurdit[ies].”83

5.4. Reconstruction of the Hegelian legal justification of punishment The grounding of punishment in Hegel consists of the verification that the punishment is a necessity of right. The keystone of the normative construction of punishment, namely the “second coercion” or that which “cancels an initial coercion” (we will see later how this might be understood), is “not only conditionally right but necessary” for right.84 Punishment, therefore, is not only in compliance with right by itself; rather, without it there would be no right. In contrast to that, the initial coercion is “contrary to right.” Hegel’s grounding of punishment does not separate the authority to punish from the duty to punish; initially, it does not attempt to demonstrate first that punishment may occur and then, after that, that punishment should occur, but instead it fulfills both tasks together. Hegel substantiates this necessity in the following way: “Because coercion destroys itself in its concept, it has its real expression . . . in the fact that coercion is cancelled . . . by coercion.”85 This thesis of the conceptual self-destruction of coercion is not to be taken as a physical thesis. It does not concern, for instance, the physical self-destruction of coercion; instead, what matters is that this coercion is “contrary to right.”86 Contrariness to right is the same as the concept of legal

83. 84. 85. 86.

GPhR } 101. Remarks. Elements, ed. Wood, p. 128. GPhR } 93. Remarks. Elements, ed. Wood, p. 120. GPhR } 93. Remarks. Elements, ed. Wood, p. 120. GPhR } 92. Remarks. Elements, ed. Wood, p. 120.

hegel’s “negation of crime”

129

impossibility.87 The coercion destroying itself means that coercion is “rightly” impossible or incompatible with the concept of right. In the self-destruction two elements must be distinguished: (1) coercion is destroyed; (2) it itself carries this destruction through: it is a selfdestruction. The first coercion is canceled by the second coercion; important here is to realize that it is canceled by nothing else other than coercion. In the second regard, coercion contrary to right differentiates itself from the infringements it creates, that is, from the “evil” or the “damage,” because “when an infringement of right occurs, it does have a positive external existence . . . but this existence within itself is null and void.”88 Even if the infringement being a consequence of the coercion has a physical existence – that is, as damage – it actually does not have any existence in right at all. This infringement is not destroyed by itself. The difference between the self-destruction of coercion and the extrinsic destruction of damage comes to expression in the form of the differentiation between the “cancellation” (in right) of coercion and the “nullity” (in right) of damage.89 As is generally known, the Aufhebung (“cancellation”) of contradictions, which is a central concept in the Hegelian dialectic, means not only destroying whatever may contain a contradiction, but also simultaneously preserving something of it. Therefore, something is preserved from coercion by its destruction. We must take note of this point for later. Many interpreters do not know how to make sense of selfdestruction of coercion or regard it as being nonsensical. This incomprehension often consists of overlooking the premises upon which the thesis of self-destruction rests. I shall provide the original text from Hegel’s Elements in order to make reference to the grammatical structures in the German original, which I also emphasize: [Because] the will is Idea or [is] actually free only in so far as it has existence . . . and the existence in which it has embodied itself is the being of freedom, [then] force or coercion immediately destroys itself in its concept. 87. There is more at stake here than either the credibility of the legal system or the honesty of the civil authorities. David Cooper believes this to be Hegel’s core argument. David Cooper, “Hegel’s theory of punishment,” in Zbigniew A. Pelczynski (ed.), Hegel’s political philosophy (Cambridge: Cambridge University Press, 1971), pp. 151–67. 88. GPhR } 97. Remarks. Elements, ed. Wood, p. 123. 89. GPhR }} 93 and 97, respectively. Elements, ed. Wood, pp. 120 and 123.

130

punishment as a means of rehabilitation

Weil der Wille, nur insofern er Dasein hat, Idee oder wirklich frei und das Dasein, in welches er sich gelegt hat, Sein der Freiheit ist, so zersto¨rt Gewalt oder Zwang in ihrem Begriff sich unmittelbar selbst.90

The fundamental premise here – as in the Elements of the philosophy of right on the whole, actually – is that free will must develop; and this development is what Hegel terms “right.” I shall here content myself with calling to mind a few central statements from the introductory paragraphs of the Elements: “The subject-matter of the philosophical science of right is the Idea of right – the concept of right and its actualization.”91 The basis . . . of right is the realm of spirit in general and its precise location and point of departure is the will; the will is free, so that freedom constitutes its substance and destiny . . . and the system of right is the realm of actualized freedom.92

And: “Right is any existence . . . in general which is the existence of the free will. Right is therefore in general freedom, as Idea.”93 Right having to be plays the role for Hegel that is played by the categorical imperative for Kant: the establishment of a state of law is a necessity. The existence of freedom required by the idea of right consists in the “abstract right” of the individual body and life as well as of personal property. Hegel notes that free will in and for itself cannot be coerced . . . except in so far as it fails to withdraw itself from the external dimension in which it is caught up, or from its idea . . . of the latter . . . Only he who wills to be coerced can be coerced into anything.94

It should be noted that “free will” must want to allow itself to be coerced by the criminal because, according to the idea of right, it must have an existence: in terms of right it cannot withdraw from externality. Right forbids the victim from a stoic retreat into internality in order to escape the criminal’s coercion. In “Ethical life” it is not the individual but the state that displays the highest existence of free will. However, the grounding of penal law in abstract right holds for those criminals whose direct victims are 90. The emphases are mine. GPhR } 92. The original text can be found in Hegel, Grundlinien der Philosophie des Rechts. Elements, ed. Wood, p. 120. 91. GPhR } 1. Elements, ed. Wood, p. 25. 92. GPhR } 4. Elements, ed. Wood, p. 35. 93. GPhR } 29. Elements, ed. Wood, p. 58. 94. GPhR } 91. Elements, ed. Wood, pp. 119–20. Regarding property, cf. GPhR } 90 (“property”) (p. 119) and } 91 (“his physical side and other external attributes”).

hegel’s “negation of crime”

131

individuals, as well as for those whose direct victim is the state.95 Now, body, life and property of the individuals are all an existence of the free will. Therefore, the body, life and property of the criminal are just as much the existence of the free will as the body and life of the victim. Therefore, self-destruction in } 92 ought to be understood in the following ways: 1 One must understand self-destruction as the thesis of the necessary destruction of that which contradicts right. Since the realization of right – and along with it as well, the existence of the free will in the body, in the life and in the property of the victim – is a necessity, the crime, being an attempt to infringe on this existence, can only fail and lead to the exercising of coercion against the criminal. 2 Self-destruction must also be understood as the thesis of the identity of the criminal’s free will with the victim’s free will. The criminal coerces or destroys the existence of free will in the person of the victim and thereby coerces or destroys him- or herself, since the criminal’s own existence is also the existence of free will. The two theses are linked to each other, since right must have its existence in every person, so that, for right, the personality of the criminal and that of the victim are equal. In Hegel, the refusal to recognize the personality of all the other humans implies a discrepancy on the part of that person who denies the recognition: on the one hand, the universal element in the person’s will is one with the universal element in the wills of the other persons; on the other hand, this universality is alien to it. Its particular will and its general will diverge. This implies that the human who does not recognize the status of the personality in other humans lacks the foundation upon which he or she can base his or her own status as a person. According to Hegel, those societies in which the personality of only some humans is recognized (for instance, Egypt or the Greece of antiquity) are societies in which the personality is something contingent. Therefore, the necessity of right requires the self-destruction of coercion. Now, this requirement is still “taken in the abstract” (that is, abstractly understood) as long as it has not come to realization: “force or coercion immediately destroys itself in its concept [emphasis 95. Cf. GPhR } 95, Remarks. Elements, ed. Wood, p. 122: “the determination which will be considered in the following paragraph also applies to the particular and further determined content [of crime], e.g. in perjury, treason, counterfeiting, forgery, etc.”

132

punishment as a means of rehabilitation

mine].” According to Hegel, punishment is the “real expression” of the conceptual destruction of coercion.96 The necessity of right – meaning, the requirement that right realizes itself because it must realize itself – demands “that coercion is cancelled . . . by coercion.”97 Hegel also terms the realization or the implementation of the requirement of right the “manifestation” of this requirement; “the punishment is . . . the manifestation of the crime.”98 How can coercion cancel coercion? Hegel gives the following definition of coercion: in coercion’s externality, the will “may either experience force in general, or it may be forced to sacrifice or do something as a condition of retaining some possession or positive being, thereby suffering coercion.”99 There are three aspects to coercion. Coercion is: 1 the exercise of force upon a person, 2 in order to make the person act differently or sacrifice something, 3 by making this action or this sacrifice a precondition for keeping his or her body, life or property. Whereas the thesis of self-destruction in } 92 relates both to coercion and to mere force, and “initial” coercion, that is, the coercion that is contrary to right (the crime), can either be coercion or force,100 Hegel in } 93 merely mentions the cancellation of coercion by coercion, and, with the mention of coercion in the Remarks to } 93, the cancellation of force by coercion. Nowhere does Hegel speak of the cancellation of either coercion or force by force. In the light of the definition of force as mere suffering, cancellation by force is explicitly excluded by Hegel by his criticism of Ernst Ferdinand Klein’s conception of punishment as being an arbitrary cancellation of an evil through an evil: If the crime and its cancellation . . . which is further determined as punishment, are regarded only as evils . . . in general, one may well consider it unreasonable to will an evil merely because another evil is already present.101

96. 97. 98. 99. 100. 101.

GPhR } 93. Elements, ed. Wood, p. 120. GPhR } 92. Elements, ed. Wood, p. 120. GPhR } 97. Elements, ed. Wood, p. 123; and } 101, Addition. Elements, p. 129. GPhR } 90. Elements, ed. Wood, p. 119. Cf. GPhR } 93, Remarks. Elements, ed. Wood, p. 120: “initial coercion, or at least force.” GPhR } 99. Elements, ed. Wood, p. 124. Regarding Klein: Hegel is referring to Ernst Ferdinand Klein, Grundlinien des gemeinen deutschen und preußischen peinlichen Rechts (Halle: Hemmerde und Schwetschke, 1796), }} 9–10.

hegel’s “negation of crime”

133

At least three consequences result from this. 1 It begs the question of the applicability of the cancellation of coercion by a coercion exercised upon the criminal, if the criminal does not allow him- or herself to be forced. We will come back to this question later. 2 The death penalty obviously does not “coerce” the criminal, that is, it does not move the criminal to perform an action or a sacrifice but exercises mere force against him or her. Hegel’s position on the question of the death penalty appears to be inconsistent. On the one hand, he maintains that murder “necessarily incurs the death penalty.”102 Admittedly he writes that it is to Beccaria’s credit that the death penalty was repealed for most crimes. Hegel himself wants to keep the death penalty for special cases, especially with murder in mind: Even if neither Joseph II nor the French have ever managed to secure its complete abolition, people have begun to appreciate which crimes deserve the death penalty and which do not. The death penalty has consequently become less frequent, as indeed this ultimate form of punishment deserves to be.103

On the other hand, Hegel really sees alternatives to the death penalty. Because punishment must coerce the criminal, whereas the death penalty does perfectly fulfill the personal wishes of some to be sentenced to death, then Hegel must actually allow some alternative punishments, even for murder. A handwritten remark in his copy of the Elements on } 99 notes that “punishment has to be sensitive . . . It even has occurred that murders [were committed] in order to lose [one’s] life, – therefore, [they are] insensitive to the death penalty; – then it came to be, that it was changed to imprisonment.” In the Lecture on the philosophy of right (1824/5) he says in a somewhat more detailed way: There have been cases in which a murder occurred in order [for one] to be sentenced to death. The murderer did it out of world weariness, contempt for life, though especially in a religious sense . . . Thus the death penalty does not affect him – he already wills to leave this life – consequently, a sentence of life has been replaced with prison in order to affect the will of the criminal.104 102. GPhR } 101, Addition. Elements, ed. Wood, p. 129. 103. GPhR } 100, Addition. Elements, ed. Wood, p. 127. 104. PhR 285.

134

punishment as a means of rehabilitation

In the Elements of the philosophy of right, there is no argument to be found against substituting alternative punishments for the death penalty for all crimes. On the contrary, an adherence to the death penalty would lead to another definition of punishment, namely to cancellation of coercion or (for example, in the case of murder) of force by (mere) force. Moreover, this new definition would contradict Hegel’s theory in some ways. I have already mentioned his criticism of Klein. Also there is his criticism of a penal theory that Fichte once took into consideration. A cancellation of coercion or of force by force, according to Hegel, amounts to the conception of punishment in which a criminal “is regarded simply as a harmful animal which must be rendered harmless.”105 This theory of punishment that Fichte deliberated upon – and that he rejected – would not treat the criminal as a rational being, according to Hegel. For this reason, he discards it. 3 The conceptual self-destruction of coercion and the real cancellation of coercion by coercion seem to imply the same conceptual selfdestruction and real self-cancellation. It is an obvious temptation to conclude that there is a self-cancellation of coercion, which could possibly be an argument in favor of retributivism. But Hegel does not actually do this; instead, he refers to a conceptual selfdestruction of force.106 Apart from this, we have just seen that a selfcancellation of mere force by mere force is out of the question, so that only a cancellation of mere force by coercion is conceivable. Notwithstanding the question whether the crime was mere force or coercion, what matters is that the cancellation of the crime happens through coercion. As we have seen, coercion is (1) the exercise of force upon a person, (2) in order to make the person act differently or sacrifice something, (3) by making this action or this sacrifice a precondition for keeping his or her body, life or property. If this definition is applied to the “second coercion” – to the coercion against the criminal – then from these three elements result three implications. The cancellation of the crime is: (1) an evil – or force – that is exercised against the criminal, (2) so that the criminal acts differently or sacrifices something, (3) by making this altered deed or this sacrifice to be a precondition for keeping his or her body, life or property. 105. GPhR } 100, Remarks. Elements, ed. Wood, p. 126. 106. Cf. GPhR } 92. Elements, ed. Wood, p. 129: “Consequently, force or coercion immediately destroys itself in its concept.”

hegel’s “negation of crime”

135

5.5. Integration of the other penal theories A threefold objection can be made against such an interpretation: first, Hegel’s criticism of Klein’s conception of punishment as being the infliction of an evil; secondly, Hegel’s criticism of the penal reform theory;107 thirdly, the fact that Hegel nowhere – at least nowhere in “Abstract right” – makes the criminal’s rehabilitation an issue. The first two objections misunderstand Hegel’s criticism of Klein and of the reform theory. Hegel’s criticism is directed exclusively against absolutizing evils and reform. Hegel writes: If the crime and its cancellation, which is further determined as punishment, are regarded only as evils in general, one may well consider it unreasonable to will an evil merely [emphasis mine] because another evil is already present.108

Hegel’s critique only maintains that the aim of punishment, which serves to ground punishment, cannot be to inflict an evil on the criminal, although the criminal has actually inflicted an evil on his or her victim. It is for this reason that Hegel speaks of a “superficial character of an evil”: the evil is not a primary but – at most – a derivative aim of punishment. As regards the reform theory, it is equally discarded only as an absolutization of reform. Hegel writes: The various considerations which are relevant to punishment as a phenomenon . . . and to its relation . . . to the particular consciousness, and which concern its effect on representational thought (as a deterrent, corrective, etc.), are of essential significance in their proper context, though primarily only [emphasis mine] in connection with the modality of punishment. But they take it for granted that punishment in and for itself is just.109

As regards the theory of rehabilitation, it is expressly found in “Ethical life.” “Retribution through revenge” is contrasted by Hegel to “genuine reconciliation of right”: “subjectively, it applies to the criminal in that his law, which is known to him and is valid for him and for his protection.”110 As early as in “Abstract right,” Hegel is saying of the cancellation of crime: “it is at the same time his will as it is in itself,

107. 108. 109. 110.

For both points one and two cf. the Remarks in GPhR }} 99 and 100. GPhR } 99, Remarks. Elements, ed. Wood, p. 124. GPhR } 99, Remarks. Elements, ed. Wood, p. 125. GPhR } 220. Elements, ed. Wood, p. 252.

136

punishment as a means of rehabilitation

an existence of his freedom, his right.”111 The punishment, being as it is just, guarantees the right, which is the existence (Dasein) of the criminal’s free will; thus, punishment is the precondition of the full membership of the criminal. It should be noted once more that the criminal’s rehabilitation as a full member of the commonwealth presupposes that punishment is just. Therefore, the rehabilitation aspect is also a secondary, derivative aim of punishment. Hegel’s theory is not absolutist, that is, he does not support the thesis that punishment is in itself good and already justified because the criminal has deserved the punishment through his or her crime. In an absolutist theory of punishment, punishment is only grounded upon the criminal and what the criminal has merited; punishment is then only seen as being a merited ill. We find good evidence in Kant’s example of the island in which the sole justification of punishment cited is: “so that each has done to him what his deeds deserve.”112 The role of this example in the Kantian theory of right will not be discussed here.113 Here I content myself with drawing attention to the fact that one searches in vain for a similar example in Hegel. Hegel’s theory of punishment thoroughly assigns an aim to the punishment: the cancellation of the crime. The cancellation of the crime by coercion has two unequal dimensions: the past and the future. The cancellation of the crime’s past dimension clearly does not mean that the crime would be made undone, that is, that everything were as if nothing had happened. At the most, some of the consequences of the crime can be eliminated by “civil satisfaction in the form of compensation (in so far as any compensation is possible).” In so far as “the damage amounts to destruction and is altogether irreparable,” “the universal character of the damage, as value, must in any case take the place of its specific qualitative character.”114 To the extent to which the criminal is thus coerced to make civil amends, the amends can be placed in the category of cancellation of coercion by coercion. Hegel, however, finds this aspect of cancellation of crime to be of marginal importance. Civil amends is non-specific to the crime. In “collisions of right” or in the case of legal disputes that arise in civil law – in cases without any penal dimension – there is also often 111. 112. 113. 114.

GPhR } 220. Elements, ed. Wood, p. 126. RL Ak vi:333. Practical philosophy, ed. Gregor, p. 474. Cf., though, Section 1.3 above. GPhR } 98 and Remarks. Elements, ed. Wood, p. 124.

hegel’s “negation of crime”

137

compensation as well.115 The criminal aspect is not situated in the evil or the harm. As I said before, Hegel indeed ascertains a “positive, external existence” in harm; it is, however, “within itself . . . null and void,” for the positive existence comprises not an existence of will but, instead, a “mere product” thereof.116 The criminal did not primarily intend harm. The criminal primarily intended the realization of his or her freedom in a specific existence – in an alien good, in a larger freedom through the elimination of other humans, etc. – by means of the exercise of a coercion on another will. At most, the criminal recklessly disregarded the harm it might cause. The compensation here is not to be confused with, for instance, the restitution of stolen goods to their owner. The question arises here of the appraisal of vandalism and of other crimes whose exclusive intent is destruction. A possible answer to this objection would be to see in these cases the criminal’s disappointed will that does not receive the existence it designed. No matter how it stands with this objection, in any case for Hegel the crime definitely belongs to the past: the harm corresponds neither to something in the will of the victim who ceased to be coerced by the criminal nor to something in the criminal who had another primary intention. The future dimension of the cancellation of the crime is the specific dimension for the crime. Hegel observes: “The positive existence of the injury consists solely in the particular will of the criminal.”117 The cancellation of the crime by coercion can now occur either during the commission of the crime or after its commission. The cancellation of the crime during its commission pertains to the right of coercion belonging to the classical ius strictum (the “abstract right – or right in the strict sense”).118 First and foremost, self-defense pertains to this immediate cancellation of the crime during its commission. By self-defense the criminal is coerced. Hegel mentions this type of coercion – admittedly only briefly – which, in the strict sense, is neither a punishment (enacted by the state) nor (private) revenge.119 It is also nearly uncontroversial. This existence of the crime is to be seen not only as being the purpose or intent of the crime during the crime’s commission, but also 115. 116. 117. 118. 119.

GPhR }} 84–6. Elements, ed. Wood, p. 117. GPhR }} 97 and 99. Elements, ed. Wood, pp. 123, 124. GPhR } 99. Elements, ed. Wood, p. 124. GPhR } 94, Remarks. Elements, ed. Wood, p. 121. Cf. GPhR } 94. Elements, ed. Wood, p. 121.

