PUNISHMENT AND THE HISTORY

OF POLITICAL PHILOSOPHY

From Classical Republicanism to

the Crisis of Modern Criminal Justice
This page intentionally left blank
Punishment and the History
of Political Philosophy
From Classical Republicanism to
the Crisis of Modern Criminal Justice

ARTHUR SHUSTER

UNIVERSITY OF TORONTO PRESS

Toronto Buffalo London
 © University of Toronto Press 2016
Toronto Buffalo London
www.utppublishing.com
Printed in the U.S.A.

ISBN 978­-1­-4426­-4728­-2

Printed on acid­-free, 100% post­-consumer recycled paper

with vegetable­-based inks.

Library and Archives Canada Cataloguing in Publication

Shuster, Arthur, 1980–, author

Punishment and the history of political philosophy : from classical
republicanism to the crisis of modern criminal justice / Arthur Shuster.

Includes bibliographical references.

ISBN 978-1-4426-4728-2 (bound)
1. Punishment – Philosophy.  2. Criminal justice, Administration
of – Philosophy.  3. Punishment – Moral and ethical aspects.
4. Punishment – Political aspects.  I. Title.

K5103.S58 2016   364.601   C2015-907071-6

This book has been published with the help of a grant from the Federation
for the Humanities and Social Sciences, through the Awards to Scholarly
Publications Program, using funds provided by the Social Sciences and
Humanities Research Council of Canada.

University of Toronto Press acknowledges the financial assistance to its

publishing program of the Canada Council for the Arts and the Ontario
Arts Council, an agency of the Government of Ontario.

Funded by the Financé par le

Government gouvernement
of Canada du Canada
To my parents
This page intentionally left blank
 Contents

Acknowledgments  ix

Introduction  3

1  The Problem of Punishment and the Limits of Reform

in Plato’s Laws  15

2  Modern Natural Right and Punishment in Hobbes’s Leviathan  51

3  Liberalizing the Criminal Law: Montesquieu and Beccaria  75

4  Retribution and Individual Autonomy in Kant’s Rechtslehre  98

5  Foucault and the Crisis of Modern Criminal Justice  116

Conclusion: Punishment and Liberalism  132

Notes  141
Bibliography  171
This page intentionally left blank
 Acknowledgments

I thank Boston College and Claremont McKenna College for their sup-
port over the years while I was completing the manuscript for this book.
This book has grown out of my doctoral dissertation at the University
of Texas at Austin, and anything of value in it I owe to my teachers
there, whose guidance and support throughout my studies, and whose
friendship since then, have been a true gift. I particularly wish to thank
Thomas and Lorraine Pangle, as well as Devin Stauffer, for sharing
with me their insights into the great questions of political philosophy,
without which this study would not have been possible. I also thank
Russell Muirhead for his valuable encouragement and sage advice.
I would also like to thank The Review of Politics for granting permis-
sion to use material (parts of chapter 4) that originally appeared in that
journal as “Kant on the Retributive Outlook in Moral and Political Life”
(Summer 2011).
A.S., August 2014
This page intentionally left blank
PUNISHMENT AND THE HISTORY
OF POLITICAL PHILOSOPHY

From Classical Republicanism to

the Crisis of Modern Criminal Justice
This page intentionally left blank
 Introduction

When Shakespeare’s tyrant Macbeth laments the meaninglessness of

human existence (as he understands it) – pronouncing it to be like “a
tale told by an idiot, […] signifying nothing” – he unwittingly reveals
to us the importance and moral significance of crime and punish-
ment. Macbeth’s haunting soliloquy is provoked by what he believes
has been a failure of cosmic justice. Early in the play, the mysterious
Weird Sisters had prophesied to him that he would gain the throne of
Scotland, which he coveted by fair means or foul. That prophecy is now
guiding (or rather, misguiding) Macbeth to his own tragic ruin, for it
has spurred him to seize the Scottish crown and thereby achieve his
tyrannical ambitions. His success proves to be short­-lived: his tyranny
is soon brought down by those who would avenge the murdered king.
He responds to these events by accusing the mysterious powers that
govern man’s fate of being false and fickle and in those two senses un-
just. We, the audience, do not share Macbeth’s bleak view of the human
condition, for unlike him, we see his defeat not as a reflection of the
cosmos’s indifference or malevolence towards human things but rather
as the fulfilment of justice: Macbeth had been tempted by the witches’
prophecy to commit his crimes, but he had also been deceived by that
prophecy and led unwittingly to his ultimate punishment. The justice
of the prophecy, and of Macbeth’s fall, seems to be confirmed by the
fact that Macbeth is killed by one of his victims: the pious Macduff. By
the play’s end, the criminal has suffered a penalty proportionate to the
gravity of his crime, his victims have been avenged, and justice seems
to have been satisfied.
Is it possible that the larger teaching of Shakespeare’s Macbeth is that
the condign punishment of crime is intimately linked with our basic
4 Punishment and the History of Political Philosophy

moral self­-understanding? If Shakespeare’s play had ended differently –

if Macbeth’s tyranny had gone unpunished – would we then have fallen
prey to Macbeth’s despair at the meaninglessness of a world devoid of
justice? Before Macbeth suffers his punishment, at a point in the play
where success for him still seems possible, it is Lady Macbeth who de-
scends into despair, then madness, and eventually commits suicide. Her
madness is brought about by her inability to live with her unexpunged
guilt (“What, will these hands ne’er be clean?”). The Macbeths’ despair
is to some extent echoed by Hannah Arendt’s struggle to come to terms
with “the banality of evil.” This striking and provocative formulation
was coined by Arendt in order to describe the infamous Adolf Eichmann,
the Nazi bureaucrat involved in administering German concentration
camps during the Second World War, who was later captured and tried
in Jerusalem. In Eichmann in Jerusalem, her report on the trial, Arendt de-
scribes Eichmann’s actions as those of a morally obtuse, petty bureaucratic
functionary incapable of appreciating the gravity of the evil he was com-
mitting. This is a source of anguish for her, for it implies that the greatest
evils perpetrated by human beings on one another may in the last analy-
sis be “banal” and thus meaningless in the sense expressed by Macbeth in
his soliloquy. In this way, at least in the context of Shakespeare’s dramatic
expression of our experience, punishment of the unjust appears as a kind
of moral necessity – required for its own sake, beyond any merely utilitar-
ian value of the penalty as a deterrent of future crime.
Yet this poetic account of our experience, powerful as it may be, is not
completely satisfying. We rightly wonder whether this poetic vision of
the moral necessity of punishment can be given systematic theoretical
articulation and grounding. We wonder how we might distinguish our
feelings of righteous indignation from the irrational, impulsive anger
that is also found in animals. (We ourselves may experience such anger
whenever we fly into a momentary rage at a chair or a door we have
bumped into.) Furthermore, we feel an urgent need to try to clarify
what we mean when we say that justice demands restoration of a right-
ful “balance” and that punishment (as distinguished from some other
intervention) is needed to accomplish this. Equally puzzling is the relat-
ed question of what determines the “proportionality” of a punishment.
We may be troubled, as well, by the complexities introduced into our
thinking about crime and punishment by certain sceptical reflections
about the nature of choice and moral responsibility, for careful study
of the human condition reveals many ways in which our choices are
constrained by forces – both external and internal – beyond our control.
 Introduction 5

The present study investigates these basic and enduring questions

about crime and punishment, and its approach is normative rather than
criminological, sociological, or historical (although criminological and
historical issues are considered when appropriate). This study is “nor-
mative” in two senses. First, it is normative because it is fundamentally
concerned with the question of the purpose or end of punishment and
with its moral justification. Second, it is normative in the sense that it
situates itself within the tradition of Western political philosophy and
connects the theme of punishment with the larger themes of virtue, law,
and the good society. This approach has the advantage of being able to
tackle directly the questions regarding punishment that we have just
raised. Insofar as criminology, sociology, and historical studies adopt a
“value­-neutral” approach to crime and punishment, aspiring to objec-
tivity, they are inclined to dismiss ordinary moral experience (such as
the experience of Shakespeare’s Macbeth) as supposedly incapable of
offering any real insights into the truth of these phenomena. This pres-
ent study assumes, on the contrary, that it is impossible to begin to un-
derstand the human significance (the significance for human beings) of
crime and punishment without paying careful attention to what we say,
think, and feel about our ordinary (i.e., not disinterested) experience in
these matters.
This is, of course, not the first time that these troubling questions
and difficulties surrounding crime and punishment have been raised.
Nonetheless, they must be raised anew, and addressed thoroughly, be-
cause it is hardly an exaggeration to say that criminal justice is today in
a state of crisis. Most readers will already be aware of various political
and legal controversies indicating that a number of aspects of modern
criminal justice systems have come under attack. The most common
target of criticism is the heavy reliance of those systems on mass in-
carceration. Large and costly prison complexes are now warehousing
thousands of inmates serving long sentences, so that the total prison
population of a modern nation can reach the hundreds of thousands,
in some cases even millions. Many observers are troubled by the sheer
enormity of these numbers; some go even further, accusing modern
prisons of being “schools of crime” that not only fail to rehabilitate and
deter but also actually “harden” criminals, who are being forced to as-
sociate only with their own like, in violent and sometimes brutal condi-
tions. This critique of modern prisons was made most pointedly and
radically by Michel Foucault in Discipline and Punish (1975) (discussed
at length in chapter 5), which continues to inspire many “abolitionists.”
6 Punishment and the History of Political Philosophy

Modern criminal justice systems are also being attacked for their sup-
posed neglect of the particular needs and circumstances of individual of-
fenders, their victims, and local communities. Indeed, correctional policy
is predominantly under the jurisdiction of remote government agencies
(federal, state, or provincial, as the case may be) and is administered by
those agencies according to uniform penal codes that cannot possibly
take into account the wide variety of local, let alone individual, factors.
This state of affairs is criticized for yielding poor results overall: pun-
ishments meted out are in some cases too harsh (to deter future crime
or satisfy victims), in others too lenient; local communities are deprived
of their autonomy in a vital function of self­-government; and as a con-
sequence of the latter, the bonds of community disrupted by crime are
not repaired.1 In response to this problem, in the past thirty years a “re-
storative justice” movement has emerged in Canada, the United States,
Australia, and other places whose principal mission is to challenge the
modern state’s monopoly over the administration of criminal justice.2
A third aspect of the crisis of modern criminal justice is the vexed
controversy that has developed regarding whether it is excessively re-
tributive and whether it ought to be retributive at all. In the 1970s the
US Supreme Court wrestled with this issue when it was faced with the
question of the constitutionality of the death penalty. In 1972, in Furman
v. Georgia, the Court placed a moratorium on the death penalty. The
question before the Court was whether the imposition of the death pen-
alty is “cruel and unusual” and thus beyond the constitutional power
of the state by virtue of the Eighth and Fourteenth Amendments of the
US Constitution.3 The five justices who made up the majority of the
Court agreed that it was, but only on the narrow ground that Georgia’s
penal laws were causing the death penalty to be imposed in an irregu-
lar and arbitrary manner; but the justices could not agree on whether
the death penalty was unconstitutional in principle. The justices’ opin-
ions revealed their awareness that whether the death penalty is “cruel
and unusual” in principle depends on what makes punishment just or
moral. And this was precisely what they could not agree on.
Perhaps the most interesting of the opinions in Furman is that of
Justice William Brennan, who notes that the Eighth Amendment’s pro-
hibition of “cruel and unusual punishments” does not define “cruel and
unusual.” If the Court is to apply the clause faithfully, it needs a princi-
pled method for determining which specific punishments fall within the
clause’s prohibition in a way that reflects the broad moral commitments
implicit in the clause itself and in the Constitution. But what exactly are
 Introduction 7

those broad moral commitments? At bottom, Brennan believes that the

“cruel and unusual” clause prohibits the infliction of punishments that
are degrading to human dignity.4 Brennan goes on to conclude that the
death penalty does indeed degrade the individual, since according to
him the prospect of an (otherwise unnecessary) death is so physically
and psychologically painful that it makes normal human functioning
impossible in the time prior to and during the carrying out of the sen-
tence; thus, it falls under the Eighth Amendment’s prohibition.5
Yet one might still wonder: Could depriving one individual of his or
her basic dignity be morally justified by society’s need for security, if it
were found that the death penalty was indispensable to effective deter-
rence of certain crimes? Moreover, could it not be argued (as some re-
tributivists do) that in some of the worst cases, death and nothing less is
deserved because of the gravity of the crime, and that to deny this pun-
ishment would amount to denying victims, and society as a whole, the
dignity belonging to them? Finally, it might be maintained that the worst
criminal has already degraded himself to the point that all a decent soci-
ety can do is to put him out of his misery, as an act of euthanasia.
Brennan himself acknowledges the existence of the possible objec-
tions we have just outlined, and his admission is illuminating:

From the beginning of our Nation, the punishment of death has stirred
acute public controversy. Although pragmatic arguments for and against
the punishment have been frequently advanced, this long standing and
heated controversy cannot be explained solely as the result of difference
over the practical wisdom of a particular government policy. At bottom,
the battle has been waged on moral grounds. The country has debated
whether a society for which the dignity of the individual is the supreme
value can, without a fundamental inconsistency, follow the practice of de-
liberately putting some of its members to death. In the United States as in
other nations in the world, the struggle about this punishment has been
between ancient and deeply rooted beliefs in retribution, atonement or
vengeance, on the one hand, and on the other, beliefs in the personal value
and dignity of the common man that were born of the democratic move-
ment of the eighteenth century, as well as beliefs in the scientific approach
to an understanding of the motive forces of human conduct, which are the
result of the growth of the sciences of behavior during the nineteenth and
twentieth centuries. It is this essentially moral conflict that forms the back-
drop for the past changes in, and present operation of, our system of im-
posing death as a punishment for crime.6
8 Punishment and the History of Political Philosophy

Although throughout most of his opinion Brennan presents as sufficient

and uncontroversial his standard – human dignity – for distinguishing
between just and unjust punishments, here he acknowledges the exis-
tence of a more fundamental controversy over the end that punishment
ought to serve. Whether or not the US Constitution’s protection of hu-
man dignity against “cruel and unusual” punishment rules out the
death penalty crucially depends on whether deterrence, retribution, or
some other goal (or a combination of certain goals) is the politically (and
constitutionally) valid end of punishment. For example, if the proper
end were general security, and if the death penalty were necessary to
effective deterrence, then it could not easily be argued that this penalty
is “cruel,” since in this case the survival of the political community
would justify this punishment (without denying that it is lamentable in
other respects). If, on the other hand, the proper aim of punishment
were retribution, then the death penalty might be justified (and thus no
longer held “cruel” in principle) as the proportional and appropriate
penalty for such serious crimes as murder. In short, commitment to de-
mocracy and individual dignity is not by itself sufficient to decide the
question of punishment.
While Brennan’s opinion shows his awareness of the existence of
the fundamental moral dispute about the proper aim of punishment, it
gives little indication as to how this dispute might be decided, and on
what grounds. His final position is that, historically, the death penalty
has been used less and less in democratic societies (although it is still
used in some), which indicates to him that the fundamental moral dis-
pute will eventually be decided against this practice. In other words, he
appears to trust in historical progress, and specifically in the progress
of general moral opinion from ignorance to wisdom (as distinguished
from technological progress). The reasonableness of this hopeful as-
sumption may be questioned, and at any rate it does nothing to ad-
dress the fundamental moral disagreement over the aim of punishment
that still exists in fact (as Brennan himself admits). That disagreement
cannot therefore be elided by reference to a possible future historical
consensus, especially since, according to Brennan, it has existed “from
the beginning of our [the US] Nation.”
That this disagreement exists became still more obvious in the four
years between Furman v. Georgia and Gregg v. Georgia (the case that lifted
the moratorium on the death penalty), when Congress and more than
thirty­-five state legislatures re­-enacted the death penalty for one or more
crimes. Partly in light of this legislative response to Furman, the plurality
 Introduction 9

opinion in Gregg acknowledged that retribution is one of the morally ac-

ceptable purposes of punishment, since this appeared to be the view of
a large portion of the country and of its representatives.7 Neither Furman
nor Gregg nor (to my knowledge) any of the Supreme Court’s major
Eighth Amendment decisions that followed have resolved the funda-
mental moral disagreement over the end of punishment – a disagree-
ment that all sides admit exists in the United States and other countries.
Present­-day scholarship has made some attempts to address funda-
mental questions regarding the morality of punishment such as those
raised by Justice Brennan’s opinion, Foucault’s book, and restorative
justice advocates. An important recent development in punishment
scholarship has been a resurgence of interest in retribution. This interest
has largely resulted from the perceived inadequacy of utilitarian punish-
ment theory, which had been predominant in the preceding generation
of scholarship. Critics observed that this theory is reductionist (and thus
unpersuasive philosophically)8 and that furthermore, its logic – which
aims to secure the greatest happiness of the greatest number – has an
alarmingly “paternalistic” tendency.9 Utilitarian punishment theories
tend to view crime as analogous to disease and to view criminals as
just a particularly dangerous class of ill human beings (although they
are not logically committed to this position). From this point of view,
the special expertise of criminologists, psychologists, and bureaucratic
administrators entitles them – not judges or juries – to dictate the na-
ture and duration of treatment (“punishment”) that offenders should
receive, according to the experts’ calculations of maximum social util-
ity. Since offenders may be pathologically incapable of making the right
choices, and may even resist attempts to cure them – and, at any rate,
since private interests may be disregarded whenever they conflict with
the requirements of social utility – criminal justice experts are entitled to
infringe on the autonomy of inmates and perhaps even on some of their
rights. Finally, guilt, innocence, and desert would lose their conventional
meaning according to this view.
Because such consequences of the utilitarian approach were unac-
ceptable to its critics, it was suggested in much of the literature that
punishment ought to aim not only at crime prevention but also at fair-
ness, proportionality, and retribution – although the precise meanings
of these commonsense notions were not always clear and much of the
“retributivist” punishment scholarship that followed was concerned
with defining these concepts and bringing them into agreement with
liberal principles. It was hoped that in this way, the coercion to which
10 Punishment and the History of Political Philosophy

convicted offenders were judged to be liable would be limited to what

they deserved, and to no more than this, regardless of how much util-
ity might accrue to society (or even to the offender himself) from an
increase in the duration or severity of punishment.
Some retributivists further argued that besides placing a limit on pe-
nal severity, retribution satisfies justice itself, which demands the equal
distribution of the burdens and benefits of civil society to all individual
members. Thus whenever one person refuses to bear his share of the
burden, which he does by violating the law, retribution is called for in
order to re­-establish the just equilibrium.10 Some retributivists also ar-
gued for retribution on the ground that it performs the important func-
tion of expressing society’s collective view of what is right and wrong,
and that penalties should be designed with the aim of communicat-
ing this view, even if this should diminish somewhat their deterrent or
rehabilitative effects.11 When the strict application of the retributivist
principle was challenged in scholarly debates, its defence by adherents
proved less than adequate.12 In response to this, another view became
prominent in the scholarly literature that was a “mixed,” or “partial”
retributivist theory of punishment inspired by the earlier work of John
Rawls and that of H.L.A. Hart, each of whom had called for a theory of
punishment that hybridized retribution and utility.13
As Rawls and Hart saw it, the problem was that both utility and ret-
ribution make legitimate moral claims upon us yet appear to be hetero-
geneous and even mutually exclusive principles. This had led some to
conclude that either one or the other principle must be true, but not both.
Against this view, Rawls and Hart argue that there is a distinction to be
made between principles that justify a practice and principles that jus-
tify particular actions falling under that practice, and that the principles
of utility and retribution fall into these two different categories, respec-
tively: the principle of utility answers the question of the fundamental
justification of punishment as a practice or institution, while retribution
justifies and regulates the administration of punishment in particular
cases. To illustrate this argument, Rawls suggests that we look at each
of the two questions as being addressed to a different public official: the
question of general justification is addressed to the legislator, who looks
to the future good, while the question of particular application is ad-
dressed to the judge, who looks backwards in time to the culpability or
innocence of the accused.14 Hart draws attention to the confused views
of actual political actors (he gives the example of a British Member of
Parliament), who will appeal at different times to different principles
 Introduction 11

to justify their favoured policy proposals. On each occasion, the official

assumes that only a single principle justifies punishment as a whole and
fails to notice that at different times he is responding to different aspects
of the institution in question – about its general purpose and about its
case­-by­-case application, respectively.15
The Rawlsian–Hartian proposed solution to the apparent tension
between retribution and utility appears to combine the best of both.
This attractive solution became influential for a time, but eventually
it too came under attack. For its critics, the Rawlsian–Hartian notion
of “mixed” retributivism seemed artificial, for it drew convenient but
spurious distinctions between questions of general justification and
questions of particular application.16 On the contrary, the critics object-
ed, retributivists are concerned precisely with the general justification of
punishment.17 As one critic argues, the fundamental problem that lib-
eral (including mixed­-retributivist) punishment theorists cannot escape
is that their attempt to justify punishment fundamentally in terms of
deterrence makes punishment a mere tax on crime, and this disagrees
with our basic moral sense that there is an important difference between
taxation and punishment.18
Thus a truly satisfying theory of punishment – one that adequately
addresses all of our basic moral concerns and questions – continues to
elude contemporary punishment theory. For this reason (and for the
other considerations raised here) it is necessary to return once more
to the fundamental questions of criminal justice. This book attempts
to do so by examining the development of the idea of criminal justice,
through its crucial transformations, in the Western tradition of political
philosophy. Thus, unlike most recent scholarly works on punishment,
this book will not provide a novel theory of punishment (in the sense of
“theory” as a programmatic set of guiding principles). Rather, it will at-
tempt to bring into sharper focus the fundamental questions of criminal
justice, in all of their complexity, in order to improve the quality of our
theorizing on this subject and in order to further elucidate the contem-
porary crisis of criminal justice.
Its main thesis (alluded to in the second part of the title) is that this cri-
sis can best be understood in light of the momentous shift in the history
of Western political thought from classical republicanism to modern in-
dividual natural right and contractarianism.19 One claim of this study is
that, unlike classical republicanism, modern natural right, in its original
formulation as the inalienable right to defend one’s life and livelihood,
necessarily implies a wholesale rejection of retributive punishment.
12 Punishment and the History of Political Philosophy

Before the full significance of this rejection can be made intelligible,

however, the meaning of retribution must first be explicated in the con-
text of classical political thought. The argument therefore proceeds first
by examining the complex classical view of punishment, as originally
elaborated in Plato’s Laws (chapter 1). This view is complex because it
contains two elements that exist in some tension with each other: the
civic­-republican commitment to retribution as a core aspect of political
justice, on the one hand, and on the other the enlightened philosophical
(Socratic) understanding of human nature, which entertains profound
doubts about the coherence of retributive punishment and of what we
would call “moral responsibility.” That chapter argues that in order to
understand properly our own concern with “just deserts,” we must
examine this notion in connection with the intensely civic spirit that
flourishes in a small participatory republic such as the ancient Greek
city­-state. My analysis in certain respects agrees with, but also signifi-
cantly departs from, the predominant “punishment as therapy” inter-
pretation of Plato’s teaching on punishment.20 My argument assumes
that the civic­-republican spirit identified by Plato, although tempered in
modern liberal­-democratic societies, nevertheless persists in some form
and continues to influence the public’s attitude towards crime and pun-
ishment. I will argue, therefore, that the logic of this civic­-republicanism,
as explicated by Plato’s Athenian Stranger in the Laws, requires retribu-
tive (non­-therapeutic) punishment and regards it as morally desirable.
The initially confusing tension in Plato’s teaching on punishment that
thereby arises – that is, the tension between its retributive and its scepti-
cal elements – reveals itself, in the course of the analysis, to be a tension
not in Plato’s thought but in political life itself.
The book proceeds, next, to examine the first theory of criminal jus-
tice in the modern rationalist tradition, elaborated by Hobbes in his
Leviathan (chapter 2). This chapter begins by analysing Hobbes’s der-
ivation of the right to punish from the fundamental natural right to
self­-preservation, and argues that this becomes the basis of all modern
punishment theories until Kant. It then discusses the aspirational or
humanizing aspect of Hobbes’s purely deterrent theory of punishment,
the rational limits of the sovereign’s punishing power, and the colli-
sion between the sovereign’s right to punish and the criminal’s right
to self­-preservation. The chapter concludes by arguing that Hobbes’s
attempt to distinguish theoretically between a punishable criminal and
an enemy in the state of nature fails. (This problem would continue
to preoccupy later punishment theorists.) While there has been a great
 Introduction 13

deal of scholarship on the penal thought of Kant and the utilitarians

(especially Beccaria and Bentham), relatively little attention has been
paid to Hobbes.21 Consequently, this chapter contributes to the growing
scholarship on Hobbes’s teaching on punishment.
The book then goes on to show how the Hobbesian idea of crimi-
nal justice came unravelled through a series of dialectical moments in
the thought of Montesquieu, Beccaria, Kant, and Foucault, respectively.
Most contemporary scholarship traces the origins of the two major al-
ternative theories of punishment – utilitarianism and retributivism – to
Beccaria and Kant, respectively. While this view is not exactly wrong, it
is potentially misleading, since it tends to obscure the fundamental con-
tinuity of the modern tradition of penal thought (broadly understood),
in which Beccaria and Kant were collaborators in spite of their obvi-
ous differences. Since Beccaria himself was an acknowledged student
of Montesquieu, the next chapter of the book (chapter 3) begins with
the crucial modifications to Hobbesian penology made by Montesquieu.
Although Montesquieu’s influence on Beccaria is well known, there
have been no sustained attempts to explicate Montesquieu’s own liberal
theory of punishment as presented in his Spirit of the Laws. This chap-
ter argues that Montesquieu’s insights into the malleability of human
nature and the effects of commerce led him to conclude that the admin-
istration of punishment in modern societies can become considerably
milder than Hobbes had anticipated. It discusses Montesquieu’s view of
the beneficial political consequences of the decriminalization of offences
against religion and mores, and of the attenuation of penal severity. The
chapter ends by examining Beccaria’s elaboration of Montesquieu’s lib-
eralizing principles of criminal law.
Modern punishment theory from Hobbes up to Beccaria never com-
pletely succeeded in expunging the appeal of retribution. I argue that
it was left up to Kant to try to reintroduce this principle into criminal
justice theory. However, he attempted to justify it on the new basis of
modern individual natural right rather than on that of classical politics.
Thus I argue (in chapter 4) that one positive accomplishment of Kant’s
teaching on punishment was to show that retribution remains a serious
concern from the point of view of ordinary moral experience, even when
that experience has been reshaped historically by the Bible and by the
idea of modern natural right. The central claim of this chapter is that
a defining moment in the development of the idea of modern criminal
justice occurs in Kant’s Rechtselehre. In that work, Kant precisely identi-
fies the threat posed to individual dignity and autonomy by modern
14 Punishment and the History of Political Philosophy

criminal justice (as articulated by Kant’s predecessors) and argues that

only the principle of retribution can protect the individual from the pa-
ternalistic tendencies of utilitarian punishment. Yet Kant’s attempt to
defend retribution on modern principles was only partly successful, I ar-
gue, and thus it left liberal criminal justice theory open to later criticism.
The final chapter (chapter 5) looks back on the development of the
idea of modern criminal justice charted in the first four chapters. It does
so in the context of a discussion of Foucault’s famous (or infamous)
critique of the modern penal system in Discipline and Punish. I defend
Foucault’s claim that the modern punishing power is in principle un-
limited (which is in agreement with Hobbes’s own account of the sover-
eign’s right to punish) and argue that in Discipline and Punish the crisis
of modern criminal justice finally comes to a head. The chapter goes on
to discuss Foucault’s critique of the penitentiary, but challenges some of
its more radical claims. By taking this historical approach, the present
book offers an alternative to the general tendency of punishment schol-
arship in recent decades – most of which is authored by legal scholars
and ethicists – to oversimplify or altogether abstract from the political
dimension of criminal justice.
In the concluding chapter, I extend my discussion of the crisis of
modern criminal justice into the present by noting certain dysfunction-
al features of actual public debates about crime, and end by suggesting
how these debates might be improved if they were informed by a better
understanding of the influence of the history of political philosophy on
the criminal justice policies and practices we have inherited.
 Chapter 1

The Problem of Punishment and the Limits

of Reform in Plato’s Laws

It is often observed that Western societies have been growing progres-

sively gentler since the dawn of the Enlightenment and the spread of
democracy, and this appears to be no less true of attitudes towards crime
and punishment. Western societies are more respectful of life, more
sympathetic to suffering, and consequently less harsh in punishing of-
fenders today than they have ever been. The matter is, however, more
complex. The passion of moral indignation aroused in any community
by crime – especially violent crime – is still a common enough occur-
rence; yet it is tempered in its effects. It seems that we allow and even
encourage ourselves to feel indignation towards offenders. Indeed, it
would be a rare person who did not feel even a little indignation in
response to the victimization of a fellow citizen, to say nothing of closer
relations. This follows naturally from the fact that we feel responsible
for (and to) our fellow citizens, neighbours, and loved ones, and there-
fore cannot help but take on some of their feelings of outrage when they
have been victimized. Accordingly, daily experience confirms that we
demand consolation for victims and assurances of the future safety of
our community; at least as emphatically, we demand “justice” for the
offenders, by which we mean some fitting and unpleasant punishment.
Yet after our initial expression of sympathy and outrage on behalf of
victims, when it comes time to put “justice” into effect, we begin to rein
in our punitive passions and qualify our demands. We scruple about
judging the offender before the due process of a legal trial; about exag-
gerating his guilt in the heat of the moment; about the harsh treatment
he might suffer at the hands of criminal justice authorities; and so on.
Provisionally, we might explain this complex response to crime in the
following way: on the one hand, our membership in a self­-governing
16 Punishment and the History of Political Philosophy

political community impels us to punitive indignation; on the other,

our humanity – our sense of ourselves as independent individuals with
private needs and fears (rather than as duty­-bound citizens) – tends to
relax our demand for “justice” and to focus our attention instead on
repairing the damage done by crime and providing for future security.
The connection, however rudimentary, that thus appears between
punishment and republicanism is echoed in the work of a prominent
contemporary advocate of the “restorative justice” movement, John
Braithwaite. A chief concern of the restorative justice movement is to
promote the reconciliation of the offending and offended parties and
the reconstitution of disrupted community bonds. Braithwaite and co-
author Philip Pettit ground this objective in republican political theory.1
They, like most other advocates of restorative justice, explicitly reject ret-
ribution, among other traditional forms of punishment. Nevertheless,
they recognize some validity in the phenomenon of moral indignation,
as evidenced by their interest in victim satisfaction (“victims’ rights”)
and in what Braithwaite calls “reintegrative shaming.”2 In order to re-
store community solidarity, restorative justice makes use of community
conferences, during which offenders are confronted with their victims
and with representatives of the community. As Braithwaite has shown,
this sort of confrontation shames offenders and brings them to express
genuine remorse for their actions. This process reintegrates offenders
and victims into a single community. Many would recognize in this pat-
tern of shaming, remorse, and reconciliation an important aim of retrib-
utive punishment, as ordinary people would understand it. Braithwaite
and Pettit’s republican approach to criminal justice sets out to satisfy a
core goal of retributive punishment even while departing from its tra-
ditional methods.
What is important for our purposes here is to consider this connection
between republicanism and punishment, and to see whether republi-
can political theory might have something to offer our understanding
of punishment that would allow us to improve on the shortcomings of
the reigning liberal punishment theory (cf. Introduction, 9–12). In the
same work, Braithwaite and Pettit echo some of the now common ob-
jections to liberal punishment theory – especially its paternalistic ten-
dency – and the republican alternative they adopt is in part a reaction
to liberalism’s failure in this regard. According to their position, “re-
publican liberty” cannot be conceived separately from civic member-
ship in a political community. Hence the republican interpretation of
liberty “equates freedom not with being left alone [as liberalism does],
 Punishment and Reform in Plato’s Laws 17

but with being given equal protection before a suitable law” (57–8). For
Braithwaite and Pettit, this definition of liberty serves as the founda-
tion for such penal policies and practices (community conferences, re-
integrative shaming) as best advance genuine community; at the same
time, it avoids the paternalistic excesses of liberal consequentialism
(such as punishment of the innocent and indeterminate sentencing).3
Yet serious questions may be raised about the passage taken by
Braithwaite and Pettit’s argument from the “republican idea” to the in-
stitutional reforms they recommend. The authors connect the idea of
equality before the law to the long republican tradition stretching back
to the classical republics, such as ancient Rome.4 This connection with
the tradition of the austere ancient republics, and with those republics’
commitment to legislating public morality, makes sense when the au-
thors endorse shaming and even legal “moralizing” as a core feature of
republican criminal justice (88). But one begins to wonder why, unlike
the historical republics, the authors stop short of imposing harsh penal-
ties to match the community’s harsh denunciation of the worst crimi-
nals. If a body of citizens can denounce – and according to Braithwaite
and Pettit ought to denounce – a murderer as having committed the
ultimate wrong against a fellow citizen, towards whom he owed a duty
of loyalty, then are we really to be surprised when those citizens de-
mand more than mere public shaming of the murderer as the condi-
tion of restoration of community solidarity?5 The authors themselves
concede that this is a potential problem in their account and suggest
that the maximum penalties established in republican law might have
to be quite severe. Yet they maintain that those harsh penalties would
never have to be used in practice – according to them, merely legislat-
ing harsh penalties without using them “may achieve the symbolic rep-
robative functions of the criminal law while toning down [its] excesses”
(176). But should we assume that an actual republican community, in-
spired by the republican idea, can be satisfied with nothing more than
mere denunciation of crime?
Thus, although Braithwaite and Pettit make the intriguing connec-
tion between republican principles and punishment as a moral phe-
nomenon, I believe they have not thoroughly weighed the qualities
of the republican spirit that would predominate in the kind of soci-
ety whose principles they articulate so well. As my exegesis of Plato’s
understanding of republican life in its full complexity will show,
Braithwaite and Pettit are mistaken in assuming that republican citi-
zens would tend to be mild in their views on punishment. It is easy for
18 Punishment and the History of Political Philosophy

us, living in a cosmopolitan, “Western” world informed by liberalism

and Enlightenment rationalism, to fall into this error, because we eas-
ily forget that our own societies are not purely republican. It cannot
be assumed, therefore, that our own comparatively mild attitudes to-
wards criminality would also be those of more purely republican soci-
eties. Nevertheless, despite our different social, cultural, and historical
contexts, we admire and to some extent still strive to affirm republican
principles, and this, as the argument of the present study will suggest,
makes us at least partly receptive to the republican attitude towards
punishment – much more receptive than Braithwaite and Pettit believe.
The re­-examination of the question of punishment in the context of clas-
sical republicanism is, therefore, not merely of antiquarian value. It is of
interest to us primarily because the problems of punishment are treat-
ed as acute moral­-political problems by classical republican thought,
and these same problems are implicated in our own debates about the
rights of criminals and victims and the purpose of punishment. In clas-
sical republican thought, these problems are raised more urgently and
radically (as I hope to show), and in this it may still be of great use to us.
Although Plato wrote dialogues on a wide range of subjects, he nev-
er wrote a dialogue with punishment as its unifying theme. The most
extensive treatment of punishment in Plato occurs in a dialogue, the
Laws, whose unifying theme is legislation – or, more precisely, the leg-
islative art required for founding a small, participatory republic. The
implication seems to be that the subject of criminal justice cannot be
understood except in light of the broader issue of law and republican
rule. Before we turn to Plato’s treatment of punishment, therefore, we
must first consider this larger context. This will require some extensive
exegesis, some of which may be challenging to readers who are not
students of the history of political thought. Readers unfamiliar with
Plato in particular may be surprised at the amount of attention paid to
dramatic elements in the dialogue, as distinguished from straightfor-
ward analysis of the Stranger’s “theoretical” arguments. This, however,
is unavoidable given the dialogue form, in which the drama provides
essential clues for correctly understanding the theoretical arguments.

•
The Laws takes as its theme not politics as a whole, but a certain kind of
politics – republican politics. It is taken for granted in a republic that
not men, but laws, must rule. This sharply sets off the Laws from Plato’s
 Punishment and Reform in Plato’s Laws 19

other two political dialogues, the Republic and Statesman, both of which
culminate in the praise of the rule of the wise, who ought not (it is ar-
gued) to be restricted by any fixed laws.6 It is not surprising, then, that
there is little occasion in those two dialogues for any discussion of penal
law, whereas in the Laws such a discussion takes up nearly one­-quarter of
the entire conversation about the institutions of a well­-ordered republic.7
The unique perspective of the Laws, as distinguished from the Republic
and the Statesman, reflects its essentially civic and legalist view of politi-
cal life (as distinguished from the fully philosophical one that might be
encountered in most of the other Platonic dialogues). It is essential to
this view that it accepts as fundamental a respect for the law and for its
authority.8 So we are not surprised that two of the three characters in this
dialogue are prominent citizens of the oldest and most famous republics
of ancient Greece – Kleinias the Cretan and Megillus the Spartan.
This is not to say that the Laws is simply conventional. It is perhaps
better to say that the fundamental attitude of respect for law is one that
admits of being either more or less reflective. For example, respect for
law is common to Kantian rational autonomy and to the beliefs of the
rustic citizens of Magnesia (the city­-state described in the Laws), al-
though it goes without saying that there is a sea of difference between
the two in terms of their critical sophistication. Similarly, while the third
interlocutor in the Laws – the nameless Athenian “Stranger” – expresses
a respect for the ancient laws of Greece, he does not do so blindly, and
in the course of the discussion we find him raising far­-reaching philoso­
phical questions about the wisdom and origins of those laws. Yet unlike
Socrates (who needles the public authorities with his sceptical ques-
tions), the Stranger proceeds cautiously, initially deferring to the most
ancient beliefs of the Greeks (even when these appear quite childish),
until the comparative political advantages of accepting or rejecting them
can be carefully weighed. In keeping with his public­-spirited mission
– the framing of a complete code of laws for a city­-state of Greek colo-
nists – the Stranger never allows his philosophical interests to take cen-
tre stage; rather, he accepts the ends of republican life, while reflecting
on those ends and promoting enlightened reform whenever possible.9
The preceding observations on the tone of the dialogue and its char-
acters are necessary for understanding the principles of the Stranger’s
legislative art, to which we now turn. A discussion of these principles
will set the stage for our account of the Stranger’s penal law.
The ends of republican politics, which the Stranger initially acknowl-
edges and allows to inform his early discussion of legislation, include
20 Punishment and the History of Political Philosophy

order and security. In the simplest sense, a city cannot be said to be well
governed if it is not stable, self­-sufficient, strong in defence against its
neighbours, and united (626c–627c). Since the “unjust” (adikoi) members
of the city cannot be allowed to prevail, only three possible alternatives re-
main, according to the Stranger: the unjust must be completely destroyed,
or they must be enslaved to the just and law­-abiding citizens, or the un-
just and the just should be reconciled and made friends (philous) through
laws laid down by a wise judge or legislator (627e–628a). Kleinias, to
whom the choice of these alternatives is addressed, picks the last option,
apparently opting for the most politically practicable course of action (as
either killing or enslaving a potentially large number of citizens would
involve great difficulties; see 628a and cf. 625c–626b).10
Although the Stranger begins by accepting as fundamental the city’s
right to preserve itself against external and internal threats, he quickly
convinces his two Dorian companions that virtue is the highest concern
of the city and that therefore, mere security cannot be the city’s sole
preoccupation. This must be true precisely from the city’s own point
of view, since the city’s highest laws and institutions aim explicitly at
the virtue of its citizens (628c–e). Virtue, in turn, can be understood not
simply as subordinate to the preservation of the city – as a mere set of
habits of heart and mind that render individual citizens serviceable to
the whole – but as an end in itself: as human excellence. The Stranger
appeals to his companions’ own sensibilities when he argues that de-
fensive war and domestic order are not ends in themselves, but rather
must be regarded as being in the service of peace and the peaceful ex-
ercise of the virtues for their own sake. Thus the martial virtues and the
virtue of loyalty to one’s regime must be regarded as only the lowest
part of virtue as a whole (cf. 628c–e with 631b–d). In short, in seeking to
reconcile the just and unjust, the law must also try to reconcile the two
distinct ends of legislation – order and human virtue.
This important introductory exchange between the Stranger and his
Dorian companions sets the tone for the remainder of the dialogue in
setting up, as a kind of methodological requirement, that enacted law
must be able to persuade rather than merely command and coerce.
Accordingly, the commandments of the law must be accompanied by
preambles that communicate to citizens the wise purpose of the laws
and encourage them to obey willingly (719e–720a, 722c–723b). The
Stranger adheres to this standard throughout the dialogue, artfully
weaving together argument, exhortation, and command (although this
has the effect of making the legal preambles somewhat complex in their
 Punishment and Reform in Plato’s Laws 21

structure). By examining the preambles specifically bearing on criminal

justice, we can hope to gain insight into what the Stranger understands
to be the proper civic view of the purpose and measure of punishment.
Two legal preambles address criminal justice. Before beginning to
legislate, the Stranger delivers a formal speech as if all the citizens of
the future colony were present (715e–734e), and this speech serves as
a general preamble to the laws as a whole. Among the topics taken up
by the general preamble is criminal justice (at 728b–d and 730d–732b).
The Stranger later returns to criminal justice in Book 9, where he elabo-
rates a special preamble to the penal laws. We will look at these two
preambles in order.

The Penology of the General Preamble

The Stranger’s Rationalist Penological Principle

The section of the general preamble containing the discussion of wrong-

doing and punishment is nested within a longer section whose subject
is the right treatment of one’s soul, body, and wealth, respectively – the
three major categories of one’s private property, as distinguished from
what belongs to the gods, parents, or the city (726a–734e; cf. 723e).11
According to the preamble, the highest and most divine possession
is the soul. The goodness of the soul consists not in having its desires
fulfilled indiscriminately, but in practising those things the lawgiver
praises as noble and abstaining from the things he condemns as base
(727b, 727c–d). Lower in rank than the soul is the body (i.e., health and
life), and lower still and in last place among one’s private possessions is
wealth (727d–728a). The preamble goes on to exhort citizens to honour
the pronouncements of the lawgiver above their own desires and gives
an argument explaining why they should do so, stressing human igno-
rance about the most important things. Those who assume that “every-
thing done in Hades is bad” judge incorrectly, since in fact “one doesn’t
know whether things aren’t just the opposite – it may be that among
the gods below are to be found the things that are by nature the greatest
goods of all for us” (727d). Therefore, one who “considers survival to be
always good” would be misled, since whether one’s own life is good for
oneself depends on what happens in Hades.
At first glance the preamble seems only to be asserting a pious dog-
ma. In fact, however, while the preamble explicitly states that we do
not know with any certainty that the things in Hades are bad for us, it
22 Punishment and the History of Political Philosophy

equally implies that we do not know whether the things there are good
either. Apparently, we must rely on the superior wisdom of the ancient
poets, who have praised the just life as good for men in both this life
and the next. Yet the veracity or authenticity of ancient reports may
reasonably be doubted (cf. 624a–625a with 636d), not to mention that
the poets themselves are notorious for their fabrications (cf. 669c–d,
700d–e, 719c). Thus, when we begin to reflect on the preamble’s seem-
ingly doctrinaire exhortation, we are surprised to discover that it in fact
establishes not a dogma but reason as its authority. More precisely, it
establishes that (1) what the law commands as good is what is good for
oneself, and (2) what is good for oneself is to be ascertained by evidence
available to the individual reasoning human being. The Stranger even
goes so far as to attribute to the ancient legislators, to fathers, and to the
gods, alike, the view that the virtuous, law­-abiding way of life is prefer-
able to the vicious, because it is more pleasant and more beneficial to
the virtuous man himself (660e–663b).12
The preamble goes on to draw the conclusion that “what is gravest is
to become similar to men who are wicked, and, in becoming similar, to
avoid good men and be cut off from good conversation, and instead to
attach oneself to the bad by seeking intercourse with them” (728b). The
preamble here seems to assume that virtue makes human beings happy
and that the laws of the city (at least if they are well ordered) lead to vir-
tue. From these assumptions it would follow that those who disobey the
laws are actually harming themselves. Apparently, the philosophically
inclined Stranger views exclusion from decent society and good conver-
sation as the worst harm that can come to one’s soul as a consequence
of injustice, and he makes no mention of shame. Yet perhaps someone
else would add that pangs of guilt will forever haunt the criminal, spoil-
ing any future enjoyment he might hope for. At any rate, the harm that
injustice does to the unjust person himself is, by this account, a much
worse consequence for him than any scourging or imprisonment that
may be imposed on him by the city as punishment. In this way, the
preamble attempts to provide a rational explanation for the public as-
sertion of the wrongness of crime as such – an explanation that is meant
to be convincing even without the threat of punishment. In doing so it
taps into, and relies on, the belief of all decent citizens that injustice is
somehow bad for us, not merely because it exposes us to the threat of a
stiff penalty, but more importantly because unjust actions are ugly and
therefore degrading to the soul of the doer of injustice.
The Stranger goes on to argue that this purportedly natural and
necessary consequence of crime is not a punishment at all, for what is
 Punishment and Reform in Plato’s Laws 23

ordinarily understood as punishment is not justice but “retribution”

(timˉoria), “since what is just [to dikaion] – including judicial punishment
[dikˉe] – is noble” (728c). We might be tempted to call the depraved con-
dition into which the criminal falls as a result of his actions a kind of
natural justice, but this would be incorrect, according to the Stranger.
Justice makes human beings better, not worse, but “both he who un-
dergoes this [i.e., retribution], and he who escapes it are miserable: one
because he does not get cured, the other because he is destroyed in
order that many others may be saved” (ibid.).13
It is quite surprising that the preamble takes for granted such a sharp
distinction between justice (dikˉe) and retribution (timˉoria), considering
that to the ancient Greeks retribution meant primarily desert­-based pun-
ishment deliberately and justly imposed by the political authority. We
are thereby further alerted to the radical character of the Stranger’s defi-
nition of penal justice, and prompted to wonder whether the preamble
is correct in proclaiming justice to be always beneficial and never harm-
ful. It is true that when punishment encourages the offender to mend his
ways, we welcome this and consider it a benefit to the offender himself.
But does it follow from this that punishment can never be called just
when it does harm, as the Stranger claims? On the contrary, do we not
often harm criminals, believing this (the harm itself) to be what they
justly deserve? To put it another way, the preamble’s claim that retribu-
tion is neither just nor noble flies in the face of the seemingly universal
experience of punitive indignation. For many – perhaps most – people,
getting “payback” against those who have done them wrong is essential
to what “punishment” means – and when we punish we do so because
it harms criminals, not despite this. To be sure, that a moral opinion is
popular does not make it true; but it does make it necessary for any
contrary opinion, such as the one advanced in the Stranger’s preamble,
to establish itself on firm grounds. Why, then, does the Stranger simply
beg the question regarding the justice of retributive punishment? The
reason for his seemingly inadequate approach to the question of retribu-
tion begins to emerge in the sequel of the general preamble.