138

punishment as a means of rehabilitation

as the present and past will of the criminal. Hegel’s presumption is that the “particular will of the criminal” continues to be a criminal will should it remain unpunished. In his lecture Natural right and the science of state (Naturrecht und Staatswissenschaft) from 1818/19, Hegel expressly says: What the will does is, according to appearance, a single, but also a universal – something temporal and something intemporal. – Thus the consequences of the deed remain even if the deed is itself hidden. Thus whoever steals remains for ever a thief, not merely in memory but actually.120

And a little later he adds: “the deed of the criminal is not merely something transient, single, but instead something at the same time universal, a law stating that infringement against somebody is allowed.”121 Therefore, if the punishment does not occur then the criminal again commits a crime. According to this, in order that right should not be harmed again, the criminal must be coerced: The positive existence of the injury consists solely in the particular will of the criminal. Thus, an injury to the latter as an existent will is the cancellation . . . of the crime, which would otherwise be regarded as valid, and the restoration of right.122

Specific deterrence is therefore an aim of punishment that necessarily belongs to the primary aim of the cancellation of the crime. General deterrence, however, is not only an indirect side effect of punishment grounded in such a way, but also necessarily itself an aim of punishment. It means: “If society is still inwardly unstable, punishments must be made to set an example, for punishment is itself a counter-example to the example of crime.”123 This general deterrent aim of the Hegelian theory of punishment is also indirectly derived from the main aim of the cancellation of the crime. With the deed, the criminal accepts all consequences that he or she can foresee. These consequences that are internal to the criminal action, under which there is the risk that the deed moves other citizens to commit a crime, pertain to his or her criminal intent. The relationship between the criminal’s will and these consequences, 120. 121. 122. 123.

NRSW 275–6. NRSW 276–7. GPhR } 99. Elements, ed. Wood, p. 124. GPhR } 218, Addition. Elements, ed. Wood, p. 251.

hegel’s “negation of crime”

139

therefore, is closer than the relationship between the criminal’s will and the harm for which he or she is held responsible. The harm is not necessarily foreseeable by the criminal: it partly pertains to the consequences that are external to the criminal action. It follows from the criminal not only being held liable for the harm caused, but also being punished for consequences internal to the criminal action, that one must coerce the criminal, to whom belong exclusively the consequences internal to the criminal action and not those external to the criminal action. The consequences internal to the deed must not necessarily occur; they do, however, pose a risk (Hegel speaks of “danger”)124 that the criminal accepts. Hegel explains this through an example in his lecture Philosophy of right: A crime is dangerous, that is, it is primarily a crime, but it has a further effect, a further context. The single action is in itself so more far reaching, more universal than it appears according to its external reality. If I hold a light to a piece of wood then the flame only touches a small part of the surface. The arsonist does the same and leaves, thus, his crime is of a lesser extent; but, the wood is in relation to other wood that together constitutes a house – this one with other houses that constitute a city, and thus the fire can destroy a city. This universality is the danger of crime.125

This danger internal to the criminal action must be strictly differentiated from danger external to the criminal action, if one wishes to avoid a consequentialist misinterpretation of Hegel: Various qualitative determinations [of crime], such as danger to public security, have their basis in more precisely determined circumstances, but they are often apprehended only indirectly in the light of other consequences rather than in terms of the concept of the thing . . . Thus, the crime which is more dangerous in itself . . . in its immediate character, is a more serious infringement in its extent or quality.126

Belonging to the particular will of the criminal is the risk that the crime comes to be an example, therefore a precedent for a third party. The coercion exercised by punishment of the will of the criminal also cancels this precedent, which has a general deterrent effect, that is, though only a secondary one. 124. GPhR } 218, Remarks. Elements, ed. Wood, p. 251. Also, cf. Ripstein, Equality, responsibility and the law, pp. 228–9. 125. PhR 279. 126. GPhR } 96, Remarks. Elements, ed. Wood, pp. 122–3.

140

punishment as a means of rehabilitation

The various amounts of danger the criminal recklessly disregards explain the various severities of punishment, depending on the era in which the crime was committed. The severity of punishment conforms to how secure the commonwealth is or what the risk is for destabilization of the commonwealth by imitation of the crime. A prestate society, in which the punishment of crimes is still a private matter, or a weak commonwealth must both punish severely in order to preserve themselves; a secure commonwealth does not have to: it should punish more mildly. Of one thing Hegel is sure, however: the commonwealth can never be so sure of itself that it can allow itself to leave a crime unpunished. Hegel writes: But whereas it would be impossible for society to leave a crime unpunished – since the crime would then be posited as right – the fact that society is sure of itself means that crime, in comparison, is always of a purely individual character, an unstable and isolated phenomenon.127

That is because even if no other citizen imitates the criminal, the criminal still will commit further crimes for as long as he or she is not coerced to do otherwise. Let attention be drawn to the fact that the aim of specific deterrence is not itself some undetermined sort of specific deterrence. Hegel’s specific deterrence is always formulated negatively: “cancellation . . . of the crime, which would otherwise be regarded as valid”; “it would be impossible for society to leave a crime unpunished – since the crime would then be posited as right.”128 Another conception of specific deterrence would consist in first assuming that the more severe a punishment is, the more deterrent it is, and thus in demanding the harshest punishment possible, in order to maximize public safety. This conception is contradicted by Hegel’s specific deterrence. On the contrary, the criminal should be sentenced to a milder punishment; punishments must simply be precisely so severe as not to contradict the cohesion of the commonwealth or directly to endanger it. In this respect, Hegel’s theory of punishment is, in my view,

127. GPhR } 218, Addition. Elements, ed. Wood, p. 251. 128. GPhR }} 99 and 218, Addition. Elements, ed. Wood, pp. 124, 251. A comment made by Hegel in his manuscript explains a “cancellation . . . of the crime, which would otherwise be regarded as valid” as follows: “that is, would have general existence, because here a single Being is universal – for everyone.” (Translation mine.)

hegel’s “negation of crime”

141

primarily a minimal specific deterrence and secondarily an equally minimal general deterrence. It is telling that Hegel criticizes various theories as being superficial, though not specific deterrence: “prevention, deterrent[s] . . . threat[s], corrective[s], etc.,” “specific equalit[ies],” and so on, are mentioned – the only thing absent is specific deterrence.129 It is characteristic of specific deterrence that the criminal is the sole, essential addressee of the punishment, whereas, in the various theories of general deterrence that operate either by threatening punishment or by making an example of the criminal, all of society is the addressee, including the criminal him- or herself, but only as one among many. This reflexive structure of specific deterrence explains why Hegel in } 101 speaks of “retribution” (Wiedervergeltung) (more properly, though, of retaliation). We have seen above in Section 5.3, however, that this retribution should not be confused with the theory of retaliation, that is, with retributivism, which is sharply criticized by Hegel. The question arises, though, how } 100, which is considered in the retributivist interpretations to be a subjective justification of retributivism and which is called an “argument from law” by Seelmann, is compatible with the interpretation that I have suggested here.130 Hegel’s } 100 is actually occupied with the subjective aspect of crime. This can be easily explained, however, through a specific deterrence interpretation. The lecture Philosophy of right (1824/5) deduces the subjective perspective expressly from the cancellation of crime in the specific will of the criminal: This existent will of the criminal . . . must be affected . . . This point is linked to the fact that a punishment must be impressive on the criminal. If punishment is not impressive then his existent will is not infringed upon by it.131

Therefore, the subjective aspect does not occur, for instance as a subjective justification without which punishment would lack a rationale, but instead the justification is deduced from the punishment’s aim of specific deterrence, which is grounded in the cancellation of a wrong. Coercion must be – by definition – sensitive. 129. Remarks on GPhR }} 99 and 101. Elements, ed. Wood, pp. 124, 128. 130. See below, Sections 5.2.1, 5.2.2 and Seelmann, “Wechselseitige Anerkennung” and “Versuch einer Legitimation.” 131. PhR 285.

142

punishment as a means of rehabilitation

Unlike Anselm Feuerbach, Hegel understands sensitivity to be not only that which moves, for instance, the appetites and fear, but also the criminal’s moral judgments. Now, talion law is, according to Hegel, what everyone, and therefore the criminal as well, can sense the most: “Jus talionis is a natural concept in perception and was seen from time immemorial as being an element of punishment.”132 This aspect of crime that belongs to a subjective moral validity is not to be confused with either a psychology of the appetites or Hegel’s justification of the right of punishment. The remark made by Hegel on } 99 makes this clear. In it Hegel clearly speaks of three aspects: 1 Coercion exercised against the appetites by an “evil” or by a “source of evil” (this correlates to Klein’s theory of punishment that was criticized by Hegel; in the Additions to } 99, Hegel also rejects Feuerbach’s rationale of punishment by the “threat” of an evil). 2 The justification of punishment (“a matter of wrong and of justice”; “the objective consideration of justice, which is the primary and substantial point of view in relation to crime”; “[the rationale that] punishment in and for itself is just”; and “we are solely concerned with the need to cancel . . . crime – not as a source of evil, but as an infringement of right”). 3 The moral feeling of the criminal (“the moral point of view, i.e. the subjective aspect of crime”). Hegel differentiates (1) from (3) by criticizing the consequences of Klein’s theory of punishment, which considers punishment as an evil: it automatically follows that the essential consideration is now the moral point of view, i.e. the subjective aspect of crime, intermixed with trivial psychological ideas . . . of stimuli and the strength of sensuous motives . . . as opposed to reason, of psychological coercion and of psychological influences on representational thought . . . (as if such influences were not themselves reduced by freedom to something purely contingent).133

The aspect of punishment that belongs to subjective morality is defined through reason and freedom. Precisely, it is both of the latter that make up both of the foundational elements of } 100. This explains why Hegel writes:

132. NRSW 277. 133. GPhR } 99, Remarks. Elements, ed. Wood, p. 125.

hegel’s “negation of crime”

143

But it is neither a question merely of an evil [which should move the appetites] nor of this or that good; on the contrary, it is definitely . . . a matter of wrong and of justice [that is, of the cancellation of the crime]. As a result of these superficial points of view, however, the objective consideration of justice, which is the primary and substantial point of view in relation to crime, is set aside; it automatically follows that the essential consideration is now the moral point of view, i.e. the subjective aspect of crime, intermixed with trivial psychological ideas.134

Hegel does indeed not recognize the retributivist dimension as being the “primary and substantial point of view,” in other words as the main justification of punishment, but he sees in it a necessary means to the end of punishment: this dimension is the “subjective aspect of crime” and of punishment though not a subjective justification. Even if the objective justification would require a subjective justification, this justification surely could not be found in the principle of retaliation (Wiedervergeltung). We have already seen how Hegel attributes it to common sense that is entangled in contradictions: “that [the crime] deserves to be punished, and that what the criminal has done should also happen to him.”135 This applies not only to abstract right: morality too considers the principle of retaliation as immediate or “superficial” and contradictory.136 Thus Hegel writes in conjunction with the principle of retaliation in “Abstract right”: It is incomprehensible how those sciences which derive their determinations from universal representations . . . should on other occasions accept propositions which contradict such so-called universal facts of consciousness.137

We find an example of that straight away in the first paragraph of “Morality,” “Purpose and responsibility,” which leads to the differentiation between the deed or the “external event,” and the action or the “purpose and knowledge of the circumstances.” According to Hegel, the consequences internal to the action can be distinguished from the consequences external to the action only “indeterminate[ly],” because the “inner necessity [consequences internal to the action] 134. GPhR } 99, Remarks. Elements, ed. Wood, p. 125. 135. GPhR } 101, Remarks. Elements, ed. Wood, p. 127. 136. Regarding morality: GPhR }} 105ff. Elements, ed. Wood, pp. 135ff. Cf. Hegel’s handwritten comment on GPhR } 118: “Superficial[:] it should go well for the good – ill for the evil.” (Translation mine.) 137. GPhR } 101, Remarks. Elements, ed. Wood, p. 127.

144

punishment as a means of rehabilitation

comes into existence in the finite realm as external necessity,”138 that is, inner necessity appears as consequences external to the action. Hegel writes: The maxim . . . which enjoins us to disregard the consequences of our actions, and the other which enjoins us to judge actions by their consequences and make the latter the yardstick of what is right and good, are in equal measure [products of the] abstract understanding.139

His conclusion is: It follows from this that the criminal stands to benefit if his action has less adverse consequences, just as the good action must accept that it may have no consequences or relatively few; and it also follows that, once the consequences of the crime have developed more fully, the crime itself is made responsible for them.140

If one separates deed and action in an abstract way then this leads one to break down (Zersplitterung) “the consequences [for penal law] minutely.”141 With this, though, the possibility of retaliation for the criminal action disappears in the judgment of common sense, and only the deed can be retaliated for: “Guilt or innocence – in relation to evils – and of evils to guilt. – Suffering in general, – punishment amongst it as well, – on the whole as a consequence of the action – or also not.”142 If the principle of retaliation only relates to the deed then it cannot perceive the subjective aspect and abstracts the criminal will, without which there is no crime, but instead only an evil. The principle of retaliation cannot be regarded, therefore, as being justification of punishment, but – as in Fichte143 – is rather justified only as a means to the end of punishment. As an external, superficial principle, the principle of retaliation is especially suited, though, to be understandable to the criminal: retaliation places an external, immediate identity between the crime and the punishment; this identity is immediate because it does not operate through the concepts of right and of law as well as through their necessary validity. Instead, this immediate identity only relates to the material consequences of the criminal’s deed for the victim and to the material consequences for 138. 139. 140. 141. 142. 143.

GPhR } 118, Remarks. GPhR } 118, Remarks. GPhR } 118, Remarks. GPhR } 118, Remarks. Elements, ed. Wood, pp. 145–6. Handwritten comments on } GPhR 118. (Translation mine.) GNR i/4 61. Fichte, Foundations of natural right, p. 228. Cf. above, Section 4.3.

hegel’s “negation of crime”

145

the criminal – in short, to the evil. Nonetheless, the principle of retaliation suggests to the criminal a regular, non-physical identity between his or her deed and punishment. Thereby, a first stage of that conscious process is reached that will lead to rehabilitation. Hegel’s rationale of punishment cannot be reduced to a mere theory of deterrence, because he justifies penal law primarily neither with specific deterrence nor with general deterrence nor with rehabilitation nor with the principle of retaliation, but instead with the necessary self-destruction of the coercion or from its self-cancellation. However, if one thinks in the usual categories, one must still observe that Hegel regards specific deterrence together with rehabilitation as the primary aim and as the primary justification of punishment, whereas general deterrence represents a secondary aim, though a nonetheless important one. Retributivism, on the contrary, delivers neither an aim nor a justification of punishment. In conclusion of this investigation of the Hegelian justification of punishment, I must observe, as in my investigation of both the Kantian and the Fichtean justifications, that if one proceeds not from the departure point of the criminal’s intention or deed, but instead from the departure point of the concept of right, which all three authors have in common, retributivism cannot then be obtained. Since the concept of right in all three authors (and correctly so) stands on a higher level than their penal law, any exponent of this concept of right will have to renounce retributivism.

PART III RETRIBUTIVIST INHUMANITY

The theory of retaliation is not only incompatible with the concept of right of Kantian philosophy (see Part i) and of German idealism (see Part ii). Contrary to Kant’s claims, it even proves itself to be a theory that respects the dignity in the criminal’s person less than the alternatives of deterrence. Most radically, Nietzsche excludes the concern for human dignity both in the citizen’s and in the criminal’s person from being a motive and a plausible aim of retributive punishment. According to Nietzsche, the idea of retaliation allows no room for the necessary internalization and reflection out of which alone bad conscience and remorse in the criminal as well as humane interaction with society could arise (see Chapter 6). In a deliberate move away from retributivism, I have made a plea in Part ii from the standpoint of a combination of specific deterrence and rehabilitation for a type of interaction that respects human dignity even with the most hardened and inhumane of criminals, that is, with those who have perpetrated crimes against humanity. Claus Roxin believes he raises a decisive objection against the “conception of specific deterrence” by accusing it of not wishing to accept its final consequences: The decisive example is at the moment the concentration camp murderer by whom some innumerable number of innocent human beings were killed due to sadistic motives. These murderers are living today mostly inconspicuously and socially integrated, so they are not in need of “rehabilitation.” Also, the danger of recidivism from which they must be deterred and from which we must be secured, does not exist with them. Should they really remain unpunished because of that?1 1. Claus Roxin, Strafrechtliche Grundlagenprobleme (Berlin: De Gruyter, 1973), p. 7.

147

148

retributivist inhumanity

Roxin is implicitly operating under the assumption that there is nobody who would answer “yes” to this question. But precisely this answer appears to me to be the right answer, as I will attempt to show by way of the punishment of crimes against humanity (see Chapter 7).

6 NIETZSCHE AND PUNISHMENT WITHOUT REMORSE

Kant does not consider that the aim of punishment is to arouse bad conscience, and thus the criminal’s remorse as well. If punishment were to be imposed in the service of such an aim, it would not occur – as Kant requires – merely because the criminal broke the law, but instead in order that a certain situation might come about. Since the arousing of bad conscience or remorse seems to be a minimalistic aim, which has moreover a clear reference to the criminal’s dignity, some interpreters of Kant develop a conception of retributive punishment directed toward this aim.1 If one proceeds under the assumption that retributive punishment especially respects the criminal’s human dignity and that it can generate – as Hegel points out (see Section 5.5) – even in the simplest mind a connection between crime and punishment, then one can expect the criminal’s remorse from retributive punishment. It is Nietzsche who best shows us that such a justification is not self-evident but instead very problematic. Nietzsche’s “polemic” writing On the genealogy of morality (1887) opposes such a moralizing conception of retaliation and of punishment. From Nietzsche’s perspective, retaliation does not stem from concern for human dignity in the criminal’s person but instead in the victim’s and other human beings’ active Schadenfreude. Also, the expiation of punishment cannot possibly be an inducement to moral conversion, but rather strengthens the criminal’s felonious will. It is punishment that is directed toward deterrence which proves to be the more humane option of penal law. 1. Cf. chapters 9 and 11 of Thomas E. Hill, Human welfare and moral worth: Kantian perspectives (Oxford: Clarendon Press, 2002).

149

150

retributivist inhumanity

In Nietzsche’s On the genealogy of morality, Essays ii 8–15 belong to the context of the second tractate whose aim is an inquiry into the genealogy of the feeling of guilt. Essay ii 8 begins with taking stock of the paragraphs regarding the genesis of promises in GdM ii 1–3 as well as about the history of the genesis of Schuld (that is, guilt or debt)2 and its acquitment in the form of making the insolvent debtor suffer: The feeling of guilt, of personal obligation, to pursue our train of inquiry again, originated, as we saw, in the oldest and most primitive personal relationship there is, in the relationship of buyer and seller, creditor and debtor: here person met property for the first time, and measured himself person against person.3

The further course of the investigation throughout these aphorisms consists of three steps. First, Nietzsche gives a description of the genesis of public punishment according to the “canon of morals relating to justice.”4 Secondly, he then delineates the genesis of public punishment in his rejection of Karl Eugen Du¨hring’s conception of justice and punishment as either ressentiment or revenge.5 Thirdly and finally, Nietzsche takes it a step further: he radically separates public punishment, whose genesis he has hitherto explained without reference to the criminal’s feeling of guilt, from this alleged feeling of guilt.6 In the following I will mainly devote myself to the first part, that is, aphorisms 8–10, because these sections depict the core of Nietzsche’s conception of punishment.

6.1. The genealogy of public punishment Out of the oldest relationship between human beings (according to Nietzsche), that is, out of the relationship between buyer and seller, arises “the habit of comparing power with power, of measuring, of calculating.”7 According to a physiological process, which was 2. In the German language, the word guilt (Schuld) is sometimes ambiguous. It can mean either a financial debt or moral guilt. Nietzsche is, of course, keenly aware of its polysemous nature. 3. GdM ii 8. Friedrich Nietzsche, On the genealogy of morality, ed. Keith Ansell-Pearson, trans. Carol Diethe (Cambridge: Cambridge University Press, 2007), p. 49. 4. GdM ii 8–10, cited here is section 8. Genealogy of morality, ed. Ansell-Pearson, p. 50. 5. GdM ii 11–13. Genealogy of morality, ed. Ansell-Pearson, pp. 52–8. 6. GdM ii 14–15. Genealogy of morality, ed. Ansell-Pearson, pp. 58–60. 7. GdM ii 8. Genealogy of morality, ed. Ansell-Pearson, p. 49.

nietzsche and punishment without remorse

151

explained by Nietzsche in The philosophers’ book,8 the habit becomes a generality. In this way, the “universalization” arises, which constitutes the “first level” of justice or its “oldest, most naı¨ve canon of morals”: “Every thing has its price: everything can be compensated for.”9 One reason why this canon of morals is “naı¨ve” is because it is still not internalized, “intellectualized” (vergeistigt) – to make use of Nietzsche’s own vocabulary. In this passage, which is based on the aforementioned principle of justice, guilt and punishment, moreover, are still investigated not in their internalized dimension, but instead only in the external dimension of a harm and of punishing the one who caused it. Not only the feeling of guilt, but also the internal guilt remains unconsidered in this explanation of the establishment of public punishment. Because this canon of morals is still not internalized, it is also naı¨ve in the sense that it openly compares “power with power” and defines itself by power relations: Justice at the first level is the good will, between those who are roughly equal, to come to terms with each other, to “come to an understanding” again by means of a settlement – and, in connection with those who are less powerful, to force them to reach a settlement amongst themselves.10

In short: justice should be comprehended merely in a positivistic sense. It is not the just exchange that is defined through the equivalence of exchanged goods, but instead the equivalence that is defined through the actual exchange of goods, which is decided, in turn, through a power relation.11 Nietzsche, regarding this point, is following Hobbes, who raises the following objection against those authors who regard fair exchanges as an arithmetical ratio: As if it were Injustice to sell dearer than we buy; or to give more to a man than he merits. The value of all things contracted for, is measured by the Appetite of the Contractors: and therefore the just value, is that which they be contented to give.12

The conclusion from this, we will have to remember later on, is that impartiality is impossible. 8. WuL i, 880. Daniel Breazeale, Philosophy and truth: selections from Nietzsche’s Notebooks of the early 1870s (Atlantic Highlands, NJ: Humanities Press, 1979), pp. 79–97. 9. GdM ii 8. Genealogy of morality, ed. Ansell-Pearson, p. 50. 10. GdM ii 8. Genealogy of morality, ed. Ansell-Pearson, p. 50. 11. Cf. Henning Ottmann, Philosophie und Politik bei Nietzsche (Berlin: De Gruyter, 1987), p. 131. 12. Part 1, chapter 15 of Thomas Hobbes, Leviathan, ed. A. R. Walter (Cambridge: Cambridge University Press, 1904), p. 102.

152

retributivist inhumanity

Just as in Hobbes’s Leviathan justice is defined by the Leviathan alone, with Nietzsche the commonwealth is limited to the tasks of protection, of peace and of ensuring social trust. According to this, the criminal is an attacker (he “also actually assaults the creditor”), a “man without peace” and “somebody who has broken his contract and his word” – therefore, someone who is not allowed to promise anymore.13 Nietzsche’s explanation of the genesis of punishment in Essay ii 9–10 of his Genealogy of morality shows us prima facie public punishment’s four periods: 1 punishment as a sort of war14 2 punishment as a “compromise with . . . anger” at those times when a “community grows in power” 3 the mitigation of punishment (“As the power and self-confidence of a community grows, its penal law becomes more lenient”) 4 impunity in times that are “not impossible to imagine” when society’s “conscious[ness] of its power” is especially developed.15 When we examine these four periods more closely then it becomes apparent not only that a public punishment exists solely in the second and third periods, but also that a legal and civil order can only exist in those two periods. In the following, I will attempt to show why this is so and to what function this fact gives rise for public punishment. Essay ii 9, in a first period, takes the logical consequence out of the determination that the criminal is an attacker, a person without peace and somebody who has broken his or her word and contract. Punished the least is the “immediate danger”; instead it is, “quite apart from this,” the word-breaking that is punished, the logical result of which ought to be that the criminal should not be allowed to promise. The criminal is namely more than merely an insolvent debtor: he or she attacks the creditor, thus not recognizing what is owed. Therefore, the question poses itself whether the creditor can reach compensation in spite of this with him or her, as is the case with the simply insolvent debtor, upon whom the creditor can inflict suffering because this gives him or her pleasure. I will return again to this question. Nietzsche prefers, however, prima facie another manner of thinking. According to Nietzsche, every member of the community is a debtor who still has to acquit the debt: he or she must contribute to 13. GdM ii 9. Genealogy of morality, ed. Ansell-Pearson, p. 50. 14. GdM ii 9. Genealogy of morality, ed. Ansell-Pearson, pp. 50–1. 15. GdM ii 10. Genealogy of morality, ed. Ansell-Pearson, p. 51.

nietzsche and punishment without remorse

153

domestic peace, to mutual protection and trust. Because the criminal cannot do this, he or she cannot be a member of the community any longer. The criminal is “eˆlend,” that is, expelled, exiled and outlawed. The same provisory conclusion was already reached, for example – as I mentioned previously (see Section 4.3) – by Fichte: the criminal would be expelled into the “desert” and everyone would be allowed to shoot him like a dangerous wild animal. The death of the criminal, and everything resulting from that exile, is no longer the concern of the commonwealth. At the same time, there are also elements of another logic to be found in the same section: 1 elements of compensation: “The community, the cheated creditor, will make you pay up as best it can”16 2 elements of mnemotechnic: the criminal “from now on, as is fair . . . is not only deprived of all these valued benefits, – he is now also reminded how important these benefits are”17 3 elements of infliction of suffering, which in Essay ii 6 counted as compensation for the unpaid debts. Nietzsche summarizes: The anger of the injured creditor, the community, makes him return to the savage and outlawed state . . . “Punishment” at this level of civilization is simply a copy, a mimus, of normal behaviour towards a hated, disarmed enemy who has been defeated . . . in fact, the rules of war and the victory celebration of vae victis! in all their mercilessness and cruelty.18

The suffering is indeed compensation for the simply insolvent debtor. If suffering is compensation for the criminal as well, then insolvent debtors and criminals are treated equally, even though the criminal’s debts fundamentally are qualitatively greater. This would point to a relative powerlessness of the commonwealth and the polity in respect to the criminal as well as to the fact that the compensation and the infliction of suffering constitute not the punishment, but instead – if at all – the mnemotechnic.19 The contrast between the “anger of . . . the community” and its reaction is obvious: first, the community “cast[s]”

16. 17. 18. 19.

GdM ii 9. Genealogy of morality, ed. Ansell-Pearson, p. 50. Also, cf. GdM ii 5f (pp. 43–6). GdM ii 9. Genealogy of morality, ed. Ansell-Pearson, p. 50. Also, cf. GdM ii 3 (pp. 41–2). GdM ii 9. Genealogy of morality, ed. Ansell-Pearson, pp. 50–1. Cf. especially GdM ii 3. Genealogy of morality, ed. Ansell-Pearson, pp. 41–2.

154

retributivist inhumanity

the criminal “away” from itself; only thereafter can “any kind of hostile act . . . be perpetrated on him.”20 It is still unclear whose anger is directed with cruelty toward the criminal. Nietzsche merely writes: “The anger of the injured creditor, the community, makes him return to the savage and outlawed state” (the logic of expulsion) and “now any kind of hostile act can be perpetrated on him” (the logic of infliction of suffering). The latter formulation is vague, just like the following and last lines of Essay ii 9. Does it concern two logical moments of public punishment or two actors of punishment? Either, the commonwealth initially expels the criminal (the first moment) and then handles him or her with cruelty like a defenseless enemy, whereupon one can imagine “all kinds of dishonour and torture” (second moment). Or, the state (the first actor) expels the criminal and the individual citizens (the second actor) perpetrate all manner of cruelties on the criminal – one might name this as a modern equivalent of lynch mob “justice.” The more plausible answer certainly lies in the middle of these. The commonwealth perpetrates these cruelties because the criminal has been excluded and because the individual citizens are angry and demand cruelties. The difference between the commonwealth as such and the sum of its members reaches its clear expression at first in the second period of punishment. There Nietzsche writes: “the wrongdoer is no longer ‘deprived of peace’ and cast out, nor can the general public vent their anger on him with the same lack of constraint.”21 The decision whether the criminal should be expelled from the community or can remain is decided by the anger of the community as such – that is, by the ruler and not by the general anger (allgemeine Zorn) of the masses. In the first period, the ruler decides for the expulsion. Whether the ruler carries out the cruelties in person or simply allows them, whereupon they would be then carried out by the crowd, is actually not the core issue: in both cases the general anger dictates the punishment – not the ruler and his or her anger. The general anger is reactive, though: it is a ressentiment that results from powerlessness.22 In this case, powerlessness results assumedly from the impossibility of 20. GdM ii 9. Genealogy of morality, ed. Ansell-Pearson, p. 50. 21. GdM ii 10. Genealogy of morality, ed. Ansell-Pearson, p. 51. 22. Cf. GdM ii 5. Genealogy of morality, ed. Ansell-Pearson, pp. 43ff.

nietzsche and punishment without remorse

155

reversing the crime’s effects. Anger is something different from a readiness to render compensation. Even if Nietzsche thoroughly regards the pleasure provided by the infliction of suffering as a compensation,23 it is possible that the angered individuals remain unsatisfied with the criminal’s suffering. For the victim, the pleasure from the suffering is only a surrogate. The fact that there is “war” between the crowd and the criminal points in this direction. Since this war does not occur between states and since the state cannot wage war with individual persons, it therefore can only be a war between the population and the criminal. A war can, however, only occur between two communities; for that reason here it concerns itself with only a “copy” and with even the “mimus” of a bellicose attitude:24 the mimus was only a farce for the Romans. In fact, in the second period, Nietzsche only talks about the avoidance of “further or more wide-spread participation [in anger] and unrest.”25 Because there is an implicit contrast to the first period in the context of this second period, one may assert that the first period is typified by a diffusion of unrest. The first period was, therefore, more a period of private justice. In the second period, punishment is no longer determined by the crowd, but instead by the commonwealth and polity as such. Only then are we dealing with a public punishment in the proper sense; in contrast to that, in conjunction with the first period Nietzsche writes “punishment” only within quotation marks. Nietzsche begins Essay ii 10 with the following words: As a community grows in power, it ceases to take the offences of the individual quite so seriously, because these do not seem to be as dangerous and destabilizing for the survival of the whole as they did earlier.26

For this reason – Nietzsche is here in agreement with Hegel – punishments can be milder. Nowadays one would essentially interpret the danger and the risk of collapse as Hegel does in } 218 of his Elements of the philosophy of right: the more stable the state, the less the risk that a crime incites further crimes and, hence, makes a stricter general deterrence necessary. Unlike with Hegel, however, with Nietzsche this 23. 24. 25. 26.

GdM ii 5ff. Genealogy of morality, ed. Ansell-Pearson, pp. 43–9. GdM ii 9. Genealogy of morality, ed. Ansell-Pearson, pp. 50–1. GdM ii 10. Genealogy of morality, ed. Ansell-Pearson, p. 51. GdM ii 10. Genealogy of morality, ed. Ansell-Pearson, p. 51.

156

retributivist inhumanity

is not what the concern is; rather, the concern is about the risk that a criminal might agitate the entire society or might bring about civil war. In which case is this a lesser risk? Perhaps it is less if the quantity and the gravity of the crimes are held as being so low or the citizens’ safety and trust are so high that the crimes are not perceived as being a direct threat by the majority of the population. This is even more probable, though, if the state possesses, in a stable manner, the real monopoly on violence, which makes it possible for the state successfully to resist the anger of its citizens. This may be the definition of a state’s “grow[ing] in power.” The state puts the criminal under its protection against the general anger, in order that the crime be totally acquittable and then truly acquitted. Unlike an enemy, with whom one remains unreconciled even after his or her defeat and death, public punishment should provide for either the complete expunction or the complete acquittability of the debt contracted by the crime. In this regard, three aspects matter. First, the state must see to it that “the wrongdoer is isolated from his deed.”27 The consequence resulting from this is that, unlike an enemy, the criminal is not to be destroyed. In this manner, the criminal cannot be reduced to his or her crime, and can later become again a member of the commonwealth. Yet one should not forget the fact that Nietzsche would in no way endorse this development. For he writes in Essay i 13 that “there is no ‘being’ behind the deed . . . ‘the doer’ is invented as an afterthought, – the doing is everything.” Were one to cite, as Michel Foucault did, a phrase from the French philosopher and politician Gabriel de Mably as being the criterion for modern penal law, in contrast to the pre-modern penal law – that says “punishment . . . should strike the soul rather than the body” – then Nietzsche more distinctly stands on the side of pre-modern penal law.28 Secondly, an equivalent has to be found for a deed isolated in this way, or, more exactly, multiple “equivalents” have to be found29 – each one probably determined by the gravity of the deed. The equivalent, therefore, can only be an illusion; (public) punishment can only be 27. GdM ii 10. Genealogy of morality, ed. Ansell-Pearson, p. 51. 28. Michel Foucault, Discipline and punish: the birth of the prison, trans. Alan Sheridan, second edition (New York: Random House, 1995), p. 16. The original quotation is taken from Gabriel de Mably, “De la le´gislation,” in œuvres comple`tes (12 vols., London, 1789), vol. 9, p. 326, and reads: “Que le chaˆtiment . . . frappe l’aˆme plutoˆt que le corps.” 29. GdM ii 10. Genealogy of morality, ed. Ansell-Pearson, pp. 51–2.

nietzsche and punishment without remorse

157

based on an illusion. Punishment is not an equivalent, but only the waiving of an equivalent. Thirdly, this should result in “a compromise with the anger of those immediately affected by the wrong-doing.” Nietzsche does not further specify the contents of this compromise. One may guess, however, that the compromise consists in inflicting suffering on the criminal and declaring him or her as being inferior to the victim. Indeed, in most of the penal theories, punishment is defined as the ¨ bel),30 not merely as some kind of answer to infliction of an evil (U the crime (a definition of punishment without a constitutive infliction of suffering is nevertheless possible; rehabilitation, for example, really goes beyond compensation, but is not thought of as being an evil). Already in Essay ii 5, Nietzsche writes that, in the case in which “the actual power of punishment, of exacting punishment, is . . . transferred to the ‘authorities,’” the victim attains “the elevated feeling of despising and maltreating someone as an ‘inferior.’”31 One should not overlook the point that the state takes no pleasure in inflicting suffering on the criminal or in his or her humiliation. The compromise can, however, partly consist in the state guaranteeing relative or increased security. The principle of acquittability (Prinzip der Abzahlbarkeit) still does not say anything about the degree of punishment. The criminal can render compensation by a mild punishment, as well as by death after manifold tortures for longer durations. The principle of acquittability only requires that every crime can be expiated or that any debt can be acquitted regardless of how serious it may be. Ultimately, therefore, reconciliation must occur along with the death of the criminal. Nietzsche addresses the question of degree in punishment for the first time in the premises of the third period of punishment: As the power and self-confidence of the community grows, its penal law becomes more lenient; if the former is weakened or endangered, harsher forms of the latter will re-emerge. The “creditor” always becomes more humane as his wealth increases; finally, the amount of his wealth determines how much injury he can sustain without suffering from it.32

30. Cf., for instance, in RL } 49e, Ak vi:331ff. Practical philosophy, ed. Gregor, pp. 472–5. Also, see above, Chapters 1–3. 31. GdM ii 5. Genealogy of morality, ed. Ansell-Pearson, p. 45. 32. GdM ii 10. Genealogy of morality, ed. Ansell-Pearson, pp. 51–2.

158

retributivist inhumanity

Nietzsche seems at this point to introduce, astonishingly, both another principle of penal law and another psychological basis of it. The principle of penal law no longer appears to be a compensation in proportion with the power relation, but deterrence instead. Likewise, the psychological motivation of the ruler no longer appears to be the will to power; instead, it appears merely to be security, that is, the merely reactive will to self-preservation. This third period comes to a climax in a fourth period: It is not impossible to imagine society so conscious of its power that it could allow itself the noblest luxury available to it, – that of letting its malefactors go unpunished. “What do I care about my parasites”, it could say, “let them live and flourish: I am strong enough for all that!”33

At this stage even deterrence is dispensed with: the “parasites” are allowed to weaken the community. The corresponding psychological premise is no longer the will to power, nor is it even the search for selfpreservation and security, but instead it is only an indifference to the world or the dissipation of luxury. The alternation in the psychological premises conforms to Nietzsche’s conception of the biological process of deployment and exhaustion of energies. In Nietzsche’s thinking, the active energies do not remain constant, but instead attempt primarily to assert themselves: they strive for power. However, at one point they will exhaust themselves and then they will only tend toward merely reactive selfpreservation or even toward disappearance – toward the nihil of nihilism. The concluding judgment of Essay ii 11 can be applied to this last stage of biological development: A system of law conceived as sovereign and general, not as a means for use in the fight between units of power but as means against fighting in general . . . this would be a principle hostile to life, an attempt to assassinate the future of man, a sign of fatigue and a secret path to nothingness.34

Not only in the contemporary debate over penal law does impunity hardly count as a serious option.35 Also, those authors who are not supporters of maximizing general deterrence being an aim of 33. GdM ii 10. Genealogy of morality, ed. Ansell-Pearson, p. 51. 34. GdM ii 11. Genealogy of morality, ed. Ansell-Pearson, p. 54. 35. For an exception, see Hermann Bianchi, “Abolition: assensus and sanctuary,” in Alexander R. Duff and David Garland (eds.), A reader on punishment (Oxford: Oxford University Press, 1994), pp. 336–51.

nietzsche and punishment without remorse

159

punishment share the view that total impunity would be a weighty incentive for crime; so that at the very least, owing to this reason of minimal general deterrence, no commonwealth can dispense with the existence of punishment without challenging the existence of the community itself. The true power of a state can never be so great that it can dispense with every kind of punishment. In situations in which a state enjoys a growing power but does not wish completely to abdicate its security, the degree of punishment may decrease; at most, impunity remains the ever unreachable horizon of this mitigation: impunity points to a utopia. Nietzsche writes, in fact, not that “it is not impossible to imagine society so increased in power,” but instead that “it is not impossible to imagine society so conscious of its power” (emphasis mine). This power consciousness relates to no realistic perception; rather, it should be understood as being a symptom of a biological or psychological condition – a condition of exhaustion. No acquittal of debt takes place in the period of impunity; in this respect, impunity is outside of the sphere of justice defined by the principle of universal acquittability and comparability. The lack of comparison and equivalence fulfills the definition of “luxury.” A luxury is something to the price of which one pays no heed. In the period in which the criminal was fought against like an enemy, however, there likewise existed no justice because there was at that time a state of war and the criminal was completely delivered to the anger of the people: there was no system of acquittability in general. According to Nietzsche, the deeper and the lower “the position of the creditor in the social scale,” the more the creditor feels a lust to inflict suffering as payback, because “through punishment of the debtor, the creditor takes part in the rights of the masters.”36 The victims of the criminal as well as the other simple citizens sharing the general anger enjoy this “right of the masters.” The rulers who hold themselves to be especially powerful and decide in favor of absolute impunity – that is, universalized pardon – also enjoy “this prerogative of the most powerful man”; however, he or she does this only on account of an illusion of power. Both the simple, angry citizen and the indifferent, allpardoning ruler, on the one hand, are powerless, but, on the other hand, they succumb to the illusion of power.37 In contrast, true power consists in “forc[ing]” “those who are less powerful [in this case, the 36. GdM ii 5. Genealogy of morality, ed. Ansell-Pearson, pp. 44–5. 37. GdM ii 10. Genealogy of morality, ed. Ansell-Pearson, pp. 51–2.

160

retributivist inhumanity

‘less powerful’ are both the criminal and his or her victim] . . . to reach a settlement amongst themselves.”38 In fact, only in the second and third periods can a compromise be reached between the interest of the criminal (impunity) and the interest of the victim (a bellicose, informal cruelty).