“Every real man must be of the spirited type”

Shortly after its initial statement on justice and retribution, the general
preamble speaks about the place of the passion of spiritedness in pub-
lic life, placing great emphasis on the political necessity of encouraging
certain punitive attitudes towards crime. According to the preamble,
citizens who merely refrain from doing injustice themselves are only
24 Punishment and the History of Political Philosophy

third in point of honour. They are less honourable than those who ac-
tively work to prevent unjust men from doing injustice by willingly co-
operating with the magistrates. “Yet the great man in the city, the man
who is to be proclaimed perfect and the bearer of victory in virtue, is the
one who does what he can to assist the magistrates in inflicting punish-
ment” (730d). “Every real man,” the preamble continues, “should be of
the spirited type [thumoeididˉe]” (731b). Since this punitive sort of person
is likely to be moved to demand retribution (cf. 731d6), it is surprising
to find such a type being praised by the preamble, shortly after it has
denied that retribution is noble or just. The preamble proceeds to ex-
plain that every real man should be of the spirited type because “there
is no way to avoid those injustices done by others that are both danger-
ous and difficult, or even impossible, to cure, except to fight and defend
oneself victoriously, in no way easing up on punishment … This every
soul is unable to do, if it lacks a high­-born spiritedness” (731b). In other
words, effective law enforcement in a small republic, where the citizens
must act as their own police force and judiciary, requires the vigilance
and enthusiastic participation of all. This may be difficult to sustain, the
preamble seems to suggest, if the citizens are not animated by a punitive
spirit, without which few people will want to get personally involved in
such dangerous business as crime control if it is possible to leave the job
to others. This is apparently why the legislator must promote spirited-
ness and praise as “perfect” those who harbour a zeal for punishment.14
Still, although the Stranger’s preamble appears to praise the punitive
citizen, it is careful not to call him “just” or to identify as “justice” the
punishment he demands.15 Moreover, what the Stranger actually says
is not that the punitive man is perfectly virtuous, but that he is to be
proclaimed perfect. The preamble does not, then, retreat from its earlier
definition of just punishment, in light of which carrying out spirited
punishment with the intention of doing harm cannot be considered vir-
tuous in the strict sense. The preamble thus appears to acknowledge
the political necessity of promoting spirited anger in a small republican
city while at the same time (and with some apparent inconsistency) in-
sisting on the lower moral status of retribution in light of some higher
standard of justice – although we must remember that we have yet to
see a demonstration of the grounds for this higher standard. The pre-
amble’s ambivalence towards retribution, and towards punitive anger,
is also reflected in its admonition regarding the proper response to two
different types of criminals. Although every real man must be of the
spirited type, he should be “also as gentle as possible.”16 “In regard to
the curable injustices men commit,”
 Punishment and Reform in Plato’s Laws 25

one must first understand that no unjust man is ever voluntarily unjust.
For no one anywhere would ever voluntarily acquire any of the greatest
evils – least of all when the evil afflicts his most honored possessions. Now
the soul, as we asserted, is truly the most honorable thing for everyone;
therefore no one would ever voluntarily take the greatest evil into his most
honorable possession and keep it for the rest of his life. So the unjust man,
like the man who possesses bad things, is pitiable in every way, and it is
permissible to pity such a man when his illness is curable; in this case one
can become gentle, by restraining one’s spiritedness and not keeping up
that bitter, woman’s raging. But against the purely evil, perverted man
who cannot be corrected, one must let one’s anger have free rein. This is
why we declare that it is fitting for the good man to be of the spirited type
and also gentle, as each occasion arises. (731c­–d)

Here, the preamble makes explicit what was only implied or assumed
earlier – that injustice somehow harms the soul of the unjust man – and
admonishes citizens not to be indignant towards curable offenders but to
pity them. Since no one would ever intentionally harm himself, the cur-
ably unjust offenders must have committed their wrongs involuntarily.
This passage in the Stranger’s preamble recalls an argument made
famous by Socrates, who held that all vice is done out of ignorance and
therefore involuntarily.17 Yet unlike Socrates’s famous argument, the
Stranger’s argument does not go far enough, and as a result it falls into
an inconsistency: if injustice always harms the soul of the unjust man,
then why ought we to pity only the curable criminal but give vent to
our anger against the incurable one? Should we not rather pity the in-
curably unjust man all the more for his condition (if it is more pitiable
to be not only sick but also incurable)? Or does the preamble mean to
imply that the incurably unjust man somehow manages to act volun-
tarily, in spite of the fact that he harms himself?18 One might argue that
a sign of the voluntariness of the incurable criminal’s injustice is that he
does not repent of his crime, even after suffering punishment. But, on
the other hand, this incorrigible contempt for justice, by which the in-
curably unjust man shows his deep ignorance of what is truly good for
him, may be a reason to pity him even more than his curable counter-
part, since he is even less capable of helping himself and thus seems to
act even less voluntarily (if voluntariness admits of degrees). Or, again,
is there some other sense in which an action can be voluntary, other
than as aiming at the good and avoiding the bad? None of these issues
is settled in the general preamble, nor can they be, since in order to
settle them a more complete account of the nature of criminality would
26 Punishment and the History of Political Philosophy

be needed. That account is postponed until the special preamble that

introduces the penal code in Book 9, when the Stranger treats serious
crimes and punishments, or, as he puts it, “the greatest matters” among
the judicial penalties (dikai).19

The Penal Preamble

Capital Crimes

The Stranger identifies three capital crimes: temple robbery, the over-
turning of the laws, and treason. The inclusion of these particular crimes
among the greatest offences – two of which are crimes against the city
and the other a crime against the gods – reflects the deep religiosity
of the Stranger’s republic and, at the same time, its intensely political
character. All three are to be tried by the same judicial procedure (de-
scribed by the Stranger at 855c–856a) and punished in the same way
where citizens are concerned (856e–857a).20 A citizen convicted of a capi-
tal crime should be punished with death, the Stranger argues, since “the
judge should think of this man as already incurable” because the civic
education he received under the best laws was not enough to prevent
him from committing one of the greatest crimes (854e). The penalty of
death should be regarded as a release from his miserable condition and
as the least of evils for him. The death penalty is thus a dikˉe, since it
benefits the criminal by helping him avoid the greater evil. The Stranger
continues in this section to refer to capital punishment as dikˉe and as
timˉoria interchangeably, maintaining the ambiguity we first saw in Book
5. Also, as before, we see an effort made in the law to invoke a sense of
anger and indignation towards crime. In order to encourage magistrates
to do their duty by punishing criminals, the preamble admonishes that
if a magistrate “holding one of the highest offices in the city lets these
things escape his notice – or, not because they escape his notice, but
because of cowardice, fails to wreak retribution (timˉoria) on behalf of his
own fatherland – such a citizen must be held to be second in evil” (cf.
856b–c with d). From this we may already conclude that, according to
the Stranger, the best republic – which is supplied by the best laws that
can be framed by a philosophical legislator – enacts capital punishment
and embraces retribution (albeit equivocally).
In contrast to the extensive treatment given to capital crime, the sub-
ject of theft, which immediately follows, gets very short shrift. “As to
stealing, whether someone steals something great or something small,”
 Punishment and Reform in Plato’s Laws 27

the Stranger says, “let there be one law and one judicial retribution
[dikˉes timˉoria] in all cases” (857a): double the value of the stolen item is
to be paid as a fine; if the convicted offender does not possess enough
property to pay the fine, he is to be imprisoned until he does pay or un-
til he persuades the successful prosecutor (857a–b). In response to this,
Kleinias objects to the Stranger’s proposal:

How comes it, stranger, that we’re saying it makes no difference to the
thief whether he’s convicted of stealing something great or something
small, whether from sacred or profane places [ex hieron eˉ hosion], or what-
ever other respects in which a theft can be entirely dissimilar? For crimes
that are thus various, is the lawgiver in no way to follow with penalties of
a similar variety? (857b)

Kleinias’s objection is provoked by the Stranger’s attempt to make the

penalties for theft proportional strictly to the material value of the stolen
goods, regardless of what other significance the stolen goods may pos-
sess, such as their religious value. By putting special emphasis on the
difference between the sacred and the profane, Kleinias seems to be in-
dicating that he thinks the significance of crime is not reducible to its
material value. He wants thieves guilty of ordinary theft, and those
guilty of sacrilege, to be punished in proportion to what seems to him to
be the different degrees of their injustice. In other words, Kleinias’s ob-
jection presupposes an understanding of criminal culpability as indexed
to some implicit scale of injustice, whose degrees of gravity can be re-
lated to the circumstances of each crime, which include, but are not lim-
ited to, its material value.
Kleinias’s notion of a scale of injustices, and their corresponding
punishments, seems intuitive. It is worth noting that many retributivists
point to this sort of intuitive sense that injustices have moral degrees as
evidence favouring their position. In response, the Stranger now wel-
comes Kleinias’s objection, telling him that it has somehow “reminded”
him of something important (857c). Could it be that Kleinias has re-
minded the Stranger of the self­-evident truth of the existence of a scale
of proportional crimes and punishments differing in kind (and not only
in degree)? This seems impossible, however, in light of the Stranger’s
earlier discussion of penal justice (in the general preamble in Book 5),
where justice was said to look to beneficial consequences and to noth-
ing else. The Stranger’s simple rule regarding the punishment of theft,
which overlooks the distinction between sacred and profane, seems all
28 Punishment and the History of Political Philosophy

the more surprising when we consider that the Stranger himself had
included temple robbery among the greatest crimes. Could it be rather
that the Stranger has deliberately provoked Kleinias’s objection?21
As becomes clear in the sequel, the Stranger has been “reminded”
not of the truth of our intuitive sense of the need for the sort of pro-
portionality of crimes and punishments alluded to by Kleinias, but of
something very different and much more strange. He now says that
Kleinias’s objection reminded him that “what pertains to the laying
down of laws has never been worked out correctly in any way, as in fact
can be said on the basis of what has cropped up now” (857c). Recalling
his earlier argument about the need for laws to be accompanied by pre-
ambles (cf. 719c–723d), the Stranger goes on to assert that human be-
ings living under all existing legislation resemble slaves being doctored
by other slaves. Slave doctors, the Stranger explains, practise medicine
on the basis of experience rather than knowledge, and prescribe treat-
ments without persuasion or explanation, whereas free doctors practise
their art by engaging in dialogue with their free patients, using argu-
ments “that come close to philosophizing, grasping the disease from
its source, and going back up to the whole nature of bodies” (857c–d).22
The Stranger now claims that if one of the slave doctors should come
across the free doctor with his patient, he would burst out laughing and
declare, “Idiot! You’re not doctoring the sick man, you’re practically
educating him, as if what he needed were to become a doctor, rather
than healthy!” (857d–e). The Stranger’s approval of the educational ap-
proach of the free doctor in the context of a discussion of penal laws
seems to imply the superiority of a similarly enlightened approach to
criminality. If injustice is analogous to disease, then it would seem to
follow that the proper response to crime – that is, what criminals de-
serve – would be some sort of individualized treatment following the
medical model, perhaps informed by some scientific art (technˉe). One
imagines this art might be concerned with the individual diagnosis of
the criminal and his spiritual disorder, and with the means that would
be adequate to reform him. Yet it is difficult to see how such a “medi-
cal” art of criminal justice could be concerned with the evaluation of
moral qualities of offences, which was the central point of Kleinias’s
objection. Is the Stranger attempting here, once again, to reassert the
superiority of the rationalist conception of justice as beneficence in the
face of Kleinias’s assertion of penal proportionality as a moral fact?
Kleinias, however, resists the Stranger’s remarkable claim here by agree-
ing with the scolding slave doctor (857e). This resistance on Kleinias’s part
 Punishment and Reform in Plato’s Laws 29

in turn compels the Stranger to change course yet again. Whereas it

seemed as if he and Kleinias had just reached an impasse, the Stranger
now asserts, contrary to our expectation, that it may be possible to com-
bine education and punishment after all – but that in order to arrive at
that solution they would first need to undertake “a precise inquiry”
into the nature of “all injustices” (859b–c). This short exchange serves to
introduce one of the climaxes of the discussion of punishment (and per-
haps of the whole work), which also happens to be the only “Socratic”
elenchus, or cross­-examination, of an interlocutor’s fundamental opin-
ions about justice in the dialogue.

Kleinias’s Opinions about Punishment

Having excited Kleinias’s interest, the Stranger recommends that they

proceed to examine the opinions of the famous Greek legislators about
the noble, the good, and the just things.23 But the Stranger does not
quite do this. Instead, he proceeds to examine how the noble and just
things stand in the opinion of the interlocutors themselves – that is, “to
what extent we now agree, and to what extent we disagree with our-
selves” regarding “the noble things and all the just things” (859c–d).
Furthermore, since Kleinias does all the answering while the Stranger
asks questions, it is really only Kleinias’s opinions about the noble, the
good, and the just that are subjected to examination.
The Stranger begins to lay bare Kleinias’s deepest beliefs about pun-
ishment by asking whether “our” opinions about the just and the noble
are consistent. He elicits Kleinias’s assent to the following:

With respect to justice as a whole [dikaiosunˉes holˉos] and just human beings,
deeds, and actions, we all somehow agree that all these are noble [kala]; so
if someone would maintain that just human beings, even if they happen to
be ugly in their bodies [aischroi ta sˉomata], were nevertheless entirely beau-
tiful [pankalous] people in respect to their very just disposition [to dikaiota-
ton eˉthos] considered by itself, in almost no case would the one who speaks
thus seem to speak off key. (859d–e)

The Stranger proceeds to argue, again with Kleinias’s assent, that if all
things are noble which partake of justice, then this would include the
things we undergo or suffer just as much as the things we do. But if this is
the case, the Stranger goes on, then they have contradicted the penal laws
they had established earlier, for “presumably we established that the
30 Punishment and the History of Political Philosophy

temple robber should die – justly – and the same for the enemy of well­-
made laws … and that while they were the most just of all sufferings, they
were also the most shameful [aischista]” (860b). In this way, the just things
and the noble things appear “to us” at one time to be all the same, and at
another time opposed (ibid.). The Stranger adds that this inconsistency or
confusion exists not only among “us,” the legislators, but also among “the
many,” who, in view of these things, “proclaim without consonance that
the noble things and the just things are separate” (860c).
It is not clear why the Stranger proceeds in this way, or in particular
why he wishes to focus on what is sound and unsound in Kleinias’s
opinions about the noble things and the just things. Perhaps the Stranger
means to imply that Kleinias’s opinions about these matters will be rep-
resentative of the opinions of the ancient legislators. Even if this inter-
pretation of the Stranger’s way of proceeding here is correct, it leaves us
to wonder about its relevance to his purported aim of combining penal
law with education. The Stranger had stated that he thinks nothing that
pertains to laying down laws had been correctly worked out in any way,
which would mean that the most that could be hoped for from a criti-
cal examination of the opinions of past or present legislators is greater
awareness of their errors. Does the Stranger believe that knowledge of
the nature of justice and injustice in the soul – the matter that would be
most important in guiding the educational (therapeutic) “punishment”
of criminals – can somehow be gleaned from knowledge of the errors of
the legislators and the many, alike, regarding the noble, the good, and
the just things?
Kleinias admits that he is guilty of the contradiction imputed to him
(860c). But was this contradiction unavoidable?24 In the first place, is it
really necessary to define justice, and thus also penal justice, as some-
thing that is entirely noble, or could it not rather be the case that some
things are just without also being noble? For example, the law com-
mands some things because they are noble and choiceworthy for their
own sake (e.g., education), but it also commands other things merely
because they are necessary (e.g., war). In this way, while all of the things
commanded by the law would be just (in the sense that the city has a
right to command them), not all would be noble. Would it not, then, be
possible to understand punishment as one of the things that are just but
not noble? Furthermore, the Stranger had just distinguished between
the beauty of the soul and that of the body. Thus, even if punishment of
criminals damaged their bodies, making them appear physically ugly
as a result of their suffering, it might still be argued that such suffering
 Punishment and Reform in Plato’s Laws 31

affects their souls in the opposite way: by undergoing just punishment,

their souls may become better (as the Stranger’s preamble earlier sug-
gested) and therefore more beautiful or noble in this more important
respect. The reader is left wondering why Kleinias did not avail him-
self of any of these possible arguments. At the very least he could have
argued, as the Stranger had done earlier, that in the case of the incur-
ably unjust man, death would be the lesser evil for him, and therefore
executing him would be a noble thing.
The fact that Kleinias does not take refuge in any of these arguments is
of the highest significance, on my reading of the dialogue. It seems likely
that Kleinias refused to accept the premise that punishment can be both
just and noble, and then did not try to avoid open self­-contradiction,
because he believes that punishment ought to be essentially harmful to
the criminal (and thus in no way noble).25 In other words, Kleinias seems
to believe that punishment is just precisely when it is retributive – that
is, when it harms criminals for the sake of harming them, since that is
what they deserve, according to Kleinias. But why is Kleinias so strongly
attached to this view in the first place, even at the cost of embarrass-
ing self­-contradiction? It is true that deterrence, as distinguished from
retribution, also requires punishments to be harmful, so one might be
tempted to think that Kleinias is not a retributivist at all, but that he
merely prefers tough penalties because they deter most effectively. This,
however, would be to misunderstand Kleinias. For deterrence presup-
poses only that punishments must appear to criminals to be bad for them
(not noble), while nothing prevents more virtuous citizens – who value
living justly more highly even than the comfort and liberty of the body
– from regarding the same punishment as being corrective and thus in
the truer sense beneficial. Yet as we saw, Kleinias denies precisely this;
he therefore cannot be said to be concerned with deterrence alone. It ap-
pears that Kleinias is firmly convinced of the justice of retribution. It is
doubtful that he is himself fully aware of the confused state of his own
opinions regarding the relation of punishment to justice and nobility,
or of the reasons behind his attachment to retribution – although as we
have seen, he does seem to express some perplexity regarding his self­
contradiction. As readers of the dialogue, however, we have leisure to
reflect on the deeper psychological reasons for this attachment. As we
will see shortly, those reasons will prove to be significant for our under-
standing of the problem of punishment in the republican context.
It is important to keep in mind that throughout the whole discus-
sion between the Stranger and Kleinias, it has been assumed (on the
32 Punishment and the History of Political Philosophy

authority of the ancient legislators) that justice is the health and good-
ness of the soul. This is what had allowed the Stranger to make the
argument that just punishment makes men good, since it makes them
more law­-abiding and therefore more just. Up to now, we have not had
any strong reason to doubt this traditional moral assumption about jus-
tice (i.e., beyond our general scepticism of the hearsay by which tradi-
tional wisdom is passed down). Now, however, Kleinias’s insistence
that punishment ought to be harmful and shameful for the unjust man,
even if it makes him more law­-abiding – a view that he expresses as a
legislator and on behalf of all legislators (see, again, 858e–859d) – forces
us to question that traditional assumption. In other words, Kleinias’s
perplexity about punishment raises the more fundamental question of
how, or whether, justice as law­-abidingness is indeed good for the soul.
It bears repeating that Kleinias does indeed believe that justice is the
health of the soul. When questioned by the Stranger regarding justice in
general, he agreed that regardless of a man’s physical condition, he is
noble to the extent that his soul is just. If he did not believe this genuine-
ly, he would have had no trouble rejecting this premise of the Stranger’s
argument, thus avoiding contradiction. Yet while Kleinias believes, and
wants to believe, that justice is the health of the soul, he does not know it
for a fact. Introducing preambles that attempt to explain, using rational
argument, the purpose of the penal laws was the Stranger’s idea, to
which Kleinias showed initial reluctance when he sided with the “slave
doctor,” who commanded patients for the sake of their health, and
against the “free doctor,” who philosophized with them. Kleinias ap-
peared to believe that just as knowledge of medicine is unnecessary for
making bodies healthy, knowledge of justice is unnecessary for making
souls healthy, since following the laws and cultivating the habit of law­-
abidingness is sufficient. From the point of view of a republican citizen,
the virtue of justice is nothing more than the habit of lawfulness.
It seems at first that this applies to Kleinias’s own case, since he has
lived a life of public service in obedience to the Cretans and their laws,
and believed this to be noble and good for him, without ever having
examined his way of life until prompted to do so by the Stranger (see
625c–627c). In fact, however, the habitual practice of justice combined
with a mere belief in its goodness are not sufficient for Kleinias. At the
very opening of the dialogue, when the Stranger first asked Kleinias
about the purpose of the ancient Cretan lawgiver, Kleinias openly ad-
mitted that the truth the lawgiver had divined was that, on every level,
there exists a war of all against all, a condition in which true goodness
 Punishment and Reform in Plato’s Laws 33

consists in advancing one’s own interest even at the expense of the in-
terest of others (626b–d). According to Kleinias’s explanation at that
time, it follows from this truth that as it would be in accordance with
nature for cities to dominate other cities rather than to treat them justly,
so it would be equally in accordance with nature for individuals to ex-
ploit one another by force or fraud. This, according to Kleinias, is where
“the argument” (ton logon) leads when it is “correctly followed … up to
its source [archˉen]” (626d). According to this argument, justice would at
best be a conditional (and thus fragile) pact, or social contract, among
individuals to refrain from mutual depredations – or, as Kleinias ex-
presses this thought, “what most humans call peace [the lawgiver] held
to be only a name” (626a). At worst, justice in this sense would be a kind
of fraud that some men practise on others in order to make them more
willing to serve others’ interests while neglecting their own. Despite
avowing this doctrine of selfishness, it is clear that Kleinias does not ful­
ly believe it himself, for if he did, he could not have lived his whole life
according to the laws of the Cretans. Therefore, as that earlier exchange
between Kleinias and the Stranger had implied, and as the later cross­­
examination of Kleinias’s opinions regarding punishment and justice
has more clearly shown, Kleinias is fundamentally confused about the
goodness of justice as law­-abidingness as a human virtue, and this con-
fusion is a source of great perplexity for him.
Kleinias has now been shown to be divided between the two sides
of his self­-contradiction: he both believes and doubts, at one and the
same time, that justice is the good condition of his soul.26 It goes without
saying that both of these things cannot be true, and this causes him to
waver regarding the goodness of justice (see, again, 859d–860b and cf.
625c–627c). Kleinias’s doubt about the goodness of justice, as revealed
through his opinions about the goodness of punishment, is one instance
of a psychological phenomenon that is a principal theme in all of Plato’s
thought.27
With this understanding of Kleinias’s self­-contradiction regarding
the goodness of justice in mind, we can now better explain his indig-
nation towards criminality and therefore his attachment to retributive
punishment. Insofar as we are in doubt regarding the goodness of jus-
tice for the just person (i.e., whether or not he is better off obeying the
law), we also equivocate regarding the badness of injustice in the soul
of the criminal. We recall that earlier, in the general preamble to the
laws of Magnesia, the Stranger had argued that the criminal does harm
to himself by becoming like the unjust and by being deprived of the
34 Punishment and the History of Political Philosophy

company of the just. This assumed that justice in the soul is good for
the just person – a belief that Kleinias holds only ambivalently. Because
of this ambivalence, Kleinias must also be less certain than the Stranger
that the unjust person harms his soul by pursuing illicit gain. On the
contrary, insofar as Kleinias believes that justice is a mere convention
or habit of law­-abidingness, and that it restrains human beings from
seeking their own advantage at others’ expense, he must think that the
unjust have gotten away with a gain (whether in the form of property,
or pleasure, or some power) and thus have gotten an advantage over
the self­-denying just men.28 Indeed, we often speak of the criminal who
escapes punishment as having “gotten away with it” – that is, gotten
away with something good. As much as a republican community (even
the best­-ordered republic conceived by the Stranger) might cultivate the
love of justice, most of its citizens will always find it easier to appre-
ciate the goodness of external goods – wealth, power, and security –
than the goodness of justice in the soul, even while affirming the higher
status of the latter. As we have seen, when criminals are apprehended
in Magnesia, the republic exacts punishments that sometimes impose
deprivation or suffering exceeding what is required for compensation
or deterrence, in a way that makes those punishments more “fitting”
or “proportional” in kind to their crimes. These terms may have been
opaque before, but we are now in a position to suggest that they are
meant to express a reciprocity by which the balance of external goods
is restored after it has been disrupted by an illicit gain. This would also
explain why retributive punishment is essentially retrospective, since a
“settling of accounts” depends on knowing how much is “owed.”
Similarly, Aristotle in the Nicomachean Ethics affirms that justice is a
human virtue, and he goes on to explain an error whereby “reciprocity
is held by some people to be the just unqualifiedly … For either people
seek to reciprocate harm for harm – if they do not, that is held to be slav-
ish – or they seek to reciprocate good for good.”29 While Aristotle denies
that reciprocity (antipeponthos) is justice unqualifiedly, he concedes that
it is justice in a certain sense, since it helps preserve the city, which can-
not exist without mutual exchange. Plato’s Stranger and later Aristotle
both teach, therefore, that the psychology behind retributive punish-
ment involves the political concern with negating the perceived illicit
gains of the crime, through reciprocal or retributive punishment, in or-
der to prevent the unjust from having (whereas, from the philosophical
perspective, they merely appear to have) more than the just. Without
this restoration of the just distribution of external goods, the dignity of
 Punishment and Reform in Plato’s Laws 35

justice must suffer in the eyes of republican citizens, for it would seem
to put the just person at a disadvantage vis­-à­-vis the unjust, contrary to
the public doctrine that justice is good for the just. If the foregoing in-
terpretation of Kleinias’s opinions about justice and punishment is cor-
rect, then Kleinias’s attachment to retributive punishment is ultimately
founded on an error or confusion regarding the meaning of justice.
Having uncovered the fundamental confusion in Kleinias’s opinions
regarding justice and punishment, the Stranger might have gone on
to complete the refutation. But unlike Plato’s Socrates (who takes ev-
ery opportunity to challenge his interlocutors’ ignorance), the Stranger
takes a more politic approach.

Voluntary and Involuntary Injustice

Kleinias being unable to find a way out of his perplexity, the Stranger
comes to his rescue. He proposes that they look together to see how
“they” (the legislators), as distinguished from “the many,” hold a view
of the just and the noble that is in fact harmonious, despite their ap-
parent confusion (860c). When Kleinias agrees, the Stranger goes on to
state a view – which he claims to have expressed earlier – “that the bad
are all bad involuntarily in every respect” (860d):

The unjust man is presumably bad, but the bad man is involuntarily so.
Now it never makes sense that the voluntary is done involuntarily. Hence
the man who does injustice appears involuntarily unjust to the one who
sets down injustice as something involuntary. This is what I must agree to
now. For I agree that everyone does injustice involuntarily. And if some-
one, out of love of victory or love of honor, asserts that the unjust are
­indeed involuntarily so, but that many voluntarily do injustice, my argu-
ment, at any rate, remains the former and not the latter. (860d–e)

We may recall that the Stranger had indeed already stated a modified
version of this (“Socratic”) thesis in the general preamble in Book 5.
While, there, he had qualified his thesis to apply to curable injustices
only (without offering any reasons for the qualification), he now pres-
ents that thesis in its unqualified form.
The Stranger does not take the time to share his deeper reasons for
believing that all injustice is involuntary;30 rather, he simply affirms
his view because, as he says, he believes it to be true, and because it
would not be pious or according to custom for him to lie (861d). Does
36 Punishment and the History of Political Philosophy

the intervening examination of Kleinias’s opinions about punishment

somehow supply the insights necessary to substantiate the Stranger’s
claim here? The most important thing we have learned from that exami-
nation is that the civic view (shared by legislators and citizens alike) is
divided not only regarding the question of the nobility of punishment
but also regarding the more fundamental question of the goodness for
the individual of the virtue of justice. Someone in doubt about the good-
ness of justice vacillates, thinking at some times that justice is better,
at other times that injustice is preferable, but never having any settled
knowledge of how the matter really stands. Now doubting something
would seem to be the same as being in ignorance about it, and if that is
so, is it not also the case that those who act in ignorance act involuntarily
(if to make a free choice is to make it with full knowledge of what one
is doing)?31 Contrary to the general preamble’s message that the incur-
able and “thoroughly evil” are appropriate objects of anger, no one who
acts unjustly in fact desires injustice knowingly or voluntarily as long
as he is in doubt about the goodness of justice. The Stranger seems to
believe that such deep­-seated doubt regarding the goodness of justice
is much more widespread than we might care to believe (we may recall
that it afflicts the legislators and “the many,” alike), and this is the most
profound psychological insight that Plato intends to teach through his
treatment of punishment.32 Such vacillating doubt may be discerned (in
the right conversational setting) even in the most public­-spirited model
citizens, such as Kleinias himself. The same doubt may also be discerned
in those who are on the opposite extreme: those who seem to be most
dismissive or contemptuous of justice. It is striking that even the most
cynical among us will still grow angry when he believes he has suffered
wrong, thereby revealing a deeper concern for justice.33 If this is right,
then it follows that all injustice is involuntary in the strict or truest sense.
The most radical political consequences of the Stranger’s view of
crime and punishment can no longer be ignored. The Stranger’s view
destroys the distinction between voluntary and involuntary injustice,
which was a core feature of all ancient legal traditions, according to the
Stranger (861b), and which, perhaps, would be essential to any workable
law code. Upon this distinction depend all laws pertaining to the mitiga-
tion of criminal liability, such as those recognizing conditions of duress,
provocation, incompetence, and so on. (Indeed, it is such mitigation that
gives our own, liberal criminal law its character of fairness and relative
mildness.) In addition, the Stranger’s thesis shakes our confidence in the
therapeutic potential of punishment. We tend to treat first­-time offenders
 Punishment and Reform in Plato’s Laws 37

more leniently than repeat offenders, in the belief that the former’s injus-
tice has more of the quality of error in it and is somehow less reflective of
the offender’s “true character.” Thus we believe that the curable crimi-
nal is one who knows sufficiently well that injustice is harmful to his
soul, and merely forgets temporarily or is swayed by powerful passions.
A gentle penalty would be all that is required, as a reminder, to “cure”
the first­-time offender and to “set him straight.” On the Stranger’s view,
however, the conventional forms of punishment – such as fines, threats,
imprisonment, and physical suffering – could not be curative in the strict
sense if all injustice is, as the Stranger claims, perpetrated in the context
of the fundamental confusion regarding the goodness of justice for the
just man. For deterrence of the first­-time offender cannot cure him of
his ignorance but can only induce him – conditionally and for a certain
time – to refrain from injustice. He is not cured, because the deepest
cause of his injustice – his ignorance – is not removed; indeed, it cannot
be removed in this way. (To expect that such ignorance can be removed
by punishment in any form would be to believe that punishment can be
more educational than educational institutions!)
More radically still, the Stranger’s thesis destroys the distinction be-
tween justice and injustice with respect to individual character. If each
of us is compelled by our nature to do what he or she believes to be
best, then just and unjust actions differ only in the sense that the former
are carried out under the unstable belief that justice is better for us,
whereas the latter are carried out under the opposite belief. It cannot
be said that the just and the unjust differ in the character of their beliefs
fundamentally, because, according to the Stranger’s view, the vast bulk
of human beings (the many and the legislators alike) vacillate, in doubt,
between the opinion that justice is the health of the soul and the opin-
ion that it is merely a useful habit of obedience. In the most fundamen-
tal sense, these are only two sides of the same state of soul: its ignorance
regarding the goodness of justice (or, more broadly, of virtue).
It goes without saying that criminal law is unimaginable without
distinctions between voluntary and involuntary and between culpable
knowledge and inculpable ignorance. Clearly, the Stranger is fully aware
of these radical implications, for he now explicitly raises the question of
how the thesis he has just uttered will affect their ability to frame a penal
code. Will they have to distinguish involuntary injustices from volun-
tary ones? Will they then have to enact greater penalties for voluntary
crimes than for involuntary ones (legislating “in defiance of the diffi-
culty”)? Or will there have to be equal penalties for all on the grounds
38 Punishment and the History of Political Philosophy

that there are no voluntary injustices (860e–861a)? The first alternative

has been ruled out by the Stranger’s thesis, while the second is politi-
cally impracticable (cf. 861b). The Stranger now claims that it may be
possible to escape their difficulty if they can show how “these things are
two [i.e., the voluntary and involuntary], and what the other difference
is [i.e., between them], so that whenever someone imposes the judicial
penalty on either of them, everyone may follow the things that are being
said and may be able to judge, somehow or other, what is fittingly laid
down and what is not” (861c). The Stranger signals thereby that what
follows may be a more politically salutary doctrine than the view he has
been developing up to this point. The alternative, “political” account of
the psychological causes of crime that follows attempts to accommodate
the punitive civic outlook, while at the same time gently moving that
outlook as much as possible in the direction of the Stranger’s enlight-
ened understanding of the involuntariness of all injustice.

Actions and Motivations

The Stranger begins by asserting that some harms committed by citi-

zens against one another in their associations have the character of the
voluntary (hekousion), while others have the character of the involuntary
(akousion). A harm done to another is not necessarily an injustice, wheth-
er great or small, nor is a benefit justice. Rather, “when a benefit comes
to pass that is not correct [ouk orthˉen], the one responsible for the benefit
is committing an injustice” (862a–b). Generally speaking, “if someone
gives something to somebody, or, on the contrary takes something away,
such a thing shouldn’t thus be called simply just or unjust, but what the
lawgiver should observe is whether someone employs a just disposition
and character in doing some benefit or injury to somebody” (862b). The
lawgiver should concern himself with both injustice and harm, but he
should keep the distinction between them straight when framing his
criminal laws. Whenever some harm has been done, the laws should
redress the injury by “preserving what may be destroyed, setting up-
right again what has fallen, and making sound what has been killed or
wounded – and once compensation has made atonement for each of the
injuries, he should always try through the laws to create friendship in
place of discord between the doers and sufferers” (862b–c).
The Stranger’s concern here with compensation and rehabilita-
tion is consistent with the rationalist penological principle he has es-
tablished. But what of the brief mention of friendship as a second and
distinct aim of penal law, especially considering that the text implies
 Punishment and Reform in Plato’s Laws 39

that compensation paid to victims may in some cases be insufficient to

restore friendship between the “doers” and “sufferers”?34 We are left to
wonder what more the legislator can be expected to do to satisfy the
victims. Does the Stranger have in mind the legal provision for solemn
ceremonies of reconciliation – perhaps involving symbolic sacrifices
to substitute for the retributive punishment of the criminal? Or does
he imagine a kind of conference, in which victims air grievances and
remorseful offenders express their shame? Perhaps this is what the
Stranger wishes for. Yet once we consider what we have already learned
from the Stranger himself about the qualities of the republican charac-
ter, and about the public attitude of the laws towards criminality, we
cannot expect such mild measures to be sufficient to effect the neces-
sary reconciliation. The Stranger does not mention retribution explicitly
here, but it is likely that he would expect citizens and magistrates in
his best republic to seek “reconciliation” (effectively, the mollification
of the law­-abiding and indignant majority) after crimes committed by
punishing retributively, in addition to whatever would be required by
way of compensation and assurance of future security. If this is what
the Stranger believes would happen in practice, in spite of his own best
wishes as legislator, then we can conclude that he has acquiesced in the
practice of retribution as politically unavoidable, even (or especially) in
a well­-ordered republic.35
So much should be done by the legislator about the harms inflicted and
suffered in the city. With respect to injustices, however, as distinguished
from harms, the legislator’s approach should be entirely therapeutic, ac-
cording to the Stranger. When harms or benefits are done unjustly – that
is, when they proceed from an unjust character – as many of these injus-
tices as are curable, “being regarded as diseases in the soul,” should be
cured in the following way. “Whatever injustice, great or small, someone
might commit,” the legislator should “teach and compel” the unjust man
“either never again to dare voluntarily [hekonta] to do such a thing or to
do it very much less” (862c–d). To accomplish this, the legislator should
employ deeds or words, pleasures or pains, honours or dishonours, and
even fines or gifts. By these means he should try to “bring about hatred
of injustice and desire, or lack of hatred, for the nature of the just” (862d–
e). As for the offender whom the lawgiver perceives to be incurable,
he should institute a “just penalty” (dikˉen) of death, because he knows,
“presumably,” that it is worse for him to go on living. Such capital pun-
ishment confers a “double benefit” by releasing the offender from the
greater evil of injustice and by emptying the city of bad men and setting
an example for would­-be criminals (862e).
40 Punishment and the History of Political Philosophy

The thrust of the passage we have just summarized is to separate

conceptually the inner motivations of our actions from their external ef-
fects (i.e., harms or benefits), as well as to define once again the scope of
penal justice proper (dikˉe – as distinguished from compensation on the
one hand, and retribution on the other) as limited to the rehabilitation
and improvement of the individual offender.
One important implication of the conceptual framework for penal leg-
islation being developed by the Stranger here is the radical innovation
of committing to individualized punishment. For the purpose of meting
out an appropriate penalty, every actual criminal law code has recourse
to one or another general classification scheme for the different types
of crime according to their effects (e.g., murder vs assault, robbery vs
burglary, etc.). This allows the classification of individual crimes as well
as their ranking in terms of severity – the implicit assumption being that
the greater the harm done, the worse (morally) was the psychological
disposition of the perpetrator. Thus, even after we account for extenuat-
ing and aggravating circumstances for each particular crime we may still
maintain that, all other things being equal, a murderer is more culpable
(and therefore liable to a more severe punishment) than a pickpocket. A
corresponding classification scheme of mandatory penalties (or ranges
of permissible penalties) allows each crime to be indexed to its appro-
priate penalty. Yet there seems to be no inherent connection between the
effect of a deed and the particular psychological state under which it
was committed – and this is part of what the Stranger’s sharp distinction
between injustice and harm is meant to draw to our attention.36 Nothing
prevents the most heinous, misanthropic person imaginable from scru-
pulously obeying the law out of a concern for his own safety, just as noth-
ing prevents an admirable character from tragically falling prey to an
impulse (perhaps even a noble impulse) that in the circumstances leads
to a great crime (consider the crime of Prometheus).37 The Stranger’s
radical separation of injustice from harm would therefore preclude any
attempt at general classification of crimes and punishments, and would
require all punishment to be individualized. Every criminal would have
to be examined by the judge, on a case­-by­-case basis and without the
aid of legal equivalencies determined in advance, as to his psychologi-
cal disposition, regardless of the harm done by his crime. Although the
Stranger does not here explicitly draw out the implications – which are
completely lost on Kleinias and Megillus – of his radical conceptual in-
novation, we must keep them in mind as we proceed.
 Punishment and Reform in Plato’s Laws 41

So much is clear regarding the task of the legislator in his approach

to crime, but how the Stranger’s distinction bears on the question of
“the voluntary” is not. We seem to be invited to conclude that the two
forms of the voluntary correspond somehow to actions and motiva-
tions, respectively – but exactly what this means remains mysterious.
Earlier, the Stranger had asserted that “the bad are all bad involuntarily
in every respect” – or, to put it differently, human beings always and by
nature pursue what they believe to be their greatest good. It is difficult
to see how this general principle can admit of any meaningful moral
distinction between actions and motivations.38 If we always intend
what is best for us then every action we choose, whether harmful or
useful to society, is done for the sake of our greatest good, and since the
voluntary is always directed at our own good in this way, there would
seem to be no room left for speaking of the “voluntary” in other senses,
unless based on distinctions that are incidental.39
Most commentators, however, understand the Stranger to be intro-
ducing a concept of voluntary action in a “legal” as distinguished from a
philosophical sense. According to this view, the voluntariness of a crime
in the legal sense would only be used to determine criminal liability,
but not its injustice, which is always involuntary.40 This legal concept of
voluntary action allows the traditional distinction between voluntary
and involuntary crimes to stand, but it does so only tendentiously. That
is because the physical aspect of crime (i.e., the harm), as the Stranger
has defined it here, would strictly no longer have any relation to guilt
or desert in the ordinary sense (the sense in which the ancient lawgiv-
ers and cities have understood it). Unlike the Stranger, actual societies
distinguish voluntary from involuntary crime so that they can separate
those who deserve punishment (the “unjust”) from those who do not
deserve it (the “just”). This is exactly how the citizens of Magnesia will
understand the pronouncements of their penal law: when that law en-
joins that voluntary crimes should be punished more harshly than in-
voluntary ones (cf., for example, 864c ff.), Magnesians will believe that
such crimes deserve harsher punishments for retrospective consider-
ations,41 whereas according to the Stranger they would “deserve” this
because the greater corruption of the voluntary criminal’s soul requires
stronger cure.42 In short, while the Stranger’s solution works on a politi-
cal level, it fails to resolve Kleinias’s moral dilemma.43 At any rate, for
Kleinias, the Stranger’s thought here remains opaque, prompting him
to ask for clarification.
42 Punishment and the History of Political Philosophy

Justice and Injustice in the Soul

The Stranger obliges Kleinias by offering an account of the parts of the

soul responsible for criminality. This account apparently has its origins
in common opinion – in what “you say and hear from one another …
about the soul” (863b).44 According to this popular view, there are three
things in the soul that can cause wrongs or faults (hamartˉemata):45 spir-
itedness, pleasure, and ignorance (cf. Republic 436b–441c). Ignorance
is itself further divided into two parts or kinds: the “simple” version
resembles Socratic ignorance in that it is aware of itself (it is conscious-
ness of knowing what one does not know), while the “double” version
“occurs when someone lacks understanding because he partakes not
only of ignorance but also of the opinion that he is wise” (863c). The
Stranger assumes that only the presumptuous kind of ignorance will
ever be responsible for grave wrongs (with respect to the extent of the
injury, and to the degree of injustice from which it flows), and directs
the lawgiver to punish it severely (presumably because reform of such
a person would require more extensive correction). Crimes stemming
from the simple kind of ignorance, on the other hand, are to be forgiv-
en or dismissed with light penalties (863d). In this way, the Stranger’s
scheme for punishing ignorance conveniently reconciles the demands
of deterrence (in punishing the more dangerous crime more severely)
with those of justice (in dispensing the right kind of reformative pen-
alty to each kind of criminal). It would even satisfy the demands of just
deserts (by punishing the arrogant, and thus apparently more unjust,
criminal more severely than the humble one) – if, that is, the Stranger’s
immediate effort to establish theoretically the voluntariness of some
crimes should ultimately prove successful.46
All three of these forces or parts in the soul, the Stranger continues,
“often turn each man in directions opposite to that toward which his in-
tention … draws him” (863e). In the face of spiritedness or pleasure, “we
… say” that we might be “stronger” or “weaker” – that is, we might be
able to conquer our passion and avoid doing wrong, or conversely we
might succumb and be conquered by it. In the case of ignorance, how-
ever, “we’ve never heard that one of us is ‘stronger,’ while another is
‘weaker’” (863d). The Stranger gives little by way of explanation of the
foregoing account, evidently because he expects it to be already famil-
iar to his audience, Kleinias and Megillus. To be sure, there is little need
to describe how pleasure, spiritedness (thumos, which primarily refers
to anger or aggression), or ignorance can lead to wrongdoing. It is also
fairly evident what the Stranger means by our being either “weaker” or
 Punishment and Reform in Plato’s Laws 43