6.2. Nietzsche’s appraisal of justice The conclusion – nearly a lesson! – that Nietzsche delivers at the end of his explanation of punishment seems at first glance to raise a direct objection to my version. He writes: Justice, which began by saying “Everything can be paid off, everything must be paid off”, ends by turning a blind eye and letting off those unable to pay, – it ends, like every good thing on earth, by [canceling itself]. The [self-cancellation] of justice: we know what a nice name it gives itself – mercy.39

With these lines of text, Nietzsche seems to see impunity as being the logical conclusion of justice or acquittability, and therefore not as being a deviation from strict justice. For Nietzsche speaks expressly not of sublimation, but of the “[self-cancellation] of justice” (emphasis mine). One would rather expect that mercy would be designated as being an extrinsic cancellation of justice, being beyond justice, and thus as it is according to Nietzsche himself, “being beyond the law ( Jenseits des Rechts).”40 Nietzsche’s charge against justice is that it seeks acquittal of debt while nevertheless allowing those unable to pay to escape punishment. There are at least two ways to understand this. Either, one sees the turning of a blind eye to those who are insolvent pertaining merely to the fourth period, that is, to absolute impunity. But then the charge of self-cancellation is incomprehensible. Or, one understands it in such a way that the turning of a blind eye indeed completely unfolds in the fourth period, though it was already – therefore, from the outset – present in the second and third periods. In this respect, even public 38. GdM ii 8. Genealogy of morality, ed. Ansell-Pearson, p. 50. 39. GdM ii 10. Genealogy of morality, ed. Ansell-Pearson, pp. 51–2. Translator’s note: in the translation, Aufhebung is rendered as “sublimation,” but in the Nietzschean sense, as in the Hegelian context, rendering it as “suppression,” “cancellation” or “sublation” might prevent any confusion. 40. GdM ii 10. Genealogy of morality, ed. Ansell-Pearson, p. 52.

nietzsche and punishment without remorse

161

punishment as acquittal of debt means the turning of a blind eye; in short, it means not truly acquitting the debt. In this context, one ought to recall two points: 1 The criminal is a promise breaker, not only a malefactor, which is why he or she loses “all rights and safeguards” and, furthermore, “all mercy as well.”41 2 Nietzsche stresses that “you live in a community, you enjoy the benefits of a community (oh, what benefits! sometimes we underestimate them today).”42 One can gather from both of these points, first, that the criminal’s debt is qualitatively especially large and, secondly, that the creditor’s assets are likewise qualitatively especially large. Now, Nietzsche never himself adopts the principle of justice that “everything can be compensated for [acquitted].”43 On the contrary, nobler things cannot be acquitted; and the institution of the state and its establishment wholly rank among these nobler things, as is demonstrated, for example, by Nietzsche’s interjection “oh, what benefits!” We are, therefore, allowed to proceed under the assumption that for Nietzsche the acquittability is not (at least not always) at hand. In the second period, the acquittal, which is then public punishment, thus means that the debt remains unpaid – at least partly: the debtor gets off relatively easily. The following explanation of this paradox seems obvious: the justice that leads to this renouncement of a real equivalent stems from the ressentiment of the victim. Nietzsche constructs a contrasting alternative theory to this theory held by Du¨hring. The latter regards the connection of crime with ressentiment and with revenge in the following manner: In our moral deliberations, we have already regarded each action which originally, inimically hurts as the object of a necessary reaction. This reaction expresses itself first internally, in a counter-sensation (Ru¨ckempfindung) that we can also name “ressentiment” and “need for retribution” or, precisely, “revenge.”44

41. 42. 43. 44.

GdM ii 9. Genealogy of morality, ed. Ansell-Pearson, p. 51. GdM ii 9. Genealogy of morality, ed. Ansell-Pearson, p. 50. GdM ii 8. Genealogy of morality, ed. Ansell-Pearson, p. 50. Eugen Du¨hring, Cursus der Philosophie als streng wissenschaftlicher Weltanschauung und Lebensgestaltung (Leipzig: O. R. Reisland, 1894–5), p. 224.

162

retributivist inhumanity

Now, Du¨hring understands justice to mean retribution (Vergeltung). Hence justice rests on the need for revenge. Why, however, ought this need to remain a ressentiment or a counter-sensation instead of leading to the action of revenge? Du¨hring explains the rejection of the “vendetta” by stating that it “yields a private war that is instigated and continued over and over again.”45 In order to avoid a civil war, one establishes a “system of conflict resolution in virtue of which one seeks appeasement of revenge by means of atonement and compensation.”46 In return for forgoing revenge, the victim receives, in addition to public peace, the guarantee that evils will be imposed upon the criminal and that he or she will feel guilt and remorse: But the high rates with which one allows one’s own bodily injury and the killing of dependants to be bought off after the fact should not lead to passing over the fact that the readiness to make an earnest material sacrifice could also vouch for the dignity of the changed will and, hence, for true remorse and for a peaceful disposition. The need for revenge does not disappear, however, only by subduing and harming the injurer, but it is also balanced out if the malefactor himself assumes his castigation honestly by humbling himself by the true confession of guilt and by the accomplishment of an apology and a punishment.47

Nietzsche refutes this explanation in two steps. First, he shows that justice and public punishment have different origins. Secondly, according to Nietzsche, the punishment does not give rise to either any feeling of guilt or any bad conscience in the criminal, but rather the opposite. Immediately, at this point, a new paradox arises. Should the punishment in accord with justice originate not from “emotional reactions” but instead from “active emotions,”48 then how can the renunciation of a real equivalent be explained on the basis of these “active emotions”? The answer lies in the aforementioned distinction between two actors, that is, between the population and those who wield power. In general, Nietzsche always responds to inquiries into the nature or into the aim of an institution with a counter-inquiry into the power relation out of which the institution arises. The “attempt” of the person in power “to localize the matter and head off further or

45. 46. 47. 48.

Du¨hring, Cursus der Philosophie, p. 225. Du¨hring, Cursus der Philosophie, p. 225. Du¨hring, Cursus der Philosophie, pp. 225–6. GdM ii 11. Genealogy of morality, ed. Ansell-Pearson, p. 52.

nietzsche and punishment without remorse

163

more wide-spread participation and unrest” is implemented by the punishment.49 This striving for civil peace, which Du¨hring also saw as the ground for public punishment as compromise and renunciation of retribution, is not a final aim in Nietzsche’s conception – unlike in Du¨hring’s conception. In Nietzsche’s conception, the rule of law and civil peace were not established for the sake of the safety of the individual citizens, but instead as a means for ensuring power for those in power. Nietzsche emphasizes this distinction: A system of law conceived as sovereign and general, not as a means for use in the fight between units of power but as a means against fighting in general . . . [requiring] that every will should regard every other will as its equal, this would be a principle hostile to life . . . a sign of fatigue and a secret path to nothingness.50

Therefore, the state’s efforts at establishing a civil peace – among other things, by means of public punishment – should be regarded as a weapon. The prevalence of power can be internal as well as external. The internal prevalence of power consists in the situation in which individual human beings – for instance, the victims – no longer inflict the “punishment,” but instead the state alone; thereby, the state enforces what has generally been later termed the “monopoly on violence.” The external prevalence of power consists in the fact that civil peace represents a considerable advantage in struggles against other states. Nietzsche’s praise of war as a principle that is favorable to life finds repeated expression in his works.51 The result of the evolution from private revenge to public punishment first appears negative, though, for the population in general and for the victim especially. This is because “[gruesome] punishment . . . has such strong festive aspects!”; and “to see somebody suffer is nice, to make somebody suffer even nicer – that is a hard proposition, but an ancient, powerful, human-all-too-human proposition.”52 At this point, the people lose the ability “to make someone suffer” (Leidenmachen) and receive as consolation only the ability “to see somebody suffer” (Leiden-sehn) in which the suffering constantly becomes milder and more discrete the more powerful the state becomes. The population 49. GdM ii 10. Genealogy of morality, ed. Ansell-Pearson, p. 51. 50. GdM ii 11. Genealogy of morality, ed. Ansell-Pearson, p. 54. 51. For instance, in Human, all too human (i } 224); Thus spoke Zarathustra (iv 312); and Twilight of the idols (} 38). 52. GdM ii 6. Genealogy of morality, ed. Ansell-Pearson, p. 46.

164

retributivist inhumanity

experiences a clear frustration because “the wrong-doer is [instead] carefully shielded by the community from this anger, especially from that of the immediate injured party, and given protection.”53 Previously, private punishment meant “tak[ing] part in the rights of the masters,”54 which disappears at this point. The fiction of equivalency or of justice is introduced as compensation for this frustration, as a “compromise” or as a “restriction . . . of the true will of life”; however, first and foremost, it is the fiction of the bad conscience or of the remorse of the criminal that provides consolation to ressentiment. This remorse, which Du¨hring holds both to be something real and to be the aim of punishment, proves to be a mere means to the prevalence of the (state’s) power for Nietzsche. Nietzsche stresses that change of perspective that plays out between the philosophy of right and the philosophy of penal law, on the one hand, and his own conception, on the other hand. According to Nietzsche, punishment has no “purpose” (Zweck); instead, the concept of punishment represents “a whole synthesis of ‘meanings.’”55 Punishment refers neither to the purpose of the population (that is losing its ability “to make someone suffer”) nor to the purpose of power (that is limiting itself), nor does it effect any remorse in the criminal, as we shall later see. Not long before Nietzsche’s On the genealogy of morality (1887) appeared, two of the most significant penal law theorists placed the purpose of punishment at the center of penal law: Rudolf von Jhering and Franz von Liszt.56 All of the great outlines of theory in the tradition, from Plato’s Gorgias to Hegel’s Elements of the philosophy of right, through Beccaria’s On crimes and punishments and Bentham’s Theory of legislation, attempted to justify the institution of punishment with an end. In this respect, Nietzsche represents a radical break in the theory of penal law; only a narrow bridge remains between Nietzsche and the tradition: both approve of the institution of punishment. Nietzsche’s conception, however, rejects every dimension of justice and any justification of punishment as being mere illusions. Admittedly one may tend toward the illusion that Nietzsche praises justice when he writes: “If it actually happens that the just man 53. 54. 55. 56.

GdM ii 10. Genealogy of morality, ed. Ansell-Pearson, p. 51. GdM ii 5. Genealogy of morality, ed. Ansell-Pearson, p. 45. GdM ii 13. Genealogy of morality, ed. Ansell-Pearson, p. 57. Rudolf von Jhering, Law as a means to an end, ed. Joseph H. Drake, trans. Isaac Husik (Union, NJ: Lawbrook Exchange, 1999). Franz von Liszt, Der Zweckgedanke im Strafrecht, (Berlin, 1883).

nietzsche and punishment without remorse

165

remains just even towards someone who has wronged him . . . well, that is a piece of perfection, the highest form of mastery to be had on earth.”57 In so doing, however, three basic aspects should not be overlooked. First, Nietzsche is referring in no case to justice, but instead to “the just man.” Secondly, the sentence is formulated hypothetically (“If it actually happens . . .”), and its condition is not fulfilled since it is “something which we would be wise not to expect and should certainly find difficult to believe.”58 Furthermore, Nietzsche never takes an idealistic position such as that: Nietzsche categorically disapproves of any idea whatsoever. Thirdly, the just character of a human being who stands the closest to justice does not stem from his or her appreciation for justice. The human who stands close to justice is only “a hundred paces nearer to justice than the man who reacts.”59 Only in so far as he or she is “the active, aggressive, overreaching man [or woman]” is this person close to justice. Therefore, he or she is primarily powerful and aggressive, and only secondarily just. We have already given the explanation for this. For Nietzsche, justice is a means to power. A justice that is an autonomous end or that is a value simply does not exist in Nietzsche’s thought.

6.3. The feeling of guilt arises not in the convict – but instead in the normal citizen We have seen that Nietzsche holds remorse to be a mere fiction. According to Nietzsche, either the convicts become later recidivists and only more skilled in crime (the punishment “strengthens the power to resist”; “we must certainly seek the actual effect of punishment primarily in the sharpening of intelligence”),60 or their wills are broken and they become dulled (“if it does happen that a man’s vigour is broken, resulting in his wretched prostration and self-abasement”).61 A third possibility is as good as excluded: “the real pang of conscience, precisely amongst criminals and convicts, is something extremely rare.”62 In order to understand this, it must briefly be mentioned where and in whom the feeling of guilt eventually arises. Nietzsche pointedly 57. 58. 59. 60. 61. 62.

GdM ii GdM ii GdM ii GdM ii GdM ii GdM ii

11. Genealogy of morality, ed. Ansell-Pearson, pp. 52–3. 11. Genealogy of morality, ed. Ansell-Pearson, p. 53. 11. Genealogy of morality, ed. Ansell-Pearson, p. 53. 14 and 15. Genealogy of morality, ed. Ansell-Pearson, pp. 59, 60. 14. Genealogy of morality, ed. Ansell-Pearson, p. 59. 14. Genealogy of morality, ed. Ansell-Pearson, p. 58.

166

retributivist inhumanity

summarizes the explanation when he speaks, in Essay ii 22, of the man or woman who has been “incarcerated in the ‘state’ to be tamed, and has [invented] bad conscience so that he can hurt himself, after the more natural outlet of this wish to hurt has been blocked.”63 We have seen how the state has monopolized the infliction of suffering by public punishment. Now the state is exercising a monopoly of power against the criminal as well. The criminal was indeed not a creditor, but instead a debtor and “someone who has broken his . . . word.” He or she therefore cannot appeal to justice against the punishment. Nevertheless, he or she is still affected like every other human being by the will to power and feels the infliction of suffering as a joy. And the convict is just as well locked up as the normal citizens, if not even more so! Then why can the convict, unlike the normal citizens, develop a feeling of guilt? Nietzsche provides the following explanation: Nor must we underestimate the degree to which the mere sight of the judicial executive procedures inhibits the criminal himself from experiencing his act, his mode of conduct, as reprehensible as such: because he sees the same kind of action practised in the service of justice and given approval, practised with a good conscience.64

At first glance, this explanation seems puzzling because it appears to confuse the question of guilt and punishment with the question of discrimination. The convict is deliberately discriminated from the rest of the citizens because he or she has committed a crime. Normal citizens are handled differently. This normative explanation is irrelevant in Nietzsche’s conception, because normative elements do not provide any explanation; instead, they themselves must be descriptively explained. Yet Nietzsche’s explanation contains two important elements. First, unlike the normal citizens, the convict sees his or her “power to resist” being strengthened most of the time, because of being more heavily oppressed than the normal citizens; unlike them, the criminal must struggle against the power of the state so that he or she has no necessity to require the internalization of what Nietzsche terms the Leiden-lassen, literally “mak[ing] someone suffer.” Secondly, unlike normal citizens – and especially unlike the victims, who are required 63. The translation renders “erfunden” as “discovered,” whereas it should be rendered as “invented.” GdM ii 22. Genealogy of morality, ed. Ansell-Pearson, p. 68. 64. GdM ii 14. Genealogy of morality, ed. Ansell-Pearson, p. 59.

nietzsche and punishment without remorse

167

to go without a real equivalent (the consoling equivalent is only a “compromise”) – the criminal experiences something that can be held to be an equivalent by him or her. The criminal is handled in the same way he or she acted; with the punishment, “debts” are cancelled. The suffering of the victim is not seen by the convict, since the state has separated the convict from the victim as well as from the revenge of the victim. Thus, there is no room left over for “debt.” The observations of such penal law specialists as George P. Fletcher, according to whom “high recidivism rates suggest that in contrast to the ideal of rehabilitation, prisons serve more to harden the antisocial inclinations of criminals than to convert them to a law-abiding way of life,”65 admittedly seem to corroborate Nietzsche’s assessment of imprisonment. Yet even if the convict is only extremely seldom reformed, then, thanks to the punishment, at least the will of a number of the criminals will be broken. At least some convicts will therefore not be recidivists, that is, they will be rendered harmless or rehabilitated thanks to the punishment after their term has been served – which is the goal of specific deterrence.

6.4. Does Nietzsche provide a theory of punishment? What does Nietzsche himself then suggest? The acquittance (Abzahlung) rests on an illusion. I also do not believe that Nietzsche’s answer would be expulsion from society. Neither the right of the masters nor pardon can be an option. Although Nietzsche does not declare himself to be against cruelty and against the joy that comes from inflicting suffering, anger as a counter-sensation, as ressentiment, is also not recommendable in Nietzsche’s perspective. One alternative remains: “mercilessness and cruelty” against the “hated, disarmed enemy who has been defeated,”66 but indeed without anger: rather with joy, without reactive feeling, without ressentiment against the person who has broken his or her word. The criminal simply attempted to prevail by his or her breach of promise. Nietzsche has no moral objections against that, because only the power relation can be considered as being a norm. The criminal, whose crime is joyfully sanctioned by a gruesome death, simply has had a stroke of luck. Only 65. George P. Fletcher, Basic concepts of criminal law (Oxford: Oxford University Press, 1998), p. 31. 66. GdM ii 9. Genealogy of morality, ed. Ansell-Pearson, p. 51.

168

retributivist inhumanity

the internalization of ressentiment, that is, the internalization of the anger of the powerless human being toward the criminal, is harmful to the whole process of the prevailing of power, meaning symptomatic for the triumph of powerlessness. In fact, Nietzsche speaks of someone who has broken his or her word not as a base, abject human being, but instead like a soldier in a war, a metaphor that most of the time carries positive connotations in Nietzsche’s works and refers to something noble. The criminal rejects the system of equivalence and as a result considers him- or herself as being incomparable. If the criminal fails, one may indeed feel joy over the defeat, though one may not judge him or her morally. Nietzsche had already mentioned this option more extensively in the Second Essay: Not so long ago it was unthinkable to hold a royal wedding or full-scale festival for the people without executions, tortures, or perhaps an auto-da-fe´ . . . To see somebody suffer is nice, to make somebody suffer even nicer – that is a hard proposition, but an ancient, powerful, human-all-to-human proposition . . . No cruelty, no feast: that is what the oldest and longest period in human history teaches us – and punishment, too, has such very strong festive aspects!67

It only begs the question whether one may even speak of a public punishment under these conditions. Provided that the criminal is punished not by private citizens – for example, in the form of a lynching, which, incidentally, displays more anger than joy – but instead by the state, one could, however, finally speak of a public punishment without at the same time approving of Nietzsche’s view. What does Nietzsche’s amoral view show us about public punishment as a feast of cruelty? Nietzsche’s argument seems bipartite and two-sided. On the one hand, he has recourse to the classical justification of punishability: the criminal did not comply with the reciprocity that is fundamental to the commonwealth, and he or she attacked the commonwealth; thus, the criminal is no longer able to be a normal member of the commonwealth. On the other hand, in Nietzsche’s conception, the further classical premise is lacking that the criminal being a human being can acquit his or her crime, that is, again become a member of society after the expiation of the crime (not only the theories of rehabilitation but also the theories of 67. GdM ii 6. Genealogy of morality, ed. Ansell-Pearson, p. 46.

nietzsche and punishment without remorse

169

deterrence and the various forms of retributivism allow – except in the highly problematic and controversial case of the death penalty – the possibility of rehabilitation, as previously stated). In its place, power and cruelty reign. My conclusion is as follows: Nietzsche’s idea of the most desirable punishment would be a mixed form of justice and cruelty. Such a mixed theory would admittedly be implausible as such. In reality, the matter looks slightly different. Nietzsche’s primary premise is not the kind of reciprocity that is a form of non-positive, natural law justice, but instead a reciprocity that is one of the forms of the “requirements of social life” that it constitutes.68 It is not the severity of the punishment, according to Nietzsche, that initially demonstrates the necessary and powerful “aid to mnemonics”;69 instead, from the outset, it is the exclusion from society or the inacquittability of the crime. Nietzsche cites as one of the means of mnemonics only the death penalties with torture: stoning, breaking on the wheel, impaling, drawing and quartering or trampling by horses, boiling in oil or wine, flaying, cutting flesh from the body, the smearing of honey on a criminal under the burning sun.70 These punishments, being death penalties, all mean expulsion from the commonwealth. The death penalty is exclusion and torture is cruelty; briefly stated, pain is “the most powerful aid to mnemonics”71 that teaches strict obedience. Nietzsche’s conception of punishment as being a mnemonic is concerned merely with the prevailing of power of the commonwealth and leaves justice and the rule of law to the respective power relations in whatever forms they may finally find themselves. A genealogy of punishment is what Nietzsche’s explanation is; by the same token, it is also a resolute recommendation for punishment. However, there is surely no inspiration for a justice oriented debate about punishment to be found in this recommendation. Apart from that, in our age it would potentially not be an efficient mnemonic aid to implement Nietzsche’s suggestion; instead, it would rather provoke – hopefully – a great deal of resistance as well as major social unrest. If our legal systems, with all their structural and functional problems, are more stable than Nietzsche’s conception then his single core argument has 68. 69. 70. 71.