“stronger” in the face of these influences. We have all had experiences

of struggling with strong passions. Sometimes we struggle successfully,
but other times we seem to surrender to, and even welcome, the illicit
passion, although at the same time we say we somehow “know” that
what we do is wrong. This is the very experience we point to when we
speak of some wrongs as being voluntary, since it seems that it is “up
to us” whether we continue to struggle against our illicit passions or
surrender to them. On the other hand, we recognize circumstances in
which a passion is so strong that resistance is altogether impossible; but
perhaps these latter cases are less common, so that for the most part our
experience with intemperate passions gives us the sense that, however
hard it might be to resist temptation, we are ultimately in control and
thus accountable.47 Lastly, we know that when it comes to ignorance,
as distinguished from passion, there is no way to struggle against it,
which is why we tend to think that ignorance extenuates culpability.
In a kind of addendum or corollary to this popular psychology of
criminality, the Stranger further elaborates the meaning of justice and
injustice. “I would clearly define for you,” the Stranger now says to his
companions, “what I say is the just and the unjust, without complica-
tion” (863e). Up to now, justice was conceived to be lawful conduct
and the habitual disposition to such conduct, while injustice was un-
derstood as criminal action and a criminal disposition. The Stranger
now elaborates on that earlier conception by defining injustice as “the
tyranny in the soul of spiritedness, fear, pleasure, pain, feelings of envy,
and desires, whether it does some injury or not.” “When, on the other
hand, the opinion about what is best (however a city or certain private
individuals may believe this will be) holds sway in souls and brings or-
der to every man, then, even if it is in some way mistaken, what is done
through this, and the part of each man that becomes obedient to such a
rule, must be declared to be entirely just and best for the whole of hu-
man life” (863e–864a). The first part of the Stranger’s claim – that the
tyranny of the passions is injustice in the soul – has a certain plausibili-
ty. Yet it is startling to hear the second part of the Stranger’s pronounce-
ment, that justice in the soul is adherence to an opinion about what is
best however a city or certain private individuals may define it – even if that
opinion is mistaken. This has been described by one commentator as
the “good conscience” view of justice.48 Everyone (even the staunchest
libertarian) would rightly be alarmed by such a public doctrine, since it
would justify even the most destructive and misanthropic ways of life,
so long as they were led consistently, in accordance with some prin-
ciple.49 One consideration goes part of the way to assuage the reader’s
44 Punishment and the History of Political Philosophy

alarm: in the Stranger’s republic there will be strict regulation of moral

opinion through extensive educational laws (see especially Books 6–7),
and heavy penalties will be imposed for heresy (see Book 10). This ef-
fectively means that the opinion about what is best that will predomi-
nate shall be the opinion of the law, which in the present context we
may assume to be wise. At the same time, it is necessary to mention that
while the foregoing may be sufficient for the ordinary citizen, there re-
mains the question of what the legislator himself – whether as founder
or as one of the constituted “guardians of the law” (see 752d ff.) – will
look to for guidance as his standard of what is best. Presumably, the
true legislator must ultimately look to reason and nature itself for guid-
ance regarding what is best, or to the gods (see 645b and context).50
With this account, the Stranger appears to have fulfilled his appointed
task of vindicating the traditional distinction, fundamental to all crimi-
nal law, between voluntary and involuntary wrongdoing, despite his
own earlier assertion that anyone who does wrong does it involuntarily.
Kleinias, at any rate, expresses his satisfaction. To repeat, it now appears
to have come to light that crime is voluntary, and thus culpable, in those
cases in which the wrongdoing individual has declined to struggle
against his illicit passions as he ought to have done (and, presumably,
could have done). The Stranger, for his part, seems satisfied as well,
as he does not now repeat his earlier objection to the law’s distinction
between voluntary and involuntary injustices. Perhaps it is enough for
him that the new distinction between the voluntary and the involun-
tary has the advantage of echoing and legally reaffirming the common
civic understanding. Yet it goes without saying that the political utility
– even necessity – of the legal definition of the voluntary hardly suf-
fices for a proof. The Stranger’s silence and apparent acquiescence in
this matter leaves many questions unanswered. As we have seen, the
Stranger had previously gone out of his way to distinguish his own
view from the conventional opinion shared by Kleinias, Megillus, and
the city as a whole, which leads us to suspect that the Stranger ultimate-
ly regards the legal distinction he has elaborated, between voluntary
and involuntary crime, as spurious. It should come as no surprise that
the author of the Republic would be willing to accommodate a certain
degree of ignorance in political life that he (or his Stranger) may regard
as ineradicable. Although the Stranger had earlier said that it would
not have been pious or lawful for him to conceal the truth (cf. 861d), it
is precisely the requirements of lawfulness and piety that now compel
him to silence his philosophical objections.51
 Punishment and Reform in Plato’s Laws 45

The Stranger ends his long discussion of the voluntary by repeating or

summarizing his threefold psychological scheme – ostensibly so that it
may be “still more firmly fixed in the memory” – but with an alteration
that proves to be significant. In the “repetition,” “the striving for expec-
tations and true opinion concerning what is best” replaces ignorance as
one of the three causes of crime (864a–b). The Stranger does not explain
this substitution and appears to regard the repetition as identical to the
original (perhaps because the striving for true opinion may be supposed
to imply ignorance about what is best). Yet the Stranger’s casual manner
of speaking cannot silence our sense of the disproportion between the
repetition and his first statement that the rule in the soul of the “opin-
ion about what is best” is nothing other than justice itself. Apparently, a
republican society will continue to look up to reason and the gods as its
highest authorities, even as in practice it acts to suppress anyone who
strives to ascend from the confused and incomplete public dogmas to
the truth about what is best – an uneasy tension, which may explain
both Socrates’s popularity in Athens and his eventual condemnation
and execution under its laws. This tension or opposition receives its full-
est expression in the Laws in what proves to be the peak of the whole dia-
logue: the theological justification of the law against impiety in Book 10
(a discussion of which is beyond the scope of the present study).
What makes this opposition all the more remarkable is that the
Stranger calls the city of Magnesia, founded by the Stranger himself,
the best practicable regime (albeit only second­-best in point of virtue;
see 739a–e). This may imply, among other things, that Plato believed
that the fundamental tension between political power and knowledge
– a tension that becomes fully palpable in the course of the attempt to
elaborate a truly just penal law – may be endemic to politics, as such,
and not merely to the city of Magnesia. It may also imply that the public
doctrine of criminal justice in a republic (and perhaps in any political
society?) is especially vulnerable to rational critique and thus consti-
tutes a sort of Achilles’s heel of the whole moral order underpinning
the political order. This, finally, would suggest that the justification of
criminal law deserves the special attention of philosophical legislators
and statesmen everywhere.

Completion of the Penal Code

Having completed his disquisition on the theoretical principles of pun-

ishment, the Stranger proceeds to the task of framing Magnesia’s penal
46 Punishment and the History of Political Philosophy

laws. This task occupies him for the remainder of Books 9 and 10 and
sections of Book 11. Plato scholars have remarked on the sharp contrast
between the radically innovative character of the Stranger’s theoreti-
cal principles and the traditionalism of his penal code – which in many
ways (albeit with certain striking exceptions) follows Athenian prac-
tice.52 In the first place, the penal code addresses crime using largely tra-
ditional categories, apparently in order of descending gravity: murder
and homicide (864d–874e); violent assault (874e–884a); insolent things
done by the young (884a–913a); infractions in business and commerce
(913a–922a); abuse of inheritances and orphans (922a–930e); neglect of
parents (930e–932d); a miscellany of other crimes (932e–942a); and mili-
tary offences (942a–945b). While this approach may seem intuitive at
first glance, it is questionable from the point of view of the Stranger’s
focus on the individual offender’s soul and character, as distinguished
from the physical damage caused by his crime. It is true that we tend to
regard the extent of the harm as a sign of the degree of injustice in the
soul – for example, we assume that a murderer is more morally corrupt
than a petty thief. Yet the Stranger has argued that injustice in the strict
sense is the rule of passion in the soul whether or not it does any damage (or
even if it confers some benefit). As we have seen, if applied consistently
this definition would require an extensive overhaul of traditional evalu-
ations of crime and of sentencing practices, in favour of an individual-
ized approach in which the focus shifts from actions to psychological
states. As a further result of this fundamental shift in approach, since the
particularity and wide variety of psychological states would be difficult
to reduce to broad categories, crime would elude its codification into
law – which, in turn, would frustrate the cause of republican govern-
ment, which crucially depends on the rule of law. This, along with the
other difficulties we have had occasion to consider, amply accounts for
the Stranger’s decision to rein in his reformist project.
Second, and more obviously, several legal provisions in the Stranger’s
penal code would, by any sophisticated standard, be considered simple­
minded and even crude. There is an abundance of provisions concern-
ing pollution and purification, omens, curses pronounced by dying
victims, exorcisms, and other such superstitions, which one would be
at pains to fit into the Stranger’s philosophical treatment of crime and
punishment.53 For example, the Stranger legislates that if a man should
be killed by an animal or an inanimate object (unless it’s a lightning bolt
“or another missile from god”), that animal or object should be formally
accused and tried by a judge and then, if found guilty, cast out beyond
 Punishment and Reform in Plato’s Laws 47

the city’s borders (873e–874a). It is clear from all this just how far the
Stranger goes to accommodate the prejudices of the Magnesian repub-
lic – the “matter” upon which the Stranger’s legislative art must work,
and to which it must be adapted, lest he lose his whole labour (cf. 707e–
709a). There is, therefore, less need to pay attention to the fine detail of
the Stranger’s penal code than to his penological arguments.54
Yet the conservative features of the Stranger’s penal code cannot
be allowed to obscure its radical innovations. The most remarkable
of these innovations is the Stranger’s dispensation for the punishment
of young heretics in Book 10. Since this constitutes a very long section
of the dialogue, a summary must suffice for our purposes. The Stranger
(on his own initiative) raises certain critical objections to conventional
piety that would have disastrous consequences for the moral order that
undergirds the political regime of Magnesia – namely, critical doubts
about the existence of providential and just gods. The Stranger attri-
butes these sceptical objections to a few precocious youths among the
citizens, whom he conjures up for the occasion. Instead of dealing with
these young sceptics harshly, he sympathizes with their perplexities
and grants their opinions a kind of formal hearing in the form of a dia-
logue among intellectual equals carried out on a highly sophisticated
level. The core of the views of these young sceptics turns out to con-
sist of the materialist doctrines of the pre­-Socratic natural philosophers,
and the ensuing discussion culminates in a refutation of these doctrines
and a provisional vindication of the theological opinions of the city.55
Following this vindication of the law against impiety, the Stranger pro-
ceeds to lay down a law against anyone who is nevertheless convict-
ed in court on charges of heresy. The heretic is to be judged first as to
whether his crime is truly motivated by his desire to know and to re-
solve his theoretical perplexity or by an unrestrained desire to live law-
lessly (i.e., by his unjust character). Depending on this determination,
the heretic is sentenced to one of two prisons. The restrained heretics
are sent to a prison called a Moderation Tank (sˉophronisterion) for no
less than five years, where they are to be individually reformed, appar-
ently through educational means, by high­-ranking magistrates of the
so­-called Nocturnal Council. The members of this Nocturnal Council,
we later learn, are an elite magistracy, a sort of conclave of philosophic
statesmen (961a–b, 964d–965a, 968a). The unrestrained heretics, on the
other hand, are to be sent to a separate prison in a wild and secluded
part of the territory to live out their lives there as incurably unjust and
dangerous individuals (907d–909d).
48 Punishment and the History of Political Philosophy

The Stranger’s reluctance to treat the perplexed heretics as most tra-

ditional societies and legislators would treat them – namely, to punish
them severely without further ado – and the great lengths to which
he goes to educate them, and even to argue with them as intellectual
equals, truly gives practical expression to the Stranger’s radical defini-
tion of criminal justice.56 It seems possible that the five years spent by
each perplexed heretic in the Moderation Tank will be an opportunity
for him to receive an uninterrupted quasi­-philosophic education from
the best­-educated and most competent elders in the city – an education
modelled after the Stranger’s own dialogue with the fictitious young
objectors. Such a “penalty” would fully meet the high standard of jus-
tice subscribed to by the Stranger himself. This is the only set of provi-
sions in Magnesia’s entire penal code that does depart from traditional
criminal law’s focus on external acts in favour of an individualized ap-
proach that looks to reason or science, rather than to law, as its guide.
The very limited scope the Stranger allows for the sort of innovations
that would be dictated by his radical principles forces us to draw a very
sobering and even somewhat disappointing lesson. Genuine moral re-
form of an offender, according to the Stranger, must involve either effec-
tive rational persuasion of the superior choiceworthiness of justice over
injustice, or the restraint of immoderate desire, or possibly both. In all
three cases, such reform must be carried out on an individualized basis,
by someone with comprehensive knowledge of the human good and
of the nature of the soul and the various workings of its drives, and on
a subject who is receptive to it. Therefore, as we can see for ourselves,
and as the Stranger’s curtailment of individualized criminal justice in
his penal code confirms, the conditions required for genuine moral re-
form – a young and agile mind with a respect for justice, on the one
hand, and a wise and patient educator wielding political authority, on
the other – would be all too rare.57 This implies that in most cases, penal
“rehabilitation” will really amount to one or another form of deterrence,
using threats, inducements, or shame to prevent the criminal from reoff-
ending, but falling short of true moral improvement.58 Moreover, the life
of a law­-revering, economically equal, and stable republic dedicated to
the cultivation of virtue must, to a significant extent, sacrifice scientific
enlightenment (cf. 818a), artistic innovation (cf. 657b), and the cultiva-
tion of the humane or “quiet” virtues that have become predominant
in modern ethical and political thought. These conditions limit the pos-
sibility of humanizing reform even further, beyond the intrinsic difficul-
ties of using punishment to effect moral education.59
 Punishment and Reform in Plato’s Laws 49

Conclusion

As we have seen in this chapter, the character of Plato’s theory of pun-

ishment is in crucial respects informed by his understanding of re-
publican politics. For citizens of the well­-ordered republic of the Laws,
punishment – particularly its proportionality to the crime – is a very
serious moral concern, much more serious than it is, for example, to the
guardian class of the Republic or to the statesman of the Statesman. In a
republic, citizens must govern themselves according to their best un-
derstanding of justice and its requirements. As Plato’s Stranger shows,
however, that understanding of justice is necessarily imperfect. Most
importantly, republican citizens, on the Stranger’s account, are prone
to a peculiar confusion regarding the goodness of justice as a human
virtue. On the one hand, the rigours and responsibilities of republican
life demand an intense love of, and belief in, justice as the highest virtue
leading human beings to happiness. Yet this love of justice is haunted
by a profound doubt – a nagging suspicion that it is, in fact, the unjust
man who truly pursues his own good, not allowing the restraints of
law and justice to stand in his way. Through the Stranger’s complex
psychological analysis, Plato is able to show how this central feature of
the civic outlook powerfully contributes to the moral appeal of retribu-
tive punishment, for insofar as citizens doubt the goodness of justice
and envy the criminal, they will believe that a harm proportional to his
offence must be exacted to negate the “benefit” of the crime. This puni-
tiveness is an ugly fact about republican life, yet it must not be confused
with mere irrational or animal vengefulness. One sign of this is that this
punitive spirit is amenable to some degree of refinement, providing the
legislator (in this case, the Stranger himself) with opportunities to mod-
erate the practice of retributive punishment.
In a political community ruled by laws rather than by enlightened
experts, the correction of offenders must come under legal regulation
no less than any other facet of social life. This means, among other
things, that a republican society must develop workable legal doctrines
of “voluntariness,” “desert,” “guilt,” and “proportionality,” without
which the practice of criminal justice according to laws would be dif-
ficult if not inconceivable. These legal concepts may not be theoretically
transparent, and they may in fact not even be consistent in the strictest
sense with enlightened principles of justice – as the Stranger’s discus-
sion reveals. Nevertheless, a republic must make efforts to reconcile
somehow these useful legal doctrines with justice understood in the
50 Punishment and the History of Political Philosophy

truest sense, and it must learn to do so through a legislative art like

the one practised by the Stranger – at once philosophic and practical.
The Stranger’s approach may serve as a model for republican attempts
to address the thorny moral problems of punishment, especially be-
cause of his unwillingness either to sacrifice open­-minded reflection
for the sake of order, or to abandon political prudence for the empty
triumph of an abstract ideal. Most importantly, the Stranger helps us
understand that the characteristic civic outlook on punishment is inti-
mately linked to its deepest moral convictions, and that the former can-
not be radically uprooted without also profoundly shaking the latter.
The Stranger’s own understanding of human virtue, as choiceworthy
for its own sake and as constitutive of happiness, leads him to radi-
cal conclusions – including the conclusion that truly just punishment
would have to meet a very high standard indeed: that of morally im-
proving the offender himself by teaching him the true value of living a
just life. Understanding this high standard to be all but impossible to
achieve in political life, the Stranger moderates his reformist zeal.
Whatever we think of Plato’s view of republican politics in the Laws,
and of how that view influences his teaching on punishment, an obvi-
ous question now looms: If a highly traditionalist approach to punish-
ment is best suited for a small, isolated, and thoroughly Greek republic,
is it reasonable to assume that it would be equally suitable to our own,
modern, scientifically enlightened, liberal democracies? It goes without
saying that Plato had no experience with modern liberal democracy
and so, arguably, could not have raised such a question for himself.
Only once the social, economic, and political conditions that would
make liberal democracy possible began to appear did serious minds
begin to consider the possibility of unprecedented, wholesale penal re-
form. The remainder of the present study examines the theories of the
greatest of these philosophical reformers and the consequences of their
ambitious reform project. Once we have completed that investigation,
we will have occasion to re­-evaluate Plato’s approach.
 Chapter 2

Modern Natural Right and Punishment

in Hobbes’s Leviathan

Punishment becomes an especially important topic in modern political

thought because the limits of the government’s right to punish serve
to define the boundary separating arbitrary violence from legitimate
political authority over naturally free individuals. This creates the need
for a normative theory of punishment that is able to reliably trace these
limits and ground them theoretically. Hobbes is generally recognized as
a foundational political thinker in the modern tradition; he was also the
first to attempt a systematic normative theory of punishment consistent
with modern natural right – the basic principle of modern politics.
To appreciate the need for a normative theory of punishment in the
modern context, we might consider Machiavelli’s advice to princes
and republics regarding “cruelty well­-used.” According to Machiavelli,
the stability of any political order requires the occasional show of over-
whelming force by the governing authority – veiled, whenever possible,
under a pretext of avenging justice. To have their desired effect of subdu-
ing or pacifying a tumultuous populace, these “punishments” must be
excessive and bloody.1 Hobbes, too, would later conclude that the prima-
ry role of the sovereign is to instill fear in his subjects by an overwhelm-
ing power: an irresistible “Leviathan”­-state, according to the Hobbesian
analysis, is the only way to restrain the politically destabilizing irrational
passions to which human nature is prone. Thus, from the outset, modern
politics’ aim has been peace and security, and its means, fear. The prob-
lem with this political principle, however, as Hobbes himself recognized,
was that its most extreme implications – peacetime curfews and deten-
tion camps, kangaroo courts, mass executions, and so forth – although
not likely to occur with any frequency in practice, would excessively
alarm most law­-abiding people. As a result, the principle itself might be
52 Punishment and the History of Political Philosophy

popularly regarded (rightly or wrongly) as much too threatening and

too degrading to human dignity to be accepted. Hobbes saw that, for the
new political principle to gain acceptance, it would be necessary to lo-
cate somewhere within the theory of absolute sovereign power minimal
normative limitations on the use of that power.
Thus, Hobbes has gained a reputation as the modern theorist of abso-
lute sovereignty, but he is also the author of the first normative theory
of punishment in the modern tradition.2 That theory of punishment, ar-
ticulated in the Leviathan, contains several serious inconsistencies or ten-
sions and, as we will see, ultimately fails. That failure is instructive and
will help us greatly to understand the enduring problems of punish-
ment in the modern context and the subsequent historical development
of punishment theory. As we will see, Hobbes would use two distinct
strategies – each with its own difficulties – to attempt to overcome the
dilemma of having to find limitations on a sovereign power that is abso-
lute. His first strategy was to argue that while the sovereign may not be
limited by any other power, he remains bound by the “laws of nature”
– that is, by precepts of reason forbidding a sovereign to punish his sub-
jects except under specified conditions. The second strategy relied on
Hobbes’s concept of “authorization” and attempted to demonstrate that
a criminal who is punished by his sovereign has actually authorized
(and therefore consented to) his own punishment, since as a subject he
authorizes all of the actions of his sovereign. The analysis in this chapter
will show where these strategies – and Hobbes’s theory of punishment
as a whole – fail, and why they cannot succeed. This ultimate failure
notwithstanding, Hobbes’s theory of punishment contains a radical cri-
tique of traditional criminal justice concepts – a theme that the present
chapter will also develop.
In Hobbes’s teaching on punishment, the most obvious and perhaps
most significant innovation is its unequivocal rejection of retribution.
Classical political thinkers had raised searching questions regarding the
justice of retribution, yet they refrained from pursuing those questions
too far and generally sought to accommodate retribution in their most
political works. In contrast to Plato’s and Aristotle’s cautious and accom-
modating treatments of retribution, however, Hobbes spells out with
the utmost clarity the furthest implications of a thoroughly rationalist
approach to criminal justice. Such an approach implies the rejection of
retributive punishment and, with it, many of the traditional concepts of
criminal justice.
 Natural Right and Punishment in Hobbes’s Leviathan 53

Punishment and the Natural Law against Revenge

To see how the more general assumptions of Hobbes’s political thought

prepare the ground for his theory of criminal justice, we may begin by
considering his denial of a universal standard of goodness, or summum
bonum (Leviathan 11.1).3 Absent an objective standard of goodness – which
Hobbes denies exists on the basis of his subjectivist psychology – our
judgments of others (i.e., of their personal merit or depravity) must be
regarded as merely partial or subjective (10.16). For what we think we are
doing when we judge others as praiseworthy or blameworthy, meritori-
ous or morally culpable, is comparing their actions, and their characters
as reflected in their actions, to some natural and universal standard of
human goodness. Accordingly, Hobbes argues that in the state of nature,
no imputation of merit (or demerit) can carry with it any moral conse-
quences; rather, it can have moral consequences only when it is agreed to
by convention and where this agreement can be enforced by a constitut-
ed sovereign authority (cf. 6.7 and 6.59 with 10.54). In the state of nature,
there is neither justice nor injustice, and therefore there is no such thing
as punishment, let alone normative principles for its regulation.
But once civil society is constituted by the recognition of a sovereign
power, the sovereign has not only a right but also a duty to define what
is good and bad, base and noble, for the regulation of his subjects’ rela-
tions with one another, with a view to their peaceful and prosperous co-
existence. After these artificial standards of goodness and wickedness
have been established, would their enforcement by the sovereign then
have to take the form of traditional criminal laws and punishments?
Hobbes argues that it would not. While it is within the scope of the
sovereign’s authority to impose a standard of human goodness akin to
the classical or Christian models – comprising all the virtues of a perfect
human being – there is no need for this. On the contrary, in opposition­
to the reigning tradition, Hobbes denies that human beings are perfect-
ed through political community and maintains that politics need look
no further than the preservation of peace.4 Furthermore, Hobbes’s ob-
servation of human beings led him to conclude that legislating a mor-
ally demanding doctrine of virtue not only does not promote peace but
in fact has the opposite effect. According to Hobbes, the powerful pas-
sions invoked by any notion of human goodness involving great self-
sacrifice are hard to control – even by threats of physical harm – and are
therefore dangerous to political unity and peace. This has important
54 Punishment and the History of Political Philosophy

implications for criminal law. To enforce a demanding standard of mor-

al virtue in the commonwealth would require sanctions against a host
of activities that in no way endanger anyone’s security or material well­-
being. What is more, when traditional criminal law took its sights by a
standard of human perfection, it invoked the indignation of the public
against those who might fail to live up to that standard. This rendered
traditional punishment retributive and harsher than it needed to be for
purposes of deterrence alone: when criminals are considered to be not
only dangerous but also base, wicked, and impious, the attitude of the
public may be expected to be more severe. On the other hand, were the
sights of politics to be lowered, as they ought to be in the Hobbesian
view, all the severity of traditional punishment would become unneces-
sary. Accordingly, Hobbes’s seventh law of nature is:

That in revenges [i.e., retribution of evil for evil] men look not at the greatness
of the evil past, but the greatness of the good to follow. Whereby we are forbid-
den to inflict punishment with any other design than for correction of the
offender, or direction of others. For this law is consequent to the next be-
fore it [i.e., the sixth law], that commandeth pardon upon security of the
future time. Besides, revenge without respect to the example and profit to
come is a triumph, or glorying, in the hurt of another, tending to no end
(for the end is always somewhat to come); and glorying to no end is vain­-
glory, and contrary to reason; and to hurt without reason tendeth to the
introduction of war, which is against the law of nature, and is commonly
styled by the name of cruelty. (15.19; emphases in original)

According to this passage, some acts of “retribution” or “revenge” would

be appropriate under the law of nature, so it appears as if Hobbes is here
affirming the rightness of retribution rightly understood. That is, rightful
retribution would aim at correction and rehabilitation of the offender, or
incapacitation, or general deterrence (punishing the criminal to set an
example for others), or restitution (compensation of victims), or any
combination of these (and perhaps other) beneficial effects. But it is clear
that in this context Hobbes is giving retribution a wholly unconventional
meaning, since – as ordinary experience abundantly shows and as more
careful analysis confirms (see the preceding chapter) – retribution looks
principally to the past and only incidentally to the future. In other words,
Hobbes here is simply ignoring the whole problem of backward­-looking
punishment. Whereas traditional penal thought had been concerned with
 Natural Right and Punishment in Hobbes’s Leviathan 55

two distinct ends – the just deserts of the criminal and the utility of the
penalty – Hobbes’s teaching decisively narrows the scope of penal jus-
tice to include only the latter.
In his comment on the natural law against backward­-looking retribu-
tion, Hobbes argues primarily from the more fundamental principle of
utility. But he also alludes to another feature of the natural law against
retribution – which might well be described as its humane and human-
izing spirit – by characterizing backward­-looking retribution not mere-
ly as imprudent or ineffective but also as “cruel” and “vain.”
There is some debate among scholars about whether Hobbes’s po-
litical teaching has a moral grounding or merely proceeds from his
materialist psychology.5 This question has some bearing on the issue of
Hobbes’s penal theory. A commonly raised objection to utilitarian pe-
nal theory – whose principle is the “greatest happiness of the greatest
number”6 – is that its attempt to justify the harming of a minority (i.e.,
the punishment of criminals) in the name of the interest of the majority
cannot succeed. The crux of this objection is that when two interests are
opposed to each other, the majority has no greater moral claim than the
minority simply because of its greater number. For punishment to be
truly justified (i.e., for it to count as moral), it would have to be shown
that the criminal somehow deserved his punishment; or that the pun-
ishment was in the common interest (the criminal’s as well as society’s);
or that the sovereign authority had some other ground for imposing the
penalty that was not reducible to partial interest. A penal theory based
solely on the materialist psychology laid out in Part I of the Leviathan
(“Of Man”) would have to be utilitarian and thus would remain open to
the objection just now raised.7
In fact, Hobbes never intended for his political teaching to rest solely on
his mechanistic account of human behaviour as fundamentally selfish.8
He makes this clear in several ways. In the Preface to the De Cive, Hobbes
writes that according to the scientific (resolutive–compositive) method,
political science would follow philosophy and physics and would have
its principles derived from those more fundamental branches of science.
But Hobbes also says there that he was able to complete his political trea-
tise independently, before completing his works of more abstract philoso-
phy, because he “saw that it did not need the preceding parts, since it
rests upon its own principles known by reason.”9 This is confirmed in
the Introduction to the Leviathan, where he says that the principles of po-
litical science derive not from abstract reasoning but rather from careful
56 Punishment and the History of Political Philosophy

reflection on the ordinary experiences of other human beings and on our

own passions – “for this kind of doctrine admitteth no other demonstra-
tion.”10 Nevertheless, these statements leave open the possibility that care-
ful reflection on our experience of human beings may still lead us to the
conclusion that men have always behaved selfishly – and indeed Hobbes
does believe this to be true. Yet Hobbes cannot simply leave it at this, since
such an account of human beings as simple self­-seekers would remain
incapable of solving the problem of vainglory, as Strauss has shown.11 The
passion of vainglory (or pride) poses a special problem because, although
vainglory makes men in society into aggressors, threatening to return
them to the violent state of nature, this fact alone is not enough to allow
Hobbes to reject the pursuit of glory and domination as irrational. That
is because, holding a subjectivist conception of the good as he does, he
must conclude that the enjoyment derived from contemplating one’s own
power (“glorying”) is no less rational an object of our desire (and thus no
less good) than any other pleasure – even when the power we delight
in is imaginary, not real (11.6). A moral basis for politics thus comes into
sight as necessary in order to adjudicate between the destructive selfish
passion of vainglory, on the one hand, and the politically edifying but still
selfish passion of the fear of violent death, on the other. Hobbes solves this
problem by introducing the concept of a natural right to self­-preservation
that is universal (not merely grounded in the subjective desire for self­-
preservation). When one’s life is in danger, even the use of force and fraud
in self­-defence is “generally allowed” – that is, acknowledged by reason
as morally permissible, and not merely as a brute fact of human behav-
iour (13.4).12 This moral insight (moral because grounded in universal or
“right” reason) becomes the foundation for Hobbes’s political teaching
and by extension for his theory of punishment.
True, this remains a peculiar kind of moral foundation. For in Hobbes’s
account, the natural right to self­-preservation is essentially selfish: it nei-
ther confers on the right­-bearer an entitlement to the active assistance of
others nor imposes a natural duty on others to provide such assistance.
Nonetheless, this radically original moral insight – albeit a low­-aiming
one – encouraged Hobbes to develop what might fairly be called a hu-
manizing normative theory of punishment, in the sense that it is com-
patible with the “humane” treatment of offenders, rather than merely
serving as prudential counsel to the sovereign authority as to the most
effective means for preserving security.
The humanizing dimension of Hobbes’s political teaching can be seen
in his argument that human beings in civil society ought to acknowledge
 Natural Right and Punishment in Hobbes’s Leviathan 57

their natural equality (15.21). It is true that this natural equality is de-
rived by Hobbes from the fact that, in the state of nature, no man is so su-
perior as to be able to secure himself against being killed by the weakest.
Yet although this necessity to acknowledge others as our natural equals
is selfish in origin, Hobbes allows for the doctrine of natural equality
to take on a morally edifying, humanizing tone. The pleasure of vain-
glory not only makes men into aggressors but also obscures in them any
awareness of their fundamental vulnerability as individuals, and hence
their equality – indeed, the passion of vainglory necessarily presupposes
the forgetting of one’s vulnerability. In the De Cive, Hobbes states that, al-
though in the state of nature “there is in all men a will to do harm,” they
do not will or intend this “for the same reason or with equal culpabil-
ity.” For, he explains, the one who acts only in self­-defence acknowledges
natural equality and is therefore “a modest man,” whereas the man who
“suppose[es] himself superior to others” and “wants to be allowed ev-
erything” is also blameworthy.13 According to Hobbes, a very common
manifestation of vainglory is punitive anger. Recall that retribution and
the punitive indignation that fuels it aim to secure no material benefit,
but only what is thought to be distributive justice: the restoration of a
certain equilibrium in the distribution of possessions, liberties, and other
goods, such that the unjust do not enjoy greater well­-being than the just.
By their own admission, angry victims and victims’ rights advocates are
concerned with restoring the dignity of victims by depriving offenders
of the equal dignity they no longer “deserve” (e.g., by shaming them).
From a Hobbesian point of view, however, this becomes nothing more
than the desire for a pointless (albeit pleasurable) “triumph” (cf. 15.19),
when considered in light of the fundamental truth of our vulnerability
as individuals. Since our demand for retribution ultimately cannot be
sustained upon the only moral basis recognized by all human beings,
this “moral” need turns out in fact to be nothing other than a desire for a
particularly intense (and for this reason deceptive) pleasure.
Criminals are indeed legally culpable, on Hobbes’s account, since they
pose a threat to peace – but this only makes them liable to punishment
that would provide “caution [i.e., security] of the future time” (15.18).
Nothing that the criminal has done in the past can ever alter the funda-
mental fact of his moral equality with every member of the law­-abiding
majority. The retributivist’s desire for a triumph over criminal offenders
is vain, on Hobbes’s view, and thus also cruel (because pointless) and
not merely imprudent. Put another way, it is cruel because it implies
an almost wilful ignorance about what makes us all human: our equal
58 Punishment and the History of Political Philosophy

vulnerability to death and suffering. In this way, the natural law against
retribution comes into sight not only as a counsel of prudence but also
as an expression of enlightened humanity. A good Hobbesian, although
motivated by nothing other than enlightened self­-interest, would still
be revolted by the barbarism implicit in retributive punishment.
It is worth emphasizing that the foregoing analysis should not be
taken to mean that Hobbes’s natural law against retribution imposes
any kind of absolute duty on the sovereign authority to refrain from all
that may be thought to be “cruel” in punishments. This does not follow
for the basic reason that, for Hobbes, the laws of nature, as such, are
not proper laws, but actually “convenient articles of peace” automati-
cally informing any enlightened reason (ch. 13, end). Nevertheless, it
is important not to overlook Hobbes’s optimism about the capacity of
human beings – selfish though they naturally are – to transform them-
selves morally on the basis of that very selfishness, rightly understood.14
Hobbes’s brief but arresting statements on the natural law against ret-
ribution exemplify this optimism and give evidence of his prescience
regarding the character of what would become the modern moral atti-
tude – the attitude of “humanity” that would become an explicit theme
of liberal political thought after Montesquieu.
Hobbes continues to develop his enlightened and humanizing un-
derstanding of criminal justice in the chapter “Of Crimes, Excuses, and
Extenuations.” The guiding question of this chapter pertains to the na-
ture of crime, and the novelty of Hobbes’s answer to that question can
best be appreciated if we first recall the classical republican conception
of law. According to Plato’s Athenian Stranger, the laws of the city must
encompass all aspects of human life, and most importantly, they must
aim at the greatest good: human virtue and happiness. Consequent to
this view of law, criminal law must be compatible with the overall aim
of improving (or at least not harming) the souls of citizens, including
of course offenders. Since the main burden of improving the souls of
citizens falls on the educational laws, criminal law functions in the sup-
portive role of enforcing morality as well as rehabilitating offenders (for
whom the educational laws were not sufficient). Another function of
the criminal law in the classical republic is to deter that relatively small
number of citizens who are not voluntarily law­-abiding. In Hobbes, by
contrast, this deterrent function becomes the primary and sole aim of
criminal law and punishment (and, indeed, of all law). The traditional
function of enforcing morality through criminal law can no longer be
sustained: since all private ends, including the pursuit of happiness,
 Natural Right and Punishment in Hobbes’s Leviathan 59

are beyond the law’s scope, individual subjects may remain at liberty
to pursue happiness as they understand it (see ch. 21). From this it fol-
lows that “where law ceaseth, sin ceaseth” (27.3).15 Since in a Hobbesian
commonwealth the laws will not be paternalistic, the scope of the crimi-
nal code will necessarily be much more limited. Hobbes’s gloss on the
Sermon on the Mount in this chapter illustrates this momentous change
of orientation. According to Jesus, we are guilty of sin (and liable to
punishment) not just when we offend against the law in deed, but even
when we merely contemplate the offence. Conversely, Hobbes goes
very far in the opposite direction: “The consideration of this has made
me think them too severe, both to themselves and others, that maintain
that the first motions of the mind (though checked with the fear of God)
be sins” (27.1).
The humanizing implications of the Hobbesian conception of criminal
law are further revealed in subsequent discussions of ignorance, the pow-
er of the passions, and the possible complicity of the criminal law itself in
the propagation of crime. In these discussions, Hobbes goes much farther
than the Socratic philosophers ever dared go in openly drawing out the
implications of a thoroughly rationalist approach to criminal justice.
Hobbes’s explicit view that “the source of every crime is some defect
of the understanding, or some error in reasoning, or some sudden force
of the passions” (27.4) is in perfect harmony with the private view of
Plato’s philosophic Stranger (although the latter was forced to qualify
his view in his public statements as legislator and teacher of legisla-
tors).16 Plato and the other Socratics had cautiously indicated that the
confines of our moral knowledge (and hence, of our competence for re-
sponsible action) and of our capacity for self­-restraint in the face of pow-
erful passions are actually much narrower than most people believe.17
Nevertheless, Plato’s Stranger encouraged lawgivers and magistrates to
try to rehabilitate offenders through measures designed to chasten their
passions, increase their capacity for self­-restraint, and instil in them a
love for justice. Hobbes, by contrast, unambiguously maintains that hu-
man nature is largely unchangeable or incorrigible. “As for the passions
of hate, lust, ambition, and covetousness, what crimes they are apt to
produce is so obvious to every man’s experience and understanding, as
there needeth nothing to be said of them, saving that they are infirmities
so annexed to the nature, both of man and all other living creatures, as
that their effects cannot be hindered but by extraordinary use of reason,
or a constant severity in punishing them” (27.18; cf. 27.1). Thus, accord-
ing to Hobbes, law must not try to transform human nature, but must
60 Punishment and the History of Political Philosophy

try to accept it and keep it within acceptable bounds by constructing in-

stitutions that will be able to work with the passions rather than directly
against them. So it is futile for the commonwealth to attempt to eradi-
cate the causes of crime in man’s naturally selfish drives. Nor would it
be wise to rely on human reason alone to teach men that their own tru-
est interests lie in upholding the laws. Rather, the commonwealth must
try to hinder the harmful effects of the passions by relying on another
powerful passion – fear – to keep men in check.
This is certainly a much less flattering picture of human nature than
the one taught by the ancients, for it implies that almost none of us
is strong enough to resist committing injustice without the threat of
punishment to restrain us: “whensoever the hope of impunity appears,
their effects proceed” (27.18). Yet this view of human nature also makes
possible the lowering of the demands we make on our criminal justice
institutions, thereby improving our chances of achieving the remaining
(much more attainable) demand for security. Since the self­-regarding
passions are natural and ineradicable (and thus should not be con-
demned as “base”), and since punishment no longer needs to concern
itself with the moral rehabilitation of criminals, the framing of criminal
codes with a view to deterrence can be done much more simply and
systematically (without the need to weave together contradictory ele-
ments, as Plato’s Stranger attempted to do in his own penal legislation).
According to the classical view, injustice in the soul has little relation to
the harm caused by the crime (see above, chapter 1, 39–41). Since the
nature of the particular injustice in the soul varies for each individual,
as does the type of punishment necessary to correct it, it is difficult to
predict – and thus difficult to codify – what punishments will be suit-
able for what crimes. But while this individualized approach to injus-
tice may be appropriate for the purpose of rehabilitation (at least on the
Platonic–Aristotelian assumptions), it is not quite suitable for the pur-
pose of deterrence. That is because deterrence depends entirely on the
predictability or certainty of the punishment (a point on which Hobbes
expands in the following chapter), which means it depends on it being
known that there is no way to escape it or to lessen its severity by ap-
pealing to the discretion or partiality of judges or assemblies. Thus the
wide discretion of judges that is implied by the rehabilitative approach
to crime actually works at cross­-purposes with the requirements of de-
terrence. The narrowing of the scope of Hobbesian criminal law, on the
other hand, allows for the restriction of judicial discretion and the fur-
ther refinement of legal categorizations of crime.18 This increase in the
 Natural Right and Punishment in Hobbes’s Leviathan 61

formal sophistication of criminal law would be advantageous both for

its effectiveness – since deterrence requires predictability – and for its
fairness – since defining a crime in the law in advance takes away from
the appearance of arbitrariness in a judicial ruling.19 This advantage of
Hobbesian criminal law is distinct from, and compatible with, the ad-
vantage already discussed of affording greater freedom from coercion
for private thoughts and feelings.
Finally, we may mention another advantage of Hobbes’s novel con-
ception of crime, as it emerges from his analysis. While classical repub-
lican theory was openly hostile to the nakedly self­-regarding desires for
physical pleasure, gain, and personal safety, since these desires were
thought to undermine the public­-spiritedness of citizens and the unity
of the city, it was more hospitable towards the passion of anger. As we
saw in the previous chapter, the laws of Plato’s Magnesia were delib-
erately ambiguous towards anger – at times even praising it outright
– since Plato’s Stranger understood this passion to be capable of oppos-
ing the selfish desires and becoming allied to the public interest. Hobbes
inverts this relationship, clearing the selfish desire for comfortable self­-
preservation of all opprobrium while reassessing the public function of
anger and judging it more harmful than useful. Hobbes thus dispenses
with the complicated and precarious rhetorical approach towards the
vengeful passions employed by Plato and Aristotle. By accommodat-
ing anger in the interest of public unity, the republican tradition was
in fact allowing a powerful cause of crime – for “there are few crimes
that may not be produced by anger” (27.17) – to become entrenched in
the commonwealth. Hobbes’s security strategy liberates political theory
and practice from this troublesome inconsistency.
In the spirit of this rationalist attitude towards crime, Hobbes shows
how the sovereign authority ought to moderate its criminal code (both
for the advantage of its own power and as a reflection of its humanity)
in a section on “excuse” and “extenuation” (27.21 ff.). “Where a man is
captive or in the power of the enemy (and he is then in the power of
the enemy, when his person, or his means of living, is so) if it be with-
out his own fault, the obligation of the law ceaseth, because he must
obey the enemy or die; and consequently such obedience is no crime.”
An enlightened understanding of the weakness of human beings and
of their natural self­-regard liberates us from the kind of indignation
that demands that individuals remain loyal to their country even when
they, or their goods, are under the control of enemies. We may observe
that this forgiving policy extends even to offenders who break the law
62 Punishment and the History of Political Philosophy

out of concern for their property (their “means of living”). This holds
true also for any man who breaks the law when “destitute of food or
other things necessary for his life … as if in a great famine he take the
food by force or stealth [possibly from another, equally hungry man],”
for then he is “totally excused.”
This accommodation of human weakness also leads Hobbes to ex-
cuse or extenuate injustices that may have been encouraged by failures
of the sovereign authority itself:

The same fact, if it have been constantly punished in other men, is a great-
er crime than if there have been many precedent examples of impunity.
For those examples are so many hopes of impunity given by the sovereign
himself; and because he which furnishes a man with such a hope and pre-
sumption of mercy as encourageth him to offend hath his part in the of-
fence, he cannot reasonably charge the offender with the whole. (27.32)20

Similarly, “those facts which the law expressly condemneth, but the
lawmaker by other manifest signs of his will tacitly approveth, are less
crimes than the same facts condemned by both the laws and the law-
maker.” Here Hobbes gives the example of duelling – condemned by
law throughout Europe, but sanctioned by well­-bred custom as an hon-
ourable way of defending one’s name. In these and many other ways,
Hobbes demonstrates what has since become a truism in modern crimi-
nology: that a significant cause of crime in traditional societies had been
inconsistency in the criminal laws themselves, and in their enforcement.
In an interesting inversion of priorities, we are made to see that what is
needed is not the reform of individual offenders – as per republican
theory – but the reform of the criminal justice system itself.
On the other hand, in contrast to much of contemporary criminology,
which locates responsibility for crime not in individuals but in their so-
cial environment (that is, in poverty, broken families, poor educational
opportunities, etc.), Hobbes is much more hopeful about the capacity
of individuals to respond rationally to credible threats from the civil
authority. Notwithstanding his view of the power of the passions and
the weakness of human reason, Hobbes assumes throughout that deter-
rence can be effective (a proposition contested by some of today’s crim-
inologists and psychologists), for “there is no suddenness of passion
sufficient for a total excuse; for all the time between the first knowing of
the law and the commission of the fact shall be taken for a time of de-
liberation, because [the would­-be criminal] ought, by meditation of the
 Natural Right and Punishment in Hobbes’s Leviathan 63

law, to rectify the irregularity of his passions continually” (27.33). This

expectation of self­-control should not, however, be confused with the
sort of self­-denying virtue praised by classical thought, since Hobbes is
not relying on reason to rule the passions, but on certain selfish passions
(viz., fear of punishment) to oppose the others with the aid of reason.21
Hobbes’s discussion of crime ends with a consideration of the sub-
jective – or what we might call cultural – component of any criminal
code. The logic of Hobbes’s political teaching implies that only attacks
on property, or on individual or collective security, ought to be criminal-
ized by the commonwealth (the latter because they endanger life, the
former because they make life onerous). But in actuality, most politi-
cal communities criminalize all sorts of activities that have little to do
with security. In some countries, public indecency (e.g., public nudity
or profanity), adultery, and polygamy are criminal offences, although it
is not clear how such activities would endanger either life or property.
In others, atheism is criminalized, while in still others certain public
expressions of religion are illegal (such as the French law against face­-
covering). While the logic of Hobbes’s analysis of crime would seem to
favour the decriminalization of all such activities by commonwealths,
he understood that even enlightened societies would likely remain at-
tached to traditional institutions like the family, and to some traditional
virtues, and that their criminal laws might have to be made to reflect
that attachment.22 Accordingly, Hobbes distinguishes crimes against the
commonwealth (i.e., against its peace, order, and authority) from crimes
against private individuals. It is clear that some offences against indi-
viduals, especially violent and property offences, also imply a danger
to the public (cf. 27.54), but there are others – such as the ones just men-
tioned – that do not. The question is how an enlightened society ought
to treat such merely perceived offences against some of its members,
when the natural law is silent regarding any matter that does not some-
how affect comfortable self­-preservation.
Hobbes’s answer to this question is simply that the sovereign should
consider the sentiments of his subjects and seek to accommodate them
as much as is possible within the bounds of the natural law by enacting
criminal laws (or tolerating existing criminal laws) that reflect those sen-
timents. Thus, when comparing crimes, the greater crime “is that where
the damage, in the common opinion of men, is most sensible” (27.41;
cf. 27.49). Therefore, “mutilation of a limb” (aggravated assault) can be
counted as a greater crime “than the spoiling of a man of his goods”
(theft), since the former seems to most people to be a more serious
64 Punishment and the History of Political Philosophy

offence than the latter (when in fact it is quite possible that certain thefts
can be more damaging to one’s future well­-being than non­-fatal bodily
injuries). Similarly, Hobbes argues that it might be popularly consid-
ered more offensive to kill one’s parent than to kill a non­-relation; to
commit a crime “in the time or place appointed for devotion” than in
any other time or place; to rape a married woman than a non­-married
one (27.48, 51, 52). In the last case, however, we may observe that most
contemporary Western societies have advanced beyond even Hobbes’s
expectations, since our criminal laws now consider rape equally bad
whether the victim is married or not, and Hobbes would certainly have
applauded such an enlightened and egalitarian development.23
We have now arrived at the second crucial moment in the develop-
ment, in the Western rationalist tradition, of the idea of criminal justice
since the original classical formulation. In our earlier discussion of pun-
ishment in the context of classical republican theory, we observed that the
gentle and self­-contented (one might even say aloof) outlook on crime
embodied by Plato’s Athenian Stranger, which inclines towards feelings
of pity for the criminal on account of his confused and disordered soul,
cannot be a model for republican citizens, who must always remain vigi-
lant and adhere to a selfless conception of virtue. By fundamentally re-
thinking what it would mean to live in a well­-ordered society, and by
abandoning the classical republican ideal, Hobbes made it possible for
rational penal reform to escape the limitations of classical republican life.
The society envisioned by Hobbes would encourage private self­-interest
and rely less on the cultivation of selfless virtue with all of its accompany-
ing harshness. This vision allows Hobbes to take the first steps towards a
thoroughly modern theory of punishment.