GdM ii 3. Genealogy of morality, ed. Ansell-Pearson, p. 42. GdM ii 3. Genealogy of morality, ed. Ansell-Pearson, p. 41. Cf. GdM ii 3. Genealogy of morality, ed. Ansell-Pearson, p. 42. GdM ii 3. Genealogy of morality, ed. Ansell-Pearson, p. 41.

170

retributivist inhumanity

miscarried. Thereby our age would differentiate itself from Nietzsche’s conception just as those two situations that Voltaire somewhat sarcastically described in respect to retaliative punishment were differentiated from one another: All these laws of blood-soaked politics have their moment in history, and one sees clearly that they are not true laws since they are short-lived. They are akin to that need to eat other people which you sometimes have in a situation of extreme famine: you eat them no more once you have bread.72 72. Chapter 14 of Voltaire, “Commentary on the book On crimes and punishments, by a provincial lawyer,” in Voltaire, Political writings, ed. and trans. David Williams (Cambridge: Cambridge University Press, 1994), pp. 244–79 (p. 264).

7 WHAT IS THE PURPOSE OF PUNISHING CRIMES AGAINST HUMANITY?

7.1. The characteristics of crimes against humanity The concept of crimes against humanity fulfills two roles. On the one hand, it is intended to remedy the loopholes in the international legal system. On the other hand, it constitutes a new kind of crime, that is, a kind of crime that entails characteristics absent from any other sort of crime. The first loophole that the concept of crimes against humanity was intended to close in the positive international legal system was the one that arose as a result of the impossibility of prosecuting not only a crime committed against the combatants and against the civilian population of the enemies, but also those committed against their own civilian populations. In this way, the concept of crimes against humanity extends the concept of war crimes to include new categories of victims. The second loophole that was supposed to be closed was the impossibility of applying this extension of the humanitarian international law to crimes already committed. The Nuremberg and Tokyo tribunals, which were charged with prosecuting war crimes and crimes against humanity committed by Germans and Japanese, respectively, during the Second World War, prosecuted crimes committed before their own institution, that is, before the concept of crimes against humanity arose prior to their own founding, thus injuring the basic legal precept of nulla poena sine lege in favor of the enforcement of a minimal natural law legal framework in international matters. Interestingly enough, the International Criminal Court (hereafter ICC) does not contain this retroactivity in its Statute, though the national penal

171

172

retributivist inhumanity

legislations contain it.1 A third loophole is only now being gradually closed, whereby the ICC was given a jurisdiction encumbered neither by a statute of limitations nor by the place where the crime has been committed, at least in those states where the statute calling it into existence was ratified. This jurisdiction is subsidiary, however, since the ICC prosecutes exclusively either those cases that the individual national judiciary authorities did not try, or those in which a clearly inappropriate decision was made. The definition of crimes against humanity seems totally able to close these loopholes, at the very least for the following three reasons. First, an international criminal tribunal such as the ICC, which is by no means a supreme court but a court that has a narrower jurisdiction, and which decides cases without any possibility of appeal, that is, as the last resort, cannot be responsible for hearing those cases related to the highest crimes that have already been heard and not grossly unfairly handled by any domestic court. Not only is the definition and the treatment of the gravest crimes (in most countries the gravest form of homicide) very different from one country to another (take, for example, American first-degree murder, German Mord and French meurtre avec pre´me´ditation, which are quite different concepts and are handled in very different ways); even if they were consistent across national borders, the numbers of cases to be handled, even if only in appeal, would widely exceed the capacity of a single court. Secondly, because crimes against humanity are considered according to their definition as “systematic attacks,” which implies that they are serial crimes, they are the most significant crimes from a quantitative point of view. Thirdly, more often than not such “systematic attacks” occur either in a failed state or in a criminal state, that is, for the former, in a state that cannot hinder such attacks, or for the latter, in a state that commits such crimes itself. As a result of that, there is a special need for the subsidiary intervention of an international institution. In short, the concept of crimes against humanity is useful as a criterion to allow only the most severe and the most unpunished crimes to be tried before such international tribunals. Were it only for these very reasons, a new category of crime would be required and it has been established in the criminal justice systems of each nation as the mere 1. Retroactivity is still the main reproach addressed to the Nuremberg Tribunal; see Larry May, Crimes against humanity: a normative account (Cambridge: Cambridge University Press, 2006), p. 211.

punishing crimes against humanity?

173

consequence of its existence in positive international law in the form of the verdicts pronounced in Nuremberg and Tokyo. The concept of crimes against humanity, being a pragmatic criterion, appears to be a necessary first step on the path toward harmonization of criminal prosecution made possible by the establishment of a global standard and of an appropriate priority rule for criminal prosecutions. Although these pragmatic reasons seem sufficient to justify the existence of crimes against humanity – a specific sort of crime in international law – I suspect, however, that some of the characteristics of this new kind of crime, as well as the justification given for its punishment, are most often misinterpreted or are inappropriate. These aforementioned characteristics are as follows: 1. According to the Rome Statute of the International Criminal Court, crimes against humanity are “widespread and systematic attacks”2 committed “pursuant to or in furtherance of a state or organizational policy to commit such attack.”3 An individual criminal destroying an entire village for personal reasons does not commit with that act any crime against humanity. A crime against humanity is not directly a crime against the institutions, as is high treason, abuse of power, bribery, etc. Yet crimes against humanity are not merely violations of the rule of law, but instead an attack against the political institutions themselves; among other reasons, because they clearly intend to exclude entire groups from obtaining political representation and from the process of political deliberation, as well as possibly to deprive them of access to due process before impartial courts. I do not agree with Antoine Garapon’s qualification of this as the violation of the “droit d’avoir des droits” (“the right to have some rights”),4 because any victim of an ordinary murder definitively loses this “right to have rights,” although he or she is not necessarily a victim of a crime against humanity. Crimes against humanity are crimes against basic political institutions. 2. Crimes against humanity abandon the usual sanction structure found in penal law between the individual crimes, as well as between the individual punishments. Unlike the domestic criminal law of each 2. Art. 7.1 of the Rome Statute of the International Criminal Court (hereafter, Rome Statute): www.icc-cpi.int/NR/rdonlyres/EA9AEFF7-5752-4F84-BE94-0A655EB30E16/ 0/Rome_Statute_English.pdf. 3. Rome Statute, Art. 7. 4. Antoine Garapon, Des crimes qu’on ne peut ni punir ni pardonner (Paris: Editions Odile Jacob, 2002), p. 134.

174

retributivist inhumanity

country, such varied crimes as homicide, kidnapping, torture and rape are subsumed under Article 7 of the Rome Statute. According to Article 26.3, committing, soliciting or facilitating the crime, or even providing the means for its commission, are all considered to be committing the crime against humanity. The “applicable penalties” as defined by Article 77 are the same for all of these deeds, whereas the amount of punishment greatly varies in the domestic legal contexts, according to which of these deeds was committed. Neither of the two options for assigning the amount of punishment (either imprisonment for a certain number of years or life imprisonment) is in any way specifically related to one of the categories of crimes that are treated separately in the basic penal codes of the individual nations. We can see one of the consequences in the case of a person A committing a crime incited by a person B. The nature of the incitement by B, however, does not amount to creating a state of duress for A. Ordinary penal law shall not always punish incitement to a crime; in such cases it punishes only incitement to the gravest crimes, and then it punishes them mostly less than the commission of the crime itself. In the Rome Statute there is no mention of a mitigation of punishment for the incitement to a crime compared with commission of a crime. On the contrary, the Rome Statute does not contain anything that would prohibit a reverse sanction structure (such as punishing someone less for a crime depending on the level of participation of that person, which goes against the usual sanction structure in most countries). In fact, a political leader who has commanded large-scale massacres without having ever killed anybody himself may be subject to a severer punishment than a soldier who has carried out his orders and is also declared by the court not to have acted under duress. 3. Crimes against humanity are not subject to the statute of limitations, whereas in most countries even the highest crimes are subject to the statute of limitations. Even in countries such as Germany, in which murder is not subjected to any statute of limitation, most of the crimes (rape, kidnapping, etc.) included in Article 7 of the Rome Statute of the ICC are subject to the statute of limitations when they are not committed in the context of a crime against humanity. First and foremost, at the time Germany decided to exclude murder from the statute of limitations, it did so precisely in order to ensure that the prosecution of crimes against humanity that were not yet part of the penal code could still be made possible; were this not the

punishing crimes against humanity?

175

intention at the time, murder would most likely have been subject to the statute of limitations. In Austria, there is no statute of limitations for murder, but there is a dramatic reduction of the amount of punishment should the case be brought twenty years after the execution of the crime.5 4. The exclusion of crimes against humanity from the statute of limitations is often considered as being morally justifiable by the monstrousness that all crimes against humanity pose. This monstrousness is usually considered as implying two consequences. First, it implies that those who commit crimes against humanity are supposed to be monsters who deserve extraordinary punishment. Secondly, it implies that the monstrousness that crimes against humanity represent should remain unforgettable, because guilt is allegedly incommensurable. Thus, it is held that punishment should have an expressive function, that is, that it should represent an expression of the inextinguishable, most deep-seated abhorrence of all humankind. In this view, no punishment can fully fit the crime, when the crime is a crime against humanity, yet an extraordinary punishment can at least be regarded as an expression of the inextinguishable guilt of the criminals. The monstrousness of some crimes against humanity, as well as the impossibility of finding any commensurable punishment to fit the gravity of the crime, is often employed as the justification for the existence of the legal category of crimes against humanity, and it implies not only the exclusion of the statute of limitations (as mentioned in the first point), but also the elements mentioned in the first and second points, though interpreting them in a peculiar way. Indeed, the absence of the usual precisely codified sanction structure (the second point) is explained in this context as the manifestation of the fact that the crime is so monstrous that it lies beyond the reach of the sanctions in ordinary penal law and of any quantitative scale of guilt and amount of punishment. But what about point one, that is, about the political significance of crimes against humanity as it is treated by the argument that the monstrousness of crimes warrants sanctions above those against conventional crimes? Those who explain the specificity of crimes against humanity as resulting from their monstrousness do not emphasize the endangerment of the public institutions brought about by the commission of the crime; 5. Austrian Penal Code (Strafgesetzbuch), Sec. 6, Art. 57(1). See May, Crimes against humanity, p. 216.

176

retributivist inhumanity

rather, they emphasize that committing the crime is a moral evil, that is, the alleged inability of the criminal ever to belong again to the political community under the rule of law.

7.2. An alternative conception of crimes against humanity In most conceptions of crimes against humanity, the preceding four characteristics create a complex that I would like to challenge. I believe the first characteristic to be true, as well as the second, but I would justify the second in a completely different way, and I will entirely reject the third and fourth ones. Therefore, I will begin my inquiry into all four characteristics with the first one. Crimes against humanity are crimes against the basic political institutions of states governed by the rule of law. I would like to introduce a distinction that is essential for determining the appropriate amount of punishment for this crime: I shall distinguish between (A) the time at which it is being committed and (B) the time after it has been committed. (A) We find in the tradition of natural law two rights that are intended to provide a remedy to a situation in which a crime against the very institutions of a legitimate political community is being committed, that is, tyrannicide and the right of resistance. By definition, this situation is one in which the rule of law is not enforced, and thus no action before a court is possible. Therefore, tyrannicide and right of resistance cannot be codified by positive law. The German Basic Law (Grundgesetz) declares: “All Germans shall have the right to resist any person seeking to abolish this constitutional order, if no other remedy is available.”6 Yet, for the reason just mentioned, the law cannot organize the exercise of this right to resistance; and it belongs not to the penal law, but to the constitutional law. Tyrannicide and right of resistance are no punishment of the tyrant, but rather provide for the tyrannicide a legal excuse for performing an action that would be, under normal circumstances, a punishable offense. I would like to discuss, however, the punishment of crimes against humanity. What would then comprise the goal of punishing a person who is at any given time in power and is committing a crime against humanity while holding office? It seems to me that the purpose of punishment can be either – most importantly – incapacitation, that is, in this case 6. Art. 20(4) of the German Basic Law.

punishing crimes against humanity?

177

the removal of the criminal who commits a crime against humanity from power, or general prevention, that is, in this case, deterrence of the members of foreign governments from committing a crime against humanity. Yet, the prosecution under the charge of crime against humanity, as it is defined not only in international law but also in internal law with its already mentioned characteristics, is either a superfluous or a useless tool. It is superfluous when it is possible to indict and to depose a member of the government or an entire government under charges derived from ordinary penal law, and the parliament is not so corrupt as either to tolerate or to support the crimes committed by the government. It is useless when either no such procedures of destitution or impeachment are foreseen by the constitution, or the criminal members of the government exercise such a power or threaten such violence that nobody will dare to prosecute. Prosecution may be impossible not only for reasons of internal dysfunction, but also for reasons of internal or international immunity. Members of governments, members of parliaments, diplomats, etc., enjoy an immunity grounded in international law that protects them even against prosecution of crimes against humanity. Since any humanitarian intervention employing military force is prohibited, except in the presence of aggression and with the consent of the United Nations Security Council, international law seems clearly not to head in the direction of a prosecution of crimes against humanity while the criminals are still in office. The concept of crimes against humanity, as it is understood by today’s human rights law, does not provide the tools necessary for the clearing of the legal obstacles on the path toward intervention against a government that violates massively the fundamental rights of its citizens. In fact, if the arguments made in favor of an unconventional punishment for crimes against humanity were to emphasize a deterrent effect at all, then it is the general deterrent effect that is supposed to result from sentencing the criminal political leader after he or she has left office. Effective prosecution and conviction of political leaders allegedly deter other political leaders from committing crimes against humanity. Yet, first, no evidence has ever been provided that political leaders who see in crimes against humanity the only means available either to promote political goals – the righteousness of which they may be truly convinced of – or simply to remain in office can be motivated to change their mind under the influence of the punishment of their foreign colleagues, although the motive for promoting

178

retributivist inhumanity

political goals they are deeply committed to or of remaining in office is certainly very strong.7 Secondly, were the aim of punishment deterrence, punishment should then be imposed directly following the commission of the crime, that is, while the perpetrator still occupies office, and not after he or she has eventually been deposed. (B) Let us now investigate the situation after the criminal has been deposed. Once the criminal has been divested of power, the preventive aim of punishment may be either: general deterrence – about which I have expressed my aforementioned doubts – or specific deterrence, which implies hindering the criminal in regaining power. Traditionally, three options are available in the case of a leader who has committed crimes against the institutions of his own state and has been deposed (for example, in cases of conspiracy, high treason, abuse of power, etc.), with each option corresponding to a characterization of the former leader’s traits as a leader, as well as of the situation in which he or she is presently. Let us assume that the former leader poses an immediate and acute threat for the institutions, because of having at his or her disposal as many supporters, as many means, as many weapons, etc., as he or she sees fit in order to have a chance to regain power. The first possible characterization implies that the former leader would lose this ability to regain power again if he were no longer to be present within the state’s territory. The classical solution for the aforementioned: banishment. This solution does not have anything to do with the suggestion made by Aaron Fichtelberg to banish the criminal from all of human society.8 In fact, banishment from all of human society ultimately amounts to condemning a person to an all-too-likely death, like that of a dangerous, wild animal, as Fichte demonstrated. The second characterization is that, even in exile, the criminal would remain as much of a danger for the state as if he or she were actually present in the state. The classical solution for the aforementioned: death. Finally, the third characterization of the former leader is that, after being deposed, the former leader no longer poses a threat to the state’s institutions. The former leader is thus condemned in order to sever the ties between the former leader and the people, as well as to demonstrate that the succeeding government does not share the same intentions as 7. See Aaron Fichtelberg, “Crimes beyond justice? Retributivism and war crimes,” Criminal Justice Ethics, 24, no. 1 (2005), 31–46 (p. 34); on p. 35 he rightly says that the threat of punishment might even be counterproductive. 8. Fichtelberg, “Crimes beyond justice?”

punishing crimes against humanity?

179

the preceding leader so that the crimes the earlier government committed can be strictly outlawed. But, in general, the condemnation is short and a pardon is rapidly granted.9 If necessary, a punishment may be added to some period of ineligibility for office. The old adage states that most of the time in politics punishment is either death (the second characterization) or a short prison term (the third characterization). It is interesting to note that in positive international law none of these three options is available to a tribunal such as those of the former Yugoslavia, of Rwanda or of the ICC. Capital punishment has been abolished by a significant proportion of the members of the UN, as well as by the ICC in The Hague, and no exception is made for crimes against humanity – unlike the exception made for the statute of limitations; the gravest crimes against humanity carry no greater sanctions than the worst instances of first-degree murder. After the Nuremberg and the Tokyo trials, capital sentences were no longer passed by an international criminal court. Were capital sentences allowed to be passed, most Western democracies would not participate in those trials, because most of them are expressly prohibited from extraditing to a court that can pass capital sentences – even in cases of crimes against humanity. Banishment is not an option because the Universal Declaration of Human Rights (Article 9), the European Convention on Human Rights (Article 7(1)) and the Final Act of the Conference on Security and Cooperation in Europe (Helsinki Accords of 1975, Principle 10) guarantee an individual’s access to his or her own country at all times. Admittedly, were the criminal leader to be judged by an international court, rather than by a domestic jurisdiction of the country of which he or she is a national, the sanction would then probably be carried out outside the country, which may be considered as de facto banishment. Finally, the sanction for the leaders of crimes against humanity cannot be a short-term sentence, but must be either a life sentence or a long period of time in prison. Two astounding points come to light when all of the aforementioned is considered as a whole: 1 The punishment for crimes against humanity radically diverges from the traditional punishment of crimes against legal and against 9. An amnesty may even be a necessary means with which to ensure a just peace; see Michael P. Scharf, “Justice vs. peace,” in Sarah B. Sewall and Carl Kaysen (eds.), The United States and the International Criminal Court: national security and international law (Lanham, Md.: Rowman and Littlefield, 2000), pp. 213–36 (p. 189).

180

retributivist inhumanity

political institutions, even though both crimes are closely related by their very nature. 2 The severity of sanctions handed down for crimes against humanity does not exceed the most severe sentences already existing for other crimes, whereas those other crimes, in most countries, are subject to the statute of limitations. I will come back to a discussion of the aforementioned observations at a later point. At this point, I will consider punishment of executants merely acting upon orders given to them by the politicians. I will again distinguish between the situation during the commission of the crime and the situation after the crime has been committed. It seems obvious that during the commission of the crime no significant number of executants, if any, can be prosecuted, if there should happen to be a systematic “state or organizational policy” in the framework in which they find themselves carrying out their criminal actions (Rome Statute Article 7.1). After the crime has been committed, the most important specific deterrent measure that must be taken is the dissolution of all criminal political organizations that had planned the crimes against humanity, guided them and, finally, carried them out. Yet, what might be the purpose of punishing the numerous individual executants? As I have already mentioned, one of the characteristics of crimes against humanity – as is also the case with war crimes – is that, on the one hand, command cannot be employed as a justification of the crime, and on the other hand, the commission of crime is not considered as being more serious than mere solicitation of an offense, in short, implying that leaders and the executants are all responsible for the crime against humanity. Yet, this is in the realm of pure theory, since only a very small proportion of the executants are prosecuted and convicted, and they are usually sentenced to less severe punishment than the leaders, moreover to sentences not any severer and often more lenient than for ordinary crimes (one thinks of a case heard before the International Criminal Tribunal for the Former Yugoslavia in 1996, in which Drazen Erdemovic´, a multiple murderer, was sentenced to only five years in prison owing to mitigating circumstances).10 Even voluntary membership in an organization officially classified as “criminal” by the Nuremberg Tribunal, such as the Schutzstaffel (SS) or the 10. See the Sentencing Judgment in the case of Prosecutor v. Drazen Erdemovic´, available at www.un.org/icty/erdemovic/trialc/judgement/erd-tsj980305e.htm.

punishing crimes against humanity?

181

Gestapo, has not been systematically punished, only the gravest cases being prosecuted. The sanctions enforced on a few executants can be really efficient neither from the point of view of specific deterrence (a few isolated individuals alone are not able to rebuild the criminal organizations to which they belonged), nor from the point of view of general deterrence (were the probability to be sentenced extremely low, the deterrence effect would not be sufficient to justify such measures). A much more efficient solution, from the point of view of general deterrence, would be a massive punishment of all executants, were it possible at all. If such a massive punishment would meet with resistance from the criminal who was indicted and sentenced, that would threaten the existence and stability of the state under the rule of law, and if the kind of very limited moderate punishment that I have described would allow it to restore the state institutions to the rule of law, all theories of punishments, whether deterrent or retributive, would prefer the latter option. However, there is a quintessential difference between deterrent and retributive theories with respect to this solution. A retributive theory of punishment considers it as an exception made to the law, which can be made only in order to protect the existence of the law in a situation in which the enforcement of the law is not possible. In the Doctrine of right, Kant evokes the situation in which “the number of accomplices” involved in a homicide “is so great that the state, in order to have no such criminals in it, could soon find itself without subjects; and . . . the state still does not want to dissolve, that is, to pass over into the state of nature, which is far worse, because there is no external justice at all in it.”11 According to Kant, who, at this point, undoubtedly belongs to the retributivist school, the malefactor in such a situation should be sentenced to a mitigated sentence, yet not through the letter of the law, but instead through a decree of pardon issued by the monarch. For a theory of deterrence, these more lenient punishments provide no exception to the law of criminal justice, but are instead part of it, because penal law must, as its sole purpose, ensure the enforcement of the rule of law as quickly as possible. In this regard, once the danger has passed that the malefactor who committed crimes against humanity might regain power under normal rule of law, the probability of recidivism will disappear. The probability of recidivism is much lower than with other 11. RL Ak vi:334. Practical philosophy, ed. Gregor, p. 475.

182

retributivist inhumanity

crimes, and because of that the punishment should likewise be much less severe than what is usually prescribed by the normal penal code. I will come back later to the retributivist position. The conclusion that I will first draw from these points may be perceived prima facie as repugnant. Because the characteristics that typify crimes against humanity are formed by their political dimension and by the lack of the usual sanction structure, they ought to lead to more lenient punishments than do ordinary individual crimes. General and specific deterrence simply do not require the same means, in cases involving crimes against humanity, as cases involving ordinary crimes perpetrated on a more individual scale. This conclusion might well become even more repugnant were we to carry out the following thought experiment. At this point, before going any further, I would like to stress the point that I have absolutely no sympathy for the person, or the ideas, mentioned below, whose policies – especially his policy advocating wholesale extermination of several ethnic, political and religious groups – rightly deserve the deepest moral condemnation and the most vehement rational rejection. But I would like to make reference to the reasoning used by the French defense attorney Robert Badinter, who finally obtained the abolition of capital punishment after he became the French Minister of Justice, after having struggled against it for his clients before numerous criminal courts. Badinter explained that the more antipathy a criminal case provokes, the more clearly pleas may be made on the grounds of principle.12 Let us begin our thought experiment by imagining that the New York Times discovered, over forty years after the close of the Second World War, that Adolf Hitler did not die in his bunker in Berlin in April 1945, but instead escaped and ever since had been living inconspicuously somewhere in the Irish countryside. Except for retributivist arguments, I see no criminal justice theoretical argument that could be presented to demonstrate that he should still be punished after forty years; at that point, he no longer poses a threat to any political institutions. In order to be absolutely sure that he would not pose any further threat, one could consider depriving him, at the very most, of his right to free speech. Drawing this repugnant conclusion, that is, pleading for the imposition of the statute of limitations, even in cases involving the most despicable criminals, contradicts one of the aforementioned characteristics of crimes against humanity, 12. See Robert Badinter, L’Abolition (Paris: Fayard, 2000).

punishing crimes against humanity?

183

but it does not necessarily mean making a plea for the lifting of the finding of guilt, as demonstrated by the Commission for Truth and Reconciliation, a body that was intended to establish the facts but not to make any condemnations.13 However, I suspect that a certain amount of time in prison would have been necessary to rehabilitate some very serious criminals of the apartheid era because they hold the most firmly rooted convictions. I will confine myself only to forwarding the thesis that in cases of crimes against humanity the conventional prescriptive period should never be exceeded, but instead even reduced.

7.3. The boundaries of our moral intuitions As I mentioned above, a rejection of this repugnant conclusion can only rely on a retributivist theory of punishment. Such a theory typically refers to the fourth characteristic of crimes against humanity that I mentioned in Section 7.1. Retributivism rejects my repugnant conclusion, because the punishment can never be equal in gravity to the crime; this is, for instance, Hannah Arendt’s thesis. Clearly, the monstrousness of all crimes against humanity is apparent, but the monstrousness of the crime cannot define a wholly new category of crime. A cannibal who has eaten his consenting victims is, in my view, more monstrous than someone who kills for stealing a million dollars. Yet, the latter will be sentenced to a more serious punishment. The life sentence will be (for instance, in Germany) the same as for someone who tortured and killed a baby before the eyes of its mother, which most of us will find more monstrous. And there are many actions that are monstrous and unpunishable. Someone may cynically declare, while eating in a luxury restaurant, that what makes him particularly enjoy the meal is the thought that so many “useless” people are dying of starvation somewhere in Africa. Or someone may declare to her dying life partner at the hospital that she never loved him, had betrayed him, and then leave without even saying good-bye. Both are monstrous actions, though fully legal and protected rights. Whether a homicide, a rape, a kidnapping, a manslaughter, etc. is monstrous is neither a criterion for ascribing it to the category of crimes against humanity, nor an element of the crime itself. The purpose of penal law, as is the purpose of the legal system in toto, is not to retribute moral 13. See May, Crimes against humanity, p. 239.

184

retributivist inhumanity

evils and perversions, but to protect society, that is, the rights of the citizens. This implies that penal law amounts to a means to restore the rule of law between all persons – including the criminals – as quickly and as efficiently as possible. If we were now to formulate the retributivist objection to my repugnant conclusion in the guise of a classical criticism, we should say that the retributivist would postulate that crimes against humanity are deserving of severer punishment than similar crimes not barring “widespread and systematic attacks”14 committed “pursuant to or in furtherance of a State or organizational policy to commit such attack.”15 However, I see several serious difficulties with such a view being utilized as a justification for disregarding the statute of limitations. 1 Were we to compare a homicide committed within the framework of crimes against humanity with an “ordinary” homicide perpetrated on the individual level, we would be hard pressed to postulate the former as being more deserving of severer punishment. Yet, should a campaign of discriminatory, forced sterilization be punished more severely than, say, an ordinary first-degree murder? The answer is not at all obvious. 2 Should a crime against humanity consisting of a massive extermination of human beings utilizing industrial methods be punished more severely than a conventional first-degree murder? Were the latter already punished by a life sentence, I see only a death sentence, or more likely a long and public torture, as a severer punishment. Yet, would that not constitute in itself a crime against humanity? The letter and the spirit of international law punishing crimes against humanity are clearly not shaping its course toward such extreme punishments. (For this reason I disagree with Larry May, who sees proportionality as requiring restraint in punishment in relation to “emotional factors”: I see it, on the contrary, as heading in the same direction.)16 3 Retributivism is not compatible with the lack of an outline of the traditional sanction structure in Articles 7 and 25 of the Rome Statute. One may object to my repugnant conclusion stating that penal law should not offend popular sentiment. Allowing Adolf Hitler refuge in 14. Rome Statute, Art. 7.1. 15. Rome Statute, Art. 7.2. 16. May, Crimes against humanity, p. 215.

punishing crimes against humanity?

185

the harbor of the statute of limitations would severely offend popular sentiment at least as much as not punishing crimes against humanity more so than conventional crimes. One can understand this argument in one of the following two ways. Either this argument posits that no moral judgment can be morally true, were it to contradict popular sentiment. Or the argument posits that such a repugnant conclusion would lead the people to overthrow the judiciary power and possibly even the rule of law. Popular sentiment cannot be allowed, however, to be a criterion for the moral judgment of punishment, among others things because it is inconsistent in many regards. Popular sentiment mixes retributive, general and specific deterrent elements. It can protest against punishments that seem too lenient, but it can allow the victim of a car accident to receive much greater compensation than the victim of rape or of serious physical injuries perpetrated by an insolvent criminal. Furthermore, common sentiment is more often prone to immoral judgments. For centuries it authorized criminal trials and sentences against animals or against witches; it authorized torture either as a punishment or as a method for interrogating a suspect; it authorized the punishment of an entire family for a crime committed by one of its members. All this and more was fully acceptable to the common sentiment. Nowadays, death by starvation of millions of people is still not perceived in common feeling as a violation of basic human rights, although such texts of international law as the Universal Declaration of Human Rights (1948) rightly recognize subsistence as a fundamental human right. Furthermore, in many countries, a majority of the citizens is still in favor of capital punishment, as well as of increasing restrictions on the right of asylum. Were one to consider the risk of the people possibly overthrowing the institutions of the judiciary, as well as the political institutions, if crimes against humanity were to be subject to the protections of the statute of limitations, there would be no evidence in favor of this view. South African institutions of state have suffered no harm to their prestige by not punishing the criminals of apartheid,17 and demonstrations against the impunity of dictators who 17. The opposite is true: without the amnesty granted to a section of the perpetrators of the apartheid regime by the new government there would have been a real threat of military upheaval in South Africa: see chapter 6 of Amy Gutmann and Dennis Thompson, Why deliberative democracy? (Princeton: Princeton University Press, 2004).

186

retributivist inhumanity

committed crimes against humanity have never endangered the democratic rule of law. For these reasons I see no valid moral objection against applying the statute of limitations and against mitigating punishment. Though the existence of the concept of crimes against humanity is a pragmatic criterion for judging the gravest crimes, and though it is the necessary first step on the path to the harmonization of criminal prosecutions by setting a global standard and an appropriate priority rule, these pragmatic reasons seem to me not to justify the existence of crimes against humanity as a fundamentally new type of crime that requires new types of punishment and goals of punishments, as well as justifying especially severe prosecutorial precepts. Treating crimes against humanity pragmatically does not imply, however, that they do less harm than conventional crimes. In fact, exactly the opposite is true: crimes against humanity are especially monstrous. This paradox reminds us that it is not the role of penal law to take revenge on criminals for our past powerlessness. Rather, the role of penal law is to protect against future crimes and to secure the rights of the victim – as well as of those of the perpetrator – within the rule of law. The victims of crimes against humanity might well be more fortunate if governments who self-sufficiently congratulate themselves for having created the ICC – which I certainly find a most useful institution of global justice – actually developed the right of asylum rather than reduced it. The central issue should be not whether or not Milosevic should be sentenced to life in prison or to only four years behind bars; instead, the central issue should be how best to wipe torture off the face of the earth.18 18. In this regard, Andrew Altman and Christopher H. Wellman, in “A defense of international criminal law,” Ethics, 115, no. 1 (2004), 35–67, confuse the role of the punishment of criminals against humanity with the role that ought to be played by military humanitarian intervention.

CONCLUSION

In Part i, we examined the Kantian, retributivist thesis, according to which the criminal is punished, because he or she merits it. Two interpretations of the Kantian concept of right are possible (see Chapter 1). On the one hand, according to the liberal interpretation, which does not refer to the internal disposition of human beings, right is defined merely as being the coexistence of the freedom of action of all human beings according to the principle of equality of right. On the other hand, according to the moral interpretation, which substantially refers to the internal dispositions of human beings, right should implement the content of the categorical imperative as far as possible through the application of coercion. In accordance with both of these interpretations, two dimensions can be distinguished in the Kantian retributivist thesis: one genuine dimension of legal ethics, which is a dimension independent of the internal dispositions of individuals (see Chapter 2), and a dimension of personal ethics (see Chapter 3). The dimension of legal ethics of the Kantian retributivist thesis contradicts the requirement for the coexistence of all freedoms, of which the freedom of the criminal is a part too, and of which it ought again to become a part – at least in the most direct way. The dimension of personal ethics contradicts the postulate of the highest good, which rather requires forgiveness toward the criminal as long as this forgiveness is compatible with the concern for the safety of the other citizens. Hence, Kant’s concept of right as well as his concept of virtue require rehabilitation instead of retribution, which ought to occur after the shortest possible period of specific deterrence, that is, after a period of incapacitation. In Part ii, I showed how this justification of punishment as a rehabilitation occurring after a period of specific deterrence was 187

188

conclusion

supported on the basis of the liberal interpretation of Kant’s concept of right by Fichte (see Chapter 4) and Hegel (see Chapter 5): the criminal is punished, in order that he or she might once again become part of the commonwealth. Fichte begins with the question of the treatment of the criminal and by a thought experiment initially draws the logical conclusion of crime. The logically consistent consequence of it would not be retaliation, but instead the exclusion of the criminal from the commonwealth and the revocation of his or her legal status. The public punishment derived from right as such emerges as an alternative to this fate that is better for the criminal so long as the alternative punishment amounts neither to the death penalty nor to torture but rather to a sentence of incarceration. Such an expiation, therefore, is not an evil, but rather a chance, which Fichte constructs in such a way that the criminal can be reformed and then finally reintegrated into the commonwealth. Hegel, in turn, chooses as his guideline not the treatment of the criminal, but the reestablishment of right as the negation of the negation – in which crime is the negation and the punishment is the negation of the negation. In so doing, the “negation” of the negation is in no case, as it is often mistakenly assumed to be, of the same sort as the simple “negation” is, and thus it should not be taken to be a response to the evil of the crime by a second evil. Contrary to many interpretations, Hegel is not a retributivist, even though he – like Fichte as well – assigns to retributivism a (merely) instrumental, psychological role in conveying to the convicted criminal the significance of the punishment. In Part iii, it finally turned out that retributivism could often demand punishment without any respect for humanity in the person of the criminal. Nietzsche offers a plausible account of the genesis of retaliative punishment (see Chapter 6), in which the motive of the institution of such punishments stems not from the respect for human dignity, but rather from a universal human cruelty toward the criminal, which hinders rather than awakens the emergence of bad conscience and remorse in the criminal. When the state establishes a monopoly on violence and guarantees public security, the people are deprived of the exercise of this cruelty and punishments become comparatively mild, such as those we have known since the end of the eighteenth century. In order to keep the promise I made in the introduction to draw conclusions resulting from the rehabilitative alternative to retributivism, I finally argue (see Chapter 7) for a treatment of even

conclusion

189

the most serious and most inhumane criminals – in other words those who have perpetrated crimes against humanity – that is no more severe than how other criminals are treated. Even these cases ought to be subject to the statute of limitations: the degree of punishment ought not to be higher – which is to say, proportional to the crime – but ought, instead, to tend to become less; the idea ought to be abandoned that the monstrousness of such crimes justifies the punishment – and indeed, the most severe possible punishment. Further conclusions can be drawn from my critique of retributivism as well as from my alternative proposal. At this point, I would like briefly to outline how the penal system should be reformed according to my alternative proposal. 1. There ought to be no fixed duration within the judgment for incarceration. Instead of designating a certain duration, the degree of punishment ought to be determined by a goal. The criminal ought to expiate the punishment for as long as he or she still poses more of a danger to the commonwealth than the other citizens do. In order for that to occur, the criterion employed to suspend punishment may not be the absolute safety of society, because every “normal” citizen who consistently remains true to the law could, at any time, commit a crime. Even now, punishments are reduced by sentences of parole, and in the case of good behavior prisoners are released before their sentences are up. Others are allowed to work outside the prison during the day. Such leeway, already customary nowadays, ought to be expanded to the general abolition of every designation of duration. 2. The principle of proportionality between crime and punishment has to be rejected. Admittedly, it is to be expected that, on average, thieves can be rehabilitated in a shorter period of time than multiple murderers who have tortured their victims. One would, therefore, probably observe a clear, statistical correlation between the gravity of the deed and the length of the process of rehabilitation. But this link should not be a matter of principle, and there would certainly also be exceptions. 3. A reform of penal law ought not to involve punishments becoming generally severer or generally milder. In effect, some punishments would become milder, others severer, all depending on the behavior of the individual convict.

190

conclusion

4. Psychological assessments of the convicts should occur regularly and they should be thorough. A process such as the following is conceivable. At the beginning of serving a sentence the criterion for a discharge would be especially strong. The criterion would weaken as time passes and would finally reverse so that gradually the burden of proof might stand on the side of the commonwealth to prove that the convict must, in fact, remain in prison. 5. Retributivism likes to contend that only retaliation is just for the victim and every other punishment would be unjust, for the criminal as well as for the victim. In the current penal system, this argument rather appears to be an empty consolation for the fate of the victim that turns out to be scandalous. Anyone who suffers grave bodily injury is better off having it occur in the course of an automobile accident for which he or she is not liable, rather than by a serial killer, because the damages awarded to victims of criminal acts are especially low. One should make it clear to the public what a criminal trial really means. Whereas potential compensation for damages, including pain and suffering, should satisfy the claims of the victims, punishment applies only to the interests of the commonwealth as a whole regarding public security. Therefore, a twofold process is conducted, even if both concerns are to be dealt with in the same trial. Instead of exclusively and systematically calling for harsher punishments as a way to win popular support, politicians should primarily think of a better and more respectful compensation for the victims. 6. An attempt ought always to be punished in the same degree as the crime itself. This is not the case nowadays. Instead, attempts at only some of the graver crimes are punishable, while carrying a lesser degree of punishment. Feinberg convincingly traced back this situation to the conjunction that retributivism wants to create between the gravity of a deed and the wickedness of the criminal: I find no intuitive plausibility at all in first basing criminal liability on moral blameworthiness, but then basing moral blameworthiness in turn upon the actual harm or absence of harm caused. The ancient view, that liability should be based not upon blameworthiness at all but instead directly upon the amount of harm caused, seems to me more honest, though no more plausible. It seems almost as if the retentionist is so fixated on actual harm that he keeps searching for the question to which it is the right answer. Not the question: “What ought to be the basis of criminal liability?” Moral blameworthiness is a plausible

conclusion

191

answer to that question. Not the question: “What is the basis of moral blameworthiness?” The traditional multiplicity of culpability conditions, motives, mitigations, aggravations, and so on, answers that question. How about the question: “What is a necessary condition for tort liability?” Now, that is more like it.1

In this, modern retributivism encounters the question of how the inner, moral guilt of a person should be punished in a legal system, if the system intends to be in charge only of freedom of action, and not of morality. Beyond these concrete suggestions for the reform of penal law, as the result of my inquiry I must address four points of more general interest. First, penal law cannot be about the (purely moral) wickedness of human beings, but instead only about the observance of the laws or about the infringement of right. Even if we were all “devils,”2 as Kant notes in Toward perpetual peace, it would still not mean that we all ought to be punished with deterrence in mind, before having committed any misdeed. Only those who cannot be hindered from committing crimes by moral or external reasons should be punished. Secondly, it is not retributivism, but the justification of punishment as a rehabilitation that is compatible with that path via which Kant wants to reach global “perpetual peace,” the condition by which the state of law would reign over the entire world. The modern tradition of natural right that Kant critically draws upon recognizes three just intentions for waging war: legitimate self-defense, recovery of one’s own goods stolen by the enemy and punishment of an unjust enemy.3 This “punishment” of the vanquished enemy is intended to mean deterrence by the theorists of just war – for instance, Hugo Grotius, Samuel von Pufendorf, Emer de Vattel, etc. The aim is a peace that is just toward both parties. In this respect, punishment is allowed only in so far as to bring about security for the victor. This theory pursues the goal of ensuring that no future injustice is carried out that would present a just ground for a further war. If this deterrent intent turns out to be efficacious, there will then be a continuously decreasing

1. Joel Feinberg, Problems at the roots of law: essays in legal and political theory (Oxford: Oxford University Press, 2003), p. 100. 2. Cf. ZeF Ak viii:366. Practical philosophy, ed. Gregor, p. 355. 3. Cf. Book 2, Chapter 1, ii.2 of Hugo Grotius, The rights of war and peace, ed. Richard Tuck (3 vols., Indianapolis: Liberty Fund, Inc., 2005), pp. 395–6.