The Aim and Limits of Punishment

The Leviathan begins the second of its two chapters devoted to criminal
justice, “Of Punishments and Rewards,” by defining punishment as “an
evil inflicted by public authority on him that hath done or omitted that
which is judged by the same authority to be a transgression of the law,
to the end that the will of men may thereby the better be disposed to
obedience” (28.1).24 This definition of punishment derives directly from
Hobbes’s earlier discussion of the natural law against retribution and
clearly restricts the aim of punishment to deterrence alone. Although
we have already discussed several features of Hobbes’s penal theory
that align with enlightened humanity, there is no doubt that the sword
of justice in a well­-ordered commonwealth will appear repressive. In a
 Natural Right and Punishment in Hobbes’s Leviathan 65

way, this is unavoidable, since in order for punishment to be effective

it must control the powerful passions in all men through fear, and this
requires that it be sufficiently severe: the aim of punishment is “ter-
ror” (28.10). This may conjure up an apprehension that such a system
of punishment would be tyrannical, arbitrary, or “Machiavellian.”
Anticipating this fear, Hobbes attempts to show that a rational penal
system would be circumscribed within certain reliable limits.
Hobbes begins his elaboration of the proper limits of the punishing
power by clarifying the origin of the right to punish:

Before I infer anything from this definition, there is a question to be an-

swered of much importance, which is: by what door the right or authority
of punishing in any case came in? For by that which has been said before,
no man is supposed bound by covenant not to resist violence; and conse-
quently, it cannot be intended that he gave any right to another to lay vio-
lent hands upon his person. In the making of a commonwealth, every man
giveth away the right of defending another, but not of defending himself.
Also, he obligeth himself to assist him that hath the sovereignty in the
punishing of another, but of himself not … It is manifest therefore that the
right which the commonwealth (that is, he or they that represent it) hath to
punish is not grounded on any concession or gift of the subjects.
But I have also showed formerly that before the institution of common-
wealth, every man had a right to everything, and to do whatsoever he
thought necessary to his own preservation, subduing, hurting, or killing
any man in order thereunto. And this is the foundation of that right of
punishing which is exercised in every commonwealth. For the subjects did
not give the sovereign that right, but only (in laying down theirs) strength-
ened him to use his own as he should think fit, for the preservation of
them all; so that it was not given, but left to him, and to him only, and
(excepting the limits set him by natural law) as entire as in the condition of
mere nature, and of war of every one against his neighbor. (28.2)

Since the natural right to defend one’s life and limb is prior to, and the
foundation of, all obligations, it is inconceivable that anyone could
grant another the right to do him physical harm for any reason.
Consequently, the sovereign’s authority to punish subjects must be
grounded not in the consent of the subject but in the absolute freedom
of the state of nature, which the sovereign retains undiminished even in
civil society, as he is not a party to the social contract (17.13).25
Yet despite this straightforward deduction of the right to pun-
ish from the unlimited right of nature, Hobbes seems to wish to avoid
66 Punishment and the History of Political Philosophy

unequivocally drawing the conclusion that the sovereign’s right to pun-

ish is equally unlimited. Everything depends on what is meant by Hobbes
when he says that the sovereign’s right to punish is absolute “excepting
the limits set him by natural law.” Hobbes repeats the idea that the natu-
ral law forbids all punishment “which is inflicted without intention or
possibility of disposing the delinquent (or by his example, other men) to
obey the laws.” According to Hobbes, any punishment that does not do
so is no punishment at all, “but an act of hostility” (28.7). But is it quite
appropriate to call this reason for refraining from imposing certain kinds
of punishment a true limitation on the sovereign? Assuming that Hobbes
is right to claim that fear of violent death and the desire for comfortable
self­-preservation can be reckoned on to preserve political order (cf. 13.14,
14.3), and assuming that the prudent sovereign wishes for nothing more
than to enjoy stable rule, Hobbes’s admonition to the sovereign not to
violate the natural law against excessive punishments is not a restraint,
but a counsel to do what is in his own best interest – and no limitations
whatsoever are said to restrict him in the pursuit of that interest.
Hobbes next draws a series of inferences from his definition of pun-
ishment and his conception of its origin in right. One of these is that a
penalty should not exceed the amount of harm stipulated in the positive
laws of the commonwealth. “For seeing the aim of punishment is not a re-
venge, but terror, and the terror of a great punishment unknown is taken
away by the declaration of a less, the unexpected addition is no part of
the punishment” (28.10). Penalties that are not announced in advance are
not advantageous for the sovereign, for the simple reason that they can-
not deter. In fact, by their very unpredictability, extralegal punishments
may even undermine respect for law by making punishment seem wilful,
while allowing would­-be criminals to hope to avoid the harshest penal-
ties by appealing to the sovereign’s mercy (cf. 27.32).26 Another inference
from Hobbes’s definition of punishment is that offenders have a right to
public trial before they can be punished (28.5; cf. 23.9). Any punishment
that is not preceded by a public conviction has no ability to deter, since
no one (besides the sovereign and his ministers) could know what the
punishment is for. For similar reasons, the law of nature forbids inflicting
punishment for an action that was criminalized ex post facto (28.11). Where
no law exists prohibiting an action, punishment cannot be anticipated and
thus can have no deterrent power. Similarly, Hobbes denies that pre­-trial
incarceration can be considered punishment, since no harm imposed be-
fore culpability has been determined can have any deterrent effect (28.20).
 Natural Right and Punishment in Hobbes’s Leviathan 67

There is a danger that modern readers (who may already agree with
Hobbes that deterrence ought to be the sole purpose of punishment)
may underestimate the provocative import of Hobbes’s arguments. In
what we have seen so far, Hobbes’s arguments lay out in general terms
what policies would serve the sovereign’s interest in maintaining his
rule, and he allows his readers to suppose that such policies would
converge with individual subjects’ concerns about fairness, justice, and
their own protection under the law. This convergence cannot be as-
sumed a priori, however. On the contrary, it becomes clear that Hobbes
is begging some very important questions in this regard when he even-
tually undertakes to show that, for all practical purposes, the natural
law would forbid the punishment of innocents.
Punishment, according to Hobbes, can only be “for transgression of
the law … and therefore there can be no punishment of the innocent”
(28.22). Yet the derivation of the right to punish has left open the logical
possibility that the harming of subjects may sometimes be justified by
the requirement of general security even when no law has been broken.
Hobbes’s commitment to an unlimited right of nature prevents him
from invoking any prior norms or duties that might restrict sovereign
authority.27 And since he cannot argue against harming the innocent in
principle, he is reduced to arguing against this on grounds of political
utility. According to Hobbes, the natural law forbids harming the in-
nocent because “there can arrive no good to the commonwealth.” But,
again, this merely assumes that harming the innocent can never be po-
litically expedient, whereas in fact it is possible to imagine a scenario
in which it is. When an angry and riotous mob believes some innocent
man to be responsible for a very serious crime he did not in fact commit,
the prudent sovereign may have to execute the unfortunate man under
a false pretext of justice, if this is the only available means for placating
the violent mob. Could Hobbes really have failed to imagine such a pos-
sibility? In fact, Hobbes openly admits elsewhere that if the sovereign
does violence to an innocent subject, he does so without injustice, as the
Biblical David did to Uriah and as Jephtha did to his daughter (21.7).
Some contemporary utilitarian punishment theorists might be in-
clined to defend Hobbes here by resorting to the “definitional stop.”
According to this argument, harming the innocent may perhaps some-
times be justified by political expediency, but it can never be justified as
punishment, since, as Hobbes says, punishment is only “for transgres-
sion of the law.” Yet this argument fails because it assumes that the
68 Punishment and the History of Political Philosophy

stated aim of punishment – “correction of the offender, or direction of

others” – is logically dependent on an actual transgression of the law,
whereas in fact there is no reason why it should be, as can be seen from
the hypothetical example I have sketched. Once Hobbes made the deci-
sion to abandon the connection between punishment and the common-
sense notion of desert, he could no longer recur to this notion as a way
of establishing a binding limit on the punishing power. Moreover, as we
have seen, Hobbes cannot avoid abandoning the commonsense notion of
desert because this is one of the premises of his wholesale rejection of
backward­-looking retribution. To put it more positively, Hobbes intend-
ed to liberate punishment from its irrational origins in ordinary political
opinion by limiting its scope to the rational aim of security – an objective
state of affairs that, according to him, has no relation to our speculative
opinions about inner “worth,” “desert,” “guilt,” or “innocence.” Yet the
price of rationalizing punishment in this way is that one must also accept
its radical moral implications. Normally, when we say that the innocent
ought not to be punished, we do not mean that “nothing useful would
be accomplished by punishing such individuals.” Rather, we mean that
if we punished them, this would somehow upset the scales of justice, or
it would conflict with our sense that good things should happen to good
people and bad things only to bad people, according to some rough pro-
portion, which we may have difficulty defining precisely, but which we
apprehend when it is violated. If these are indeed the thoughts behind
the ordinary objection to punishment of the innocent, then the objection
cannot be maintained within a Hobbesian framework, from which the
principle of retribution has been expunged.28
Perhaps it is because he was aware of the morally extreme implica-
tions of his principles that Hobbes still tried his best to argue against
punishment of the innocent on commonsense, non­-utilitarian grounds
– although this would make such arguments merely rhetorical. In the
same chapter, Hobbes claims that in addition to being imprudent,
punishment of the innocent is also forbidden by the natural law com-
manding gratitude (rendering good for good) and by the natural law
commanding equity (equal distribution of justice). Punishing inno-
cents amounts to rendering evil for good, which is ingratitude, and it
amounts to a violation of equity, since it imposes harm on some inno-
cents but not on all (28.22). We can see just how misleading these argu-
ments are if we try to relate them to Hobbes’s theoretical framework.
In an earlier part of the Leviathan, Hobbes had established gratitude as
being due to benefactors because “no man giveth but with intention
 Natural Right and Punishment in Hobbes’s Leviathan 69

of good to himself,” since when that intention is frustrated “there will

be no beginning of benevolence or trust; nor, consequently, of mutual
help, nor of reconciliation of one man to another; and therefore they
are to remain still in the condition of war, which is contrary to the first
and fundamental law of nature, which commandeth men to seek peace”
(15.16; emphasis in original). The natural law commanding gratitude
is, therefore, itself justified as a consequence of the fundamental right
to self­-preservation. Similarly, “equity” is defended with a view to the
consideration that, without impartial treatment of persons, “the contro-
versies of men cannot be determined but by war” (15.23).
The preceding does not, however, mean that we should dismiss
Hobbes’s suggestion that a deterrence­-based criminal justice system
would be more humane on the whole (in the sense that it would be
milder than the alternative). For it seems plausible that a criminal justice
system informed by Hobbesian principles would dispense with many
punishments (all of those that in no way contribute to security or that are
purely retributive) and that it would be receptive to adopting the mild-
est possible modes of punishment, so long as they are effective (since
suffering is no longer itself the end of punishment). To anticipate later
chapters of the present study, it is fair to say that the impressive accom-
plishments achieved by the modern criminal justice system were first
made possible by Hobbes’s unleashing of the punishing power from its
traditional moral restraints. This has led to the transformation of crimi-
nology into a technical science, which in turn has vastly increased the
sophistication of modern criminal justice policy analysis. Nor can we
overstate the significance of Hobbes’s new conception of criminal jus-
tice for the many modern innovations in criminal procedure (in investi-
gation, trial, appeal, etc.), all of which have minimized the miscarriage
of justice due to human partiality. Having said this, we should not be
too quick to attribute these happy developments to a corresponding im-
provement in the moral understanding of criminal justice. The extent to
which Hobbes’s theory of punishment succeeds in convincing us of the
justice and nobility of the kind of criminal justice system he proposed
can only be judged after we have traced further historical developments
in the practice and theory of modern criminal justice.

Punishing the Guilty and the Doctrine of Authorization

The preceding analysis has shown that Hobbes’s theory of punish-

ment implies that it is permissible to punish the innocent in certain
70 Punishment and the History of Political Philosophy

circumstances, in conformity with natural law. This, in turn, led us

to observe that in Hobbes’s framework, words like “innocence” and
“guilt” lose their ordinary meaning. Admittedly, for reasons stated by
Hobbes, eventualities such as punishing the innocent would be exceed-
ingly rare for practical purposes. We are inclined simply to ignore (as
merely “academic”) the thorny moral problem of justifying such a step
should it ever come to pass, and this is just what Hobbes himself does
by concealing the problem under rhetorical obfuscations. Yet the same
problem re­-emerges in another context in which it cannot be ignored:
the much more common case of punishing the guilty.
The requirement of justifying punishment of the guilty is itself often
ignored. While we have immediate and strong reservations about pun-
ishing the innocent, we have no such objections to punishing the guilty.
On the contrary, when the guilty suffer punishment, we say they had
it coming without thinking to ask why we believe this to be true. Thus
when we were confronted with the difficult case of punishing the inno-
cent in a political emergency, we demanded justification, but there does
not seem to be the same urgency to look for a justification for punishing
convicted criminals. Rather, the controversy surrounding punishment
of criminals is commonly about how much and what kind of punishment
they should receive. The need to justify the punishment of criminals
cannot, however, be escaped in the context of Hobbes’s thought,29 where
we find him once again trying to soften the shock of the paradoxical
implications of his political rationalism.
There is nothing inconsistent in Hobbes’s claim that the sovereign has
the right to punish convicted offenders even while the offender has an
equal right to resist. It is true that the criminal, as Hobbes understands
him, is completely guiltless in the strict sense (he is no more morally
blameworthy than any other human being, since we are all motivated
by the same desires that are beyond our control) – yet this does not
make the criminal immune from punishment. What is more, the crimi-
nal is entirely justified in resisting his punishment, as his right to defend
himself from harm is the first and most fundamental right (cf. 14.8, 29).
The subject’s obligation to his sovereign becomes void as soon as his
life or means of existence become threatened, which is the case when
he is sentenced to punishment (21.12).30 None of this poses any theo-
retical difficulty for the punishment of criminals: when the criminal’s
obligation to his sovereign becomes void, he returns to the state of na-
ture, in which nothing can be just or unjust (since “notions of right and
wrong, justice and injustice, have there no place” – 13.13), from which it
 Natural Right and Punishment in Hobbes’s Leviathan 71

follows that, just as the criminal has a right to defend his life and limb,
the commonwealth has a right to restrain or kill the dangerous criminal
as an enemy in the state of nature. Thus, there exists a clash of rights in
the conflict between the criminal and civil society without any injustice
being committed on either side. This is because, for Hobbes, a right is
essentially a liberty – nothing more or less (14.3) – and as such it does
not impose on the sovereign the obligation to respect the subject’s right
to resist (just as in the state of nature no one is obliged to respect any
other’s right to preserve himself).31
Despite the unimpeachable consistency of Hobbes’s arguments in de-
fence of society’s right to coerce anyone who poses a danger to it, this
doctrine remains extreme from the point of view of ordinary morality.
Hobbes himself saw this as a problem to be addressed.32 One way in
which Hobbes shows his awareness of this as a problem is by insisting
on characterizing the coercion of criminals by the sovereign as “pun-
ishment,” explicitly distinguishing this from an act of “hostility.” He
attempts to substantiate this distinction by arguing that, although the
criminal has a right to resist his punishment, that punishment is never-
theless not an act of hostility because the criminal himself authorizes his
own punishment, just as he authorizes all of his sovereign’s actions (28.3
and 6; cf. 18.3). Accordingly, the first and fourth inferences of the natu-
ral law regarding punishment state that any evil inflicted by anyone
other than the sovereign is an act of hostility “because the acts of power
usurped have not for author the person condemned,” as (it is implied) they
ought to have if they are to be considered punishment in the strict sense.
One difficulty with this doctrine of contractarian authorization of
punishment is that Hobbes has already shown that the sovereign’s right
to punish is not a grant and thus does not rest on the consent of the
punished, but on the right of nature. It might be argued that Hobbes
was forced to conclude that the criminal authorizes his own punish-
ment as a consequence of his more basic argument (elaborated earlier in
the Leviathan) that each subject authorizes all of his sovereign’s actions,
without exception, by virtue of the nature of the original social contract
(17.13, 18.6, 21.10, 14). But this only compels us to ask why the subject
must authorize all of his sovereign’s actions in the first place. If the end
of civil society, and thus of sovereignty, is to secure the safety of each
– since this is the condition of the obligation of each to his sovereign
– then does it not follow from this that subjects should authorize the
sovereign’s actions only up to the point that they actually serve this pur-
pose?33 Hobbes admits, however, that punishments – or at least severe
72 Punishment and the History of Political Philosophy

corporal and carceral punishments – seriously threaten the well­-being

of the criminal and thus cannot reasonably be construed as having the
criminal’s preservation as their end. Furthermore, how are we to make
sense of the strange consequence that the criminal authorizes his own
punishment yet at the same time rightfully resists that same punishment,
as if he could both will and not will his punishment simultaneously?
Hobbes’s most direct attempt to make sense of the latter difficulty
runs as follows.

Again, the consent of a subject to sovereign power is contained in these

words I authorize, or take upon me, all his actions, in which there is no restric-
tion at all of his own former natural liberty; for by allowing him to kill me,
I am not bound to kill myself when he commands me. It is one thing to say
kill me, or my fellow, if you please, another thing to say I will kill myself, or my
fellow. (xxi.14; original emphases)

This argument turns on what seems to me to be a specious assumption

that the subject can authorize his own death without authorizing the
means to it. It is specious because it is hard to imagine how one can truly
consent to the end (the punishment) without also consenting to the
means to that end (e.g., turning oneself in to the authorities instead of
fleeing the country). It seems to me that one must will either the end and
the means, or neither. That being so, it would make Hobbes’s attempted
solution here meaningless, or merely formal.34 I am consequently in-
clined to agree with those who regard Hobbes’s doctrine of authoriza-
tion as a rhetorical strategy, since his theoretical framework does not
actually depend on this doctrine.35
By holding that the criminal authorizes his own punishment, Hobbes
makes punishment appear to take place entirely within the civil order,
rather than in the state of nature as a conflict between two hostile pow-
ers. This conforms better to the commonsense point of view that crimi-
nals remain members of civil society despite having offended against
its laws, and are still entitled to respect as fellow members of that so-
ciety. In actual practice, while we condemn criminals as unjust and
blameworthy, we still consider them our fellows for whom we cannot
help caring and to whom we still believe we owe some obligations. It
seems impossible to accept that as soon as a citizen has been convicted
of a punishable crime, he is immediately expelled from the community.
What is more, as Hobbes’s new account of criminality points to the fun-
damental guiltlessness of the criminal’s motives, there remains even
 Natural Right and Punishment in Hobbes’s Leviathan 73

less reason in modern societies influenced by Hobbes to treat criminals

with contempt than there had been in traditional ones informed by the
retributive outlook. The rhetorical assertion that the criminal autho-
rizes his own punishment was, then, essential for Hobbes to be able
to present his new conception of punishment persuasively, as not only
effective but also humane.

Conclusion

Hobbes’s theory of punishment was the first moral theory of punish-

ment constructed entirely on modern principles. What makes it re-
markable, and what enables it to set the tone for all subsequent modern
thought on criminal justice, is its attempt to show that a penal system
aiming at deterrence could be not only highly effective but also hu-
mane.36 So it would be a mistake to classify Hobbes as a teacher of a
purely utilitarian conception of crime and punishment.
Hobbes also contributed to the improvement of modern punishment
by making it possible to increase the effectiveness of deterrence by en-
couraging the scientific study of crime and by sketching the outlines of
a more rational criminal law aimed at deterrence. It is noteworthy that
his definition of deterrence – “that the will of men may thereby be better
disposed to obedience” – is capacious enough to include rehabilitation
(as long as it serves the ultimate goal of crime prevention). This, as we
will see, cleared the way for later rehabilitation movements in Western
countries (the reform of criminals through prison labour, drug rehabili-
tation programs, community reintegration programs). Hobbes himself,
however, was content to leave men as they were – with their selfish pas-
sions largely unchanged – and thus did not feel compelled to stress the
state’s potential role in dispensing non­-punitive preventive care to past
offenders who were predisposed to crime by psychological illness, drug
addiction, or dysfunctional family settings. But how successful was the
Hobbesian marriage between utility and humanity? How persuasive
is he when he assures us that the humane treatment of criminals never
conflicts with social utility? This question fundamentally informs the
subsequent development of modern penal theory.
As we have seen, Hobbes had no qualms about spelling out, in the
clearest possible terms, that one particular consequence of his concep-
tion of natural right was that retribution could no longer be justified.
On Hobbes’s principles, the traditional demand for retributive pun-
ishment could only be understood as rooted in cruelty and vainglory.
74 Punishment and the History of Political Philosophy

Nevertheless, while his moral psychology had relativized the content

of the good, this relativism did not extend to the content of justice. The
meaning of justice is not subjective but universal, according to Hobbes’s
analysis. According to that analysis, making covenants and enforcing
them through rationalized punishment is the only means by which
human beings can secure their lives and livelihoods, and this remains
true despite the diversity and mutability of the objects of men’s desires.
There is therefore much in Hobbes’s thought on crime and punishment
that is promising – and, we might even say, morally edifying – but there
is also much left to be desired. His argument against retribution is con-
sistent with his fundamental political principles, but from the point
of view of ordinary moral sensibilities it merely begs the question by
means of a simple consequentialism. This leaves us to wonder: Is it not
possible to make some room in the practice of modern criminal justice
for a retributive element, even if this should require sacrificing some of
its logical coherence? Hobbes’s refutations of moral positions he oppos-
es often emphasize the intolerable evil that is the state of nature, and the
fragility of civil society, which is always threatened by the anti­-social
passions of individuals. Yet considering the overwhelming power of
the state in comparison to the weakness of the individual in civil society,
it would not seem to be excessively dangerous to make some effort to
accommodate commonsense notions about what the criminal deserves.
Is it not worth considering whether we could not find in retributive
punishment the solution to the nagging Hobbesian problem of the ugly
conflict between society’s right to punish and the natural right of the
criminal to self­-defence? Is it not possible that corporal and carceral
punishment, and perhaps even capital punishment, might be justified
as what a criminal might deserve, precisely as a naturally free being and
a bearer of rights? This line of argument would eventually be pursued
by Kant (and later Hegel), who would go on to reject Hobbesian pun-
ishment in order to complete the task Hobbes had set for himself: to ar-
ticulate a moral theory of punishment grounded in modern principles.
 Chapter 3

Liberalizing the Criminal Law:

Montesquieu and Beccaria

Among Montesquieu’s many contributions to liberalism is his teach-

ing on criminal law. Nothing indicates more clearly the importance of
this subject for Montesquieu than his pronouncement in The Spirit of the
Laws that “the citizen’s liberty depends principally on the goodness of
the criminal laws.”1 Before Montesquieu, Locke’s critique of absolute
sovereignty had already made institutional restraint of sovereign pow-
er a theme of liberal political thought, but Locke had not specified the
proper limits of sovereign power in the sphere of criminal justice in par-
ticular. Montesquieu would therefore be the first to take up this project.
The Spirit of the Laws follows and extends the Lockean argument that the
separation of powers is essential to free government, but it teaches also
that even in a system of separated powers of government, the liberty of
individual citizens will not be secure until it is protected from arbitrary
or oppressive criminal laws.2
Montesquieu’s liberal theory of criminal law follows Hobbes in im-
portant ways. Hobbes had taken pains to commend to the sovereign
such moderate practices as transparency and publicity in the criminal
law, and fairness and regularity in punishment, as well as the limita-
tion of penal severity (which was not to exceed the requirements of
deterrence). All of these innovations are echoed and reaffirmed in
Montesquieu. Yet Hobbes’s principle of absolute sovereignty left his
doctrine of criminal justice in a precarious position, since it could not
but undermine his efforts to trace moderating limits on the punishing
power. In Montesquieu, by contrast, we find no evidence of a doctrine of
absolute sovereignty or even of a social contract. Indeed, Montesquieu
denounces in no uncertain terms the sort of “justice” dispensed by des-
potic governments, and for this reason he is celebrated as the father of
76 Punishment and the History of Political Philosophy

classical liberal criminal law, in the sense that we now understand it as

belonging to a larger theory of limited government and constitutionally
protected individual rights.
The teaching on criminal law evolves gradually over the course of
several books of The Spirit of the Laws, culminating in the explicit pro-
posals for reform in Book 12. Montesquieu departs most markedly from
Hobbes by positing that a government whose fundamental purpose is
to provide individual security will rely less than any other form of gov-
ernment on the threat of punishment itself: a political society committed
to individual liberty is the least punitive of all. Concretely, this means
that free governments must learn to depend less on instilling fear and
more on cultivating mild mores (popular sentiments cultivated by
education and civil institutions) as a means to sap those destructive
urges that are the common causes of criminality. While Montesquieu’s
intentions are easily stated in this way, it is not a simple matter to ex-
plain the philosophical insights behind them. An adequate understand-
ing of The Spirit of the Laws’s teaching on criminal law reform requires
some careful reflection about the place of the subject of criminal law in
Montesquieu’s political teaching as a whole.

Fear, Human Nature, and Despotism

According to Hobbes, the overwhelming and irresistible power of the

absolute sovereign is entirely suitable for maintaining order because
man is by nature a rationally self­-seeking being. Concerned above all
with his private good, each individual will be law­-abiding and peace-
able only as long as he believes that general fear of the sovereign’s sword
will keep all in check. The principle and task of criminal justice, accord-
ing to this view, may therefore be straightforwardly reduced to deter-
rence. This cannot suffice for Montesquieu, however, since he departs
from Hobbes’s understanding of human beings as by nature rationally
selfish. While human beings in the state of nature selfishly seek to sat-
isfy basic needs, they are not rational. Moreover, and perhaps more im-
portantly, Montesquieu denies that the fearfulness, timidity, and crude
self­-regard that characterize man in the state of nature continue to be his
dominant motivations after he has entered into civil society.
Montesquieu’s account of human nature introduces the important
premise of our essential malleability. Man is “that flexible being who
adapts himself in society to the thoughts and impressions of others,
 Liberalizing the Criminal Law: Montesquieu and Beccaria 77

[who] is equally capable of knowing his own nature when it is shown

to him, and of losing even the feeling of it when it is concealed from
him” (“Preface” [xliv–xlv]).3 It is this malleability or adaptability that
– more than either our rationality or selfishness – explains the charac-
ter of human beings in society.4 This insight accounts for the shift of
emphasis in Montesquieu’s political thought to the amazing variety in
the particular circumstances of actual societies, which can have a great
deal of influence on passions, beliefs, and institutions. This directly in-
forms Montesquieu’s great maxim that the laws and form of govern-
ment should relate not so much to abstract principles of natural law, or
natural right, as to the particular disposition of the people for whom
they are established (1.3).
From this new theoretical vantage point, Montesquieu is able to agree
with Hobbes’s claim that the fear of an overwhelming power could serve
as the principle of government, while denying that this is the only possi-
ble stable political arrangement. According to Montesquieu’s threefold
typology of governments, fear is the principle that informs despotism,
as distinguished from republics and monarchies (3.9). Fear of violent
death can become the dominant passion in society only after a certain
way of life debases man’s soul and renders it slavish (4.3). Hobbes was
right, then, when he argued that in a political society founded on the
principle of fear, the criminal laws must be simple and strict (cf. 6.2),
but he was wrong to have assumed that every political society must be
founded on fear. It is not necessary here to expound the reasons why
Montesquieu regarded despotism as an inferior form of government
compared to the two “moderate” forms of government – monarchy and
republic.5 For our purposes it is enough to acknowledge that, accord-
ing to Montesquieu, wherever the particular circumstances of a nation
permit liberty and moderate government, this course will be preferred
by wise legislators to the alternative of despotism.6 Therefore, since the
criminal laws of each society must agree with the principle of its form of
government, it would be a grave mistake for moderate governments to
try to simplify their criminal laws, as Hobbes would have them do. For
such laws would at best be in permanent contradiction with the spirit
of moderate government and at worst would tend to undermine such
government, leading it in the direction of despotism.
There is an additional reason, besides its theoretical simplicity, why the
argument advanced by Hobbes for a simplified and rationalized crimi-
nal law is tempting, and that is that it appeals not only to our concern for
78 Punishment and the History of Political Philosophy

security but also in a way to our sense of justice. It is not uncommon for
the victim of a crime to feel that, above all, justice must be swift, and this
is exactly what a rationalized criminal law promises – for to rationalize
law is to simplify it in accordance with a single principle, such as utility
(the greatest happiness of the greatest number) or reciprocity (an eye
for an eye). We are most struck by the inconvenience of the innumer-
able formalities and delays that seem to pervade criminal proceedings,
and we want these to be lessened as much as possible and brought into
conformity with a clear principle of justice. Montesquieu acknowledges
the moral appeal of immediacy and simplicity in criminal justice and
admits that this is more likely to be found in absolute states. “It is con-
stantly said,” he writes, “that justice should be rendered everywhere as
it is in Turkey” (6.2). Yet Montesquieu also stresses the unseen dangers
of simplifying criminal justice. The delays and formalities of criminal
laws – frustrating as they may be from the point of view of our insistent
demand for simple and immediate justice – serve as indispensable safe-
guards of individual liberty, inasmuch as they provide the accused with
a range of legal remedies to challenge any suit lodged against them.7
This is particularly important when a public accusation is serious, when
conviction might place the life and household of the accused in mortal
danger. Thus Montesquieu (like Locke) was more sensitive than Hobbes
to the problem posed by the weakness of the individual in the face of his
powerful sovereign. Even if the sovereign should be so enlightened as to
avoid unnecessarily oppressing his subjects, to them this would still be
much too flimsy a guarantee of their safety. The citizen can feel secure
only if he knows that legal protections exist of which he can avail himself
– those same innumerable avenues of appeal and delay at all stages of
the criminal process that can be so frustrating to victims of injustice. For
these reasons, the drive to rationalize the criminal law by removing what
appear to be inconvenient and unnecessary formalities can be very dan-
gerous to liberty. Thus from a Montesquieuean perspective, retributiv-
ism and utility – two very different theoretical principles – pose the same
problem for moderate government if applied systematically to criminal
law, since either would lead to the law’s oversimplification.
On the other hand, if the legal remedies available to the accused are
to be multiplied, just as Montesquieu recommends, would this not
make it easier for the innocent and the guilty alike to escape conviction,
and would this not thereby encourage crime and vitiate the most basic
function of government, namely, the security of the innocent?
 Liberalizing the Criminal Law: Montesquieu and Beccaria 79

Shame, Liberty, and Moderate Government

Montesquieu’s emphasis on human malleability enabled him to elabo-

rate an alternative way in which moderate governments might maintain
peace without always having to resort to the threat of force. In moderate
governments a decent sense of shame becomes a much more effective
deterrent against criminality than the fear of death, whereas in despotic
states men are so oppressed and miserable that one has little hope of
influencing them except by threatening them with very severe punish-
ments. Montesquieu argues that individuals living in moderate states
not only enjoy the security of their lives and possessions but also feel
the esteem of their fellows and develop a passionate love of their home-
land. Consequently, they fear losing these social goods even more than
they fear pain and death. Since a bad reputation is enough to cause one
to lose one’s enjoyment of social goods, in moderate states where “one
fears the loss of life more than one dreads death as such … punishments
that merely suppress life are sufficient.” In such societies, the wise legis-
lator “will insist less on punishing crimes than on preventing them; he
will apply himself more to giving mores than to inflicting punishments”
(6.9).8 Insofar as a criminal conviction in itself carries with it the infamy
of the crime, this is sufficient to deter many crimes even if no punish-
ment should follow. Accordingly, in moderate states “the greatest pen-
alty for a bad action will be to be convicted of it” (ibid.). Thus, even
though in moderate states the guilty may successfully appeal, delay, or
mitigate their deserved punishment, the threat of conviction, with its
accompanying opprobrium, could well be expected to prevent crimes
just as effectively as the threat of severe penalties does in despotic states.
Perhaps even the prospect of being accused of a crime would be enough
in many cases to check wrongful behaviour. In this way, Montesquieu
indicates how the important political problem he has in view might be
solved – that is, how it may be possible to maintain conditions in which
individual citizens could feel truly secure from arbitrary legal prosecu-
tion, without sacrificing public security to the extensive legal protec-
tions guaranteed to the accused.
In addition to this argument for decreasing reliance on punishment,
as such, as a means of crime prevention, Montesquieu goes on to argue
for a decrease of the severity of punishment. According to him, increas-
ing severity in punishment is not just contrary to the spirit of moderate
government – it is self­-defeating. Once again, Montesquieu’s view of the
80 Punishment and the History of Political Philosophy

natural adaptability of human beings provides the key to his analysis.

“Experience has shown that, in countries where penalties are gentle, the
citizen’s spirit is struck by them as it is elsewhere by heavy ones” (6.12).
Where some law is frequently violated, the simple legislative response
is to impose a stiffer penalty, and this does the job for a short time. Yet
over the long term “the imagination becomes inured to this heavier
penalty as it had to the lesser, and as fear of the lesser penalty dimin-
ishes, one is forced to establish the heavier in every case” (ibid.). All
that the escalation in penal severity accomplishes is to brutalize citizens
by rendering them insensitive to extreme suffering, thus making them
harsher towards one another and less (not more) fearful of punishment.
Eventually, even capital punishment becomes only a marginally effec-
tive deterrent of crime, although once society has reached such a pass,
capital punishments become the only possible method of maintaining
order. Thus, Montesquieu calls the death penalty “the remedy … for a
sick society” (12.4). Such considerations lead Montesquieu to conclude
that punitive criminal laws corrupt moderate governments: contrary to
Hobbes, moderate government and security can and do go together.
But if the escalation of penal severity can corrupt citizens and render
them slavish, then perhaps the moderation of punishments through en-
lightened penal reform can have the opposite effect by instilling gentle
mores and encouraging mild sentiments, while improving security and
strengthening the institutions of moderate government (6.13).
We must now complicate the simple account of Montesquieu’s teach-
ing presented so far. Montesquieu illustrates his thesis about the nature
of criminal law in moderate governments with his customary use of
detailed historical examples. Yet a careful reading of the chapters on
criminal law makes one doubt that Montesquieu’s historical examples
fully bear out his conclusions. Most dubious is Montesquieu’s claims
about the mildness of the criminal and penal laws of certain exemplary
republics and monarchies of which he speaks. The surface argument
about the connection between moderate forms of government and
mild criminal laws carries some plausibility: in despotic states subjects
are mere possessions of the despot, but in moderate states – that is,
monarchies and republics – citizens are endowed with substantial legal
rights and are relatively free in their intercourse with one another; as
a result, in moderate governments individuals could be expected to be
treated with greater respect and consideration, especially when their
guilt or innocence is in question in a criminal trial. Yet the historical
proof Montesquieu marshals in support of this argument as it applies
 Liberalizing the Criminal Law: Montesquieu and Beccaria 81

to actual republics and monarchies is ambiguous at best. Let us take the

case of republics first.
Montesquieu gives evidence of some mild penal practices in early
Rome, Sparta, and Athens (all of them republics and therefore mod-
erate governments according to Montesquieu’s schema). Yet that evi-
dence is elsewhere (and sometimes in the same context) contradicted by
descriptions of practices carried out by those same cities that can only
be described as harsh and punitive.9 Montesquieu’s extensive discus-
sion of the penal laws established by the Roman republic may serve as
our case in point. He concedes that the laws of the Twelve Tables, en-
acted sixty years after the birth of the republic, were “full of very cruel
provisions” – including capital punishment for many crimes as well as
“punishment by fire.” He attempts to excuse this appearance of repub-
lican cruelty by arguing that those severe penal laws were authored by
a few individuals – the Decemvirs – empowered with extraordinary au-
thority, whose rule would later become despotic (6.15). Yet Livy, who is
Montesquieu’s source on this subject, relates the same events somewhat
differently. According to Livy’s account, the plebs were allowed to re-
view and amend the legislation of the Decemvirs before it was enacted,
which seems to suggest that the plebs were largely satisfied with the
legislation as proposed by the Decemvirs. Moreover, the criminal laws
Montesquieu identifies as cruel remained in force in Rome, according to
Livy, even after the tyrannical Decemvirs were deposed.10 Montesquieu
tries to argue further that the gentleness of the republican spirit still
shines through in the fact that those remaining harsh provisions of the
penal laws in the Twelve Tables were eventually modified or abrogated
by the Roman people. Yet he himself makes clear that those changes
did not come until very much later – approximately 250 years after
the founding of the republic.11 All of this makes the case in favour of
Montesquieu’s thesis seem tenuous at best, at least when it comes to the
Roman republic.
Even more telling, Montesquieu’s thesis – that republican penal jus-
tice is moderate – is contradicted by his own theory of the nature of re-
publican government. He repeatedly stresses the unnatural harshness of
the republican way of life. That way of life demands so much repression
of one’s most natural drives in the name of the public good that it ren-
ders human beings prone to excessive anger and cruelty (3.3–4, 4.5–6).
Thus despite leaving the comforting surface impression that republics
are gentle and humane (perhaps in order to avoid disappointing readers
with a deep admiration for classical politics), Montesquieu quietly but
82 Punishment and the History of Political Philosophy

clearly indicates that this form of government will nearly always tend
to cultivate harshness in citizens and in the criminal laws themselves.
In a similar way, Montesquieu deliberately overstates the gentleness
of criminal laws in monarchies, the other form of moderate government.
The extent of this overstatement would have been quite palpable to
thoughtful readers in Montesquieu’s time. He claims that in moderate
states “the head of even the lowest citizen is esteemed” (6.2), and while
this may be true relatively speaking (when compared with despotism),
he does not deny that the lowest­-ranking individuals in a monarchy are
not accorded the same esteem as the highest­-ranking. What is more, the
highest orders in monarchies are believed to possess honour – that is, to
be sensitive to it and to value it above even life itself – whereas no such
sensibility is attributed to the common people. This implies that, from
the point of view of the higher classes in a monarchy, shame cannot be
used as an effective measure to deter criminality in the common people,
as it can in the nobility (6.10). Since, according to Montesquieu, honour
is prized as the highest good in monarchies, the lack of it among the
common people draws upon them the contempt of the nobility and the
court, along with all the accompanying cruelty in the administration of
justice that may be expected. Contrary to Montesquieu’s general claim
about criminal law in monarchies, therefore, his arguments actually im-
ply that it would be perfectly natural for monarchical laws to impose
corporal punishments on the common people. Indeed, Montesquieu
himself later tacitly concedes that punishments imposed in monarchies
on the highest orders tend to be significantly milder than those im-
posed on the lowest.12 Over the course of his discussion, Montesquieu
observes that under the European monarchies, forensic and penal prac-
tices included torture (or “the question”); the wheel, and other excru-
ciatingly painful and disfiguring punishments; the death penalty for a
wide range of crimes; and corporal punishment of the lower orders.13
It is important in this context to add another dimension to Montesquieu’s
ambiguous treatment of the supposedly gentle criminal laws of republics
and monarchies. The European monarchies that Montesquieu had before
his eyes were, of course, Christian ones. Elsewhere Montesquieu praises
Christianity’s moderating effects on mores,14 yet he says very little about
Christianity’s influence on criminal laws. Is it possible that Montesquieu
maintains a respectful silence about Christianity in this connection be-
cause it had not had a moderating effect on criminal law in Europe – and
perhaps even the opposite effect? The only mention of Christianity in all of
Book 6 (the book treating criminal law thematically) is found in a chapter
 Liberalizing the Criminal Law: Montesquieu and Beccaria 83

on the powerlessness of Japanese laws. There, Montesquieu attributes

that powerlessness to the harshness of the Japanese manner of punish-
ing offenders. Montesquieu’s explanation for why that nation punishes so
harshly has recourse to certain opinions peculiar to the Japanese:

In Japan almost all crimes are punished by death because disobedience to

such a great emperor as Japan’s is an enormous crime. It is not a question
of correcting the guilty man but of avenging the prince. These ideas are
drawn from servitude and derive chiefly from the fact that the emperor is
the owner of all the goods and so almost all crimes are committed directly
against his interests. (6.13)

It is striking how much the idea that the Japanese have about their em-
peror resembles the idea that the Bible has of its God (cf. 12.4).15 If the
parallel is granted, then could the biblical idea have had a similar effect
on criminal law in Christian Europe, rendering it harsher and more pu-
nitive (and therefore less effective – see 6.13) than it might otherwise
have been?
Therefore, notwithstanding the unreserved praise Montesquieu some­
times flatteringly bestows on criminal law in the so­-called “moderate” gov-
ernments, his fully developed view is much more complex. Paradoxically,
the criminal laws of a people will not be moderate simply by virtue of
the fact that their government happens to be moderate – that is, mo-
narchic, republican, or Christian. It seems that some additional condi-
tions must exist to allow for moderation in criminal law. What are these
conditions? Notwithstanding Montesquieu’s allusions to some less than
gentle aspects of the criminal justice practised in the European monar-
chies, it remains true for him that the European states are fertile soil for
moderate politics because their laws, institutions, and climate are hospi-
table to liberty (understood as the security of the citizen and the consti-
tutional separation of powers). The arguments for this thesis – so central
to the Spirit of the Laws as a whole – are complicated and are developed
at length in other parts of the work. One highlight is Montesquieu’s
presentation, in Books 11 and 19, of his model of liberal society – an
idealized England – a nation that, unlike other moderate nations and
governments, has liberty as its chief aim. According to that well­-known
account, England’s fortuitous mixture of commercialism, restlessness of
spirit, and a certain kind of equality (19.27), when combined with a fa-
vourable political system (11.6; Montesquieu was, of course, speaking of
England after the Glorious Revolution of 1688), created conditions for an
84 Punishment and the History of Political Philosophy

unprecedented pursuit of liberty. Such liberal politics is uniquely hospi-

table to moderate criminal law reform, because moderation in criminal
justice means nothing other than putting the emphasis on protecting
the liberty (i.e., security, or the “opinion of … security”) of the citizen
– whether he be a defendant or plaintiff, whether an offender or a law­-
abiding member of society.
The most distinguishing quality of human beings is their ability to
adapt to their physical and social circumstances, which includes the
capacity to conform to the opinions of the society into which we they
are born, taking up its predominant beliefs and practices and regulating
their lives in accordance with them. It stands to reason that this quality
makes us sensitive to shame, since we are loath to incur a bad reputation
with our fellows. Thus, by the same token, we normally avoid wrongdo-
ing not only because we fear a penalty, but also out of a decent sense of
shame (12.4 [pp. 190–1]). This decent aversion to injustice may be un-
dermined wherever tyrannical rulers terrorize and brutalize their sub-
jects, but since the chief aim of liberal societies (such as Montesquieu’s
England) is the protection of life and private property, individuals there
are left free to engage in all sorts of voluntary commerce (broadly un-
derstood), so that they come to value their membership in their chosen
associations. With such positive incentives to social membership, liberal
societies are uniquely well suited – better than any other form of moder-
ate government, to say nothing of despotism – to practise non­-coercive
and indirect methods of law enforcement.16 Hobbes was correct when
he defined the aim of punishment as deterrence, but his view of human
nature prevented him from appreciating that effective deterrence in lib-
eral societies should be sought in the celerity, regularity, and ignominy
of punishments, not in their severity. One could even go so far as to say
that, up to a point, in liberal states the severity of punishments is in-
versely proportional to their effectiveness.
Although it is true that not all eighteenth­-century European states
were as committed to liberty as Montesquieu’s England (and perhaps
not even England itself),17 they possessed most of the elements dis-
cussed by Montesquieu as being crucial for liberalism, and therefore
they had at least the potential of successfully carrying out liberalizing
reforms, including the important reform of criminal law. This set of his-
torical circumstances provided Montesquieu with the opportunity to
outline the principles that such reform might follow if it was to achieve
its full potential.
 Liberalizing the Criminal Law: Montesquieu and Beccaria 85

Montesquieu on the Idea of Criminal Justice

There remains, however, a further difficulty with the picture we have

painted so far. Montesquieu teaches that laws in general flow from, and
are informed by, the structure and nature of the established government.
This would lead one to expect English criminal law in particular to be
milder and more humane than the more traditional criminal codes of
other European nations, in proportion to the degree to which the English
constitution is more liberal than the other national constitutions. Yet
Montesquieu mentions English forensic practices only to criticize them
(such as the English Parliament’s practice of issuing bills of attainder).18
In fact, the English criminal law of this period (the “Bloody Code,” as it
is referred to now) embodied many punitive ideas and enacted many
harsh practices – a fact that could scarcely have escaped Montesquieu’s
notice. According to Blackstone, writing in 1769, English law recognized
no fewer than 160 capital crimes.19 Contrary to what Montesquieu has
prepared us to expect, eighteenth­-century England was very far from
achieving enlightened criminal law reform; indeed, it was no better off
in this respect than the other European monarchies. But how could this
be if the structure of the English constitution was exceedingly hospitable
to liberty?
Montesquieu could not explain the severe character of England’s
criminal code by reference to its constitution, nor does he ever suggest
that this penal severity was merely the product of inept legislation or
that it was an artefact of the imperfect state of the technical sciences of
criminology and penal administration. Rather, he indicates that these
excesses were the result of incorrect ideas about punishment – in par-
ticular, traditional notions about the scope of wrongdoing to which
punishment properly applies.20
Montesquieu apparently follows the classical and Christian traditions
in affirming that punishment is by its nature retributive or desert­-based:
“punishments” rightly understood are “a kind of retaliation, which
causes the society to refuse to give security to a citizen who has deprived
or has wanted to deprive another of it” (12.4 [p. 191]). Montesquieu ap-
pears to draw this view of punishment from his understanding of natu-
ral law or justice broadly conceived, as presented earlier in The Spirit of
the Laws. According to that earlier presentation, natural laws are certain
necessary relations in the nature of things, which, as such, do not owe
their being to human convention. In particular, natural justice dictates
86 Punishment and the History of Political Philosophy

that “if there were intelligent beings that had received some kindness
from another being, they ought to be grateful for it,” and, conversely,
“that one intelligent being who has done harm to another intelligent
being deserves the same harm in return” (1.1). We are surprised to find
Montesquieu expressing this endorsement of the traditional concept of
reciprocity and desert in punishment, especially considering how much
he follows Hobbes in other respects.
It seems that Montesquieu finds fault not with the traditional idea
of retribution as such but rather with the scope of its application.
Traditionally, punishments were held to apply to all sorts of wrongdo-
ing, which Montesquieu groups into four categories of crime: crimes
against mores; those against religion; those against public tranquility;
and, finally, those against security. According to Montesquieu, however,
punishments (supplices) – as distinguished from penalties in the broader
sense (peines) – ought properly to apply only to this last class. Thus, it
would be in accordance with natural justice for society to express its
anger and indignation against those citizens whose violence threatens
public security and to punish them with a reciprocal harm, as they de-
serve. But this would not be appropriate in the case of other sorts of
crimes. Montesquieu thus implies that criminal justice in traditional so-
cieties – including biblical Israel and the ancient republics, to say noth-
ing of despotic regimes – violated natural justice.21
Montesquieu supports this view of a more limited scope for retribu-
tive punishment for the following reason. Retribution or vengeance
may apply to crimes of the other three classes, but it is not the place of
the civil authority to punish them. Since the ultimate purpose of civil
society is or ought to be security, only crimes against security are crimes
against civil society, which may and ought to avenge itself by natural
right. A crime against religion, on the other hand, is a crime against God
and should be left for God himself to avenge.22 It follows more gener-
ally that each penalty should be drawn “from the particular nature of
the crime” – that is, from the sphere of activity in which the crime was
committed (12.4 [p. 189]). Montesquieu seems to regard collective reli-
gious practice, and the collective practice of moral virtue, as the sphere
of what Locke had called “voluntary associations.” Such associations
(as Locke had understood them)23 are often useful to civil society as
sources of moral and civic education, and as such, they can safely be
tolerated and even encouraged by it. But insofar as these associations
have eternal salvation as their goal – a goal that is beyond the scope
of political life – they lie outside of the civil authority’s jurisdiction in
 Liberalizing the Criminal Law: Montesquieu and Beccaria 87

all matters that do not affect security. This means both that the state
should not attempt to regulate the spiritual beliefs of religious or moral
associations on the one hand, and on the other that it is absolved of any
responsibility for enforcing the doctrine of any particular faith.
Montesquieu does not deny that civil society may have a legiti-
mate interest in maintaining and regulating religion, mores, and pub-
lic tranquility, since all of these can have an effect on security. But in
this regard, legislation need only correct; it need not punish. The chief
mode that correction of breaches of religion, mores, and public decency
should take in a liberal polity is loss of reputation and social standing
– as Montesquieu had previously suggested and as he now repeats in
Book 12. Sacrilege may be sufficiently corrected (i.e., either deterred or
negated in its consequences) by expulsion from the congregation of the
faithful. Breaches of “public or individual continence” – that is, offences
against the virtue of moderation – should be curbed by public infamy,
but fines might also be imposed (perhaps because someone willing
to offend against common decency might not be deterred by shame).
Finally, breaches of public tranquility or order may be corrected by exile
or by “penalties that restore men’s troubled spirits and return them to
the established order” (12.4). It would appear that many of these cor-
rective measures could be carried out by society itself (as distinguished
from its public officials), especially in cases of offences against religion
or mores. When an individual insults what is held sacred by his fellows
or offends their accepted standards of decorum, no one is in a better
position than that community itself to censure or, if necessary, cast out
the offender – all without having to resort to coercive measures, which
are usually reserved for the civil authority.
Montesquieu’s conception of liberal criminal justice thus appears to
combine political goods that had been irreconcilable in traditional soci-
eties. It seems that the happy circumstances of liberal societies allow for
mildness in the penal laws; for effective deterrence without a dangerous
centralization of power; and for retribution to be adequate to the prin-
ciples of natural justice without having a brutalizing effect on society.24
But liberal criminal justice, as outlined by Montesquieu, faces two
problems. First, as we have seen, Montesquieu fundamentally follows
Hobbes in making individual security the primary goal of politics;
but Hobbes had shown that the fundamental natural right to self­
preservation, if consistently followed in its political implications, is in-
compatible with the traditional concepts of moral responsibility, desert,
and punishment. Montesquieu affirms that retribution is in accord with
88 Punishment and the History of Political Philosophy

natural justice, but nowhere does he prove this. On the contrary, he him-
self had argued that retributive punishment is at home in despotic re-
gimes, which are an inferior form of government.25 Perhaps something
led Montesquieu to believe that the concern for reciprocity is connected
to a human phenomenon that transcends even the differences among
the three types of government – namely, the family. After all, do we not
initially learn reciprocity in the context of the family – namely, in the debt
of gratitude we believe we owe our parents?26 Contrary to both Hobbes
and Rousseau, Montesquieu regarded the family as in some sense natu-
ral and thus believed that the part of morality originating in the family
must be accommodated by wise legislation.27 Furthermore, according to
Montesquieu the decency of a society depends on the maintenance of
its mores – even in liberal polities, where mores may be somewhat less
pure.28 For all of these reasons it may be politically desirable for the laws
to express in an attenuated way – to echo, as it were – certain core ele-
ments of purer mores, including righteous indignation, when this can
be done without compromising the liberty of the citizen. For example,
since the severe punishment of violent crimes is in accord with liberal-
ism’s emphasis on security, the criminal law would not be in contradic-
tion with the spirit of liberty if it were to impose the death penalty on
very serious violent crimes such as murder, for that punishment would
simultaneously be in conformity with the spirit of retribution. Having
said all this, however, we must add that such political considerations do
not amount to a proof of the justice of retribution.
The second problem confronting Montesquieu’s liberal conception
of criminal justice runs in the opposite direction. If we grant that the
concern to return harm for harm must be accommodated politically, we
might go on to wonder whether the scheme of criminal law sketched by
Montesquieu would be sufficient for this purpose. We might wonder
whether the restriction of punishment to crimes against security, and
whether the relatively mild modes of correction proposed for offences
against mores and religion, would fully satisfy justice as reciprocity,
especially in the case of what are considered to be the most egregious
wrongs against “mores.” Would it be enough merely to enact deterrent
penalties against horrendous crimes such as rape, sadism, and pedo-
philia, and then to allow society informally to censure rapists, sadists,
and pedophiles? Or would such penalties fall short of what retributive
justice, as ordinarily understood, would demand?
These difficulties arise in the context of Montesquieu’s discussion
in Book 12. He names rape and kidnapping as two crimes that attack
 Liberalizing the Criminal Law: Montesquieu and Beccaria 89

security and mores at the same time; he then argues that punishments, in
the strict sense, rightly apply to such crimes only as offences against secu-
rity (12.4 [p. 191]). He does not mention exactly what punishment he be-
lieves rapists deserve, although we know well enough that the criminal
laws of most actual governments impose very serious penalties against
this crime with the intention of expressing society’s moral outrage. In
fact, it could be argued that the seriousness of this crime has more to
do with its violation of morality or mores than it does with the physical
harm it causes.29 That is because rape seems to be a serious crime in our
eyes not only because it damages the victim’s body (damage which may
in some cases be minimal) but also because it damages her dignity.
We might question in yet another way Montesquieu’s grounds for nar-
rowing the range of punishable crimes to crimes against security alone.
While some basis may be found for this in the Christian distinction be-
tween the spiritual and the temporal, it is not so clear that Christian
political thinkers always drew the same conclusions for criminal justice
as Montesquieu has done. According to both Calvin and Augustine, the
ultimate authority to punish is God’s alone, yet God can and does del-
egate this authority to his human representatives and especially to the
civil authority.30 According to these authors it follows that, since God
punishes all injustices (and not merely crimes against public security),
the civil authority may well take it upon itself to punish moral wrongs,
especially those that offend public decency, modesty, and the family. It
seems unlikely that Montesquieu – who was well aware of the enduring
human phenomenon of righteous indignation – would have favoured
the erosion of the legal expression of society’s anger towards morally
serious crimes like rape: such a thing would fly in the face of any decent
society’s most deeply cherished beliefs. Rather, he seems to have antici-
pated a gradual decline in the absolute (but not the relative) severity of
penalties imposed on morally egregious crimes, as the severity of pun-
ishments prescribed by liberal penal codes gradually declined overall.31
As we have seen, Montesquieu’s endorsement of retributive pun-
ishment was qualified by his commitment to liberty and by his appre-
ciation of the affinity between retribution and despotic government.
Consequently, he seems to have attempted to continue Hobbes’s rhe-
torical project of transforming the popular idea of punishment by con-
triving to limit the scope or application of the principle of retribution as
much as practicable. Whereas in the classical and biblical traditions the
principle of retribution had applied to all wrongdoing, in Montesquieu’s
teaching righteous anger and retribution are (to repeat) reserved for
90 Punishment and the History of Political Philosophy

crimes against security alone. Other moral wrongs were to be attributed

merely to human frailty and, as such, pitied. By elaborating a set of pro-
grammatic principles for criminal reform (while blurring most of the dif-
ficult and radical moral–theoretical questions they raised), Montesquieu
seems to have intended to contribute to the vast cultural project of po-
litical Enlightenment that was already well under way. For this large­-
scale project, the transformation of popular ideas about criminal justice
would be crucial because its effect would be to decrease, over time, soci-
ety’s serious concern for the reciprocal return of harm – a major obstacle
to the softening of mores that Montesquieu viewed as necessary for the
spread of commerce, liberal institutions, and scientific enlightenment.

Montesquieu’s Legacy and Beccaria’s Treatise

Montesquieu’s new ideas about criminal law reform in liberal societies

were astonishingly successful in Europe and North America, although
much of that success had to wait for his ideas to be popularized by the
Italian philosophe Cesare Beccaria, in the latter’s short treatise On Crimes
and Punishments (1764). This was perhaps to be expected, for the daunt-
ing size and complex structure of Montesquieu’s work must have made it
less accessible to the wider reading public. Beccaria himself did not hide
the debt his work owed to the “immortal President de Montesquieu.”
Not long after Beccaria’s treatise was published, enthusiasm for pe-
nal reform spread throughout Europe and the United States. Instantly,
criminal law became a favourite subject of political pamphlets as well
as legal and philosophical treatises. Beccaria’s admirers included other
theoreticians, such as Voltaire, D’Alembert, Blackstone, and Bentham, as
well as influential sovereigns and statesmen like Catherine II of Russia,
Maria Theresa of Austria, and Thomas Jefferson.32
Beccaria’s work is perhaps best known for its opposition to capital
punishment, but in this he followed in the footsteps of Montesquieu,
who had argued that the death penalty was the remedy resorted to by
a “sick society.” Actually, Beccaria’s most significant theoretical con-
tribution was his attempt to ground the right to punish in the social
contract of liberal theory.33 In this he departed from both Hobbes and
Montesquieu, who found the origin of the right to punish in nature
rather than in human convention. Beccaria’s argument is as follows:

It was necessity … that constrained men to give up part of their personal

liberty; hence, it is certain that each man wanted to put only the least
 Liberalizing the Criminal Law: Montesquieu and Beccaria 91

possible portion into the public deposit, only as much as necessary to in-
duce others to defend it. The aggregate of these smallest possible portions
of individual liberty constitutes the right to punish; everything beyond
that is an abuse and not justice, a fact but scarcely a right. (ch. 2)

On this liberal­-contractarian basis, Beccaria proceeds to argue against

capital punishment:

By what right can men presume to slaughter their fellows? Certainly not
that right which is the foundation of sovereignty and the laws. For these
are nothing but the sum of the smallest portions of each man’s own free-
dom; they represent the general will which is the aggregate of the indi-
vidual wills. Who has ever willingly given up to others the authority to
kill him? How on earth can the minimum sacrifice of each individual’s
freedom involve handing over the greatest of all goods, life itself? … Thus,
the death penalty is not a matter of right, as I have just shown, but is an act
of war on the part of society against the citizen that comes about when it
is deemed necessary or useful to destroy his existence. (ch. 28)

Beccaria’s argument in this passage invokes the Hobbesian authoriza-

tion thesis in order to reject it. This is perfectly reasonable, for we have
already seen that Hobbes had failed, despite his best efforts, to square
the circle, as it were, of showing how the criminal might retain his right
of self­-preservation while at the same time authorizing his sovereign
(by way of the social contract) to put him to death.
Beccaria’s refutation of the state’s right to impose capital punishment
is misleading, however, because the absence of a right to punish does
not, in Beccaria’s own view, constitute an absolute moral prohibition
against executions. On the contrary, the death penalty would continue
to be appropriate in two cases, according to Beccaria. First, if a crimi-
nal threatens the very existence of civil society itself – because “he still
has sufficient connections and such power that he can threaten the se-
curity of the nation even though he be deprived of his liberty” – then
“his death is required” (ch. 28). Second, even when there is no imminent
catastrophic threat to the existence of the civil order, the execution of a
criminal would still be necessary “if his death were the one and only de-
terrent to dissuade others from committing crimes.” Beccaria goes so far
as to call this last use of capital punishment “just and necessary.” In order
to soften the jarring effect of this apparently self­-contradictory conces-
sion, Beccaria suggests that in “the experience of all ages … the ultimate
92 Punishment and the History of Political Philosophy

punishment has never deterred men who were determined to harm so-
ciety.” In other words, although it is theoretically possible that the death
penalty might be justifiable in some circumstances, in practice, according
to Beccaria, it is such an ineffective deterrent that it may safely be retired
as an obsolete form of punishment. But is this claim about the useless-
ness of the death penalty in fact true? It may be that Beccaria does not
really mean that the death penalty has no ability to deter, but rather that
it is not the only effective means of deterrence. While the death penalty is
a better deterrent than no punishment at all, it is, according to Beccaria,
a weaker deterrent than milder but more enduring punishments; it is in
this sense that the death penalty can be said to be useless. The claim that
mild but long­-lasting punishments are the most effective depends on a
psychology that emphasizes the strength of habit and the weakness of
reason’s control over the passions. Thus, Beccaria argues that “it is not
the severity of punishment that has the greatest impact on the human
mind, but rather its duration, for our sensibility is more easily and surely
stimulated by tiny repeated impressions than by a strong but temporary
movement. The rule of habit is universal over every sentient being, and,
as man talks and walks and tends to his needs with the aid of habit, so
moral ideas are fixed in his mind only by lasting and repeated blows”
(ch. 27). By this view, hard labour and imprisonment would be better
punishments than the death penalty, since their effects are more continu-
ous. Beccaria further argues (drawing on Montesquieu) that whatever
effectiveness the death penalty may possess when it is first instituted
eventually decays over time; indeed, it may even indirectly contribute to
crime as people become accustomed to the sight of violence and suffer-
ing inflicted by human beings. “To the degree that punishments become
more cruel, men’s souls become hardened, just as fluids always seek the
level of surrounding objects” (ch. 27).34
Beccaria’s argument that the death penalty is ineffective as a deter-
rent in most – but not all – practical circumstances still leaves those rare
cases to be accounted for, in which harsh punishment presents itself as
the only expedient. Punishment in such cases must also be accounted
for as right or just, if we are to have a fully satisfying moral theory of
punishment. This, of course, is precisely the problem that Hobbes had
faced, a problem that he attempted to solve by recourse to contractarian
authorization – which, as we have seen, Beccaria rejects. Although in the
final analysis, the social contract is incapable of authorizing capital pun-
ishment, this failure is not fatal to Hobbes’s theory of punishment, as we
saw, because Hobbes located the origin of the right to punish in the state
 Liberalizing the Criminal Law: Montesquieu and Beccaria 93

of nature (i.e., in the natural right of self­-preservation). Beccaria, on the

other hand, makes no room for such a right, and this puts his theory of
punishment in jeopardy.
Perhaps for this reason Beccaria elevates general utility to the sta-
tus of a moral principle. To be precise, while Beccaria argues that the
social contract is the origin of the right to punish, he makes general
utility – the “greatest happiness shared among the greatest number” –
the purpose and measure of punishments, and indeed of all laws and
public policies generally (see “Introduction”). We learn from Beccaria,
however, that general utility is “just” in a sense that is not distinguish-
able from the force and advantage of the majority. “Note that the word
‘right’ is not a contradiction of the word ‘force’; the former is, rather, a
modification of the latter – namely, the modification most useful to the
greatest number … By ‘justice,’ moreover, I do not mean anything but
the bond necessary to hold private interests together” (ch. 2). But what
makes the advantage of the majority just or right from the point of view
of the criminal? Certainly not any duty to sacrifice his own interest to
that of civil society, since Beccaria explicitly denies the existence of any
such duty (see ibid.).35 At least in the Hobbesian picture the criminal
would have to acknowledge that the sovereign was proceeding against
him with the same right of nature as he himself claimed. Thus, gen-
eral utility does not prove to be any more successful than contractarian
authorization in avoiding the morally problematic implications of the
modern punishing power.36
Perhaps Beccaria believed that the unresolved theoretical difficulties
in his treatise would not prevent its practical proposals from gaining
a popular audience; perhaps, that is, he anticipated that his resound-
ing condemnation of severity in punishment in the name of humanity
would appeal to popular sensibilities that had been softened by the in-
fluence of liberalism. Thus, in the context of his claim that capital pun-
ishment is a poor deterrent, he writes that “if [the argument] does not
persuade men, who always suspect the voice of reason and heed the
voice of authority, then one needs only to consult human nature in or-
der to feel the truth of my assertion.” Similarly, his treatise occasionally
appeals to a certain kind of benevolence stemming from “kindly virtues
born of an enlightened reason” (ch. 3). He seems to be suggesting that
even when we have some reason to doubt that mild penalties would be
as effective as severe ones, we might still listen to our hearts and opt
for humane (although somewhat less effective) measures, regardless of
what reason may tell us. The problem of the unaccounted­-for case of the
94 Punishment and the History of Political Philosophy

sovereign treating the criminal as a mortal enemy in the state of nature

may be forgotten by those of us who have confidence in the goodness
of our hearts.37
The generation of penal reformers that followed Beccaria in Europe
and the United States agreed almost unanimously on theoretical prin-
ciples.38 What remained was to start putting these principles into prac-
tice by drawing up large­-scale programs for modernizing criminal
justice institutions in light of the discoveries of modern medicine and
the burgeoning fields of psychology and criminology. Above all, the
reformers began expressing enthusiasm for the newest technical inno-
vation: the penitentiary.39 Credit for this innovation is to be given espe-
cially to the American penal reformers at the time of the founding and
in the early nineteenth century. The penitentiary movement began in
Philadelphia, where it was supported in equal measure by the Quakers,
who abhorred all forms of violence, and by the Enlightenment rational-
ists. Montesquieu and Beccaria had shown the world how punishments
could become milder without losing their effectiveness; the American
reformers then followed this principle to its end: incarceration is the
ideal punishment, since it is the least physically painful and degrading
(or so it was thought) while still being sufficiently threatening to serve
as a deterrent. What is more, varying the length of prison terms would
modulate their severity, and this would allow incarceration to replace
virtually all other forms of punishment.
If these advantages were not enough to recommend it, penitentiary
punishment offered a further benefit: it promised to rehabilitate prison-
ers, who could engage in useful trades and educational programs while
serving their time. This would prepare them to lead more productive,
independent lives upon their reintroduction to society. Some of the more
pious reformers even believed that the penitentiary would provide an
opportunity for inmates to meditate on their past actions, to repent of
their crimes, and perhaps even to embrace religious salvation. (Indeed,
this supposed aspect of modern carceral punishment is what gave pris-
ons their popular name: “penitentiaries.”) When he visited the United
States in 1831, the French statesman and political philosopher Alexis de
Tocqueville’s official business was to investigate the American peniten-
tiary system on behalf of the French government. In the course of their
travels, he and his friend Beaumont expressed reservations about the
prospects for genuine moral and religious reform of prisoners in the
new penal institutions; indeed, Tocqueville described Americans’ preoc-
cupation with penal reform as a kind of “monomania.”40 Nevertheless,
 Liberalizing the Criminal Law: Montesquieu and Beccaria 95

both men spoke highly of the institutions themselves as forming the

core of a humane and effective new system of punishment.
The influence of Montesquieu’s criminological and penological ideas
on the religious understanding of these matters can be seen in the writ-
ings of some of the more pious penal reformers, such as Benjamin Rush,
who established the Philadelphia Prison Society during the founding
era. Rush published two influential essays on punishment in which
he championed the new, scientific approach to crime advocated by the
European philosophers, while seeking to justify that approach in the
light of scripture. According to Rush, the Christian obligations of for-
giveness and universal benevolence – which included the promotion
of repentance for the salvation of men’s souls – were binding on indi-
vidual Christians as well as on state governments. The penitentiary, he
argued, served the ends of the Christian faith beautifully.41 By contrast,
there was no religious duty to impose the death penalty; in fact, a duty
not to impose it was incumbent on all Christians and their governments.
Following a Montesquieuan strategy of apologetics, Rush argued that
only God can take life, since he alone gives it (15).
Rush’s scriptural arguments can be characterized (and not unfairly)
as somewhat tendentious, and they are worth restating here as evi-
dence of the extent to which the Montesquieuan–Beccarian influence
on penal thought went unquestioned. Rush admits that some might
object to his appeal to the authority of scripture, observing that there
are passages in the Bible in which God directly commands the Israelites
to punish murder with death. In response to this objection, he provides
three arguments (21). First, he claims that nothing in scripture is against
reason, although some things in it are above reason. Since capital pun-
ishment is against reason, it cannot possibly be the will of God. Second,
the “order and happiness of society” is agreeable to the will of God,
and since capital punishment destroys the order and happiness of so-
ciety, it must again be contrary to the will of God. One might wonder,
though, whether Rush is not avoiding the real question here, since no-
where in the Gospels or in the Hebrew Bible does it say that the will of
God cannot be contrary to reason, or that the “order and happiness of
society” must be understood in terms of rational utility, as Rush seems
to understand them. Rush cites two well­-known Biblical passages on
punishment: the first from Leviticus, in which God utters that “he
that killeth a man shall surely be put to death”; and the second from
Genesis, addressed to Noah, in which He says that “whoso sheddeth
man’s blood, by man shall his blood be shed.”42 Rush claims that in
96 Punishment and the History of Political Philosophy

both these passages the divine utterance is not a command or precept,

but rather a statement of fact about the habitual practice of human be-
ings – namely, that societies tend to punish murder with death, whether
or not this is in accordance with God’s will (14–15, 21–2). Yet one might
wonder at Rush’s confidence in his interpretation, for the context of
each of these utterances seems to indicate that they are commands from
God to his subjects (the Israelites and Noah, respectively) to execute
a particular guilty party. In both cases God appears to be delegating
the punishment of the guilty to a human agent. One wonders whether
Rush is not intentionally misrepresenting these passages.
Rush’s third argument against capital punishment on scriptural grounds
is more persuasive. He argues that the laws and commandments given
by God to the Jews were only suitable to them and that the retributive
penal laws of the Jews ought not to be binding on Christian societies.
Appealing to the authority of Paul, he argues that the imperfection and
severity of the divine dispensation for the Israelites was intended to pro-
vide a contrast to the mildness of Jesus, who presented his own teaching
as to some extent eclipsing the Mosaic law (8, 22). This does not, howev-
er, affect the commands given by God to Noah, which are understood to
apply to all survivors of the Flood. This is as far as Rush goes in his scrip-
tural arguments – which is to say that he does not go very far in address-
ing the many difficulties posed by scriptural exegesis on this subject.
Instead, Rush seems to take the reasonableness and moral uprightness
of the penal reformers’ principles for granted – as evident to all – and
devotes the bulk of his discussion to showing the expediency of peniten-
tiary incarceration and the ineffectiveness of capital punishment.43

Conclusion

Montesquieu accepted Hobbes’s basic political teaching on the primacy

of individual security, but he also undertook to correct Hobbes’s pre-
scriptions for criminal justice practices in light of his own insights into
the nature and conditions of actual liberal societies. Much more clearly
than Hobbes, the French philosopher recognized liberalism’s potential
for softening mores and moderating passions, and he concluded that
criminal laws should gradually accommodate this salutary change, and
reinforce it by lessening the severity of criminal justice and by rely-
ing more on the positive incentives of social membership than on na-
ked threats. In accordance with this greater emphasis on the role of the
historical change in popular passions in humanizing Western criminal
 Liberalizing the Criminal Law: Montesquieu and Beccaria 97

justice practices, Montesquieu did not make it his priority to develop a

systematic theoretical defence of the right to punish based on modern
natural right; instead, he turned his attention to the task of outlining a
long­-term program for liberal criminal law reform. Montesquieu thus
ignored, without ever resolving, the troubling perplexities that had
plagued Hobbes’s efforts to justify punishment on modern principles.
Beccaria followed Montesquieu in many respects, while popularizing
and drawing out the practical implications of the latter’s reformism.
But whereas Montesquieu had a statesmanlike appreciation for the lim-
its of rational reform – which included a willingness to accommodate
some vestiges of the popular attachment to retribution – Beccaria’s little
treatise attempted to synthesize the rationalist zeal of Hobbes (includ-
ing the latter’s psychological reductionism and contractarianism) with
a lofty appeal to the nobler sentiments of his age. This synthesis was
as popular as it was incoherent. Thus, an unsatisfactory utilitarianism
quickly came to dominate criminological and penological thought in the
West, and that utilitarianism would soon come under attack by Kant.
 Chapter 4

Retribution and Individual Autonomy

in Kant’s Rechtslehre

From its beginnings, and on through its various transformations up un-

til the end of the eighteenth century, modern rationalist penal theory
had justified punishment exclusively as a means of preventing crime.
All other benefits of punishment – such as to rehabilitate offenders, or
to placate popular indignation – were viewed as ancillary to this pri-
mary one. Since the purpose of retribution, as commonly understood,
was essentially backward looking, it could in no way be admitted as an
aim of punishment, whether primary or ancillary. As noted earlier in
this book, Hobbes’s seventh law of nature commanded that “in revenges
(that is, retribution of evil for evil) men look not at the greatness of the evil
past, but the greatness of the good to follow.”1 Kant was the first modern
thinker to oppose this movement in punishment theory, regarding it as
undue lowering of the sights of criminal justice. Because there has been
considerable scholarly dispute about the extent to which Kant actually
championed retribution – and about the extent to which he was willing
to qualify his support for retribution in order to accommodate compet-
ing political concerns such as expediency, necessity, or honour2 – we
must allow the scholarly debates to intrude on the development of our
argument more than has been the case in preceding chapters. While I
will make some concessions to the objectors in what follows, I will show
that Kant remains the modern champion of retribution.
Kant argued that retribution is primarily what we mean by punish-
ment, and that contrary to earlier modern thought on this subject, it is
possible to think through and make sense of retribution as the primary
aim and justification of punishment. According to Kant, the true mean-
ing of retribution is accessible through notions already familiar to us
from everyday usage. Yet while the notions of “fit” and “proportionali-
ty,” “debt” and “desert,” are recognizable to ordinary moral experience,
 Retribution and Autonomy in Kant’s Rechtslehre 99

they are still far from being theoretically transparent. Accordingly, Kant
set out to explicate these common notions (primarily in his Rechtslehre)
and to show how they are connected to practical principles of reason,
while also showing why a purely utilitarian (or deterrence­-based) con-
ception of punishment is inadequate. According to Kant, it is precisely
our reason – as distinguished from the mere passion for revenge – that
requires retribution in appropriate circumstances.

Retribution as a Moral Duty of the State

Kant follows Hobbes in constructing civil society and government out

of a lawless state of nature, via a social contract.3 Since the civil order
is our sole bulwark against a return to the state of nature, and the fun-
damental condition of all right, the first duty of the constituted state is
to preserve itself and its subjects. On this basis, it would seem that the
principle of punishment would have to be deterrence. It thus comes
as a great surprise to readers of the Rechtslehre that when Kant turns
to discuss punishment, he denies that deterrence – or any other mate-
rial benefit – can be its primary concern: “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” (MM 6:331 [105]; emphasis in
original). The essence of punishment is not deterrence but retribution,
for “only the law of retribution … can specify definitely the quality and
the quantity of punishment; all other principles are fluctuating and un-
suited for a sentence of pure and strict justice” (ibid., 6:332 [105–6]; em-
phasis in original). And what exactly is the principle of retribution, as
dictated by strict justice? “None other than the principle of equality (in
the position of the needle on the scale of justice), to incline no more to
one side than to the other … whatever undeserved evil you inflict upon
another within the people, that you inflict upon yourself; if you steal
from him, you steal from yourself; if you strike him, you strike yourself;
if you kill him, you kill yourself” (ibid., 6:332 [105]). Punishment, then,

must be inflicted upon [the criminal] only because he has committed a crime,
for a man can never be treated merely as a means to the purposes of an-
other or be put among the objects of rights to things: His innate personal-
ity protects him from this, even though he can be condemned to lose his
civil personality. He must previously have been found punishable before
any thought can be given to drawing from his punishment something of
use for himself or his fellow citizens. The principle of punishment is a
categorical imperative. (Ibid., 6:331 [104–5]; emphases in original)
100 Punishment and the History of Political Philosophy

As this passage indicates, Kant’s view is that retributive punishment is a

requirement of the categorical imperative – the foundational principle of
all morality. The criminal, by virtue of his status as an end in itself, is im-
mune from any punishment based on a purely utilitarian (or consequen-
tialist) calculus, since such a punishment would amount to his being
treated not as a person, but as a thing; not as an end, but as a mere means.4
We must begin by briefly reviewing what the categorical imperative is
and why it requires human beings to be treated as ends in themselves.
Kant’s famous “Copernican” revolution precipitated a parallel revolu-
tion in moral philosophy. The crucial transformation was in the status of
physical security (or of material goods more generally). Earlier modern
thought held that civil society came into being in order to safeguard in-
dividual well­-being, but according to Kant, subjective individual goods
can be good only conditionally; for without a good will even happiness
itself becomes only an empty enjoyment. Through his critical philoso-
phy Kant claims to have shown that the goodness of any enjoyment
that human beings can experience depends upon some unconditional
good, and that that good can only be the good will, which performs duty
for duty’s sake (Gr 4:393 [61]). Since every person, as a rational being,
possesses such a will, each must be treated as an ultimate end in itself.
Accordingly, one of Kant’s formulations of the categorical imperative is:
“act in such a way that you always treat humanity, whether in your own
person or in the person of any other, never simply as a means, but al-
ways at the same time as an end” (ibid., 4:429 [96], second illustration).
Treating others as ends in themselves means, more specifically, acting
towards them in such a way that our actions are always compatible with
others’ choices, and thus with their ends, whether the content of those
ends is laudable or deplorable.5 Since each individual is an autonomous
being, I must treat my fellow not as a mere passive object of my benevo-
lence, but rather as an end “in all his actions” as well, by making his ends
“as far as possible my ends” (ibid., 4:428, 430 [95, 98]).
In light of this new vision of human goodness and dignity, Kant finds
the principles of punishment formulated by Hobbes and Beccaria inad-
equate, because they are incapable of respecting the humanity inherent
in each criminal – a humanity that must be respected no matter how
heinous the crime. It is not just that those earlier thinkers had failed to
draw the proper conclusions for criminal justice from the social contract.
Rather, the social contract itself is insufficient to ground any moral the-
ory of punishment. For that, one must begin from different premises,
namely, from the categorical imperative.
 Retribution and Autonomy in Kant’s Rechtslehre 101

Unfortunately, Kant did not leave a clear argument deducing the prin-
ciple of retribution from the categorical imperative, so we must construct
such an argument for ourselves based on Kant’s indications. Whenever
he speaks about punishment (including his discussion in the Rechstslehre),
Kant appeals to what might be called enlightened popular moral opin-
ion, taking his premises from that authority rather than deriving them
from first principles. While critical philosophy is crucial for elucidating
the rational grounds of morality, it is not always necessary for deciding
every practical question. That is because, according to Kant, critical phi-
losophy is in harmony with ordinary moral opinion – so long as it is
enlightened, for having cast off ignorance and prejudice in the age of
Enlightenment, modern societies have arrived at a correct understanding
of man’s duties, although very few people actually call themselves phi-
losophers.6 Thus, instead of recapitulating the derivation of the categori-
cal imperative, the introduction to the Rechtslehre (“Preliminary Concepts
of the Metaphysics of Morals”) presents a kind of glossary of basic termi-
nology from everyday use, including notions such as “person,” “choice,”
and, indeed, “crime.”7 In this summary of the basic terms of enlightened
popular morality, Kant defines a person as someone whose actions, as
well as their consequences, can be imputed to him as their author (MM
6:223 [16]). Thus we say that someone is responsible not only for his ill
behaviour (e.g., throwing a tantrum), but also for its consequences (e.g.,
damaging the property of others while so enraged). Yet Kant goes further
than this to argue that the consequences of one’s actions include their
due reward or punishment.8 If this is true, then respecting the autono-
mous choices of others involves punishing them whenever they should
transgress the law.
The notion that punishment is a form of respect for persons may
sound odd. Kant does not deny that as a selfish (“natural” or “empiri-
cal”) being, the criminal does not wish to be punished. But he neverthe-
less insists that we are not merely selfish beings, but also at the same
time moral and rational agents whose concerns transcend our private
interests – and in this he purports to echo ordinary moral experience.
When viewed as a moral and rational agent, the criminal necessarily
wills his deed along with all of the effects it entails, including any pun-
ishment due for voluntary wrongdoing.9 To repeat, as a selfish being he
will surely resist his punishment; but as a rational being – and for his
own sake we must assume that he has not irretrievably lost his ratio-
nality – the criminal necessarily accepts that punishment must follow
crime, in his own case no less than in another’s, as a dictate of universal
102 Punishment and the History of Political Philosophy

law. Thus, where Hobbes’s doctrine of authorization failed to show why

a criminal should accept his own punishment as just, Kant’s distinction
between the empirical self and the rational­-autonomous self manages
to do just that (assuming we accept the premises). Moreover, Kant does
not introduce any principle of authority other than rational autonomy:
in having a duty to undergo punishment, the offender is obligated to
sacrifice his private interests, but he is thereby obligated to no higher
authority than his own rational agency.
In this sense, to punish a criminal would be to pay him the highest
compliment, whereas to let him off the hook would be to condemn him
as childish, irresponsible, and something less than a full human being.10
Rewards and punishments, then, are imputable effects of our deeds just
as much as are any of the other effects of our actions, which is to say that
we are the authors of our own punishments and rewards. According to
Kant, only retribution satisfies the moral requirement of treating crimi-
nals as autonomous agents, since only it, as distinguished from utilitar-
ian punishment, defines the imputable rightful effect of the criminal’s
action. Utilitarian penalties have nothing to do with the grounds of a
criminal’s choice, since they look only to future conditions, such as se-
curity or well­-being. Retribution, on the other hand, establishes a pro-
portionality or likeness, in both quantity and quality – according to the
principle of “like for like” or lex talionis – between the offender’s action
(i.e., the crime) and its rightful effect (i.e., his punishment). In this way,
retribution respects and fulfils the offender’s autonomous choice.11 It is
for these reasons that Kant presents the right to punish in the Rechtslehre
as civil society’s moral duty to its criminals (what we would call a “posi-
tive” right). Kant illustrates the strictness of this obligation when he com-
ments on the hypothetical example of a society about to dissolve itself by
the consent of its members – at which point punishments will no longer
have a deterrent effect at all. In such circumstances even the last mur-
derer remaining in prison, according to Kant, must still be executed “so
that each has done to him what his deeds deserve” (ibid., 6:333 [106–7]).12
Kant concluded from the failures of Hobbes and Beccaria that any
theory of punishment simply cannot be made coherent if punishment is
understood as being concerned exclusively with deterrence. Yet insofar
as those earlier thinkers were aware of this insoluble difficulty, they
still refused to resort to commonsense moral opinions about desert and
retribution as a possible solution to their quandary, for that would have
amounted to reintroducing the harsh, politically destabilizing passions
that modern political philosophy had worked so hard to temper. Kant
 Retribution and Autonomy in Kant’s Rechtslehre 103

certainly did not wish to return to a pre­-modern conception of punish-

ment. Nonetheless, he saw a new possibility: the violent passion for re-
venge that Hobbes had denounced as a species of vanity and a vestige
of an older, less enlightened political tradition was, Kant believed, dis-
tinct from the concern with the dignity of the criminal that had inspired
a retributivist like Kant himself. Whereas the former was the product
of ignorance, selfishness, and violent habits, the latter was born of an
impartial concern with fair relations among equal human beings under
a universal law. In other words, Kant (in agreement with Montesquieu)
believed that commerce and liberal institutions had exerted a historical
humanizing effect on popular sensibilities and were continuing to do
so,13 and he deduced from this observation that whatever concern with
retribution remained in post­ -Enlightenment popular opinion must
have its origin in practical reason itself. Kant could now invoke popular
notions about punishment as his authority because public opinion had,
for the first time, ceased to be vengeful.
There is, however, a serious problem with the account I have provid-
ed so far – one that has led some Kant scholars to reject his view of pun-
ishment as incoherent.14 The difficulty stems from the fact that morality
and right are not identical for Kant; rather, the latter is understood to be
contained within the former. Moral laws, as such, are the unconditional
dictates of pure practical reason, which are in turn divided into two
kinds, juridical and ethical, according to the manner of their incentives.
Juridical laws, or laws of right, can have external incentives, meaning
that conformity with these laws can be coerced; whereas ethical laws
must serve as their own incentives, in the sense that conformity with
them requires that we obey them for their own sake.15 From the point
of view of juridical legislation it does not matter what motivates us to
act lawfully – whether it be a noble respect for the moral law or a base
desire for profit. Thus, for example, I may fulfil a contract merely be-
cause I hope to engage in profitable commercial dealings in the future.
Although it cannot be said that I have thereby acted ethically, I never-
theless have done all that is required of me by juridical law. For its part,
the state can do everything necessary to constrain me to obey my con-
tractual obligations, but it has no right to inquire into my motivations
for complying. But if this is the case, then if I should actually break the
law, what business does the state have in punishing me for my “inner
wickedness,” as Kant would have it? How can criminals be punished
for their immoral motives if juridical legislation is indifferent to inner
motivation?16 There is, in addition, the problem related to the fact that
104 Punishment and the History of Political Philosophy