192

conclusion

number of wars. Since the victor has already obtained supremacy, the punishment is often mild. Herein lies the strategy of the theory of just war for the preservation of peace. Let us now imagine what consequences a retributivist conception of punishment might have for this theory. If the vanquished enemy should fail to understand that the victor’s case is just, the retributive punishment by the victor will be perceived by the vanquished as being unjust. Unlike in national law, in international law there are no higher-ranking judges, so that every sovereign is his or her own judge, and although there cannot be two parties whose suits are ultimately equally just, there can be two bona fide parties waging war whose reasons for war appear completely legitimate and prima facie just. Unlike a punishment that is meant to deter and to protect against the recurrence of the grounds that led to war, retributivism requires the vanquished not only to assume liability for all damages caused in the war (compensation, reparations, etc.), but also to serve a severe punishment. If the vanquished enemies waged the war in good faith, this retributive punishment will then be perceived by them to be unjust and thus may consitute a possible causa iusta for a future war. For this reason, such a retributive punishment stumbles into a contradiction with Kant’s primary objective, that is, with perpetual peace. Kant’s first preliminary article deprives every retaliation of its prerequisite by requiring that possible discoveries about the past should not be taken into consideration: Causes for a future war, existent even if as yet unrecognized by the contracting parties themselves, are all annihilated by a peace treaty, no matter how acute and skilled the sleuthing by which they may be picked out of documents in archives.4

Even though Kant had the material claims of the warring parties in mind in this passage, this remark certainly also applies in the realm of penal law. Even in the Doctrine of right, Kant recommends the same strategy for (civil) peace on the national level, as the following repeatedly cited passage shows: Accordingly, every murderer – anyone who commits murder, orders it, or is an accomplice in it – must suffer death; this is what justice, as the idea of judicial authority, wills in accordance with universal laws that are grounded a priori. – If, however, the number of accomplices (correi) 4. ZeF Ak viii:344f. Practical philosophy, ed. Gregor, p. 317.

conclusion

193

to such a deed is so great that the state, in order to have no such criminals in it, could soon find itself without subjects; and if the state still does not want to dissolve, that is, to pass over into the state of nature, which is far worse because there is no external justice at all in it . . . then the sovereign must also have it in his power, in this case of necessity (casus necessitatis), to assume the role of judge (to represent him) and pronounce a judgment that decrees for the criminals a sentence other than capital punishment, such as deportation, which still preserves the population. This cannot be done in accordance with public law but it can be done by an executive decree that is, by an act of the right of majesty which, as clemency, can always be exercised only in individual cases.5

Here abiding by the law is set aside in favor of preserving civil peace for the very reason of preserving it, because civil peace is the sole prerequisite for a state of law in which laws can be issued and enforced. This problem, to which retributivism is exposed within the Kantian philosophy of right, is actually related, thirdly, to a fundamental problem: retributivism is not compatible with an important principle of the Kantian philosophy of right, that is, with the permissive law (see Sections 2.4 and 4.4). The permissive law allows an exception to be made to law’s implementation, so long as this exception only concerns what has occurred up to then and not that which either is occurring in the present or will occur in the future. Such an exception should provide for the peaceful establishment of a state of law where the application of the law to the existing situation would lead to regression to the state of nature or into civil war. However, the implementation of the law in the present and in the future is not allowed to be postponed ad calendas graecas,6 but instead should occur in the most direct way possible.7 Now, Kant defines right as being the “sum of the conditions under which the choice of one can be united with the choice of another in accordance with a universal law of freedom,”8 by which the power of choice (Willku¨r) of all human beings – therefore, also of the criminal – is meant. Thus, if there is in fact a way enabling one to reintegrate the criminal into society in the most direct 5. RL Ak vi:334. Practical philosophy, ed. Gregor, p. 475. 6. ZeF Ak viii:344f. Practical philosophy, ed. Gregor, p. 321. 7. Cf. RL Ak vi:247. Practical philosophy, ed. Gregor, pp. 401–2. Also, cf. Reinhard Brandt, “Das Erlaubnisgesetz, oder: Vernunft und Geschichte in Kants Rechtslehre,” in Brandt (ed.), Rechtsphilosophie der Aufkla¨rung (Berlin: De Gruyter, 1982), pp. 233–85. 8. RL Ak vi:230. Practical philosophy, ed. Gregor, p. 387.

194

conclusion

way, then this way is commanded. If this way actually exists at all, then it is, per definitionem, rehabilitative punishment. The categorical imperative of right consists not only in an ideal legal system, but also in the steps that are necessary for either the establishment or the reestablishment of this legal system. Writings such as Toward perpetual peace and Idea for a universal history with a cosmopolitan purpose belong to the latter dimension of the categorical imperative of right. Even though } 49e devoted to penal law is an editorial section of the Doctrine of right, which has to do primarily with the ideal legal system, and therefore with the first dimension, penal law can be categorized into the second dimension. In this respect, Gustav Radbruch’s dichotomy between right as a realm of justice in compensation (commutative justice) and an extralegal realm of distributive justice and social utility is to be resolutely rejected.9 Penal law, which incontrovertibly constitutes an essential component of right, pertains neither to commutative justice nor to corrective justice. The former presupposes the agreement of all parties over the conditions and the items of the trade. Be it only for this reason, punishment cannot rest on commutative justice. Not even retributivism can be described as a trade. “Eye for eye, tooth for tooth”10 or “the retribution of what one merits” does not mean trading eyes or teeth, or trading the deserts of a human being for the infliction of evils by another human being. Now, the return to the status quo ante pertains to corrective justice. However, how can the status quo ante of a murder victim be reestablished even to a small extent? In this respect, Cesare Beccaria’s rhetorical question is irrefutable: “Can the wailings of a wretch, perhaps, undo what has been done and turn back the clock?”11 The punishment, rather, has to do with distributive justice, because it assures every human being of integration or of reintegration into the commonwealth – regardless of his or her achievements or actions up to then – for the simple reason that every human being as a being capable of reason has an innate, unforfeitable right to such.

9. Cf. Gustav Radbruch, Rechtsphilosophie, ed. Erik Wolf, seventh edition (Stuttgart: K. F. Koehler, 1970), p. 265. 10. Lev. 24:20 (KJV). 11. Chapter 12, entitled “The purpose of punishment,” in Cesare Beccaria, On crimes and punishments, in Beccaria, On crimes and punishments and other writings, ed. Richard Bellamy, trans. Richard Davies (Cambridge: Cambridge University Press, 1995), p. 31.

conclusion

195

Fourthly, the categorical legal imperative of integration or reintegration into the commonwealth likewise derives a deeper grounding in the Kantian foundation of morality, namely in the empirical circumstances of the exercise of autonomy in a limited being capable of reason who can also act against his or her duty, irrationally. On the one hand, a human being should bear the responsibility for his or her actions because he or she has freedom of action. On the other hand, the consequences of actions carried out in the past should not lead to the consequence that a person can no longer act freely in the future. These two dimensions do not limit one another. The consequences of the actions hitherto done are limited by the commandment that requires treating every human being as a being capable of reason with an inalienable freedom of action. Now, this inalienable freedom of action is not limited by the consequences of past actions. Rather, it is limited by the same inalienable freedom of action of the other human beings. It is only for this reason that the Kantian concept of right is an object of the categorical imperative, which we may also term the “categorical imperative of right,” to borrow Otfried Ho¨ffe’s expression.12 We therefore find a twofold hierarchization of moral goods that can collide with one another. First, the commonwealth would receive precedence over individual freedom in the case of a collision, which is why it is allowed provisionally either to limit or to suspend the freedom of the criminal. Secondly, in the case of a collision, individual freedom receives precedence over the consequences, which the actions of individuals should otherwise imply; this is why crime should be forgiven by society – with consideration for the first priority, that is, with consideration for the security of the commonwealth – in the most direct manner possible.13 At the center of the Kantian idea of autonomy, under the guidance of which we always ought to act and to which, therefore, the legal system is subject, is the forward-looking project of the full

12. Cf. chapter 5 in Otfried Ho¨ffe, Categorical principles of law: a counterpoint to modernity (University Park: Pennsylvania State University Press, 2002). 13. This very arrangement of priority holds equally, by the way, for other realms in the Kantian doctrine of right, for example property: cf. Jean-Christophe Merle (ed.), “Eigentumsrecht,” in Merle (ed.), Johann Gottlieb Fichte: Grundlage des Naturrechts (Berlin: De Gruyter, 2001), pp. 159–72; and Section 2.3 in Jean-Christophe Merle, Justice et progre`s: contribution a` une doctrine du droit e´conomique et social (Paris: Presses Universitaires de France, 1997).

196

conclusion

development of the human being as a being capable of reason, and not the painfully exact accounting of the misdeeds of humanity in the epoch of the – as Kant terms it – “baby walker.”14 As Albert Camus observed in the debate over the death penalty, one could equally say of retributivists that: “we all know today’s executioners are humanists.”15 It is just that we do not hold the same concept of humanity as they do. 14. The epoch of the Ga¨ngelwagen, a device used by children learning to walk. Was ist Aufkla¨rung? Ak viii:35. Practical philosophy, ed. Gregor, p. 17: “walking wagon.” Also found in KrV B 174. Critique of pure reason, ed. Guyer and Wood, p. 269: “leading-strings.” 15. Albert Camus, Reflections on the guillotine: an essay on capital punishment by the 1957 Nobel Prize winner, trans. Richard Howard (Michigan City, Ind.: Fridtjof-Karla Publications, 1959), p. 50.

BIBLIOGRAPHY

Primary texts Beccaria, Cesare. On crimes and punishments, in Beccaria On crimes and punishments and other writings, ed. Richard Bellamy, trans. Richard Davies (Cambridge: Cambridge University Press, 1995), pp. 1–113. Bentham, Jeremy. An introduction to the principles of morals and legislation, reprint of 1823 edition (Oxford: Oxford University Press, 1907). Camus, Albert. Reflections on the guillotine: an essay on capital punishment by the 1957 Nobel Prize winner, trans. Richard Howard (Michigan City, Ind.: Fridtjof-Karla Publications, 1959). Cicero, Marcus Tullius. On duties, ed. M. T. Griffin and E. M. Atkins (Cambridge: Cambridge University Press, 1991). Du¨hring, Eugen. Cursus der Philosophie als streng wissenschaftlicher Weltanschauung und Lebensgestaltung (Leipzig: O. R. Reisland, 1894–5). Fichte, Johann Gottlieb. Foundations of natural right, ed. Frederick Neuhouser, trans. Michael Baur (Cambridge: Cambridge University Press, 2000). Grotius, Hugo. The rights of war and peace, ed. Richard Tuck (3 vols., Indianapolis: Liberty Fund, Inc., 2005). Hegel, G. W. F. Elements of the philosophy of right, trans. H. B. Nisbet (Cambridge: Cambridge University Press, 1991). Naturrecht und Staatswissenschaft (1818/19), in Hegel, and Vorlesungen u¨ber Rechtsphilosophie, ed. Reinhard Lauth (4 vols., Stuttgart and Bad-Cannstatt: Frommann-Holzboog, 1973–4), vol. 1, pp. 217–351. Philosophie des Rechts (1824/25), in Hegel, Vorlesungen u¨ber Rechtsphilosophie, ed. Reinhard Lauth (4 vols., Stuttgart and Bad-Cannstatt: FrommannHolzboog, 1973–4), vol. 4, pp. 67–752. Principes de la philosophie du droit, trans. Jean-Franc¸ois Kerve´gan (Paris: Presses Universitaires de France, 1998). Hobbes, Thomas. Leviathan, ed. A. R. Walter (Cambridge: Cambridge University Press, 1904). Jhering, Rudolf von. Law as a means to an end, ed. Joseph H. Drake, trans. Isaac Husik (Union, NJ: Lawbrook Exchange, 1999).

197

198

bibliography

Kant, Immanuel. Kants gesammelte Schriften (Berlin: Ko¨niglich Preußische Akademie der Wissenschaften, 1902). Lectures on ethics, ed. Peter Heath and J. B. Schneewind, trans. Peter Heath (Cambridge: Cambridge University Press, 1997). Klein, Ernst Ferdinand. Grundlinien des gemeinen deutschen und preußischen peinlichen Rechts (Halle: Hemmerde und Schwetschke, 1796). Krause, Karl Christian Friedrich. Abhandlung u¨ber die Idee und die Eintheilung der Philosophie und der Mathematik und den innigen Zusammenhang beider, in Paul Hohlfield and August Wu¨nsche (eds.), Philosophische Abhandlungen: aus dem handschriftlichen Nachlasse (Leipzig: Schulze, 1889), pp. 5–40. Grundlage des Naturrechts oder philosophischer Grundriß des Ideals des Rechts, (Jena: Gabler, 1803). Mably, Gabriel de. “De la le´gislation,” in œuvres comple`tes (12 vols., London, 1789). Mill, John Stuart. “April 1868 speech on capital punishment,” in Mill, Utilitarianism, ed. George Sher (Indianapolis: Hackett, 2001), pp. 65–70. Nietzsche, Friedrich. On the genealogy of morality, ed. Keith Ansell-Pearson, trans. Carol Diethe (Cambridge: Cambridge University Press, 2007). Plato. Laws, trans. Trevor J. Saunders, in John M. Cooper and D. S. Hutchinson (eds.), Plato: Complete works (Indianapolis: Hackett, 1997), pp. 1318–1616. Pufendorf, Samuel von. De jure naturae et gentium, ed. James Brown Scott, trans. C. H. and W. A. Oldfather, Classics of International Law (2 vols., Oxford: Clarendon Press / London: Humphrey Milford, 1934). Schopenhauer, Arthur. The world as will and idea, trans. R. B. Haldane and J. Kemp, seventh edition (4 vols., London: Kegan Paul, Trench, Tru¨bner and Co., 1909?). Schroeder, Friedrich-Christian (ed.). Die Carolina: die Peinliche Gerichtsordnung Kaiser Karls V. von 1532 (Darmstadt: Wissenschaftliche Buchgesellschaft, 1986). Seneca. De ira, in Moral essays, trans. John W. Basore (3 vols., London: Heinemann/ Cambridge, Mass.: Harvard University Press, 1928), vol. 1, pp. 106–355. Sidgwick, Henry. The elements of politics, reprint of 1891 edition (New York: Cosimo Classics, 2005). Voltaire. “Commentary on the book On crimes and punishments, by a provincial lawyer,” in Voltaire, Political writings, ed. and trans. David Williams (Cambridge: Cambridge University Press, 1994), pp. 244–79.

Secondary texts Altman, Andrew and Wellman, Christopher Heath. “A defense of international criminal law,” Ethics, 115, no. 1 (2004), 35–67. Ataner, Attila. “Kant on capital punishment and suicide,” Kant-Studien, 97, no. 4 (2006), 452–82. Badinter, Robert. L’Abolition (Paris: Fayard, 2000). Bedau, Hugo. “Retribution and the theory of punishment,” Journal of Philosophy, 75, no. 11 (1978), 601–20. Benn, Stanley. “An approach to the problems of punishment,” Philosophy, 33 (1958), 321–41.

bibliography

199

Bianchi, Hermann. “Abolition: assensus and sanctuary,” in Alexander R. Duff and David Garland (eds.), A reader on punishment (Oxford: Oxford University Press, 1994), pp. 336–51. Bosanquet, Bernard. The philosophical theory of the state (London and New York: Macmillan, 1899). Braithwaite, John. Crime, shame and reintegration (Cambridge: Cambridge University Press, 1989). Braithwaite, John and Pettit, Philip. Not just deserts: a republican theory of criminal justice (Oxford: Clarendon Press, 1990). Brandt, Reinhard. “Das Erlaubnisgesetz, oder: Vernunft und Geschichte in Kants Rechtslehre,” in R. Brandt (ed.), Rechtsphilosophie der Aufkla¨rung: Symposium Wolfenbu¨ttel 1981 (Berlin: De Gruyter, 1982), pp. 233–85. Breazeale, Daniel. Philosophy and truth: selections from Nietzsche’s Notebooks of the early 1870s (Atlantic Highlands, NJ: Humanities Press, 1979). Brugger, Winfried. “Darf der Staat ausnahmsweise foltern?” Der Staat, 35 (1996), 67–97. “Vom unbedingten Verbot der Folter zum bedingten Recht auf Folter?” Juristenzeitung, 35, no. 4 (18 February 2000), 165–73. Byrd, Sharon. “Kant’s theory of punishment: deterrence in its threat, retribution in its execution,” Law and Philosophy, 8, no. 2 (1989), 151–200. Clark, Michael. “A non-retributive approach to punishment,” Ratio, 17, no. 1 (2004), 12–27. Cohen, Hermann. Schriften zur Philosophie und Zeitgeschichte, ed. Albert Go¨rland and Ernst Cassirer (Berlin: Akademie Verlag, 1928). Deith, John. “On the right to be punished: some doubts,” Ethics, 94 (1984), 191–211. Feinberg, Joel. Problems at the roots of law: essays in legal and political theory (Oxford: Oxford University Press, 2003). Fichtelberg, Aaron. “Crimes beyond justice? Retributivism and war crimes,” Criminal Justice Ethics, 24, no. 1 (2005), 31–46. Flechtheim, Ossip Kurt. “Die Funktion der Strafe in der Rechtstheorie Hegels,” in Flechtheim, Von Hegel zu Kelsen: rechtstheoretische Aufsa¨tze (Berlin: Duncker und Humblot, 1963), pp. 9–20. “Zur Kritik der Hegelschen Straftrechtstheorie,” Archiv fu¨r Rechts- und Sozialphilosophie, 54 (1968), 539–48. Hegels Straftheorie (Berlin: Duncker und Humblot, 1975). Fletcher, George P. Basic concepts of criminal law (Oxford: Oxford University Press, 1998). Forster, Wolfgang. Karl Christian Friedrich Krauses fru¨he Rechtsphilosophie und ihr geistesgeschichtlicher Hintergrund (Ebelsbach: Aktiv Druck und Verlag, 2000). Foucault, Michel. Discipline and punish: the birth of the prison, trans. Alan Sheridan, second edition (New York: Random House, 1995). Garapon, Antoine. Des crimes qu’on ne peut ni punir ni pardonner (Paris: Editions Odile Jacob, 2002). Guillarme, Bertrand. Penser la peine (Paris: Presses Universitaires de Paris, 2003). Gutmann, Amy and Thompson, Dennis. Why deliberative democracy? (Princeton: Princeton University Press, 2004).