Kant says elsewhere that while God can see into our souls and know
our true motivations, human beings have no such insight.17 How, then,
can the state punish a criminal retributively “in proportion to his inner
wickedness” if it cannot know his true motivations?
I will address the second difficulty first. There is one passage in the
Rechtslehre that wrestles with the tension between the moral require-
ment to punish criminals as they deserve and the difficulty we have dis-
cerning one another’s true motivations. There Kant advances a tentative
(and perhaps not completely satisfactory) solution. He tries to show that
in the case of capital crimes, the “fitting of punishment to the crime” will
always occur by “imposing the death sentence in accordance with the
strict law of retribution,” since “only by this is a sentence of death pro-
nounced on every criminal in proportion to his inner wickedness” (MM
6:333 ff. [106 ff.]). To explain his point by way of illustration, Kant gives
the example of two rebels: one who acts from the honourable (though
ultimately mistaken) motive of wishing to depose a government he re-
gards as illegitimate, and another who acts only for the sake of his pri-
vate gain. Kant argues that both would receive what they deserve if
they were sentenced to die. “Since the man of honor is undeniably less
deserving of punishment than the other, both would be punished quite
proportionately if all alike were sentenced to death; the man of honor
would be punished mildly in terms of his sensibilities [i.e., valuing hon-
our more than life] and the scoundrel severely in terms of his [i.e., valu-
ing life more than honour]” (ibid., 6:334 [107]). Thus there would be no
need for the state to inquire into the two criminals’ actual motivations;
whatever their motivations happened to be, a sentence of death (as dic-
tated by the principle of retribution) is the fitting punishment!
This would indeed be an ingenious solution to the problem were
it to hold generally in all cases of possible crimes and motivations.
Unfortunately, it is doubtful that it would hold in all cases, and thus
Kant’s solution is partial at best. Nevertheless, what is important is that
this passage shows that Kant refused to abandon his position that pun-
ishment must requite deeds in terms of what gives them their moral
worth – namely, their motivations.18 Perhaps Kant believed that, ulti-
mately, we have to accept a certain amount of uncertainty in judging
the motives of others, given the morally undesirable alternative. For
Kant (as later for Hegel), not punishing for fear of making a mistaken
judgment about motivations or circumstances would have morally
worse consequences than abolishing retribution, since a mistaken judg-
ment causes undue (and admittedly regrettable) physical harm to the
 Retribution and Autonomy in Kant’s Rechtslehre 105

criminal, without touching his inner dignity, while abolishing retribu-

tion entirely would amount to denying to all criminals their rationality,
thus dehumanizing them.19
In order to address the first difficulty – that juridical legislation’s in-
difference to incentives of action seems to rule out retribution – we must
first specify more precisely what Kant meant by “right” as this concept
is developed in the Rechtslehre. Because of his status as an end in itself,
man is the highest being in nature. In relation to the rest of nature, man is
an absolute master who possesses an unlimited right to use, transform,
and even destroy non­-rational beings according to his will.20 In actuality,
however, there are many rational beings coexisting at the same time in
a finite physical environment, each with an equally rightful claim to un-
limited external freedom. This inevitably brings human beings into con-
flict over external possessions. As a result of the equality of all rational
beings, men are required to observe the law of external reciprocity, ac-
cording to which any action is right as long as it can coexist with every-
one else’s equal external freedom.21 Reason authorizes enforcement of
this law, since this can be done without impinging on anyone’s freedom:
any asymmetrical coercion of another is a hindrance to equal freedom,
and thus the enforcement of the law of reciprocity can be justified “as a
hindering of a hindrance to freedom” (ibid., 6:231 [25]). This universal
authorization of reciprocal coercion then becomes the source of all right-
ful or juridical – that is to say, external – legislation. In civil society, the
entire authority to enforce the law of reciprocal coercion belongs to the
state, by virtue of which it can arbitrate conflicts and compel those who
defy the authoritative arbitration to submit to it by force.22
We must also keep in mind that while our juridical obligations may
be externally coerced, they are also at the same time morally incumbent
upon us, since duties of right are also necessarily duties of ethics.23 As
such, they ought to be obeyed even without external compulsion. This
implies that when I disobey a juridical law I not only disrupt the equi-
librium of an external system of reciprocal rights, but also act immorally
and thus incur moral guilt. As we have already seen, Kant understands
punishment essentially not as the enforcement of laws, but as recom-
pense for moral guilt, since punishment is the categorically (rather than
conditionally) necessary consequence of wrong. As such, punishment
does not derive from the state’s authority to enforce the system of equal
external freedom – that is, from right – but rather from the categorical im-
perative that enjoins us to treat others as ends by respecting their autono-
mous choices – that is, from the moral law in a more fundamental sense.
106 Punishment and the History of Political Philosophy

We can now see how Kant could have conceived of punishment as

essentially retributive without contradicting his distinction between ju-
ridical and ethical lawgiving. In this we have a prime example of how,
in Kant, morality not only coexists with but also in certain cases quali-
fies and restricts the administration of right. While the principle of right
requires deterrence as a means of enforcing equal external freedom, the
administration of punishment cannot be purely deterrent (i.e., deter-
rence cannot be its only justification), because that would violate the
moral requirement that we treat human beings as ends. Deterrence can
be a rightful aim of punishment, but only on the condition that it has
met the requirements of the principle of retribution.
Some commentators have tried to bring Kant back in line with classi-
cal liberalism by challenging, on his own grounds, his conclusion that
retribution is a strict duty of the state. Most recently, Thomas Hill has
claimed that Kant’s “crucial thesis concerns our liability to suffer in the
recognition of our own misdeeds, not our right or duty to make oth-
ers suffer for theirs.”24 Hill believes that according to Kant there is no
duty, but only an authorization, to punish the guilty, for his “retribu-
tive policies” are “not based on … intrinsic desert; nor do they stand as
fundamental moral requirements … Rather, they are best understood
as derivative features of a practice that requires independent justifica-
tion.”25 On Hill’s reading, Kant’s fundamental justification for punish-
ment is deterrence, which Hill takes to be implicit in the state’s function
as the enforcer of civil society’s laws. Those passages in the Rechtslehre
in which Kant appears to endorse a retributivist position of the sort I
have laid out above should, according to Hill, be interpreted in light
of the purely deterrence­-based justification of the institution of punish-
ment as a whole. Retribution, therefore, serves only as a necessary con-
dition – but not a sufficient one – of punishing the guilty; rather than
being a strict duty, it is only a restraint on the state’s enforcement power
keeping it from punishing the innocent and from punishing the guilty
disproportionately.26 In short, there is no moral imperative to punish
the guilty when no beneficial consequences can be expected to follow.27
Tunick has attempted to defend a similar reading of Kant.28 He points
to two passages in the Rechstlehre where examples are given of justifi-
able exceptions to the law of retribution. He reads these as providing
evidence that, for Kant, punishment is fundamentally consequentialist.
In one passage Kant relates a hypothetical case in which a shipwrecked
and drowning man saves himself by pushing another shipwrecked man
off a plank, thus keeping himself from drowning.29 Kant says about this
 Retribution and Autonomy in Kant’s Rechtslehre 107

case that a court should not punish the first man for murder, and the
reason he gives is that “the punishment threatened by the law could
not be greater than the loss of his own life” by drowning. Tunick inter-
prets Kant as implying that, since there would be no point in having a
penal law in this instance, for it could have no deterrent effect, the pur-
pose of penal laws as such must be deterrence.30 Tunick attributes the
same meaning to two more examples, in which Kant says that a soldier
who kills his opponent in a duel and a mother who kills her illegitimate
child ought to be exempted from capital punishment (as required by the
principle of retribution), because in both cases the law against murder
conflicts with an overriding incentive of honour (of a brave soldier in
the one case, of a chaste woman in the other), albeit one created by a
“barbarous and undeveloped” custom.31
I believe that Hill’s conclusion goes too far. As I have tried to show
in my analysis of Kant’s statements on punishment in the Metaphysics
of Morals, his clear intention was to understand punishment as a moral
category, and this led him to try to connect it with the categorical im-
perative and to argue for retribution as its essential purpose. Thus, I
do not think it is accurate to say that deterrence is the only justification
for punishment and that retribution is not a duty to the criminal. I do,
though, think that Tunick’s interpretation of Kant’s exceptions to the
law of retribution has some merit. At one point, Kant seems to want
to suggest that the exceptions he mentions are not really dispensations
from the law of retribution. For example, he says regarding the cases of
the duellist and the mother who kills her illegitimate baby that these
two individuals find themselves faced with two competing and equally
binding imperatives – the honour of their station and the prohibition
against murder – and that this clash of imperatives causes each to be
thrown back into the state of nature, where killing does not count as
murder.32 Thus, according to Kant, retribution would not have to be
exacted since in the state of nature there was no law against which to
transgress in the first place. But Tunick’s interpretation of these excep-
tions seems better than Kant’s, on the latter’s own terms. Kant’s appeal
to the state of nature to explain these exceptions seems rather contrived.
Is the conflict between a just law (i.e., against murder) and a “barbaric”
sense of honour really an insoluble quandary? It would seem less con-
trary to morality to require a man to defend his honour in court (or in
the court of public opinion) than to allow him to take another’s life un-
necessarily. Similarly, although it might require a hard choice, it seems
less contrary to morality to hold a woman responsible for not having a
108 Punishment and the History of Political Philosophy

child she would be unwilling to raise than to allow her to kill that child
with impunity once that child is already born. Here, Kant seems to be
trying to invent an ad hoc theoretical justification for a practical solu-
tion to the conflict between morality and competing political forces –
namely, between the moral demand for retribution and the entrenched
traditional sense of honour. The same seems true of Kant’s discussion
of the right of clemency, where his concern is clearly with the stability
of the civil order.33
Thus, although I do not think that Tunick’s ultimate conclusion – that
for Kant, the purpose of punishment is deterrence – is warranted, I do
think that he helpfully points to the fact that in certain details, Kant
failed to adhere consistently to his basic position regarding the categori-
cal character of the duty to punish retributively. On the one hand, Kant
wanted to affirm that retribution is a strict duty; on the other, he wanted
to balance this moral imperative against other concerns – although pre-
sumably a moral duty cannot really be thought of as categorical if it is
subject to exceptions.34 Nevertheless, it seems to me that despite this in-
consistency Kant does articulate a basis upon which it may be possible
to justify retribution in the strong sense, understood as a duty rather
than as a mere side­-constraint.
We have now traced the arguments by which Kant expounds the
meaning of retribution and defends it as the essential purpose of pun-
ishment in a morally serious civil society. Nevertheless, the talk of “pro-
portionality” and “likeness,” as that which truly respects the criminal’s
choice, remains rather vague, however much it may appeal to com-
monplace assumptions. The cogency of Kant’s argument for retributive
punishment depends on whether or not his claim that retribution pre-
serves the inner dignity of the offender as an autonomous being can
be explained in clear terms. But so far, we have only accepted this as a
formulaic premise. What proof is there that the offender himself really
does experience his punishment as dignifying and morally elevating?
We now turn to this question.

Conscience and the Basis of the Feeling of Guilt

What distinguishes human beings, according to Kant, is our capacity to

act as beings subject to universal legislation. Thanks to this capacity, I
may hold myself in the highest esteem and may claim a right to be re-
spected by others on an equal footing, regardless of the status in which
I have been placed by society, nature, or chance. Although Kant tends
 Retribution and Autonomy in Kant’s Rechtslehre 109

to emphasize this egalitarian basis of personal worth, he also acknowl-

edges another source of worth for which we can be held in esteem: the
goodness of our actions. An example of the latter may be found in the
familiar case of Kant’s world­-weary and reluctant philanthropist who,
“no longer moved by any inclination … tears himself out of this deadly
insensibility and does the action [i.e., helping others in need] without
any inclination for the sake of duty alone.” At that moment and “for
the first time his action has its genuine moral worth” (Gr 4:398–99 [66]).
Although, as a human being, the reluctant philanthropist already de-
serves respect, his difficult and rare self­-conquest elevates still higher
his standing in our eyes and in his own, and it is responsible for the
moral satisfaction he experiences. Kant’s characterization of this feeling
of moral satisfaction echoes unmistakably classical and biblical accounts
of virtue. According to the Rechtslehre, “there is a subjective principle of
ethical reward, that is, a susceptibility to being rewarded in accordance
with laws of virtue: the reward, namely, of a moral pleasure that goes
beyond mere contentment with oneself … and that is celebrated in the
saying that, through consciousness of this pleasure, virtue is its own re-
ward” (MM 6:391 [154]; emphasis in original). Although virtue is itself
intrinsically choiceworthy, it also makes us deserving of reward, one
form of which is the feeling of moral pleasure, because it essentially
involves a painful struggle against one’s natural inclinations.35
The case of self­-reproach is analogous to that of self­-esteem. When
we believe ourselves to have done wrong in some way, we judge our-
selves deserving of self­-reproach. We experience this as “pangs of con-
science.” Kant seems to interpret this inner experience of self­-reproach
not as a mere habituated emotional response (or a biological reflex) but
as an act of reason. For Kant, our pangs of conscience are the voice of
our own reason as it judges us, refusing to allow us to forget our des-
erts and even forcing us to suffer the pain of remorse as a kind of inner
judicial punishment. Granted, we are distressed by the pain of remorse
and in a sense wish to be rid of it. But from the premise, Kant seems to
infer that the pangs of conscience are the result of an act of reason and
the fact that we cannot get rid of these pangs, no matter how much we
may resist them – that we subject ourselves to such a trial voluntarily.
Kant uses the analogy of the judge quite extensively to describe the
workings of the reproachful conscience. When the conscience passes
judgment, it “pronounces the sentence of happiness or misery, as the
moral results of the deed” (ibid., 6:439n [189n]). The conscience does
not merely identify certain types of actions as morally good or bad; its
110 Punishment and the History of Political Philosophy

judgment is also a “sentence” bearing consequences: it “either acquits

or declares us guilty and deserving of punishment” (LE, “Conscience,”
129). What is more, it is not enough that self­-reproach and repentance
follow this judicial pronouncement of the conscience; the conscience is
satisfied only “if [its judicial verdict] is felt and enforced” (ibid., 131).
According to Kant,

the first effectual expression of this judicial verdict which has the force of
law is moral repentance; the second, without which the sentence is inop-
erative, is action in accordance with the judicial verdict. If it does not result
in practical endeavor to do what is demanded for the satisfaction of the
moral law, the conscience is but an idle conscience, and however penitent
we may be the penitence is vain so long as we do not satisfy the debt we owe
to the moral law; for … a debt is not satisfied by penitence, but by payment.
(Ibid., my emphases)

By “payment” of the “debt” we owe to the moral law, as distinguished

from mere “penitence,” Kant seems to be referring to judicial punish-
ment. Although he does not spell it out, his meaning seems to be that a
person with a healthy conscience must be willing not only to accuse
himself but also to submit to punishment, knowing this to be the mor-
ally necessary consequence of his actions. If he does not so submit, then
he loses the basis for self­-respect and is degraded in his own eyes.36 To
preserve his dignity as a moral being the offender must submit himself
to retributive punishment.
We can see from Kant’s remarks on the reproachful conscience that
there is indeed some support in ordinary moral experience for his view
of the dignifying effect of retributive punishment on the criminal him-
self. That there is no parallel to this account in the works of Hobbes,
Montesquieu, or Beccaria – the classic defenders of deterrence­-focused
punishment – might be regarded as a serious weakness in their penal
thought. Still, one wonders whether Kant’s assumption that a rational
basis exists for the reproachful conscience can be defended by critical
reason itself, as distinguished from common practical experience. This
work is done by Kant’s concept of the highest good.

Retribution and the Highest Good

Kant’s account of the highest good (or summum bonum) functions to re-
store harmony between the two aspects of our being: the rational and
 Retribution and Autonomy in Kant’s Rechtslehre 111

the empirical. As empirical beings, we are driven by the desire to fulfil

our needs, the total satisfaction of which we call happiness. As moral
beings we hold that we ought to pursue virtue – a good higher than
happiness and choiceworthy for its own sake – above all else, regardless
of whether it also contributes to our well­-being. On the one hand, there
is nothing intrinsic to the concept of virtue that requires the sacrifice of
this­-worldly happiness – Kant was not an ascetic. On the other, Kant
acknowledges that there is no necessary correlation in nature between
virtue and well­-being, as both experience and natural science teach us.37
It is all too common to find virtuous individuals suffering undeservedly
and the unjust prospering. According to Kant, we cannot help but reject
this “antinomy” of practical reason, which threatens to undermine mo-
rality’s coherence. In order to reject that antinomy, reason demands that
we postulate the conditions that would allow the convergence of hap-
piness and morality – the highest good. Besides the postulate of the effi-
cacy of rational freedom, there are two such conditions: (1) the existence
of an omnipotent, intelligent, and just being that always acts in accor-
dance with the laws of morality to reward the virtuous and punish the
vicious, and (2) the immortality of the soul.38 For, if it were otherwise,
then “the moral law … must be fantastic, directed to empty imaginary
ends, and consequently inherently false” (ibid., 5:114 [118]).
Contrary to the biblical view, the highest good cannot be construed as
a gift of divine grace. Kant’s God is not the mysterious God of the Bible,
who will be gracious to whom he will be gracious and whose ways are
beyond our ken, but the perfect agent of practical reason. If the highest
good were to depend on grace, then virtue might not always be reward-
ed – an implication that practical reason rejects. If “virtue and happi-
ness are thought of as necessarily combined, so that the one cannot be
assumed by a practical reason without the other belonging to it,” then
this combination must be “as the connection of cause and effect” (ibid.,
5:113 [117]). The only way we can conceive of the connection between
virtue and happiness as causal is if we postulate the immortality of the
soul and the existence of a just and omnipotent God, who could never
will anything other than the highest good, as reason understands it.39
Such a God would guarantee the “effect” of happiness as a consequence
of moral goodness by distributing well­-being to each in exact propor-
tion to his desert, “for to be in need of happiness and also worthy of it
and yet not to partake of it could not be in accordance with the complete
volition of an omnipotent rational being” (ibid., 5:110 [114–15]). Only
on the assumption of such postulates is morality possible, since only
112 Punishment and the History of Political Philosophy

then can we allow ourselves to hope that human beings will be reward-
ed and punished in accordance with their worthiness by an unerring
judge, if not in this life then in the next. According to Kant, rational faith
in such postulates is unavoidable; we always affirm them when we are
acting morally, even when we are not aware of it.
The notion that each person will be held to account by an omniscient
and all­-powerful judge for every one of his actions is at the same time
terrifying and comforting. Kant’s God is first and foremost not a com-
passionate and merciful God, but a just one. For in distributing happi-
ness in accordance with the concept of the highest good, God does not
will that all men should necessarily be perfectly happy, but rather that
they should enjoy as much happiness as they have deserved through
their deeds. In Kant’s Critique of Judgment, he says that the final purpose
of nature, as created by God, “can only be man under moral laws” (CJ
5:445 [334]; emphasis in original). But, Kant continues,

I say deliberately: under moral laws. The final purpose of creation is not
man [acting] in accordance with moral laws, i.e., a man whose behavior
conforms to them … And this agrees perfectly with the judgment that hu-
man reason makes when it reflects morally on the course of the world.
Even in evil we believe we perceive the traces of a wise reference to a
purpose, provided we see that the wanton villain does not die until he has
suffered the punishment he deserves for his misdeeds … The highest wis-
dom in the government of the world we posit in this: that the opportunity
for good conduct, but the consequence of both good and bad conduct, is
ordained according to moral laws. In the latter consists, properly speak-
ing, the glory of God, and hence it is not unfitting if theologians call it the
ultimate purpose of creation. (Ibid., 5:449n [338n]; emphases in original)

According to Kant, then, the ultimate end of divine providence is not

necessarily men acting in complete conformity with moral laws. Rather,
the end of providence is perfect justice: the proportional distribution of
happiness in accordance with moral worthiness.40
This ringing endorsement of divine retribution is striking, and signifi-
cant for our understanding of Kant’s view not only of divine retribution
but also of the significance of retribution, as such.41 Recall that in the
Rechtslehre, Kant had argued that judges duly constituted by the state
must embody the principle of retribution in order to protect the dignity
and autonomy of criminals. To that account of the political function of
punishment – as imposing a strict duty on the state – we now add the
 Retribution and Autonomy in Kant’s Rechtslehre 113

present account of the wider moral significance of retributive punish-

ment as it emerges in the context of Kant’s second and third critiques. If
one focuses on the Rechtslehre alone, it is easy to see how one might get
the impression that deterrence is the fundamental aim and justification
of punishment as an institution and that the principle of retribution is
only a necessary, but not a sufficient, reason for punishing crime. We
have now seen that for Kant, retribution is an aspect of the ultimate end
of practical reason, the highest good.
It can now be said, after a review of the evidence, that Kant under-
stood retributive punishment of the guilty as an intrinsic moral good.
This helps explain why he regarded it as a strict moral duty of the state
(as distinguished from a limitation or side­-constraint).42 Furthermore,
we can now also better appreciate the extent to which the principle of
retribution resonates with ordinary moral experience, even after that
experience has been shaped historically by the idea of modern natural
right. Finally, and most importantly, we can now see how Kant under-
stood retribution to be grounded in pure reason, as an extension of the
highest good.
What, then, can be said about the merits of Kant’s demonstration of
the rational basis of the principle of retribution? Some commentators
have pointed out several difficulties with Kant’s linking of virtue and
happiness in the concept of the highest good. Shell, for example, has ob-
served that Kant’s argument seems to blur together two distinct issues:
moral worth as an unconditioned good, on the one hand, and moral wor-
thiness as the condition of entitlement to happiness, on the other.43 How
exactly does reason make the transition from the former to the latter?44
Furthermore, in our earlier discussion of Plato’s penal thought we
saw the Athenian Stranger raise the question of the coherence of retrib-
utive punishment precisely from the point of view of the commonsense
moral opinions to which retributivists like Kant try to appeal. There
we saw that the question of retributive punishment points to the more
fundamental question of the goodness of moral virtue itself. Retributive
punishment, as Kant and most people understand it, assumes that an
unjust action diminishes the unjust person’s worthiness to be happy,
since we say that the unjust person “got away” with some portion of
happiness to which he was not entitled (e.g., stolen property, or illicit
self­-gratification). The punishment that follows is thought to restore the
imbalance between worthiness and well­-being. But if moral virtue is
something intrinsically good and self­-sufficient, and if it brings its own
rewards of pleasure and inner well­-being (in the conscience), while guilt
114 Punishment and the History of Political Philosophy

is accompanied by pangs of self­-reproach, then retributive punishment

must be irrational, since the unjust man would have to be considered
already unhappy in spite of his ill­-gotten gains, and thus need not be
subjected to further suffering. Likewise, it would follow from the impli-
cations of this commonsense view that just or law­-abiding individuals
are already so completely fulfilled that they ought not to feel any indig-
nation or envy towards the criminal, but should rather pity him. Kant
insists that moral virtue is an unconditional (self­-sufficient) good and
that it is accompanied by inner well­-being, and he also wishes to affirm
the justice and reasonableness of retribution. Yet I do not believe that
Kant’s arguments in support of retributive punishment are sufficient
to defend its coherence in light of the Platonic–Socratic critique.45 That
happiness ought to be dependent on virtue, and that we must believe
in the existence of a divine dispenser of retribution, had been judged
by Kant to follow from the facts of the moral consciousness. But is the
moral consciousness as unequivocal about what constitutes just pun-
ishment as Kant claims? Rather, does not the moral consciousness itself
under certain circumstances approve of the departure from strict justice
in favour of justice in another sense – as expressed, for example, in the
principles of equity and mercy? In other words, is it not possible that
the moral consciousness itself is much more divided as to the goodness
of retribution than Kant had admitted? Whereas Kant had ascribed our
concern with happiness exclusively to our empirical natures, is it not
possible that he did not sufficiently consider whether, or to what extent,
this concern is essentially bound up with our concern with justice itself?

Conclusion

One of the most important implications of Kant’s penal thought is that

modern societies continue to find the connection between punishment
and desert deeply compelling, notwithstanding their abiding concern
with security. For Kant, this meant that just deserts and retribution must
be accommodated within any modern theory of punishment. Kant was
able to formulate a retributive theory of punishment on modern prin-
ciples – that is, without returning to the classical analysis of punishment
– and in this respect Kant’s achievement may be admired. As a retribu-
tive theory of punishment, Kant’s account avoids the morally shocking
implications of the essentially consequentialist theories of earlier mod-
ern thinkers, especially the implication that the interests of criminals
must bow to the interest of the greater number. The sort of retributive
 Retribution and Autonomy in Kant’s Rechtslehre 115

punishment Kant imagined would not, therefore, be the kind of venge-

ful, harsh, and even brutal punishment that one might have observed
practised in the austere classical republics. Rather, Kantian retribution
would serve as a measure of proportionality that would, in many cases,
limit the severity of the penalty in a manner that should make it more
fitting to the gravity of the crime, its circumstances, the intention of the
criminal, and so on. In short, in the modern context the principle of retri-
bution could be expected to moderate punishment by helping to avoid the
excesses of utilitarian calculation. At the same time, the retributive aim
of Kantian punishment is also intended to do honour to our enduring
concern with justice, understood as both a political and a moral virtue.
Although we had occasion to raise questions about the ultimate co-
herence of the popular concern with retribution to which Kant appeals,
drawing on certain Socratic reflections that we have gleaned from our
examination of Plato, there is also much in Kant’s solution to the prob-
lem of punishment to recommend it as a political compromise. With
the benefit of hindsight, however, we can say that the historical success
of Kant’s theoretical achievement has been mixed. On the one hand,
many Western democracies continue to express their support for retri-
bution as an appropriate aim of punishment, distinct from deterrence,
rehabilitation, and other accepted aims. Some countries – particularly
the United States – even acknowledge retribution as an aim of their
criminal justice system in their statutes and in the judicial opinions
of their courts. On the other hand, the laws and governments of most
Western countries have ceased to regard retribution as an appropriate
aim of punishment, and there are currently several activist movements
(such as Amnesty International and the Restorative Justice movement)
that are opposed to retribution on principle. Moreover, there is more
intellectual opposition to retribution in Western societies than there is
principled support for it. In the next chapter we will turn to perhaps the
best­-known attack on the modern criminal justice tradition, one that
criticizes not only its retributive element but indeed the whole edifice:
Foucault’s Discipline and Punish.
 Chapter 5

Foucault and the Crisis of

Modern Criminal Justice

All of the modern thinkers we have examined up to now shared a

sense of optimism about the modern penal reform project as a whole.
The discoveries made by Enlightenment science and philosophy con-
cerning man’s nature and his rights, when applied to criminal justice,
promised to purge all of the prejudice and barbarism that had plagued
traditional criminal justice institutions. According to Foucault, howev-
er, the optimism of these thinkers, and of the penal reformers inspired
by them, was misplaced. While the Enlightenment had indeed ushered
in an age of unprecedented order and security, the new “penality” that
would come to enforce and sustain this order could not be described
as humane (mild and benevolent), despite being technically sophisti-
cated. Foucault’s account of the institutional and intellectual history
of modern punishment is intended to show that the Enlightenment’s
penal reform project took an unexpected turn, ushering in an unprece-
dented mode of social control that Foucault calls “discipline.” This new
form of correction, according to Foucault, does not rehabilitate, does
not deter, and is not a milder form of punishment than its historical
precedents. Indeed, the new mode of correction has the appearance of
mildness, having exchanged corporal punishments for incarceration,
when in fact it has only adopted new methods that are the more op-
pressive for being insidious.
The characteristic techniques of modern discipline have been firmly
entrenched in our institutions for a long time, according to Foucault,
but we are only now beginning to understand their nature and dimen-
sions and to appreciate their relation to the original modern idea of
punishment. It is this insidious modern penality that Foucault purports
to analyse, and it is in relation to this historical examination that he first
announces the crisis of modern criminal justice.
 Foucault and the Crisis of Modern Criminal Justice 117

Contractarianism in Theory vs Practice

Foucault acknowledges that to a significant extent the ambitions of the

early penal reformers were fulfilled. Towards the latter part of the eigh-
teenth century, violent crimes substantially declined, while punishments
became milder (Discipline and Punish, 75). In the popular understanding,
as among the intellectual and political elites of the time, it became a
widely accepted principle that no matter how horrific the crime, society
must continue to respect the humanity inherent in the criminal. “In the
worst of murderers,” the spirit of Enlightenment proclaimed, “there is
one thing, at least, to be respected when one punishes: his ‘humanity’”
(74). The new humanizing spirit in criminal justice was not so much a
theoretical doctrine, according to Foucault, as “a cry from the heart of
an outraged nature,” one that was expressed in a “lyrical” rhetoric by
the Enlightenment philosophes (74, 90–1). This is not to say that crime de-
clined altogether; rather, the drop in violent crime coincided with a rise
in property crime, according to contemporary reports cited by Foucault.
Nor did law enforcement become less strict in every respect, according
to this account: although it became less corporal in most cases, it also be-
came more intrusive and far­-reaching as more categories of crime were
established in law – particularly those against property, reflecting the
“bourgeois” class’s conception of justice – and as the police apparatus
began to grow and take on more of a surveillance role.
The “lyrical” rhetoric in terms of which the appeal to humanity as
the limit of punishment was framed tended, however, to obscure the
theoretical paradox underlying that noble sentiment. According to “the
general theory of the contract,” citizens are presumed to have consent-
ed to the established laws, since these protect the lives and property
of all. In light of this principle, any violation of the lawful order is a
sure sign of the offender’s dangerousness and thus gives society a right
on contractarian grounds (“since the defense of each individual is in-
volved” in the defence of the whole) to use any means it may deem
necessary to protect itself against the threat (89). In his discussion of
the theoretical basis of the modern punishing power, Foucault limits
himself to a quote from Rousseau’s On the Social Contract:

Every malefactor, by attacking the social rights, becomes, by his crimes, a rebel
and a traitor to his country; by violating its laws he ceases to be a member of
it; he even makes war upon it. In such a case the preservation of the state is
inconsistent with his own, and one or the other must perish; in putting the
guilty to death we slay not so much the citizen as the enemy. (quoted at 90)1
118 Punishment and the History of Political Philosophy

“Every malefactor” (however great or small his crime) gives the society
whose laws he has violated an “absolute right” over his life, making him-
self liable to the full range of punishments, according to their discretion.
Since every citizen is assumed to consent to everything that is enacted in
the laws, the criminal himself must logically accept even the extreme
penalty as the legitimate consequence of his deed, by virtue of which he
becomes “worse than an enemy” and “nothing less than a traitor” and “a
monster,” even in his own eyes. The full consequence of contractarian
punishment theory, then, according to Foucault, is that “the least crime
attacks the whole of society; and the whole of society – including the
criminal – is present in the least punishment” (ibid.). Where could one
possibly find a moderating principle in such a theory of punishment?
But this very brief exposition of the foundational argument for a con-
tractarian right to punish is incomplete, for it relies only on the most
extreme, anti­-liberal version of contractarianism – namely, Rousseau’s.
Recall that Hobbes denied that individuals transferred all of their rights
to the sovereign and argued that they may, and must, retain their natu-
ral right to preserve their own lives. Hobbes even conceded to individu-
als in civil society a corresponding right to judge for themselves what is
necessary for their preservation, and even a right to the means required
for a comfortable existence. On the basis of these inalienable rights
(which are not to be confused with entitlements), Hobbes deduced a
substantial range of excuses and extenuations that may be reasons to
exempt the criminal from the most extreme punishments if he commit-
ted his crime under some duress or restraint. In his exposition, Foucault
does not take into account this (or any other) alternative version of con-
tractarianism, and this weakens his conclusion that even the least crime
makes the offender a rebel and a traitor so that “the preservation of the
state is inconsistent with [the preservation of the criminal], and one or
the other must perish.”
Our investigations in preceding chapters can help us understand that
Foucault’s conclusion – that according to modern contractarianism the
least criminal becomes the property of the state – is true only as a formal
possibility, but it is not formally necessary, nor is it necessary for it to
become common practice, given modern conditions. We will be able to
confirm this if we focus on two elements of the Hobbesian deduction of
the right of sovereignty. According to Hobbes’s deduction (which the
Rousseau passage quoted by Foucault closely parallels), the only secure
way for individuals to escape the violent state of nature is to recog-
nize (1) that the sovereign authority established by the social contract is
 Foucault and the Crisis of Modern Criminal Justice 119

absolute, and (2) that the fundamental guarantee of the political union’s
continued existence is the continued consent of its members. The con-
sent of citizens – be it motivated by fear of violent death (as in Hobbes)
or by respect for the general will (as in Rousseau) – is of the utmost im-
portance in practice because, if it should ever lapse, there would be no
other power on earth to take over the task of maintaining the civil order.
It is formally true that any crime, however minor, may be regarded as
a threat to the unity of the political will, since in every criminal action
the offender takes it upon himself to decide what is permissible, which
is the definition of the state of nature, where each is his own judge.2
However, nothing in this contractarian deduction requires that the least
criminal be put to death or transformed into property of the state, since
in the state of nature there is no justice, and therefore no guilt or liability
to punishment. It is therefore not true that the least criminal loses all his
rights against civil society and is to be dealt with at society’s absolute
discretion. What is more, it is not true given actual conditions (at least
in a Hobbesian political system) that civil society would find it useful to
punish a minor crime with any great severity, since the lapse of consent
on the part of one subject is practically negligible in the face of a power-
ful sovereign who enjoys the support of the vast majority of his subjects.
Nevertheless, Foucault’s conclusion remains correct as a formal pos-
sibility, since civil society retains an absolute right to defend itself, and
since it (or its constituted representative) remains the sole arbiter of
what is necessary and useful to its preservation. Hobbes, as we saw,
had unambiguously affirmed this formal possibility again and again.
Everything then comes to depend upon the actual political conditions
that inform criminal law and administration in a given contractarian
society. Hobbes, as we know, preferred monarchy as a form of govern-
ment, and therefore in an ideal Hobbesian state we might expect the
sovereign to favour an enlightened policy where severe punishments
are reserved for only the most dangerous offences. But conditions are
significantly different in a regime governed by democratic majorities
of small property owners. Simply put, a democratic majority consist-
ing of small property owners living under a regime that is concerned
with preserving individual security and protecting commerce will tend
to favour stricter law enforcement and more punitive corrections – not
because the property­-owning majority is vengeful, but because its in-
tense fear for the security of private persons and property will incline
it to exaggerate the threat of crime and the means thought to be neces-
sary to contain it. As we know from experience, when crime rates rise
120 Punishment and the History of Political Philosophy

– or are perceived to rise (cf. 76–7) – an apprehensive public will often

demand the enactment of laws that are “tough on crime,” even if this
goes against the enlightened opinion of policy­-makers. And this fearful
preoccupation with the maximum possible security of our persons and
property will continue to predominate in liberal societies, at least ac-
cording to Foucault’s analysis.
The most massive proof of this that is adduced by Foucault is the
steady and seemingly inevitable expansion of carceral punishment. In
every modern nation today, hundreds of thousands – and in some coun-
tries, millions – of criminals have been warehoused for long, sometimes
indefinite terms. This is a punishment whose conditions can hardly be
considered humane and that has little hope of correcting or returning
offenders to law­-abiding society. (More on this below.) What is more,
Foucault claims that this outcome was inevitable and that it could have
been foreseen even in the eighteenth century, since, according to his anal-
ysis, at that time it was the recently empowered economic middle class
that had the greatest influence over the character of the new criminal
law reforms, so that the economic (rather than humanitarian) concerns
of the bourgeoisie were “the essential raisons d’être of penal reform” (89).
Against the perversely harsh modern criminal justice systems, the
voice of humanity asserts itself as a “discourse of the heart.” It is the voice
of intellectuals, of political figures, of lawyers and judges, and of human-
itarian activists. Yet Foucault argues that the impassioned character of
this humanitarian objection of the heart, however sincere, is inherently
unstable, since it depends not on the feelings of the one who is liable
to punishment, but on “those of the men who … have the right of exer-
cising against him the power of assembly” – namely, the judges, jurors,
and legislators. While the sentiments of judges and political leaders may
prove to be magnanimous on some occasions, they may just as easily be
influenced by “all the hardness of heart induced by familiarity,” as well
as by the voting public to which they are accountable (91). The unflatter-
ing truth is that democratic majorities and their leaders can easily afford
to wax eloquent about the humanity of the criminal; yet when it comes
time to make policy decisions, fear of crime generally wins out against
magnanimity and enlightened policy.
It is all to the good when the disciplines of criminology, sociology, and
psychology make efforts to inform the public that the impact of “tough
on crime” legislation may be less than anticipated and that punishment
is less effective at preventing crime than less coercive methods, such as
educational programs for juvenile offenders and psychiatric treatment
for past offenders with psychological problems. In the final analysis, the
 Foucault and the Crisis of Modern Criminal Justice 121

contractarian theory of punishment – especially in democratic societies –

implies that the ultimate judgment of the danger posed by crimes, and of
the appropriate methods of crime prevention, rests with the public. For
the entire edifice of the liberal state stands or falls with the public’s opin-
ion of its security – which, as we recall, Montesquieu took to be the core
meaning of “liberty.” To use Hobbes’s formulation, “the greater crime is
that where the damage, in the common opinion of men, is most sensible.”3
Thus, no matter how much I, as the offender, might appeal to the mercy
of my fellows, or beg them to consider how negligible was the harm I
caused, or how pressing my circumstances, or, again, how unreasonable
is the law that I violated, I may justly be treated “as a lion or a tiger, one
of those wild savage beasts, with whom men can have no society nor
security,” to use Locke’s formulation.4 Foucault is therefore right to this
extent: that in modern liberal democracies, criminal justice practices will
tend to be informed by the most fearful policy.
This in no way alters the fact that retribution remains invalid in the
new correctional systems, since it remains true that by definition, ret-
ribution has no social utility. In this respect, one might argue, the new
systems are humane, at least when compared with traditional penal
practices. Yet perhaps the decline of retribution did not contribute to
an advance towards moderation in punishment after all. The principle
of retribution requires proportionality in punishments, and this means
not only that the severity of punishments may not be decreased below
what the crime deserves, but also that it may not be increased above
that level. In modern penal theory, in which retribution has been dis-
credited,5 the way has been opened to disproportionally heavy punish-
ment, which is justified on grounds of general utility and society’s right
to self­-defence. This increased severity is reflected not so much in the
length of individual prison sentences, but rather in the mode of punish-
ment itself – namely, mass incarceration. It may be that incarceration
was originally introduced as the primary mode of punishment for its
mildness (as a non­-corporal form of punishment) as well as for its util-
ity, but it continues to be heavily relied upon even now that it has be-
come clear that mass incarceration is often quite degrading and difficult
to endure psychologically and even physically. Thus Foucault writes
that while “the right to punish has been shifted from … vengeance …
to the defense of society,” it “now finds itself recombined with elements
so strong that it becomes almost more to be feared” (90).
So we must therefore grant that Foucault has struck a serious blow
against modern punishment theory and that in elucidating its contra-
dictions he has made explicit the tensions that have existed in modern
122 Punishment and the History of Political Philosophy

punishment theory since Hobbes. Foucault’s analysis reveals that the

supposed fortuitous harmony between general utility and moderation,
which the penal reformers had taken for granted, has proved to be il-
lusory. The Enlightenment philosophes believed that both of these goals
could be achieved by the same reforms, but according to Foucault they
had overestimated the bourgeois class’s ability to judge its own best in-
terests, rightly understood, as well as its capacity to act on its humanitar-
ian aspirations.

Foucault on the Prison

We now turn from Foucault’s analysis of modern punishment theory to

his examination of carceral punishment itself. Foucault’s historical ac-
count of the emergence of the prison is intended to show that what was,
in the eighteenth century, only a theoretical possibility has over time
become a reality, and is today an urgent problem – namely, the increas-
ingly despotic character of the administration of punishment.6 Foucault
does not, however, regard the emergence of carceral punishment as a
natural outgrowth of modern punishment theory. In the first place, he
points out that the early penal reformers had maintained that the de-
terrent effect of punishment would be maximized if the penalty for a
given crime resembled it in quality and quantity as much as possible,
so that the association of the one with the other in the minds of citi-
zens would be as clear as possible. According to that view, theft would
best be punished by confiscation, abuse of public office by loss of civil
rights, murder by execution, arson by burning, and so on. This resem-
blance of the punishment to the crime “is an advantage for the stability
of the [mental] link, an advantage for the calculation [i.e., on the part
of the criminal] of the proportions between crime and punishment and
the quantitative reading of interests” (104–5; 114).7 Such a scheme obvi-
ously could not be carried out if every punishment was reduced to one
form – incarceration – with only the length of the prison term varied
according to the severity of the crime. Foucault further observes that
the early penal reformers had many of the same reservations about in-
carceration as the later critics of the prison would have: it is costly; it
allows vices to be communicated from hardened criminals to petty and
first­-time offenders; it exposes inmates to the arbitrary will of guards
and prison administrators; it hides convicts behind walls, thus prevent-
ing them from serving as a salutary example to the public; and so on
(114–15). But “within a short space of time, detention became the es-
sential form of punishment. In the [French] penal code of 1810, between
 Foucault and the Crisis of Modern Criminal Justice 123

death and fines it occupies, in a number of forms, almost the whole

field of possible punishments.” How did this come about?
According to Foucault, part of the reason for the rapid growth of pris-
ons was enthusiasm about a technological advance of the nineteenth
century – what Foucault calls “penitentiary science” (120–6). This was
an effort to apply modern medical and psychological theories, as well as
certain principles of architectural design (championed by Bentham in his
Panopticon), to the creation of new prisons – penitentiaries – that would
not only detain and punish but also cure and reform. The key innova-
tions of penitentiary science were isolation of the individual prisoner and
his constant observation. These innovations were immediately put into
practice in a handful of model prisons. The most famous of these were
in Philadelphia (the Walnut Street and Cherry Hill prisons) and Auburn,
New York.8 They were built to house inmates in individual cells, in or-
der to remove them from the harmful influence of their fellow inmates
and to encourage reflection and repentance. An inmate’s life was to be
strictly regimented, and his time was to be divided into periods of labour,
isolation, and religious instruction, so that no idle habits or corrupting
influences could interfere with his rehabilitation. This benefit was to be
further ensured through constant surveillance by prison officials occupy-
ing conveniently located observatories. Foucault admits that these tech-
nical innovations were important, but he denies that they are sufficient
by themselves to explain the rapid expansion of carceral punishment.
The deeper reason for this rapid expansion, according to Foucault,
was the development of what he calls modern discipline (to be dis-
tinguished from punishment proper). In Foucault, this is a technical
concept that is meant to capture a very broad range of practices and
techniques that were already in use in the late eighteenth century in the
military, hospitals, schools, monasteries, and workhouses. Punishment,
as traditionally understood, and as the eighteenth­-century penal re-
formers continued to understand it, was meant to operate on the
minds or souls of the convicted through “signs” and “representations.”
Whether the emphasis of traditional punishment was retribution or de-
terrence, it was to be accomplished by communicating a message (a
threat, a legal judgment, the assertion of a norm) to the criminal and to
society as a whole. In contrast, modern discipline, which increasingly
becomes the new “penality,” targets not the soul, but the body:

The point of application of the penalty is not the representation, but the
body, time, everyday gestures and activities; the soul, too, but in so far as it
is the seat of habits. The body and the soul, as principles of behavior, form
124 Punishment and the History of Political Philosophy

the element that is now proposed for punitive intervention. Rather than on
an art of representations, this punitive intervention must rest on a studied
manipulation of the individual … As for the instruments used, these are no
longer complexes of representation, reinforced and circulated, but forms of
coercion, schemata of constraint, applied and repeated. Exercises, not signs:
time­-tables, compulsory movements, regular activities, solitary meditation,
work in common, silence, application, respect, good habits. (128)

The prison presented itself as an ideal opportunity to apply these tech-

niques, and according to Foucault, this is what really drove the rapid
increase in the use of incarceration as the primary mode of punishment,
despite its inconsistency with the principles of punishment theory.
Once the penitentiary became established in the public understand-
ing as a great advance in criminal justice, the wedge that had been
driven between modern punishment theory and penitentiary practice
began to widen, and it continued to widen as the logic of penitentiary
science took its course. Whereas criminal justice theory requires judicial
sentencing and punishment to follow clear legal standards known to all,
disciplinary science calls for absolute discretion for prison administra-
tors, which militates against determinate sentencing in accordance with
fixed legal standards.9 For while justice proclaims that all individuals
are equal in the eyes of the law, and that they must be judged by their
deeds alone, disciplinary science teaches that the effectiveness of cor-
rection is directly related to the extent to which disciplinary techniques
are individualized – that is, the extent to which the qualities of body
and soul of each individual, to the last minute detail, are taken into ac-
count when determining an appropriate corrective regimen.10 Despite
this contradiction, a fundamental change in orientation was soon ac-
cepted by courts and legislative assemblies, which were persuaded of
the social utility of the penitentiary technique, and which began to cede
more and more responsibility for penal administration to prison offi-
cials, who over time acquired a “despotic” power in this domain (108).11
As a result, “what is now imposed on penal justice as its point of appli-
cation, its ‘useful object,’ will no longer be … the juridical subject of an
ideal contract; it will be the disciplinary individual” (227).
Foucault’s discussion of the emergence of modern discipline goes
quite far, and much of it is speculative and unlikely to convince the most
sceptical of readers. It denies that modern discipline – taken as a whole
in all its practical effects – was ever intended as a constitutive element
of any of the large­-scale political projects conceived by seventeenth­- or
 Foucault and the Crisis of Modern Criminal Justice 125

eighteenth­-century political philosophy. He credits the mechanistic view

of man introduced by Descartes as only partly responsible for opening
up this new possibility of social organization and control (136–7). He
goes on to argue that the genesis of what he calls modern discipline is in
large part a mysterious process: a convergence of apparently unrelated
institutions and techniques that together make up “a new micro­-physics
of power” that had for some time “constantly reached out to ever broad-
er domains, as if they tended to cover the entire social body” (139).12
Many questions can be raised about this general thesis, and many more
can be raised about specific features of Foucault’s argument.
In the first place, Foucault focuses his investigation of the history of
modern criminal justice on France, and we may wonder whether all that
applies to a state famous (or notorious) for its administrative centraliza-
tion would also apply to countries with stronger traditions of federalism
and local self­-government. Moreover, when he does turn to examine
prominent American penal reformers, Foucault reads them tenden-
tiously. When discussing Benjamin Rush, for instance, he stresses Rush’s
interest in a medical approach to prisoner rehabilitation (Rush had been
a physician) while neglecting his genuine piety and his concern for the
moral reform and spiritual salvation of prisoners.13 One could argue
more generally that Foucault’s analysis does not take sufficiently seri-
ously the great hopes of many penitentiary advocates that their new
prisons would have an unprecedented capacity to transform seeming-
ly hardened criminals into responsible citizens and good Christians.14
Foucault is strangely cynical about this aspiration of the penitentiary
movement, dismissing it as mostly empty rhetoric.
Like some other observers, Foucault finds that the expectations of the
early penal reformers for the penitentiary were excessive.15 As a rule,
penitentiary inmates have had high rates of recidivism throughout the
long history of this institution ­– a fact that has been and continues to be
a great embarrassment for the modern criminal justice system. What is
worse, few if any of the costly subsequent reforms intended to address
this basic problem have had any positive effect.16 The ideals of peni-
tentiary science could not be attained in the model prisons of Auburn
and Cherry Hill, and they appear to be as far out of reach today as ever.
Penitentiaries today hardly even deserve the name, since the primary
purpose they continue to serve is incapacitation and deterrence (and
even the effectiveness of the latter is challenged by some criminolo-
gists). Carceral punishment has thus largely failed to meet the goals
that were its explicit original justification. Yet despite this, it continues
126 Punishment and the History of Political Philosophy

to be the primary – indeed, all but exclusive ­– mode of punishment in

the Western world. Some now take for granted that the prison system
(one can hardly continue to call it a penitentiary system) is necessary,
believing that one can no longer imagine any other way of dealing with
convicted populations that in some Western countries numbers in the
hundreds of thousands, and even in the millions. Those who are not
convinced of the inevitability of mass carceral punishment, however,
remain at a loss to justify it. The continued existence of this morally
unaccountable institution thus looms as a massive indictment (perhaps
not fatal, but embarrassing nonetheless) of liberalism itself. For all the
questionable claims of Foucault’s account, it is difficult to deny this, its
main thrust, just as it is difficult to ignore the problem of carceral pun-
ishment once one becomes aware of its dimensions.

The Present State of Modern Criminal Justice

The greatest contribution of Discipline and Punish is its analysis of the

collapse of the modern theory of criminal justice at its very heart: its
attempt to justify punishment in light of a rational standard of natural
right. As we argued in earlier chapters, this collapse was always inevi-
table. Although it took some time (and several transformations in, and
iterations of, the theoretical framework) for this process to work itself
out, its conclusion could no longer be ignored or denied after the histor-
ical emergence and failure (from the point of view of its original inten-
tion) of mass carceral punishment. “In becoming a legal punishment,
it [i.e., carceral punishment] weighted the old juridico­-political ques-
tion of the right to punish with all the problems, all the agitations that
have surrounded the corrective technologies of the individual” (235).
The “problems” and “agitations” of which Foucault speaks are the
troubling signs of unregulated coercion in the midst of liberal society –
signs that threaten to undermine our feeling of individual freedom and
security.17 In spite of his very different philosophical orientation, in his
concern with this phenomenon Foucault may be said to continue in the
spirit of the Montesquieuan liberal­-republican tradition.
The most urgent question that remains concerns the present status and
future prospects of the embattled carceral penal system. For Foucault,
the outlook is bleak, since he regards that system both as indispensable
(at least within the limits of the prevailing political order) and as inca-
pable of being salvaged from his critique, given that for him, the insinu-
ation of modern disciplinary techniques into the carceral penal system is
 Foucault and the Crisis of Modern Criminal Justice 127

a consequence of the logic of liberal institutions. Foucault impugned the

intentions of the original penal reformers as tied up with the economic
interests of the bourgeois class. In his words, “humanity [was] the re-
spectable name given to … economy and to its meticulous calculations”
(92). Similarly, Foucault claims that the penal system as it now exists is
a means to perpetuate the political domination of that same class over
the working class, and the exploitation of the latter by the former. The
“peasant illegality” that developed after the French Revolution was in
fact, according to Foucault, a political protest against the new regime of
bourgeois property holders (273–4). But instead of acknowledging this
illegality as a political challenge, the new regime criminalized it as a “de-
linquency” (a term that, for Foucault, captures the sorts of offences typi-
cal of the poor classes: “petty theft, minor acts of violence, routine acts of
law­-breaking” – 278–9). This “legal” strategy of neutralizing the greatest
political threat to the new regime had to be perpetuated, and this was ac-
complished by means of the prison, according to Foucault. Here, the fail-
ure of the penitentiary technique to lower recidivism rates turned out to
be an unexpected political advantage to the ruling classes. “Delinquency”
initially seems ideally suited for penitentiary correction, and any failure
of such correction is likely to be attributed to extrinsic causes (imperfect
institutional design, maladministration, etc.). In fact, however, “the pris-
on, apparently ‘failing,’ does not miss its target; on the contrary, it reaches
it, in so far as it gives rise to” delinquency, which is “secretly useful” since
it is at once “refractory and docile” (276–7). In short, Foucault’s provoca-
tive claim is that the carceral penal system continues to exist – not despite
its failings, but because of them – in order to create a permanent illegal
underclass whose offences are tolerable and whose continued existence
serves the interests of the property­-owning majority.
The argument we have just summarized is nothing if not provocative,
albeit also speculative and, it must be admitted, somewhat theoretically
extravagant. Nevertheless, aside from calling further into question the
true motives of political majorities in liberal­-democratic countries, it
is in fact somewhat misleading, since, on the terms of Foucault’s own
analysis, it gives the ruling economic classes too much credit. That is
because, as we have already seen, Foucault’s basic claim is that the de-
velopment of modern disciplinary techniques is a historical process that
has always escaped human control.18 The ultimate cause of this histori-
cal development is not human agency but technology, understood as a
quasi­-autonomous force in human affairs. Thus when the penitentiary
came to dominate criminal justice, it did so in spite of the wishes of the
128 Punishment and the History of Political Philosophy

public and its representatives (whether those wishes were noble or, as
Foucault alleges, base). As Foucault himself puts it, “if the penitentia-
ry, in so far as it went well beyond mere detention, was able not only
to establish itself, but to entrap the whole of penal justice and to im-
prison the judges themselves, it was because it was able to introduce
criminal justice into relations of knowledge that have since become its
infinite labyrinth” (248–9). Foucault seems to be saying that the “rela-
tions of knowledge” into which criminal justice was thus fatefully intro-
duced were none other than the technical sciences of the psychological
and physiological mechanisms that underlie illegal behaviour, which
Foucault associates with penitentiary science.
But how does the discovery of a mechanistic science of criminal be-
haviour “entrap” the criminal justice system, and in such a way as to
elude human control? Foucault’s answer to this question is twofold. In
the first place, penitentiary technique is indispensable because of its
utility. The utility Foucault has in mind cannot be tied to the effective-
ness of penal correction, since according to him, the penitentiary not
only fails to correct but also creates delinquency. Rather, the utility of
the carceral penal system is evidently political in nature. For Foucault
this means, as we have seen, that the creation and maintenance of a
delinquent class blunts the political threat that the propertyless classes
would otherwise pose to the regime of the propertied majority. Even if
we doubt the truth of this particular imputation, however, it is hard not
to suspect that some of the public’s inertia regarding the carceral penal
system (despite its dissatisfaction with that system) has some political
dimension. Critics point out that modern criminal justice’s exclusive
reliance on mass incarceration could be replaced with some combina-
tion of incarceration and community­-based correction (especially for
less hardened offenders).19 Essentially, this would mean transferring a
substantial portion of the correctional and crime prevention functions
from government agencies to local communities, where they might take
the form of community reintegration programs, neighbourhood watch
groups, or even informal mechanisms such as neighbourly vigilance
and the sharing of parenting duties.20 This strong shift of responsibility
for crime prevention from governmental institutions and processes to
communities has been, and may remain, politically unpopular because
of its many inconveniences from the point of view of modern societies,
which are accustomed to extensive protections (supported by statutes
and by cultural consensus) of individual privacy and autonomy. All of
 Foucault and the Crisis of Modern Criminal Justice 129

us hesitate to mind our neighbour’s business, lest the next time our own
private doings become the focus of neighbourly scrutiny. The existence
of this laissez­-faire cultural consensus seems to favour mass incarceration,
for all its problems, over more communal methods of crime prevention.
This point would seem to receive support from our earlier discussion of
Plato’s analysis of the limits of criminal justice in the republican context.
According to Plato, what one might call informal community­-based
methods of crime prevention all require a vigilant, public­-spirited citi-
zenry whose moral outlook is informed by a demanding conception of
self­-denying virtue. Yet the qualities that make human beings fit for an
austere republican life are precisely the same ones that also make them
less fit for modern liberal institutions, which presuppose an overriding
concern for individual security, property, and freedom of expression.
For these deep­-seated reasons, community­-based methods of crime pre-
vention would be difficult to introduce into liberal society.21
The second part of Foucault’s answer to the question of the intrac-
tability of penitentiary technique has to do with what the discoveries
of the sciences of criminal behaviour imply for the “juridical” status of
crime and of the criminal. The individualizing knowledge of peniten-
tiary technique focuses less on the criminal deed and its circumstances
than on the criminal himself – on his psychology, social position, and
upbringing. This leads to “a curious substitution: from the hands of
justice, [the penitentiary system] certainly receives a convicted person
but what it must apply itself to is not, of course, the offence, nor even
exactly the offender, but a rather different object, one defined by vari-
ables which at the outset at least were not taken into account in the sen-
tence, for they were relevant only for a corrective technology” (251–2).
In fact, juridical determinations of an offender’s culpability (the grav-
ity of his offence, and so on) are not simply irrelevant to penitentiary
technique – they actually contradict it by the terms by which they are
understood. The psychological reductionism that is characteristic of
modern criminology leaves no room for stock legal or moral concepts
such as premeditation, responsibility, free will, and the conscience; nor
can it be in any way reconciled with these concepts (unless, that is, one
fully subscribes to Kantian compatibilism). In the face of this positiv-
ist scepticism, a retreat from the carceral penal system in the direction
of any non­-penitentiary approach to criminal justice would be difficult
to justify. Foucault calls this dilemma a “criminological labyrinth from
which we have certainly not yet emerged” (252 and context).
130 Punishment and the History of Political Philosophy

Conclusion

Foucault’s critique of modern criminal justice theory brings its vari-

ous tensions and contradictions to a crisis point. As we saw in earlier
chapters, the major theorists of that tradition were sufficiently aware
of its problems. Even when their attempted solutions were not all that
might have been desired, the earlier penal theorists made every effort to
“soften” the remaining difficulties through doctrinal or rhetorical means
(thus, Hobbesian “authorization” and Beccaria’s humanitarianism).
Foucault, by contrast, explodes all such stopgap devices. He directs his
attack especially against the equivocation of liberals about the true aim
of the modern criminal justice system by asking whether its fundamen-
tal aim is just and humane or whether it amounts to a project for extend-
ing the political and economic advantage of the propertied majority. The
defenders of modern criminal justice have always claimed that justice,
humanity, and the security of life and property always go together un-
der a regime of enlightened criminal laws; Foucault rejects this stance
on the basis of his understanding of the character of liberalism and of
bourgeois society.
The strongest piece of evidence that Foucault advances in support of
his thesis about modern criminal justice in crisis is his analysis of carcer-
al punishment. According to that analysis, penitentiary technique indi-
vidualizes punishment, replaces the juridical subject with the object of
disciplinary science, and reduces all punishment to a single mode (thus
destroying any semblance of proportionality). These consequences run
contrary to the explicit principles of modern penal theory, yet they were
made possible by that theory. Since the principles of the original penal
reformers are defied by the carceral penal system we have today in its
operation, we must somehow explain that system’s rapid emergence as
well as the near­-universal acceptance of its necessity. The explanation
Foucault offers is at times extravagant, yet some of its chief points can-
not be denied. In the first place, despite the public’s dissatisfaction with
mass incarceration, its political utility for contemporary liberal societies
makes it difficult to abandon for otherwise more attractive methods of
crime prevention. Furthermore, the discoveries of the behavioural sci-
ences, which are so integral to modern criminology, have helped under-
mine the credibility of traditional moral and juridical concepts, thereby
making the status quo appear unavoidable.
Discipline and Punish has its fair share of critics, and indeed, there is
much that deserves to be criticized. Yet once one gets beyond the general
 Foucault and the Crisis of Modern Criminal Justice 131

objections to Foucault’s opaque formulations, his sometimes unnec-

essary neologisms, and his occasionally tendentious use of historical
facts, one begins to see that his critics are not actually challenging his
critique of the principles of modern criminal justice – and it is not clear
whether they always fully understand that critique. (It bears repeating
that the critique of the theory of modern criminal justice should not be
confused with Foucault’s broader arguments and claims about modern
discipline.) Given its faults, Discipline and Punish has been easy to ig-
nore for contemporary penal theory and scholarship, especially in the
English­-speaking world. But perhaps the key reason why Foucault is
often too easily dismissed by criminologists and punishment theorists
is that he makes politics central to his analysis. Contemporary crimi-
nology and punishment theory have tended to treat their respective
subject matters as independent of politics, perhaps with the intention
of forestalling further controversy in an already controversial field. Yet
it has been one of the claims of this study that the political context is
essential for understanding the various problems of criminal justice.
Although no longer himself in the tradition of modern political phi-
losophy, Foucault remains our most direct link to that tradition’s teach-
ing on crime and punishment. Having done the work of connecting
Foucault to the development of Western rationalist penal thought, and
having traced the key moments in that development leading up to its
present crisis, this study has perhaps made possible a rethinking of the
most challenging criminal justice issues facing us, as citizens of liberal­
democratic societies, in ways that are more fully informed about what
is truly at stake.
 Conclusion

Punishment and Liberalism

The discipline of criminology currently understands itself as a value­-free

science, like economics and sociology, with which it shares many of its
methods. While examining the rise and fall of crime rates and their un-
derlying causes, modern criminology has thus far left questions of policy
aims to political theorists, and ultimately to legislators and democratic
majorities. Public debates, however, have been largely unsuccessful in
sorting through some of the difficult normative issues implicit in ques-
tions about the competing aims of punishment. In fact, it has become
routine for the public simply to look away. There is little reason to hope
there will be a spontaneous change in the lamentable public inaction re-
garding the fundamental problems of modern criminal justice. Political
theory for its part has so far ignored criminal justice. As political theo-
rist Albert Dzur recently wrote, “Anglo­-American political theory has
hardly registered mass incarceration and has done little to analyze any
incongruity with core democratic commitments.”1 The present study has
attempted to remedy this gap in the scholarship by uncovering the origi-
nal debates in Western political thought surrounding the principles of
punishment. This, however, is still only a beginning, one that clarifies
the terms of the debate but that by itself cannot give rise to real policy
decisions. An extensive empirical knowledge of the dimensions of crime
in a large society, which is the proper sphere of criminology, is indispens-
able to any serious attempt to formulate practical proposals for criminal
justice policy. Such expertise must inform the practical application of the
normative principles and must in turn be informed by those principles.
Criminologists cannot afford to keep neglecting the essentially politi-
cal task of weighing and adjudicating among competing aims of crimi-
nal justice. Criminologists often trace the origin of their discipline to
 Conclusion: Punishment and Liberalism 133

Beccaria’s On Crimes and Punishments, which they understand as hav-

ing established the empirical study of the causes of crime and the ef-
fects of punishment. But as we have seen, unlike later criminologists
Beccaria was no less concerned with the normative questions of pun-
ishment than he was with the empirical ones. In fact, all of the intellec-
tual forerunners of modern criminology we have discussed – Hobbes
and Montesquieu, as well as Beccaria – believed that the question of the
proper aim of punishment was a matter for serious dispute and debate,
having serious moral implications, and that this question could not be
resolved simply by a popular vote for one or another set of preferred
policy objectives. Criminologists, and not just political theorists, must
concern themselves with the ends of punishment.
The present study might help criminologists identify some of the out-
standing questions their discipline must begin to answer. First, up to
what point is the unprecedented security we enjoy worth the corrosive
effect that mass incarceration has on our sense of civic unity? Given
that security must rank quite high as a priority in a liberal political or-
der, how much must we moderate our noblest aspirations to apply the
institutions of punishment to the rehabilitation of criminals and the re-
constitution of civic community? Importantly, our investigations have
implied that some such tradeoff among the competing ends of criminal
justice must be accepted – since no perfect harmony of ends is possible
in this sphere – and that it is better to accept a reflective compromise of
our aspirations than to delude ourselves either with a policy of utility
that poses as morality or with a moral formalism that remains deaf to
political necessities (both physical and psychological). Second, crimi-
nology must be able to tell us what kinds of offences can be criminalized
or decriminalized in a liberal political order and to what extent, given
that liberal societies are naturally quite tolerant of what Montesquieu
called crimes against mores and public tranquillity, but only up to a point.
Can recreational drug use be decriminalized without seriously under-
mining some minimum level of public decency and self­-respect? Can
punishment of certain severe crimes – such as rape and murder – be
carried out without any accommodation of public sentiments of moral
indignation and retribution? To be sure, these are difficult questions, the
answers to which would depend on extensive knowledge of many com-
plicated dimensions of society, including demography, the family, and
what sociologists call “social capital.” But if such answers are to have
any value, they must come from criminologists who have thoroughly
informed themselves about the normative aspects of these questions.
134 Punishment and the History of Political Philosophy

Specific questions about mass incarceration (with various moral and

political implications) would also need to be addressed. The many griev-
ances typically levelled against mass incarceration include the follow-
ing: that it is ineffective in the performance of its corrective function, as
evidenced by high rates of recidivism; that it is ineffective in the perfor-
mance of its deterrence function, due to low or unpredictable conviction
rates; that it results in inhumane conditions for inmates, especially over-
crowding and rampant violence; that the courts rely too heavily on prison
sentences, as opposed to community­-based corrections; and that prison
sentences are excessively punitive (a charge made by critics on the left)
or excessively short or arbitrary (a consequence of indeterminate sen-
tencing and early release laws, according to critics on the right).2 Certain
other charges commonly levelled against the modern penal system – for
example, that it discriminates by race in sentencing, and that it crimi-
nalizes poverty and mental illness – are strictly speaking not criticisms
of mass incarceration per se; nevertheless, it is sometimes suggested or
implied that incarceration somehow exacerbates these injustices. Some
of these criticisms presuppose that crime prevention is the sole proper
aim of punishment and demand the reform of carceral punishment with
a view to enhancing its effectiveness; other criticisms on this list presup-
pose other ends, such as humanitarian benevolence, equality, rule of law,
justice, or civic solidarity. To address these criticisms squarely, it will be
important for criminologists to make all such presuppositions explicit.
Because of the empirical orientation of contemporary criminology,
whenever current debates do touch on normative issues they have a
strikingly disoriented character. One prominent debate may serve as
an illustration. Over the past two generations, the discipline of crimi-
nology has been preoccupied with questions surrounding an unprec-
edented crime wave in many Western countries. In the 1960s and 1970s
alone, the rates of homicide in Canada and the United States more than
doubled (according to Statistics Canada and the FBI’s Uniform Crime
Reports), while property offence rates approximately tripled in both
countries. This sharp increase in crime drew much public attention,
which paved the way for batteries of studies of many aspects of crime
and criminal justice policy. At that time, various explanations for the
sharp increase in crime were offered.3 The most common and least con-
troversial of these focused on the sudden increase in the numbers of
young people following the postwar baby boom. Since young males are
overwhelmingly the group most at risk of falling into crime, the sud-
den influx of young men in the postwar decades was in itself a highly
 Conclusion: Punishment and Liberalism 135

significant factor in the rise of crime rates, according to this explanation.

The discovery of this age effect did not by itself satisfy criminologists,
however, since it did not provide any insights as to what might be done
to curb soaring crime (short of getting rid of young men). Furthermore,
as a theoretical explanation of crime, it was obviously incomplete, since
youth by itself cannot explain criminality. It is true that a large pro-
portion of “career” criminals (those with high offence rates and long
criminal careers) start their careers as young men, but overall, most
young men never commit a serious offence in their lives. So the ques-
tion of what could lead certain young men (and individuals from other
groups) to take the fatal step of committing a serious crime remained.
From the criminologists’ point of view, there was yet another reason
to be dissatisfied with the age­-effect explanation of the crime boom.
They noted that while this explanation was consistent with similar
spikes in crime rates in many other Western countries, it could not ac-
count for the relatively low crime rates during the same postwar period
in some non­-Western countries, such as Japan. If culturally conservative
nations like Japan could keep crime under control when the age of their
populations fluctuated, could it be that Western culture, or certain char-
acteristic institutions of Western societies, were somehow responsible
for the rising crime rates?
Numerous attempts were made to come up with “structural” explana-
tions to supplement the age­-effect theory of crime. Among conservatives,
the eminent criminologist James Q. Wilson pointed to the “countercul-
ture” movements of the 1960s and 1970s, which attacked “middle­-class
values” and promoted self­-expression, free love, and recreational drug
use.4 If young people – Wilson’s argument ran – had embraced the stan-
dards of responsibility and industry inherited from their parents, they
would have been less tempted to turn to crime; but flouting these stan-
dards became acceptable and even admirable among Western youth,
with the result that internal controls on illicit behaviour were weakened.5
Yet Wilson proposed this explanation only as a plausible account, not
as a demonstrable scientific theory; indeed, he took other criminologists
to task for seeking “fundamental causes” of criminal behaviour, which
Wilson himself regarded as impossible to find. He advocated, instead,
taking a “citizen’s” perspective when analysing criminal justice policy.
Such a perspective would be sceptical of extravagant or radical theo-
retical explanations and would favour commonsense assumptions about
individual agency and motivation. (Wilson evidently believed that his
own account, which emphasized cultural factors in the crime boom of
136 Punishment and the History of Political Philosophy

the 1960s and 1970s, would be compelling to a commonsense, civic per-

spective.) Notwithstanding this scepticism towards fundamental causal
explanations, Wilson later proposed a kind of fundamental theory of
criminality in Crime and Human Nature (which he co­-authored with psy-
chologist Richard Herrnstein). In this landmark work, he presented an
extensive review of criminological and psychological studies and con-
cluded that individual biology (especially factors such as body type and
a genetic predisposition to psychopathology) contributed just as much to
criminal behaviour as did an individual’s social environment. On these
grounds, Crime and Human Nature favoured more traditional methods of
crime prevention, arguing in support of deterrents such as incarceration
and stronger policing. Wilson was particularly influential in the United
States, where he popularized his “Broken Windows” theory of crime
control, with its emphasis on stricter policing of minor offences – “order
maintenance” – as an indirect means to prevent more serious crimes.6
He was also a strong advocate for the expansion of the prison system
that has been under way since the 1980s – an expansion that, according
to some, has substantially lowered crime rates (although this remains a
hotly debated issue among criminologists).
Wilson became a controversial figure because of his efforts to defend
the traditional approach to crime. Some at the time viewed that approach
as having failed to prevent the crime boom of the 1960s and 1970s; others
contended that it exaggerated the crime boom. A very different expla-
nation for violence and criminality than Wilson’s was championed by
liberals and progressives. According to President Lyndon Johnson’s at-
torney general, Ramsey Clark, the traditional approach to crime, which
for the causes of criminality in the individual, was misguided, and the
true causes were to be found in deeper, systemic social ills. Among those
systemic ills were Americans’ individualism, their national predisposi-
tion to violence, and their selfish acquisitiveness. High crime rates could
never, according to Clark, be reduced sufficiently by getting “tough on
crime”; this could only happen through sweeping social, economic, and
educational reforms. According to Clark’s Crime in America, only relief
from poverty and mutual fear could prevent young men living in urban
slums from falling into a life of crime, and this could only be accom-
plished if reason and science were brought to bear on the “old instincts”
of the nation as a whole so that the “reflex to violence” and the “acquisi-
tive instinct” (responsible for the injustice of steep economic inequality)
were “conditioned out of the American character.” Clark’s call to ad-
dress the systemic causes of crime was meant to appeal to Americans’
 Conclusion: Punishment and Liberalism 137

interest in their own security, but at the same time it clearly and elo-
quently spoke to his audience’s nobler sensibilities. Clark was implying
that the modern criminal justice system’s methods were failing because
they were both ineffective and unjust. But in spite of its high moral tone,
the book contained not even the rudiments of a theory of what justice
demands (or any reference to any such existing theory). Thus Clark’s
vague assumptions that justice, security, social equality, and political
solidarity had always gone together could not be considered more than
mere speculations. Clark’s book paid almost no attention to the higher­
order political questions of criminal justice that it implicitly raised; and
it remained entirely oblivious to the tradition of Western penal thought,
which had for a long time wrestled with those very questions. Without
an adequate understanding of those questions, it is impossible to grasp
the political status of security or its relation to other political priorities
in a liberal­-democratic order. Thus, Clark’s call for wholesale criminal
justice reform seemed at once confused, utopian, and partisan.7
Although Wilson found himself on the opposite side of the debate
over the appropriate response to rising crime rates, his work did not
quite avoid the problems that affected his opponents. Whatever the
merit of his preferred policies from the narrow point of view of effective
crime control, his arguments for those policies rarely if ever addressed
his opponents’ more radical contention that the traditional approach to
criminal justice was fundamentally unjust. Moreover, Wilson did not
realize that his appeal to the civic perspective (which he made partly in
order to avoid the hard systemic questions and to counter the sociologi-
cal theories of crime of some of his colleagues) required an answer to the
basic question of the justice of the traditional approach to crime, since
as citizens we demand fairness, proportionality, and humanity from our
penal institutions no less than we demand security. The criminological
studies of recent decades are of no help here; indeed, as Wilson himself
argued in Thinking About Crime, we cannot reasonably hope that such
studies will ever establish the “fundamental causes” of crime,8 which
means that as citizens, we ought not to abandon our collective respon-
sibility of thinking about the basic questions of criminal justice on the
pretext that such questions might safely be left to value­-free positivist
science. As Thacher has recently put it, “strong causal arguments [in
criminology] rarely succeed to the degree needed,” since “causal analy-
sis that aims to identify large indirect effects produced through com-
plex causal chains is unlikely to offer clear­-cut advice for policy.”9 We
have no choice but to use our theoretically informed political judgment
138 Punishment and the History of Political Philosophy

to decide whether to continue to rely heavily on traditional criminal jus-

tice approaches, or whether to begin to emphasize welfare spending,
medical treatment programs, community correction, and various other
methods as means of crime control.
Unfortunately, the current theoretical debates about punishment,
which tend to be conducted in legal and ethics journals, are drawn large-
ly from the sorts of political arguments that had originally been used to
found modern criminal justice theory. When present­-day liberals draw
on the tradition of modern political theory, they limit themselves to two
attenuated strands of that tradition: a utilitarianism handed down from
Beccaria and Bentham on the one hand, and a neo­-Kantianism favoured
by leading liberals like John Rawls, H.L.A. Hart, and Joel Feinberg on the
other.10 As we noted in our introduction, in recent decades neo­-Kantian
theories of punishment have won prominence, now that utilitarian pun-
ishment theories have fallen out of favour due to what has generally
been regarded as their moral reductionism. Liberal thinkers who have
helped redefine the recent punishment debate tend to agree with Kant
that there is a real need to restore to punishment its retributive dimen-
sion in some form, in order to account for our respect for the dignity
and rights of individuals in the face of the frequently overwhelming
popular pressure to advance the greatest happiness of the greatest num-
ber. These liberal punishment theorists have tried to argue that retribu-
tive punishment is fully compatible with liberalism and that there is
no real tension between the backward­-looking concern with desert that
characterizes retribution and liberals’ forward­-looking commitment to
external liberty and physical security.11
Yet a recent friendly critic of liberalism, Stanley Brubaker, has argued
to the contrary – that contemporary liberals cannot justify retributive
punishment on their principles and that they do not even fully under-
stand what punishment is.12 Like Kant, Brubaker believes that propor-
tionality is central to what most people think makes punishment just and
that ordinary, practical knowledge of the appropriateness of retribution
is as common in liberal societies as it is in any healthy political com-
munity. As one explanation of the preponderance in political life of the
retributive outlook, Brubaker suggests that retributive punishment per-
forms an important expressive function: through it, a political commu-
nity emphatically affirms its shared conception of the human good and
the rank­-ordering of virtues and vices by rewarding men it admires and
punishing those whose actions it condemns as immoral. Contemporary
liberals, by contrast, cannot justify retributive punishment because of
 Conclusion: Punishment and Liberalism 139

their basic theoretical commitments, according to Brubaker’s account.

Most prominent strands of contemporary liberalism (Brubaker cites
Lockean, neo­-Kantian, and utilitarian versions in particular) insist on
maintaining a strict neutrality towards the various conflicting views of
the human good, as a result of their scepticism regarding that good.
Constrained by this neutrality, liberals can impose “penalties” (which
Brubaker distinguishes from “punishment”), which are intended to de-
ter crime and to compensate victims for their losses, since this can be
accomplished without requiring individuals to obey the law for what
some might hold to be the morally right reasons. But liberalism’s neu-
trality towards the good denies the right to punish in order to express
moral condemnation of criminal acts, since doing so would presuppose
some standard of the human good.13 Thus, according to Brubaker, while
liberals may deter, compensate, and rehabilitate, they cannot punish in
the sense that most resonates with ordinary moral experience.
On Brubaker’s account, when contemporary liberals try to make room
for retribution in their penal schemes, they do so either inconsistently
or by way of emptying the notion of retribution of its essential mean-
ing. When societies punish, they surely do so for the sake of deterrence,
but they also punish for the sake of retribution because they see it as a
legitimate and independent end of punishment. On this basis, Brubaker
criticizes Hart and Rawls, who deny that we ever punish because we
want the guilty to “pay” for their actions. They claim that we only pun-
ish for the sake of deterrence and that when we invoke the notion of
just deserts we only mean to set an upper limit, or side­-constraint, on
public coercion in order to forestall punishment of the innocent and dis-
proportionate punishment of the guilty. While such a limiting principle
could well be desirable on humanitarian grounds, it would not, accord-
ing to Brubaker, entirely satisfy our concern for retribution, since, con-
trary to Hart and Rawls, we do in fact tend to believe that punishing the
guilty is an end in itself.14 Since liberal societies do punish retributively
(in Brubaker’s sense), liberalism’s failure to justify retribution creates a
troubling dissonance between liberal theory and practice.15
This failure of neo­-Kantian liberalism to make sense of the enduring
appeal of retribution, or even to take seriously the suggestion that ret-
ribution may constitute a proper aim of punishment, is, I would argue,
a reflection of its failure to think through the questions and problems
of punishment in light of the tradition of political philosophy. Thus
contemporary liberals would do well to give more serious thought to
the classic liberal approach to retribution found in Montesquieu. Like
140 Punishment and the History of Political Philosophy

contemporary liberals, Montesquieu is sceptical of retribution, but be-

cause he eschews the formalism of Hobbes, he is more sensitive to the
fact that even thoroughly liberal societies partly depend on institutions
that do not neatly harmonize with liberalism, such as the family and re-
publican virtue (or an attenuated form thereof). Montesquieu helps us
see how these institutions themselves contain the origins of the concern
for retributive justice and how a moderated retributive element might
(and must) be accommodated by a liberal criminal law. Nevertheless,
for the deeper meaning of the concern for retributive justice, contem-
porary liberals would have to turn not to Montesquieu, or Hobbes, or
even Kant, but to Plato. Our discussion of Plato’s Laws showed that the
appeal of retribution could be explained as originating most powerful-
ly in the moral demands of republican political life. In the tradition of
Western political philosophy, classical republicanism was understood
to be one of several fundamental alternative ways of organizing poli-
tics – and indeed human life as a whole – that would always remain a
profoundly appealing possibility. The way of life represented by classi-
cal republicanism – its commitment to the public good and at the same
time to individual human excellence – must therefore be reckoned with,
even by its opponents.16 Plato himself was in some ways critical of clas-
sical republican life and of the intellectually confused punitive attitude
to which it was prone, yet he found the latter to be inseparable from the
former when he considered the nature and limits of political life.
This study has admittedly posed more challenges than it has offered
concrete solutions, and some of these challenges are daunting. Continuing
to ignore these challenges, however, is not an option, while facing them
squarely promises great rewards in the form of a richer, more engaged
civic life in which punishment is not an injustice to the offender and an
embarrassment to society, but a reconciliation of the two.
 Notes

Introduction

1 Braithwaite, Restorative Justice, 11 ff., 45, 74, 100–3, 129; see also Hahn, Emerg­
ing Criminal Justice; Christie, “Conflicts as Property”; Schneider, Refocusing
Crime Prevention, ch. 1; Sutton et al., Crime Prevention.
2 Thus restorative justice advocates propose to return some of the control over
the administration of criminal justice to local communities by introducing
“community conferences” into the criminal justice process. Such conferences
would deter and correct at the local level through “reintegrative shaming”
rather than through traditional punishments, thereby healing frayed social
bonds and reviving a sense of community. Advocates of restorative justice
claim to embrace traditional goals of incarceration – namely, rehabilitation,
deterrence, and victim satisfaction – while rejecting it as a means to that
end. See Braithwaite, Restorative Justice, 80–8, 129–32. See also 95 ff.
3 Furman v. Georgia, 408 U.S. 239.
4 Ibid., 270.
5 Ibid., 286.
6 Ibid., 296.
7 Gregg v. Georgia, 428 U.S. 183–7 (1976).
8 As Emile Durkheim shows, a purely utilitarian understanding of wrong-
doing and punishment can be rejected even on purely empirical grounds.
See Durkheim, Sociology and Philosophy, 40–9.
9 See Morris, “Persons and Punishment”; C.S. Lewis, “The Humanitarian
Theory of Punishment”; Murphy, “Marxism and Retribution,” 242–3;
Feinberg, Doing and Deserving, ch. 5. The position against which these crit-
ics are reacting is defended by Smart in Part I of his essay in Smart and
Williams, Utilitarianism. See also Benn, “An Approach,” 331–4.
10 See Dagger, “Playing Fair”; Morris, “Persons and Punishment.”
142 Notes to pages 10–16

11 Feinberg in Doing and Deserving, ch. 5, provides a clear account of the

“­expressivist” conception of retribution. See also Primoratz, “Punishment
as Language.”
12 For a review of the objections made against the various types of retributiv-
ism and the rebuttals to those objections, see Boonin, The Problem of
Punishment, ch. 3.
13 See Rawls, “Two Concepts of Rules,” and Hart, “Prolegomenon.” The differ-
ence between the two is that while Rawls tries to defend the principle of retri-
bution as a rule within a rule­-utilitarian framework, Hart defends retribution
as independently grounded in the principle of fairness. Later, though, Rawls
adopts the principle of justice as fairness and explicitly follows Hart’s view
of punishment. See Rawls, A Theory of Justice, 211–12 and 276–7.
14 Rawls, “Two Concepts of Rules,” 5–6.
15 Hart, “Prolegomenon,” 1–4, 10–11. Hart calls this political actors’ “over­-
simplifying tendency.”
16 See Corlett, “Making Sense of Retributivism,” 81–3. See also Boonin, The
Problem of Punishment, 63–77, for a critique of Rawls’s rule­-utilitarian ver-
sion of the argument.
17 Moore, Placing Blame, chs. 3–4; Corlett, “Making Sense of Retributivism”;
Brubaker, “Can Liberals Punish?,” 822–5 and 827–31.
18 Brubaker, “Can Liberals Punish?,” 829–30. But see also Brubaker and
Terchek, “Punishing Liberals?,” 1315–16. On the distinctions we common-
ly draw between punishment and other kinds of coercive measures, see
Feinberg, Doing and Deserving, ch. 5.
19 With the exception of Koritansky’s The Philosophy of Punishment, I know
of no scholarship that makes its theme the connection between criminal
justice theory and political philosophy. The present book goes beyond
Koritansky’s (a collected volume) in that it not only links penological
ideas to political principles in the works of individual thinkers (as the
essays in Koritansky’s volume do) but also attempts to map the historical
transformations of criminal justice theory onto the parallel history of polit-
ical philosophy in the West.
20 See Mackenzie, Plato on Punishment; Saunders, Plato’s Penal Code; L.
Pangle, “Moral and Criminal Responsibility.”
21 Important exceptions are Cattaneo, “Hobbes’s Theory of Punishment”;
Norrie, “Thomas Hobbes”; Schrock, “The Right to Punish.”

Chapter 1

1 Braithwaite and Pettit, Not Just Deserts: A Republican Theory of Criminal Justice.
See also the Introduction to this volume and the sources cited there in note 1.
 Notes to pages 16–22 143

2
See Braithwaite, Restorative Justice, 17–18, 73–8.
3
Braithwaite and Pettit, Not Just Deserts, ch. 6, esp. 88–92.
4
Braithwaite and Pettit, Not Just Deserts, 54, 59–60.
5
This, of course, is not to say one way or the other whether such a demand
on the part of republican citizens would be right.
6 Thus, in my view, the relation between the two dialogues is complementa-
ry, and their respective teachings cohere to comprise a greater whole (as
argued by Bruell in “On Plato’s Political Philosophy”). I am less inclined
than some scholars to regard the Laws as a “late” work in Plato’s intellec-
tual development that gives evidence of a break with his teacher, Socrates,
or of a great mind in decline (or both).
7 Adkins, in Merit and Responsibility, ch. 14, esp. 293–5, gives a somewhat
different explanation for why Plato’s thematic discussion of wrongdoing,
moral responsibility, and punishment belongs in the Laws rather than in
one of Plato’s “earlier” political works.
8 By contrast, philosophy asks “What is law?,” thus proclaiming its igno-
rance of its nature and power. Cf. beginning of the Minos.
9 At 702b–702e, Kleinias announces that he has been appointed by his na-
tive city of Knossos to serve on a committee of legislators to draw up a
comprehensive legal code for a new Greek colony, Magnesia, and it is on
his behalf that the Stranger elaborates his system of laws, as well as its un-
derlying principles. For the original Greek text of the Laws, I have relied
on the Oxford critical edition and consulted the Loeb and E.B. England
editions. Some of the shorter translations are my own, but for longer trans-
lations I have relied on T. Pangle.
10 The other two alternatives are much too radical to be implemented. The
first would require that the few “just” should somehow overpower the
many “unjust” (see 627b), while the second option (the enslavement of the
vicious to the virtuous) might be possible only in the city in speech – not
in deed – presented in the Republic, where philosophers rule autocratically
over willing subjects.
11 The ancient Greek word timˉoria (one of the common words for punishment)
is related to timˉe, which translates as “honour,” “status,” or “wealth” – all
of which may be understood as types of property. Hence timˉoria comes to
mean the restoration of timˉe to one who has lost it by taking it away from
the offender responsible for inflicting the loss. Cf. Saunders, Plato’s Penal
Code, 3–4, 20. See below, 49 (and n37), for a comment on Aristotle’s discus-
sion of reciprocity, a branch of corrective justice that is believed to be the
return of harm for harm, and that is modelled on commercial exchange.
12 The Stranger seems to speak with some truth here. Hesiod writes (Works
and Days, 270–2): “I would not myself be just, nor have my son be just
144 Notes to pages 23–4

among bad men: for it is bad to be an honest man where felons rule; I trust
wise Zeus to save me from this pass.”
13 There is some controversy over the correct interpretation of this ambigu-
ous passage. I am inclined to agree with England (in Plato, Laws of Plato,
vol. 2, 476–8), who reads timˉoria as referring only to the criminal’s acquisi-
tion of a bad character and bad associates, not to the destruction of the
criminal that the city may deliberately cause “in order that many others
may be saved.” Mackenzie’s reading of timˉoria as referring to both the cap-
ital punishment imposed by the city for purposes of deterrence and the
wretched condition of the soul of the criminal as a consequence of his as-
similation to bad men seems logical, insofar as both fail to make the crimi-
nal better. See Mackenzie, Plato on Punishment, 196n62. Yet the Stranger
may have reason for his reluctance to call deterrent punishment timˉoria.
The general preamble clearly intends to disparage retribution, on the
ground that it produces no positive effects to redeem the harm it does.
Deterrence, on the other hand, does benefit others, though it does nothing
good for the offender himself, and therefore from the point of view of po-
litical necessity can claim to be a certain kind of justice. Yet the Stranger re-
fuses to call deterrence a dikˉe in the strict sense, and so allows it to occupy
a kind of moral middle ground. This seems to be in keeping with the
Stranger’s civic or statesmanlike orientation, alluded to earlier. As we will
see later, the penal code elaborated in Book 9 contains a number of features
explicitly designed with a view to deterrence.
14 Lewis goes so far as to write that “[t]he nerve of the problem of punish-
ment in the Laws is the tension between reason and spiritedness.” Lewis,
“The Limits of Reform,” in Koritansky, The Philosophy of Punishment, 29.
The passion of spiritedness (thumos) plays a very important political role
in Plato’s dialogues. Its central character – the concern with one’s own –
lies at the heart of a number of distinctively human emotions and mental
states, some noble and some base, that include love of honour, civic and
familial loyalty, courage, shame, and indignation. Cf. T. Pangle, “The
Political Psychology,” 1062–4. Spiritedness is the political passion because
without it political life would not be possible: one cannot have a city with-
out loyal citizens, brave soldiers, and attentive parents. But despite, or
rather because of, the intense concern with one’s own that it fuels, spirit-
edness can too narrowly focus our view on what is our own – especially
our honour or dignity – so much so as to blind us to reason and to what is
truly good for us, for our friends, and for our city (cf. 731e–732b). The self­-
assertive tendency of spiritedness that causes us to demand the respect we
are owed and to lose sight of the good leads us to want to harm those who
 Notes to pages 24–6 145

have dishonoured us, even if this would not advance our true interests in
any way. From the spirited point of view, every unlawful or undeserved
injury is viewed as a loss of honour for the victim and as a “gain” of hon-
our for the offender. Spirited indignation is guided by the thought that it
can restore this loss of honour by inflicting an equivalent injury, and thus
dishonour, on the original offender. From the point of view of spirited in-
dignation, it is no matter that the future interests of the two parties are lost
sight of in the course of this retributive settling of accounts. By indicating
that the man declared by the city to be virtuous will be punishing out of
spirited anger, the prelude shows its awareness of the fact that its authori-
tative affirmation of the absolute benevolence of justice will not be able to
transform the essentially spirited character of the citizenry. It thereby ac-
knowledges the need to tolerate to some extent the retributive manifesta-
tion of spiritedness, even while it tries to temper this manifestation as
much as possible.
15 From 730d1 the Athenian never uses dike, but only kolasis (or a cognate)
to designate punishment (730d8, 731b7). This word can mean to punish,
chastise, or correct, and does not carry with it the sense of being neces­
sarily in accordance with right.
16 For an extensive discussion of the various ways in which the Stranger’s in-
stitutions in the Laws are designed to temper spirited anger, see L. Pangle,
“Moral and Criminal Responsibility in Plato’s Laws.”
17 Cf. Apology 25c­–26a, as well as Socrates’s long discussion of the “measur-
ing art” in Protagoras 351b–360d, esp. 358c–d.
18 Mackenzie, in Plato on Punishment, at 134–6, 197–8, overlooks this inconsis-
tency in the text and reads the general preamble as simply repeating the
Socratic refutations of voluntary wrongdoing that can be found in the
Protagoras and the Gorgias.
19 Regarding the greatest crimes, the Stranger proposes to show “in each sin-
gle case what retribution [timˉoriai] should be attached” (853a). This is a
noteworthy mention of retributive punishments (timˉoriai) as a subset of
judicial penalties (dikai). It is peculiar because, as we recall, the general
preamble had clearly denied that timˉoriai were dikai (on the ground that
justice, dike, always confers a benefit, whereas retribution only intends
harm). This blurring of the difference between timˉoria and dikˉe echoes the
equivocal status of retribution that we observed in our discussion of the
general preamble.
20 If a domestic servant or a resident alien is caught committing one of the
three mentioned capital crimes, he is to be whipped as much as the judges
should decide is appropriate and thrown naked beyond the borders of the
146 Notes to pages 28–33

country, but not killed, since it is possible that by undergoing punishment

he will become better, “for no judicial punishment that takes place accord-
ing to law aims at what is bad” (854d).
21 As Saunders’s Plato’s Penal Code, 139–40, suggests.
22 This is the first instance of the word “philosophy” or one of its cognates in
the whole work, which further marks the dialogue’s ascent from the more
mundane subjects of legislation to a theoretical plane. Cf. Strauss, The
Argument and Action, 128–9.
23 Examining the opinions of the celebrated Greek legislators is not quite the
same thing as examining what is best concerning legislation, which is what
the Stranger had actually proposed he and Kleinias do. The Stranger ap-
pears to assume that the famous legislators had left behind them written
preludes explaining the wisdom of their laws, a possibility he had ruled
out moments earlier when he asserted that human beings living under all
actual legislation live as slaves being doctored by slaves (cf. again 857c).
24 Cf. Strauss, The Argument and Action, 130.
25 This may explain the reason for the Stranger’s implausible argument that
actions are always suffered in the same manner as they are performed. In
proceeding this way, the Stranger makes his own position (i.e., punishment
is both just and noble) more awkward, and thus easier for Kleinias
to dispute. He thus makes Kleinias’s refusal to dispute it even more puz-
zling, inviting us to reflect on the reasons for that refusal. On these grounds
I disagree with Saunders’s interpretation of this passage as merely a prepa-
ratory “lesson in predication.” See Saunders, “The Socratic Paradoxes,” 422.
26 See L. Pangle, “Moral and Criminal Responsibility,” 459.
27 Plato’s Socrates’s knowledge consisted of his awareness of our ignorance
about the “noble and good” things (cf. Apology 21b­–22e). Through his
many conversations with men and women in all walks of life, Socrates
was able to learn that the greatest obstacle preventing most people from
gaining a greater awareness of their ignorance about the most important
things is not their indifference about these things, but rather the great
power they hold over our hearts, a power that often prevents us from sub-
jecting our conflicted opinions about these things to thorough examination
for fear that such an examination might lead to the disappointment of our
fondest hopes. Thus most men, according to Socrates’s view, remain deep-
ly conflicted throughout their lives about justice, the noble, and the good,
while only a few, like Socrates himself, are able to acknowledge to them-
selves the seriousness of their condition of ignorance, even if only for a
few brief moments in their lives (cf. Alcibiades I 116e–118b). See also Bruell,
“On Plato’s Political Philosophy,” 97–102. This view is echoed by the
 Notes to pages 34–9 147