200

bibliography

Hampton, Jean and Murphy, Jeffrie. Forgiveness and mercy (Cambridge: Cambridge University Press, 1988). Hart, H. L. A. Punishment and responsibility (Oxford: Clarendon Press, 1968). Hassemer, Winfried. Einfu¨hrung in die Grundlagen des Strafrechts, second edition (Munich: C. H. Beck, 1990). Hill, Thomas E. “Kant on punishment: a coherent mix of deterrence and retribution,” Annual Review of Law and Ethics, 5 (1997), 291–314. “Kant on wrongdoing, desert, and punishment,” Law and Philosophy, 18 (1999), 407–41. Human welfare and moral worth: Kantian perspectives (Oxford: Clarendon Press, 2002). Ho¨ffe, Otfried. Gibt es ein interkulturelles Strafrecht? Ein philosophischer Versuch (Frankfurt a.M.: Suhrkamp, 1999). (ed.). Immanuel Kant: Metaphysische Anfangsgru¨nde der Rechtslehre (Berlin: De Gruyter, 1999). “Vom Straf- und Begnadigungsrecht,” in O. Ho¨ffe (ed.), Immanuel Kant: metaphysische Anfangsgru¨nde der Rechtslehre (Berlin: De Gruyter, 1999), pp. 213–33. “Ko¨nigliche Vo¨lker”: zu Kants kosmopolitischer Rechts- und Friedenstheorie (Frankfurt a.M.: Suhrkamp, 2001). Categorical principles of law: a counterpoint to modernity (University Park: Pennsylvania State University Press, 2002). Gerechtigkeit: eine philosophische Einfu¨hrung (Munich: C. H. Beck, 2004). Holtman, Sarah. “Toward social reform: Kant’s penal theory reinterpreted,” Utilitas, 9 (1997), 3–21. Honderich, Ted. Punishment: the supposed justifications (New York: Harcourt, Brace and World, 1969). Honneth, Axel. The struggle for recognition: the moral grammar of social conflicts, trans. Joel Anderson (Cambridge, Mass.: MIT Press, 1996). Ho¨sle, Vittorio. “Was darf und was soll der Staat bestrafen?”, in Ho¨sle, Rechtsphilosophie des deutschen Idealismus (Hamburg: F. Meiner, 1989). Kaufmann, Matthias. Rechtsphilosophie (Freiburg i.Br.: K. Alber, 1996). “Zwangsrecht (}} 13–16),” in Jean-Christophe Merle (ed.), Johann Gottlieb Fichte: Grundlage des Naturrechts (Berlin: Akademie Verlag, 2001), pp. 125–37. Kersting, Wolfgang. “Sittengesetz und Rechtsgesetz: die Begrundung des Rechts bei Kant und den fru¨hen Kantianern,” in Reinhard Brandt (ed.), Rechtsphilosophie der Aufkla¨rung: Symposium Wolfenbu¨ttel 1981 (Berlin: De Gruyter, 1982), pp. 147–77. Wohlgeordnete Freiheit: Immanuel Kants Rechts- und Staatsphilosophie, second edition (Berlin: De Gruyter, 1993). Kleinig, John. “Punishment and moral seriousness,” Israel Law Review, 25, no. 3 (1991), 401–21. Klug, Ulrich. “Abschied von Kant und Hegel,” in Ju¨rgen Baumann (ed.), Programm fu¨r ein neues Strafgesetzbuch: der Alternativ-Entwurf der Strafrechtslehrer (Frankfurt a.M.: Fischer, 1968), pp. 36–41. Koller, Peter. “Probleme der utilitaristischen Strafrechtfertigung,” Zeitschrift fu¨r die Gesamte Strafrechtswissenschaft, 91 (1979), 45–95.

bibliography

201

Ko¨stlin, Christian R. Neue Revision der Grundbegriffe des Kriminalrechts, reprint (Aalen: Scientia, 1970). Ku¨hl, Kristian. Die Bedeutung der Rechtsphilosophie fu¨r das Strafrecht (Baden-Baden: Nomos, 2001). Landau, Peter. “Karl Christian Friedrich Krauses Rechtsphilosophie,” in KlausMichael Kodalle (ed.), Karl Christian Friedrich Krause (1781–1832): Studien zu seiner Philosophie und zum Krausismo (Hamburg: Meiner, 1985), pp. 80–92. Lazzari, Alessandro. “Eine Fessel, die nicht schmerzt und nicht sehr hindert: Strafrecht,” in Jean-Christophe Merle (ed.), Johann Gottlieb Fichte: Grundlage des Naturrechts (Berlin: Akademie Verlag, 2001), pp. 173–86. Lesch, Heiko Hartmut. Der Verbrechensbegriff: Grundlinien einer funktionalen Revision (Cologne: Heymanns, 1999). Liszt, Franz von. Der Zweckgedanke im Strafrecht (Berlin, 1883). Ludwig, Bernd. Kants Rechtslehre (Hamburg: F. Meiner, 1988). May, Larry. Crimes against humanity: a normative account (Cambridge: Cambridge University Press, 2006). McTaggart, John E. Studies in Hegelian cosmology (Cambridge: Cambridge University Press, 1918). Merle, Jean-Christophe. “Il punto di vista educativo e religioso dei Contributi destinati a rettificare il giudizio del pubblico sulla Rivoluzione francese: la dimensione politica del Saggio di una critica di ogni rivelazione,” in Aldo Masullo and Marco Ivaldo (eds.), Filosofia trascendentale e destinazione etica (Milan: Guerini, 1995), pp. 303–25. Justice et progre`s: contribution a` une doctrine du droit e´conomique et social (Paris: Presses Universitaires de France, 1997). (ed.). “Eigentumsrecht,” in Merle (ed.), Johann Gottlieb Fichte: Grundlage des Naturrechts (Berlin: Akademie Verlag, 2001), pp. 159–72. Mohr, Georg. “Unrecht und Strafe,” in Ludwig Siep (ed.), G. W. F. Hegel: Grundlinien der Philosophie des Rechts (Berlin: De Gruyter, 1997), pp. 95–124. Murphy, Jeffrie G. “Does Kant have a theory of punishment?” Columbia Law Review, 87 (1987), 509–32. Neumann, Ulfried and Schroth, Ulrich. Neuere Theorien von Kriminalita¨t und Strafe (Darmstadt: Wissenschaftliche Buchgesellschaft, 1980). Ottmann, Henning. Philosophie und Politik bei Nietzsche (Berlin: De Gruyter, 1987). Pelczynski, Zbigniew A. (ed.). Hegel’s political philosophy (Cambridge: Cambridge University Press, 1971). Piontkowski, Andrei A. Hegels Lehre u¨ber Staat und Recht und seine Strafrechtstheorie, trans. Anna Neuland (Berlin: De Gruyter, 1960). Pogge, Thomas W. “Is Kant’s Rechtslehre comprehensive?” Southern Journal of Philosophy, 36, supplement (1997), 161–87. “Is Kant’s Rechtslehre a ‘comprehensive liberalism’?” in Mark Timmons (ed.), Kant’s Metaphysics of morals (Oxford: Oxford University Press, 2002), pp. 133–58. Primorac, Igor. “Punishment as the criminal right,” Hegel-Studien, 15 (1980), 187–98.

202

bibliography

Primoratz, Igor. “Banquos Geist: Hegels Theorie der Strafe,” Hegel-Studien, 29, supplement (1986). Pugsley, Robert A. “A retributivist argument against capital punishment,” Hofstra Law Review, 9 (1981), 1501–23. Quinton, Anthony M. “On punishment,” in Harry Burrows Acton (ed.), The philosophy of punishment (London: Macmillan, 1969), pp. 55–64. Radbruch, Gustav. Rechtsphilosophie, ed. Erik Wolf, seventh edition (Stuttgart: K. F. Koehler, 1970). Ricœur, Paul. Le Juste (Paris: Editions Esprit, 1995). Ripstein, Arthur. Equality, responsibility and the law (Cambridge: Cambridge University Press, 1999). Rosen, Fred. “Utilitarianism and the punishment of the innocent: the origins of a false doctrine,” Utilitas, 9, no. 1 (March 1997), 23–37. Roxin, Claus. Strafrechtliche Grundlagenprobleme (Berlin: De Gruyter, 1973). “‘Schuld’ und ‘Verantwortlichkeit’ als strafrechtliche Systemkategorien,” in Roxin(ed.), Grundfragen der gesamten Strafrechtswissenschaft (Berlin: De Gruyter, 1974), pp. 171–97. Scala, Klaus. “Hegels Begriff der Strafe und die moderne Strafvollzugsproblematik,” Hegel-Jahrbuch (1987), 164–70. Scharf, Michael P. “Justice vs. peace,” in Sarah B. Sewall and Carl Kaysen (eds.), The United States and the International Criminal Court: national security and international law (Lanham, Md.: Rowman and Littlefield, 2000), pp. 213–36. Scheid, Don E. “Kant’s retributivism,” Ethics, 93 (1983), 262–82. Schild, Wolfgang. “Ende und Zukunft des Staatsrechts,” Archiv fu¨r Rechts- und Sozialphilosophie, 70 (1984), 71–112. Schwarzschild, Steven S. “Kantianism and the death penalty,” Archiv fu¨r Rechts- und Sozialphilosophie, 70 (1985), 343–77. Seelmann, Kurt. “Versuch einer Legitimation von Strafe durch das Argument selbstwiderspru¨chlichen Verhaltens des Strafta¨ters,” Jahrbuch fu¨r Recht und Ethik, 1 (1993), 315–26. “Wechselseitige Anerkennung und Unrecht: Strafe als Postulat der Gerechtigkeit?” Archiv fu¨r Rechts- und Sozialphilosophie, 79, no. 2 (1993), 228–36. Shue, Henry. “Torture,” Philosophy and Public Affairs, 7, no. 2 (1977–88), 124–43. Sorell, Tom. Moral theory and capital punishment (Oxford: Blackwell, 1987). Stillman, Peter G. “Hegel’s idea of punishment,” Journal of the History of Philosophy, 14 (1976), 169–82. Tunick, Mark. Hegel’s political philosophy: interpreting the practice of legal punishment (Princeton: Princeton University Press, 1992). van den Haag, Ernest. “Why capital punishment?” Albany Law Review, 54, nos. 3–4 (1990), 501–14. Willascheck, Marcus. “Why the Doctrine of Right does not belong in the Metaphysics of Morals,” Annual Review of Law and Ethics, 5 (1997), 205–27. Wolf, Jean-Claude. Verhu¨tung oder Vergeltung? Einfu¨hrung in ethische Straftheorien (Freiburg i.Br.: Alber, 1992).

bibliography

203

“Strafe als Widerherstellung eines Gleichgewichts,” Jahrbuch fu¨r Recht und Ethik, 11 (2003), 199–216. Wood, Allen W. Hegel’s ethical thought (Cambridge: Cambridge University Press, 1990). “Kant’s Doctrine of right,” Introduction to Otfried Ho¨ffe (ed.), Immanuel Kant: Metaphysische Anfangsgru¨nde der Rechtslehre (Berlin: Akademie Verlag, 1999), pp. 19–39. Zaczyk, Rainer. Das Strafrecht in der Rechtslehre J. G. Fichtes (Berlin: Duncker und Humblot, 1981).

INDEX

abolitionism, 20; see also impunity acquittal of a debt, see debt adiaphoron, moral, 27 agreement, 21 Altman, Andrew, 186 Arendt, Hannah, 183 Aristotle, 21, 100 Ataner, Attila, 59 attempt, 92, 190 autonomy of the will, see freedom of action Badinter, Robert, 182 banishment, 178, 179; see also exclusion from society Beccaria, Cesare, 2, 18, 20, 70, 90–1, 164, 194 Benn, Stanley, 108 Bentham, Jeremy, 18–19, 20, 164 bestiality, 42 Bianchi, Herman, 1, 158 Bosanquet, Bernard, 107 Braithwaite, John, 5, 7, 13 Brandt, Reinhard, 59, 103, 193 Brugger, Winfried, 2 burden of proof, 69 Byrd, Sharon B., 44, 45–9 Camus, Albert, 196 cancellation (Aufhebung), 115, 124, 126, 129, 134–5, 136–8, 160 categorical imperative, 25–43, 58–9, 72, 90, 119, 130, 194, 195 child, 64–5, 66, 67 Christianity, 7 Cicero, Marcus Tullius, 100 Clark, Michael, 102 coercion, 30–1, 33, 38, 39–41, 42, 46, 67, 74, 91–4, 125, 128–45 Cohen, Hermann, 59–60

Commission for Truth and Reconciliation, 183 common sense, 113, 122–4, 143, 149, 184–5 compensation, 21, 54–5, 73, 92, 127, 136–7, 151, 152–3, 155, 156–8, 161, 192 conscience, bad, see remorse consciousness, common, see common sense consent, 113, 114 consequences, internal to the act, 138–9, 143–4 contract, 29–31, 54, 152 contribution, 57 conversion, 79–80 Cooper, David, 129 corporal punishment, see torture crime against humanity, 149 cruelty, 153–4, 160, 167, 168 death penalty, 1, 5, 20, 31, 47–8, 49–50, 53, 54–5, 59–60, 60–1, 68, 70, 94, 95, 106, 117, 133–4, 157, 169, 178–9, 182, 184, 185, 193, 196 debt, 21, 54, 74, 150, 152–3, 156, 157, 159–61, 167, 168 Deith, John, 80 desert, see merit deterrence: general, 2–3, 8, 18, 19, 57–71, 94, 97–8, 101–2, 104, 107, 110, 119, 135, 138, 140–1, 177–8, 180–1, 181–2, 191; specific, 4 Du¨hring, Karl Eugen, 150, 161–4 duty: of right, 32–3; of virtue, Ebbinghaus, Julius, 26 enslavement, see slavery Erdemovic´, Drazen, 180 evil, see wickedness

204

index example, deterrence through, 93, 95, 104, 138 exclusion from society, 94, 96, 99, 100, 102–3, 105, 153–4, 167; see also banishment expiation contract, 94, 99, 102 expressivism, 72, 175 Feinberg, Joel, 191 Feuerbach, Anselm, 2, 142 Fichte, Johann Gottlieb, 14, 42, 83, 85–106, 116–18, 134, 144, 153, 178, 188 Fichtelberg, Aaron, 178 Flechtheim, Ossip K., 108, 111–13 Fletcher, George P., 167 force, 132–4 forgiveness, 78, 81 Forster, Wolfgang, 105, 106 Foucault, Michel, 1, 156 freedom of action, 195 freedom of the will, 35, 62–4, 65 Garapon, Antoine, 173 goal of punishment, 72–3, 162, 164, 165, 189, 191 Grotius, Hugo, 19, 21, 191 Grundgesetz (German Basic Law), 176 Guillarme, Bertrand, 111 guilt, 150, 175, 183, 191 Gutmann, Amy, 185 Hampton, Jean, 72, 127 happiness, 55, 75–6, 77–8, 80 harm principle, 37 Hart, H. L. A., 70, 73, 82 Hassemer, Winfried, 4, 5, 10, 107 Hegel, Georg Wilhelm Friedrich, 10, 13, 14, 42, 83, 85, 105, 107–45, 149, 155, 164, 188 highest good, 55, 75–6, 90 Hill, Thomas E., 44, 55, 72, 80, 149 Hitler, Adolf, 182, 184 Hobbes, Thomas, 2, 18, 19, 151–2 Ho¨ffe, Otfried, 4, 6, 25, 34, 44, 51, 56, 107, 108, 195 Hohfeld, Wesley Newcomb, 92 Holtman, Sarah, 44 Honderich, Ted, 108 Honneth, Axel, 126 Ho¨sle, Vittorio, 42 Hufeland, Gottlieb, 26 humanity, crime against, see crime against humanity humanity as an end in itself, 32–3, 40, 46, 61, 64, 93, 119 humiliation, 157

205

impunity, 158–60; see also abolitionism imputability, 62–4 incapacitation, 3, 19, 71, 80–1, 94, 104, 176 incorrigible prisoners, 94–5 innocents, punishment of the, 19, 22, 48, 74 insight of the criminal, 114, 145 internalization, 147, 151, 168 International Criminal Court (ICC), 171–5, 179 International Criminal Tribunal for Former Yugoslavia (ICTFY), 180 intervention, military humanitarian, 177 ius strictum, 137 Jhering, Rudolf von, 164 Joseph II, Holy Roman Emperor, 133 justice: commutative, 21, 151, 159–62, 164–5, 194; corrective, 135; distributive, 21, 174; Hobbesian, 152; in general, 169; natural law, 169 Kant, Immanuel, 10, 13–14, 21–3, 24–43, 44–71, 72–83, 90, 92, 101, 118–20, 136, 149, 181, 187–8, 191, 192–6 Kaufmann, Matthias, 88, 108 Kersting, Wolfgang, 25, 26, 28, 39–41 Kerve´gan, Jean-Franc¸ois, 108 Klein, Ernst Ferdinand, 132, 134, 135, 142 Kleinig, John, 127 Klug, Ulrich, 107 Koller, Peter, 3 Ko¨stlin, Christian Reinhard, 107 Krause, Karl Christian Friedrich, 105–6 Ku¨hl, Kristian, 3, 10, 30 Landau, Peter, 3, 106 Lazzari, Alessandro, 88 Lesch, Heiko H., 110 liberalism, 25, 34–43, 106 Liszt, Franz von, 164 Locke, John, 34 Ludwig, Bernd, 28, 29 lynching, 168 Mably, Gabriel Bonnot de, 156 McTaggart, John Ellis, 107 marriage, 42 maxim, 28, 33, 56, 75 May, Larry, 172, 175, 183, 184 mercy, 160–1, 167 merit, 74, 91 Merle, Jean-Christophe, 43, 87, 90, 195 Mill, John Stuart, 20, 34, 37 Milosevic, Slobodan, 186 Mitnick, Kevin, 69 mixed theories, 5, 12, 44–53, 88–9, 104, 107–20, 169

206

index

mnemotechnic, 153, 169 Mohr, Georg, 108–9, 118–19 monstrous crimes, 175, 183 morality, 23–4, 25–34, 38, 42, 44, 45, 142–3 Moses, 19 Murphy, Jeffrie G., 72 natural law, 18, 169, 176, 191 necessity, see right of necessity negation: of the crime, 107–45; of the negation, 112, 122 Neumann, Ulfried, 6, 9 Nietzsche, Friedrich, 14, 147–70, 188 nulla poena sine lege, 171 nullity of crime, 115, 116, 117–18, 129, 137 Nuremberg, tribunal of, 171, 179, 180 Ottmann, Henning, 151 pardon, 31, 68, 159, 167, 179, 181 parole, release on, 69 Paul, Saint, 80 peace, civil, 163, 193; see also war, civil permissive law, 103, 193 personality, 61–4 Pettit, Philip, 13 Pfaelzer, Marianna, 69 Piontkowski, Andrei A., 107 Plato, 17, 38, 73, 116, 164 Pogge, Thomas W., 34, 62, 65, 66 positivism, legal, 38 Primorac (Primoratz), Igor, 108, 111, 115, 122 promise, 29–31, 152, 161 property right, 41, 60, 65 proportionality principle, 5, 19–20, 21, 43, 45, 49, 70, 74, 82, 113, 120–2, 124, 127, 159, 168, 174, 183, 184, 189 Pufendorf, Samuel von, 2, 19, 21, 191 Pugsley, Robert A., 59 Quinton, Anthony M., 108 Radbruch, Gustav, 194 recidivism, 9, 10, 13, 57, 68, 69, 95, 147, 165, 181 recognition, 99–100, 116–18, 131 reconciliation, 113 reform, see rehabilitation rehabilitation, 3, 5, 7, 9, 13, 57–8, 61–71, 81, 85–106, 108, 117, 135–6, 145, 147, 167, 189, 194 remorse, 14, 147, 149, 162, 164, 166 republic, 25 resentment, see ressentiment resistance, 176

responsibility, 6–7, 13 ressentiment, 154, 161 restoration of right, 112, 113, 127–8, 194 retaliation, 6, 72, 90, 112, 115, 117, 120–1, 125–7, 141, 143–4 retribution, 6, 8, 49, 50–7, 74, 92, 127, 141, 161 retributivism, 2, 3–4, 6, 7, 21–4, 43, 51–7, 72, 81–2, 110, 119, 141, 143, 147, 181, 183, 192 revenge, 56, 78, 81–2, 114, 125–6, 135, 137, 161–2, 163; see also vengeance Ricur, Paul, 5 right, concept of, 85–7, 89, 90, 97, 99, 101, 116, 144–5 right of necessity, 46, 52, 94, 193 Ripstein, Arthur, 80, 108, 139 Rosen, Fred, 3, 74 Roxin, Claus, 7, 9, 11, 12, 13, 107, 147–8 savage, 67 Scala, Klaus, 108 Schadenfreude, 149 Scharf, Michael P., 179 Scheid, Don E., 44, 51 Schild, Wolfgang, 108 Schmalz, Theodor, 26 Schopenhauer, Arthur, 2, 107 Schroth, Ulrich, 6, 9 Schwarzschild, Steven S., 59 security, public, 95, 140, 157–9, 190, 191, 195 Seelmann, Kurt, 108, 111, 116–17, 141 self-defense, right of, 91, 137, 191 Seneca, 17, 73 Shue, Henry, 2 Sidgwick, Henry, 17, 20, 23 slavery, 56, 60–1, 66, 95 social contract, 85, 94, 102 Sorell, Tom, 60 sovereign, 91 state of nature, 52, 53, 57, 58–9, 101, 154, 181, 193 statute of limitation, 174–5, 184–6, 189 Stillman, Peter G., 108 suffering, see torture suicide, 37, 59, 61 talion law, 17, 19–20, 21–3, 51–2, 56–7, 87, 95–6, 101, 113, 121–2, 127, 142 teleology, 39–41, 65 Thompson, Dennis, 185 threat of punishment, deterrence through, 2, 46–8, 70, 93–4, 98, 104–5, 119 Tokyo, tribunal of, 171, 179

index torture, 1, 5, 20, 48, 96, 106, 128, 152–3, 154, 157, 163, 166, 168–9, 184, 185, 186 Tunick, Mark, 108 tyrannicide, 176 universalization, 56 usefulness, 99, 100–1 utilitarianism, 18, 44, 51, 68, 100 van den Haag, Ernest, 2 Vattel, Emer de, 191 vengeance, 76–7 victim, 9, 81, 92, 127, 130, 137, 155, 157, 159–60, 162, 163, 171, 173, 186, 190, 194

207

virtue, 74–6, 77 Voltaire, 170 war: civil, 156, 159, 193; international, 191–2; see also peace, civil Wellman, Christopher H., 186 wickedness, inner, 43, 55, 73–4, 75, 76, 77–8, 79, 80, 101, 102, 122, 176, 184, 190–1 Willascheck, Marcus, 34 Wolf, Jean-Claude, 55, 107 Wolff, Christian, 2, 39 Wood, Allen W., 36, 108–9, 114 worthiness of punishment, 73, 81–2 Zaczyk, Rainer, 93