Stranger in the passage we are presently considering when he attributes

the confusion about the identity of the just and the noble things not only
to themselves, as legislators, but to “the many” as well.
28 Cf. again 626b: “For … nothing is really beneficial, neither possessions nor
customs, unless one triumphs in war. For then all the good things of the
defeated belong to the victors.”
29 Aristotle, Nicomachean Ethics, 1132b21–1133a2.
30 For Socrates’s full reflections on this subject, see again the texts sited in
n17, above.
31 See Protagoras, 352c–357d; Gorgias, 466e ff. For a very thorough discussion
of the Socratic critique of retributive punishment in the context of the
Gorgias, see Stauffer The Unity of Plato’s Gorgias. See also the commentary
of Mackenzie, Plato on Punishment, ch. 9.
32 Other commentators would be inclined to classify the psychological state I
have described as just one of many kinds of ignorance, which would not be
functionally different from the other types – for example, a non­-vacillating,
erroneous conviction that injustice is superior to justice. Both these forms of
ignorance would look functionally the same, from their point of view, since
both would equally exculpate an offender of any guilt. There is a significant
difference, however, between the Socratic argument that all injustice is in-
voluntary, and the results of the Stranger’s examination of Kleinias’s opin-
ions about punishment. The former is a purely hypothetical deduction
– depending on several assumptions (that all men choose what is good for
them; that justice is the greatest good for us) – that does not necessarily re-
flect what human beings living in political society actually believe. On the
other hand, Plato’s depiction of the Stranger’s examination of Kleinias does
point to a plausible view of what most people actually believe. One of the
ways in which this is helpful is by making it possible to explain some of the
phenomena we are investigating: for example, why criminals can appear
to commit injustice deliberately, and then later repent of their actions.
33 Examples of such characters in Plato are Thrasymachus, Polus, and
Callicles. See Republic 336b; Gorgias 484a–b, 485b–c.
34 C. Zuckert, Plato’s Philosophers, 117, notes that restoring friendship is a dis-
tinct aim of punishment, but does not go far enough in drawing out the
implications of this.
35 Cf. Saunders, Plato’s Penal Code, 356. The traditional distinction between vol-
untary and involuntary crimes, which the Stranger is about to re­-establish,
may – to the extent that it is persuasive – make it possible in practice to give
a kind of publicly convincing justification for adding a desert­-based dimen-
sion to penalties.
148 Notes to pages 40–1

36 See also Shorey, What Plato Said, 342–3.

37 One might also consider the example of the Roman Coriolanus. The
Stranger’s view of the problematic connection between motivations and
deeds recalls the theme of the Theaetetus regarding the problematic con-
nection between our opinions about things and the things themselves
(i.e., the problem of knowledge).
38 See Shuchman, “Comments,” 36–7.
39 To draw distinctions based on any effects other than one’s own greatest
good would be extrinsic to what the Stranger means by voluntary action.
The Stranger seems to anticipate that such incidental considerations could
not affect what he means by the voluntary when he suggests (in perfect
Socratic fashion) that the voluntary in one sense cannot differ from the
voluntary in some other sense by virtue of being voluntary: “But in what
way are they two, if they don’t differ from one another by being … volun-
tary?” – 861d. Compare Socrates’s arguments in the Protagoras about the
“calculating art,” where considerations of pleasure and the noble are ex-
cluded as extrinsic to that calculation of one’s greatest advantage, which
saves our lives. See also the Hipparchus, as well as commentary on this
dialogue by Bloom in T. Pangle, The Roots of Political Philosophy, 32–52.
40 See O’Brien, “Plato and the ‘Good Conscience,’” 195n17; Mackenzie, Plato
on Punishment, 144, 201–2; Saunders, “The Socratic Paradoxes,” 422–4; but
cf. Strauss, The Argument and Action, 132 and context. Aristotle, too, rejects
on political grounds the Socratic thesis regarding voluntary injustice. See
Nicomachean Ethics 1099b12–25; 1114a31–b25. In his discussion of the legal
conception of “intention” as a requirement for criminal liability, Duff ob-
serves that “courts and commentators seem to agree” on “a central notion
of … intention or ‘purpose’; but they differ on whether that notion need be
defined at all; and on whether, if it needs defining, it should be defined in
terms of ‘desire’ (or ‘want’), or of ‘decision,’ or of acting ‘in order’ to bring
a result about”; yet such theoretical confusion apparently does not prevent
juries or judges from applying the concept of intentionality in practice.
See Duff, Intention, Agency and Criminal Liability, 27, 31–2.
41 The Stranger’s law itself makes it explicit that justice demands that the
worst criminals suffer in return exactly that harm which they wrongfully
inflicted: “If someone ever kills his father, he must endure to suffer vio-
lently this same thing at the hands of his children at some time; if he
should kill his mother, it is necessary that he himself be born partaking of
female nature, and having become such, must at a later time depart life at
the hands of the offspring. For there is no other purification for the shared
polluted blood, nor is the pollution willing to be washed away, until the
 Notes to pages 41–2 149

soul that perpetrated the deed pays for murder with murder, like for like,
and thus, by appeasing, lays to rest the spiritedness of the entire family”
(872e–873a).
42 See also L. Pangle, “Moral and Criminal Responsibility,” 467: “But where-
as most people see [voluntary murders] as deserving more harm, the
Athenian sees them as calling for more help. And although most people
view them as in every respect voluntary, the Athenian’s argument taken
as a whole suggests that crimes committed “voluntarily” are at the deep-
est level even more involuntary than harm inflicted accidentally, for in cor-
rupting one’s own soul one acts more against one’s true wishes than in
making a simple mistake.”
43 In this respect, the Stranger’s approach is similar to Aristotle’s. Like
Plato’s Stranger and Socrates, Aristotle defines justice as directed to some
good – or more precisely, the common advantage of the city – but he dif-
fers from Plato in defining the common advantage as consisting in mutual
exchange among citizens. Aristotle argues that for mutual exchange to
­occur, all the things exchanged must be somehow made commensurable.
Thus “in communities concerned with exchange, the just in this [i.e., cor-
rective] sense … holds them together … For either people seek to recipro-
cate harm for harm – if they do not, that is held to be slavish – or they seek
to reciprocate good for good. And if they do not do this, there is no mutual
exchange, and people stay together through mutual exchange.”
Nicomachean Ethics 1132b31–1133a2.
44 Mackenzie, Plato on Punishment, 173–5, and Saunders, Plato’s Penal Code,
145–7, do not take this important qualification into account, and thus take
the tripartite psychology presented by the Stranger unproblematically as
his own view.
45 Note that the Greek word hamartˉema is ambiguous, in that it could mean
either a (morally culpable) wrong, or a failing or error that occurs through
no fault of our own. The use of this ambiguous term (instead of the less
ambiguous adikia) forces us to recall the Stranger’s own understanding
of all injustice as involuntary, which has now been all but abandoned.
46 We might already wonder, however, whether the Stranger is right to as-
sume that the simple, self­-aware sort of ignorance will always be less dan-
gerous to society than the presumptuous sort of ignorance. Is it not possible
that a consistent sceptic – who knows that he does not know whether jus-
tice is or is not human virtue – could choose to publicly dissimulate his
knowledge of ignorance for the sake of his private ends? Later (in Book 10),
when the Stranger addresses the crime of heresy, he appears to come close
to the latter conclusion. There, he argues that the men who do not hesitate
150 Notes to pages 43–8

to speak their impious and unjust opinions publicly, and who openly ridi-
cule conventional piety and morality, would be less harmful to the city than
the “ironical” men who (perhaps because they are more consistent in their
opinions about justice) hide their true opinions and deceitfully adopt the
forms of conventional piety and become diviners, tyrants, demagogues,
and sophists (908a–e).
47 This seems consistent with the Stranger’s image of the human being as a
puppet of the gods, who is pulled in different directions by certain cords
(reason and the passions), but who can also “assist” the pull of certain cords
against others (644d–645b). See also Adkins, Merit and Responsibility, 302–3.
48 “If a man’s reason and desires are not in conflict, whatever his basic view of
life, he is to be termed dikaios [just], provided that his actions are based on
reason, not passion or desire.” Adkins, Merit and Responsibility, 307–8; ital-
ics in original.
49 For this reason, O’Brien, Saunders, and Mackenzie all deny Adkins’s thesis
and favour the interpretation of kan sphalletai ti (at 864a) as referring only to
errors regarding circumstances that lead to harm, not to fundamental errors
in moral understanding. See Saunders, “The Socratic Paradoxes,” 430–2;
Mackenzie, Plato on Punishment, 247–9; O’Brien, “Plato and the ‘Good
Conscience,’” 84–5.
50 This seems to imply that the ultimate and highest authority in the city will
not be the law itself (to say nothing of consent), but either reason or divine
revelation.
51 We must still wonder why the Stranger would ever state his “true” opin-
ion at all, and in such an unqualified manner that the tension between it
and the spurious public doctrine becomes palpable and troubling.
Consider, in this regard, Strauss, The Argument and Action, 128–9.
52 See Morrow, Plato’s Cretan City, 241–96; Saunders, Plato’s Penal Code,
chs. 7–14.
53 Adkins, Merit and Responsibility, has shown to what extent the traditional
(heroic and pre­-Socratic) notions of Greek morality – including pollution,
shame, and heroic virtue – are in tension with the Socratic or Platonic em-
phasis on moral responsibility. See chs. 5 and 8, with chs. 13–14.
54 For an excellent discussion of the various details of Magnesia’s penal code,
and their relation to the Stranger’s theoretical views, see Saunders, Plato’s
Penal Code.
55 Regarding some crucial difficulties with the Stranger’s refutation of the
heresies, see Strauss, The Argument and Action, 146–9.
56 Even the unrestrained heretics are punished less severely than they would
be in a more traditional society. Although they are ruled by unrestrained
 Notes to pages 48–55 151

passion and are said to be incurable, they are not executed to prevent
them from falling deeper into psychic corruption, as one might have ex-
pected from earlier discussions. Furthermore, the claim that they deserve
“neither one nor two deaths” is ambiguous. See Strauss, The Argument and
Action, 156.
57 Given the possibility that even after five years the perplexed heretic may
remain unreformed, the Stranger must legislate that those convicted of
heresy a second time must be punished with death (909a).
58 Saunders, Plato’s Penal Code, 168–78, provides an ingenious account (based
on passages in the Timaeus) of how conventional punishments might still
effectively serve the cause of moral reform. Briefly, he argues that, since the
underlying condition of an unjust character is an unhealthy arrangement of
physiological forces (whether due to habit or to strong passion), the penalty
creates a physical disturbance that “unsettles” the unhealthy arrangement,
making moral reform possible. In other words, this sort of reform seems to
follow the model of electroshock therapy. I remain unpersuaded of the ef-
fectiveness of such a process to produce genuine moral reform, as it seems
to me more likely that, after the disturbance, an unhealthy soul would lapse
back into its unhealthy habits than that it would spontaneously take on
healthy ones. It seems to me that the Stranger is right, that genuine moral
reform would require a much lighter touch, informed by extensive knowl-
edge of the individual case – like the sort of treatment that would take
place in Magnesia’s sˉophronisterion.
59 See, again, Adkins’s excellent work on the differences between classical
Greek and modern morality.

Chapter 2

1 See Machiavelli, The Prince, ch. 7; Discourses on Livy, III.1.

2 The fundamental importance of Hobbes for modern punishment theory
is acknowledged by Cattaneo (“Hobbes’s Theory of Punishment”) and
Norrie (“Thomas Hobbes and the Philosophy of Punishment”).
3 All references to the Leviathan (given as chapter and paragraph) are to the
Edwin Curley edition and its pagination.
4 See De Cive 1.2, first note.
5 One view, championed by Taylor and by Warrender, is that Hobbes’s politi-
cal teaching is entirely independent of his philosophical materialism and
that the former is in fact grounded upon a kind of proto­-Kantian conception
of duty. (See Taylor, “The Ethical Doctrine of Hobbes,” and Warrender, The
Political Philosophy of Hobbes.) Strauss (The Political Philosophy of Hobbes, chs. 2
152 Notes to pages 55–8

and 7) maintains that the moral attitude informing Hobbes’s political teach-
ing developed before his turn to modern natural science and that it (the
moral basis) consists in Hobbes’s view of the antithesis between the rational
passion of fear and the irrational passion of vanity. Hampton (Hobbes and
the Social Contract Tradition), Gauthier (The Logic of Leviathan, ch. 2), and
Macpherson (The Political Theory of Possessive Individualism, 70–5) argue
in various ways that Hobbes intended to ground his political teaching on
a moral basis, but failed. Goldsmith (Hobbes’s Science of Politics, 93–109)
and Watkins (Hobbes’s System of Ideas, ch. 5) have argued that Hobbes is a
straightforward materialist whose political science is a descriptive account
of the consequences of human psychology. My view is closer to that of
Strauss, and I briefly indicate the textual evidence for it in the discussion.
6 Utilitarian punishment theory has its origins not in Hobbes but in Cesare
Beccaria’s On Crimes and Punishments, which followed the Leviathan by
about a century. See chapter 4 of the present study.
7 Hence, Heyd (“Hobbes on Capital Punishment”), who views Hobbes’s po-
litical teaching as grounded in materialism, assumes that his argument for
a subject’s right to resist punishment rests on naturalistic premises, which
I will show it does not. Or, at any rate, Hobbes did not intend for it to rest
merely on naturalistic premises.
8 It is helpful in this regard to contrast Hobbes’s theory of natural right with
that of his contemporary Spinoza. Although the two may seem similar, only
the latter unambiguously developed a theory of right that was thoroughly
naturalistic. See Spinoza’s Theologico­-Political Treatise, ch. 16, beginning.
9 De Cive, Preface to the readers (18–19).
10 See also De Corpore (in Molesworth, The English Works of Thomas Hobbes,
vol. I) part I, ch. 6.7: “Civil and moral philosophy do not so adhere to one an-
other, but that they may be severed.” “For the causes of the motions of the
mind are known, not only by ratiocination, but also by the experience of
every man that takes the pains to observe those motions within himself.”
Italics in original.
11 Strauss, The Political Philosophy of Hobbes. See, again, the chapters cited in
note 5 above.
12 See also De Cive 1.7: “Amid so many dangers therefore from men’s natural
cupidity, that threaten every man every day, we cannot be blamed for look-
ing out for ourselves” (my emphasis).
13 De Cive 1.4.
14 The issue of the status of the laws of nature in Hobbes’s works has been
discussed at length by other commentators. I follow Johnston (The Rhetoric
of “Leviathan”) in reading Hobbes’s occasionally contradictory statements
 Notes to pages 59–64 153

on this subject as rhetorically useful for advancing the moral transforma-

tions Hobbes wished to bring about. Consider, for example, the marked
difference in tone between chs. 13 and 15 of the Leviathan. Whereas in the
former the laws of nature are called “convenient articles of peace,” in the
latter they are connected to virtue and the conscience (see especially 15.36–
40). Similarly, in ch. 17 Hobbes goes so far as to write concerning the fun-
damental covenant by individuals to constitute the sovereign authority
that “it is more than consent, or concord; it is a real unity of them all”
(17.13). Although it is possible to give such statements too much weight –
as both Taylor and Warrender do – it is important not to dismiss them
simply as failed arguments.
15 Contrast with Aristotle, Nicomachean Ethics 1138a6: “Whatever the law
does not command, it forbids.”
16 Recall that the Stranger’s thesis about the involuntariness of all injustice
was in some tension with the traditionalist penal legislation he enacts for
the republic of Magnesia.
17 Consider, for example, Laws 853c–d.
18 This has indeed been the trend historically in the development of modern
criminal law.
19 That publicity is an important desideratum of democratic governance has
been argued – for example, by Habermas, in The Structural Transformation
of the Public Sphere, ch. 25.
20 In this, Hobbes may be said to anticipate later criminological theories that
locate the causes of crime in the institutional structures of the surrounding
society.
21 One might still question whether Hobbes has underestimated the impul-
siveness of individuals who are prone to crime (such as drug addicts and
psychotics) and who therefore cannot help giving in to immediate self­-
gratification, even while they fear severe, but remote, punishment. Perhaps
Hobbes had not anticipated the truly dysfunctional social conditions in
some of the Western world’s most crime­-ridden cities.
22 Compare with Hobbes’s own individualistic conception of the family
in Leviathan ch. 20 and De Cive ch. 9, and see Hampton’s commentary
in Hobbes and the Social Contract Tradition, 9–11.
23 It is an interesting sign of Hobbes’s rigorous thinking, or perhaps of his
egalitarianism, that he even suggests that robbery of a poor man might be
considered a greater crime than robbery of a rich one, since “it is to the
poor [one] a more sensible damage.”
24 Thus, if we understand the rightful end of government as Hobbes does –
as the preservation of peace – then we cannot accept the argument
154 Notes to pages 65–8

commonly advanced by retributivists like Morris (or quasi­-retributivists

like Rawls) that while political society might come into being for the sake
of mutual benefit and defence, the laws and institutions created to enact
this aim can have an independent justification of their own, such as fair-
ness. See Rawls, “Two Concepts of Rules,” 4–8; Morris, “Persons and
Punishment,” 477–9. If the right to self­-preservation is the foundation and
source of all other rights and obligations, then any duty of fairness would
have to be understood in just these terms. Hobbes shows with perfect clar-
ity that retributive punishment cannot be understood to be just in a society
founded on a pact of mutual defence and benefit. To attempt to establish
retribution on contractarian grounds is already to move in the direction
of Kant (see chapter 4).
25 Hence Heyd, “Hobbes on Capital Punishment,” 122, is clearly mistaken
when he claims that the Hobbesian right to punish is founded on the so-
cial contract.
26 Hobbes was therefore the founder of yet another truism for criminologists:
that the effectiveness of punishment depends on its certainty just as much
as, if not more than, on its severity (see 27.18).
27 Cf. Thomas Aquinas, Summa Theologiae II–II, Q. 64, A. 6 (in Baumgarth et al.,
On Law, Morality, and Politics), where it is maintained that one ought to dis-
obey one’s sovereign when he commands the killing of an innocent. See also
the Nicomachean Ethics 1110a26–8, where Aristotle suggests (though he does
not absolutely affirm) that it might be better to die rather than to be com-
pelled to do very shameful things; Plato, Laws 869b9–c7, has the Athenian
stranger legislate that one may not kill a parent, even in self­-defence.
28 Jeremy Bentham indicates his awareness of this tension between utilitarian
punishment on the one hand and the proportionality or “fittingness” of
penalties according to the ordinary moral understanding on the other.
“Authors of celebrity,” Bentham writes, “have … said that the greatness of
temptation is a reason for lessening the punishment; because it lessens the
fault; because the more powerful the seduction, the less reason is there for
concluding that the offender is depraved. Those, therefore, who are over-
come, in this case, naturally inspire us with commiseration. This may all
be very true, and yet afford no reason for departing from the rule. That it may
prove effectual, the punishment must be more dreaded than the profit of
the crime desired. Besides, an inefficacious punishment is doubly mischie-
vous; – mischievous to the public, since it permits the crime to be commit-
ted; – mischievous to the delinquent, since the punishment inflicted upon
him is just so much misery in waste. What should we say to the surgeon,
who, that he might save his patient a small degree of pain, should only
 Notes to pages 70–1 155

half cure him? What should we think of his humanity, if he should add to
his disease the torment of a useless operation? It is therefore desirable that
punishment should correspond to every degree of temptation” (my em-
phasis; original paragraph divisions have been omitted for the sake of sim-
plicity). Bentham, The Rationale of Punishment, ch. 6 (70; cf. 74, “Rule IX”).
29 Pitkin, “Hobbes’s Concept of Representation,” I and II; Orwin, “On the
Sovereign Authorization”; Tuck, Natural Rights Theories, ch. 6; Hampton,
Hobbes and the Social Contract Tradition, chs. 5 and 8; Martinich, Hobbes, ch. 4.
30 Hobbes conceived of the natural right to defend one’s life as quite expan-
sive. We can no more alienate our right to resist wounds or imprisonment
than we can alienate our right to self­-defence, “because a man cannot tell,
when he seeth men proceed against him by violence, whether they intend
his death or not,” and since “the motive and end for which this renouncing
and transferring is introduced, is nothing else but the security of a man’s
person, in his life and in the means of so preserving life as not to be weary of
it” (14.8). This is as much as to say that the criminal who is sentenced to any
substantial corporal, carceral, or in some cases even pecuniary punishment
immediately regains his original natural right to all things and so returns to
the state of nature vis­-à­-vis his former sovereign and fellow­-subjects.
In the earlier Elements of Law (see I.19.10 and II.20.7), Hobbes had main-
tained that each subject does renounce his right to defend himself against
his sovereign; this allowed Hobbes to avoid the conflict between natural
right and the right to punish. But, in a different way, that strategy was even
more damaging to Hobbes’s overall purpose, since it compromised the
privileged moral status of self­-preservation. Hobbes was therefore forced
to revise his view in the later De Cive and the Leviathan. With this expansive
definition of the right to self­-preservation, Hobbes anticipates Locke.
31 I therefore believe that Schrock (“The Right to Punish”) and Norrie
(“Thomas Hobbes”) go too far in concluding that this clash of rights under-
mines Hobbes’s whole theoretical framework. In Norrie’s words, it “leads
to the immanent collapse of, and implicit denial of the possibility of, the
social state and the institution of punishment” (307). In my view, Gauthier
and Martinich are more correct to argue that the clash of rights (of the
criminal and of civil society, respectively) implied by punishment does not
create an inconsistency in Hobbes’s political theory, although it does reveal
the limits of the social contract. See Gauthier, The Logic of Leviathan, 146–9;
Martinich, Hobbes, 115–18. Hampton, Hobbes, 197–207, argues that although
this does not cause Hobbes’s theory to fail theoretically, it does make it im-
practicable or utopian: “Hobbes’s argument does not fail because he cannot
establish the rationality of creating an absolute sovereign, nonetheless it
156 Notes to pages 71–2

fails because he cannot establish, given his psychology, that men and wom-
en are able to do what is required to create a ruler satisfying his definition
of an absolute sovereign” (197; emphasis in original).
32 Johnston, The Rhetoric, 80–2, points out that the Hobbesian sovereign’s au-
thority is “curiously negative” in that it is established only by the subjects
divesting themselves of their natural rights. It would be “difficult to con-
vince oneself,” Johnston continues, that this “rather hollow conception of
power is really adequate to do the work Hobbes’s theory requires it to
do.” Accordingly, “Hobbes’s increased sensitivity to the importance of
public opinion as an element of sovereign power had led him to be more
concerned than ever before that the rights of sovereignty should be recog-
nized, in a widespread and public manner, as legitimate. The concept of
authorization helped meet this concern.”
33 Thus Hampton (Hobbes, 206) draws the conclusion that “Hobbesian people
empower a ruler by obeying his punishment commands, and they do so
whenever they decide such obedience is conducive to their best interests”
(emphasis in original).
34 See also Schrock, “The Right to Punish,” 853–68. Cattaneo and Norrie both
believe that the subject’s unconditional authorization of his sovereign’s ac-
tions is necessitated by Hobbes’s principles, but I do not think that either
gives sufficient reasons for this. Both rely too much on Hobbes’s assertions
but do not consider whether those assertions are substantiated on Hobbes’s
own terms. See Cattaneo, “Hobbes’s Theory of Punishment,” 293; Norrie,
“Thomas Hobbes,” 304–6. Tuck gives a somewhat stronger defence of the
necessity of unconditional authorization. According to Tuck, the sovereign’s
natural right only permits him to act in his own defence, and this appears to
Tuck not to give him the right to punish in order to defend civil society as a
whole. Unconditional authorization therefore becomes necessary to justify
punishment. Tuck, Natural Rights Theories, 129–30. But the problem Tuck fo-
cuses on is not as great as he believes, for the sovereign’s personal safety
and the security of civil society are mutually dependent; thus, punishing
the guilty is also an act of self­-preservation by the sovereign.
35 Orwin and Johnston make the best case for the view that Hobbes’s theory
of authorization is essentially a rhetorical device that emerges only in the
Leviathan – the most rhetorical of Hobbes’s three major political works –
and is absent in the earlier Elements of Law and De Cive, which already con-
tain Hobbes’s political thought in its mature form. See Orwin, “On the
Sovereign Authorization,” 28–32; Johnston, The Rhetoric, 80–2. As we will
see in the next chapter, this particular inconsistency in Hobbes’s position
was successfully attacked by Beccaria.
 Notes to pages 73–8 157

36 The humanizing features of Hobbes’s punishment theory continue to in-

form modern penal thought and practice. As an illustration of this, one
might consider the passionate appeals to the decency and humanity of
modern democratic societies by recent penal reformers like psychologist
Karl Menninger and former US Attorney General Ramsey Clark, whose
moralizing eloquence appears side by side with their calls for modern
societies to consider their own safety and security. See Ramsey Clark,
Crime in America; Karl Menninger, The Crime of Punishment.

Chapter 3

1 Montesquieu, The Spirit of the Laws, trans. Cohler, 12.2 (emphasis added).
Henceforth all references unaccompanied by a title are to this text. I have
consulted the Oeuvres complètes de Montesquieu (ed. André Masson) for the
original French.
2 Prior to Montesquieu, modern thinkers had focused on those threats to in-
dividual security that stemmed from religious dispute, disputes about the
best regime, or natural scarcity. Montesquieu himself was far from deny-
ing the importance of these issues as fundamental political problems. In
fact, it is the success of his predecessors (consider Montesquieu’s praise
of the “great men” of France, England, and Germany in the “Preface,” xlv)
that made it possible for him to turn his attention to such specific ques-
tions as the criminal law. See also Carrese, The Cloaking of Power, 18–25.
3 Thus, for example, although family life is natural to human beings, the
structure of the family is not necessarily (that is, by nature) monogamous.
See 26.3–7 and 14.
4 For a more extensive discussion of human nature as malleable and non­-
teleological, see T. Pangle, Montesquieu’s Philosophy of Liberalism, 28–37.
5 Nevertheless, in the decisive respect despotism remains a legitimate form
of government for Montesquieu, since it satisfies the minimum requirement
of human nature, which is preservation. In this respect, he is in complete
agreement with Hobbes. See 1.2: “Peace would be the first natural law.”
6 See T. Pangle, Montesquieu’s Philosophy of Liberalism, 107: “In his presenta-
tion of the traditional forms of government [which takes up the first eight
books of The Spirit of The Laws], Montesquieu has revealed the inadequa-
cies of their particular principles and the aims that derive from those prin-
ciples. He has thereby pointed to the desirability of a government having
no other purpose than the security and comfort of its citizens.”
7 As Carrese puts it, “a complex judging power becomes the crux of
[Montesquieu’s] moderate constitutionalism because its rules and
158 Notes to pages 79–85

procedures stymie excesses of raw power, or of high idealism, that could

damage individual interests and tranquility.” The Cloaking of Power, 35.
8 Montesquieu was not so naive as to believe that no harsh punishments of
any sort would ever be needed. His intention was not to call for the com-
plete abolition of corporal punishments. See 6.10, 18.
9 For examples of mild republican penal practices, see pp. 83, 84, and 88; for
examples of harsh republican practices, see pp. 11, 16, and 93; for ambigu-
ous cases, see pp. 85–6, 89, and 92.
10 See Livy, 3.34.
11 Cf. 12.11n29 and context. In the text cited by Montesquieu, Livy’s point
seems to be that the Valerian and Porcian laws were introduced in order to
protect the political rights of the plebs rather than to moderate the severity
of punishments for its own sake (as Montesquieu claims).
12 See 6.15, where Montesquieu describes the types of penalties imposed on
the different orders when Rome under the emperors “drew somewhat
nearer to monarchy.” See also 6.21, where Montesquieu remarks that it is
in the nature of monarchy to take an interest in the security of “important
men” but remains silent about men of common status.
13 See pp. 83, 84–5, 90, and 92. Consider also the ambiguity of 6.9: “It would
be easy to prove that in all or nearly all the states of Europe penalties have
decreased or increased in proportion as one has approached or departed
from liberty.”
14 See Book 24; but cf. 25.13.
15 According to both T. Pangle and Schaub, Montesquieu uses the theme of
despotic (human) government extensively in The Spirit of the Laws and in
the Persian Letters as an allegory for the government over men by the God
of the Bible. See Pangle, The Theological Basis, ch. 2; Schaub, Erotic
Liberalism, ch. 5. See also Krause, “Despotism in The Spirit of Laws,” in
Montesquieu’s Science of Politics, ed. Carrithers, Mosher, and Rahe, 252–3.
16 Montesquieu distinguishes the passion of shame – which has in it an ele-
ment of self­-love – from the much more self­-denying love of virtue that
one finds in republics. For this reason, the love of honour and the feeling
of shame that accompanies loss of reputation are not anti­-liberal passions.
In discussing Montesquieu’s account of honour, Krause writes that, al-
though there are tensions between democracy and the love of honour, and
although “democratic forms of honour are bound to be weak … still dem-
ocratic societies need some form of honor” to sustain “the capacity for po-
litical agency.” “The Politics of Distinction and Disobedience,” 490–1.
17 See below.
18 12.19. See also 12.10 and 22.
 Notes to pages 85–8 159

19 Blackstone, Commentaries, IV.1 (pp. 18–19).

20 See, again, 12.4 (p. 190); cf. Hobbes, Leviathan, ch. 15. Thus, like Hobbes,
Montesquieu believed that the character of criminal laws is deeply con-
nected to prevailing popular ideas about justice and desert. To regard tra-
ditional penal practices as merely limited by technological backwardness
is to misunderstand the phenomenon. Modernization of criminal law
depends fundamentally on popular moral enlightenment.
21 Cf. 19.16, especially n16.
22 See also Carrese, The Cloaking of Power, 64–9.
23 According to Locke, a church is “a voluntary Society of Men, joining them-
selves together of their own accord, in order to the publick worshipping
of God, in such a manner as they judge acceptable to him, and effectual
to the Salvation of their Souls” (A Letter Concerning Toleration, 28).
24 This might seem paradoxical from the point of view of today’s liberal­-
democratic societies, which experience very high levels of violent crime.
However, the pronounced increase in violent crime is a relatively recent
phenomenon in modern democracies (dating to the 1960s or 1970s in most
Western countries) and appears to be connected to the erosion of the tradi-
tional family. See, for example, Francis Fukuyama, The Great Disruption.
For evidence of the statistical relationship between family dysfunction and
crime, see Farrington, “Families and Crime,” in Wilson and Petersilia, eds.,
Crime and Public Policy. It is an important question whether Montesquieu’s
understanding of the conditions of liberalism could have allowed him to
anticipate this problem.
25 See 6.19, and again 6.2 and 12. See also 5.14.
26 Consider, again, the formulation of the principle of retribution at 1.1.
27 Consider these statements at 1.3: “Since nature has established paternal
power …,” and “Political power necessarily includes the union of many
families.” On the character of the passions and moral qualities established in
the family – especially, the shame and love of dependants on the one hand,
and the jealousy of fathers on the other – see 10.10, 12. In the context of his
endorsement of the law of retaliation against violent crimes, Montesquieu
states that this law is “drawn from reason and from the sources of good and
evil” 12.4 (my emphasis). This statement may imply that “the sources of
good and evil” may be distinct from reason and thus not necessarily ratio-
nal. Since the earliest – and in a sense the foundational – moral education we
receive is in the family (cf. 4.4), it may well be said to be a “source of good
and evil.” In European societies, the paternalistic morality that originates in
the family seems to be reinforced by biblical religion. The virtues natural to
dependent members of the family listed in 16.10 are also Christian virtues.
160 Notes to pages 88–93

28 See, again, 20.1.

29 Later liberals like Joel Feinberg argue that all of the most serious crimes can
be shown to harm the individual (thus violating the “harm principle,” as
originally conceived by John Stuart Mill) and that this obviates the need for
criminal law to appeal to the alleged inviolability of any independent moral
standards, cherished for their own sake by the community, apart from the
harm they cause to individuals. Yet Feinberg is able to do this only by defin-
ing “harm” to the individual (the “thwarting of an interest”) in a very broad
sense that goes well beyond physical harm (“interests in the continuance
of a foreseeable interval of one’s life and the interests in one’s own physical
health and vigor”) to include any actions that would undermine “emotional
stability” by producing “groundless anxieties and resentments,” and even
actions that would thwart “ultimate goals and aspirations” held by the indi-
vidual. See Feinberg, Harm to Others, ch. 1; cf. his Harmless Wrongdoing,
ch. 30 (esp. his discussion of James Fitzjames Stephen at 155–9).
30 See Calvin, Institutes, VI (in the F.L. Battles translation, see 212, 216–19);
Augustine, The City of God, I.21.
31 Thus, rape is now generally punished by prison terms of ten or more years
– with about half of this time regularly waived for good behavior – which
is arguably less severe than traditional penalties for the same crime, such
as death or castration.
32 Beccaria’s treatise also had its opponents among both jurists and the
Catholic clergy. See Maestro, Cesare Beccaria, ch. 3.
33 Before Beccaria, Rousseau had elaborated a contractarian argument for the
right to punish (in On the Social Contract). But Rousseau’s social contract is
not a liberal one, since for Rousseau each party to the contract must alien-
ate the entirety of his natural right to the sovereign (the general will) and
thus retains no rights against the sovereign.
34 Beccaria’s psychology also leads him to reject as erroneous the view – typi-
cal of retributivism – that the true measure of crimes is to be found in the
evil intention or intrinsic injustice of the persons who commit them.
Human beings are not ultimately responsible for their intentions, which
are actually “contingent on the impression caused by the [external] objects
at the time and the preceding disposition of the mind”; these dispositions,
in turn, “vary from man to man and in the same man according to the
very swift succession of ideas, emotions and circumstances.”
35 Cf. Beccaria’s brief comments (in his preface, “To the Reader”) on “natural
justice,” which is, along with divine justice, “of [its] essence constant and
unchangeable,” but which, Beccaria insists, has no more bearing than the
latter on actual political life, since political life is infinitely variable.
 Notes to pages 93–4 161

36 I have not encountered any commentary on Beccaria that shows awareness

of this serious theoretical difficulty, although in contemporary punishment
debates it is very common to find utilitarianism being criticized for subor-
dinating the interests of the individual to those of the majority. Thus, for
example, Hostettler in Cesare Beccaria (72) somehow reads Beccaria as hold-
ing that “justice has nothing to do with physical power,” when in fact it is
said to be nothing but the power of the majority used in the service of its
advantage. Crimmins notes that Beccaria’s assumption that individuals
enter into the social contract for the sake of self­-preservation (a right they
never lay down) “sounds much like Hobbes’s justification of resistance to
a punishing sovereign.” Yet Crimmins never even suggests that this may
potentially be in tension with Beccaria’s principle of general utility. See
Crimmins, “The Principles of Utilitarian Penal Law in Beccaria, Bentham,
and J.S. Mill,” in Koritansky, The Philosophy of Punishment, 141–4.
37 But how reliable are the enlightened sentiments to which Beccaria appeals
when it comes to precisely those cases in which the interests of the majori-
ty are believed to conflict with the integrity of the life and limb of the of-
fender? As we will see in chapter 5, below, Foucault rightly discerned that
the principle of compassion as a limit on the punishing power is essential-
ly a selfish principle, since it respects the sentiments of the law­-abiding
majority and thus is quite susceptible to being overridden by the casuist-
ries of prudential reason, especially when the fear of danger is compelling.
38 William Bradford, a leading penal reformer in the United States at the time
of the founding, testifies to this in his report to the governor of
Pennsylvania (“An Enquiry”). The report opens with the claim that “the
general principles upon which penal laws ought to be founded appear to
be fully settled.” “Among these principles some have obtained the force
of axioms, and are no longer considered to be the subjects of doubt or
of demonstration.” These axioms include the principle that prevention of
crimes ought to be the sole purpose of punishment and that every punish-
ment that is not absolutely necessary for this “is a cruel and tyrannical
act.” All that remained to be done, according to Bradford, was to show
how these principles might be applied effectively in practice. Cf. Thomas
Jefferson, The Complete Jefferson, 61.
39 For comprehensive historical accounts, see Hirsch, The Rise of the
Penitentiary; Rothman, The Discovery of the Asylum.
40 They observe that “there are in America as well as in Europe, estimable
men whose minds feed upon philosophical reveries, and whose extreme
sensibility feels the want of some illusion. These men, for whom philan-
thropy has become a matter of necessity, find in the penitentiary system
162 Notes to pages 95–100

a nourishment for this generous passion.” Beaumont and Tocqueville,

On the Penitentiary System in the United States.
41 See Rush, A Plan for the Punishment of Crime, 6–7, 12.
42 Lev 24:10–23; Gen 9:6.
43 Pestritto (Founding the Criminal Law, chs. 1–3) likewise argues that the re-
markable consensus then existing regarding the overall program of penal
reform allowed many unresolved theoretical problems and questions to
remain obscured.

Chapter 4

1 In addition to the thinkers whose penal thought we have already dis-

cussed, see Locke, Second Treatise of Government, s. 8; Mill, On Liberty, 9.
2 See Merle, German Idealism, chs. 1–3; Sussman, “Shame and Punishment”;
Hill, “Kant on Wrongdoing”; Tunick, “Is Kant a Retributivist?”; Byrd,
“Kant’s Theory of Punishment”; and Scheid, “Kant’s Retributivism.”
3 Kant, Metaphysics of Morals (henceforward MM), 6: 305–13 (84–90).
Whenever possible, the pagination of the standard German edition of
Kant’s works will be given as the first set of page numbers (along with the
volume number); the numbers in parentheses that follow refer to the cor-
responding pages in the cited English translation. Other works by Kant
frequently cited in the text have been identified by the following abbrevia-
tions: Critique of Judgment (CJ); Critique of Practical Reason (CPr);
Groundwork of the Metaphysics of Morals (Gr); and Lectures on Ethics (LE).
4 Cf. Hegel, The Philosophy of Right, s. 100. In the decisive respect, Hegel fol-
lows Kant in his understanding of retribution as essential to punishment.
Hegel’s derivation of the principle of retribution differs somewhat from
that of Kant. According to Hegel, a crime is a “negatively infinite judg-
ment” that negates not only the particular – that is, the property right of
the victim – but the universal as well – namely, right as such. To allow a
crime to go without retribution is to allow the negation of right, and of
freedom along with it, to stand. Retributive punishment is therefore to be
understood as the negation of a negation. See ibid., ss. 95, 99–101.
5 Consider Kant’s illustration of the formulation of the categorical impera-
tive to treat human beings as ends in themselves in the case of making a
false promise. Kant argues that a false promise could never be understood
as being compatible with the choice of the person whom we have de-
ceived. Even if lying were necessary for the other’s own good, it would
still be incompatible with his autonomy, and thus would amount to treat-
ing him as a mere means to our ends. See Gr 4:429–30 (97).
 Notes to pages 101–3 163

6 See Gr 4:403–5 (71–3).

7 See Uleman, “External Freedom in Kant’s ‘Rechtslehre,’” 585–91.
8 6:227–8 (19–20). If one’s actions fall short of what one can be constrained
by law to do – that is, if they are morally culpable – then their “rightful
effect” is punishment. Conversely, if one does more than what the law
requires, this is meritorious, and the rightful effect of meritorious action
is reward.
9 Cf. MM 6:335 (108).
10 This view has been revived more recently by neo­-Kantians like Morris
(see “Persons and Punishment”), and Murphy (see “Marxism and
Retribution”).
11 In the context, Kant anticipates the objection that equal retribution “is not
possible in terms of the letter” in every single case. No one can demand an
eye for an eye from a blind man. Kant argues, however, that it is possible
to remain true to the spirit of the principle, if not always to its letter. In his
example, Kant suggests that when an innocent person has been insulted,
the punishment of the offender can satisfy the demand of retributive jus-
tice, even if it does not take the exact same form as the original insult, as
long as the offender is made to feel shame in proportion to the outrage he
caused his victim. See MM 6:332–3 (105–6).
12 Cf. Kant’s note at Gr 4:430 (97), where he clearly implies that punishment
is a strict duty to others.
13 See Kant’s essay, “An Answer to the Question: ‘What is Enlightenment?’”
in Political Writings.
14 See, for example, Jeffrie Murphy, “Does Kant Have a Theory of Punishment?”
15 MM 6:218–19 (20); cf. MM 6:214 (14).
16 One might object that in the Rechtslehre, retributive punishment is con-
cerned not with the ends of a criminal’s choices, but merely with their
“form” as they relate to the choices of others, as Kant appears to state in
the context of his discussion of the principle of reciprocity (MM 6:230 [23–
4]). But in that context, Kant seems to be speaking only about reciprocal
relations between persons insofar as their actions, “as facts,” can have in-
fluence on each other – that is, wholly as external phenomena – and does
not intend to say anything about imputation, a subject with which he had
already dealt earlier. (Cf. translator Mary Gregor’s note c, on page 24.) It is
hard to imagine what Kant could have in mind, if not the criminal’s im-
moral ends, when he speaks of his “inner wickedness.”
An issue to which I cannot give adequate treatment here is why retri­
bution must follow some immoral actions but not others. The answer, I
­believe, has to do with Kant’s distinction between perfect and imperfect
164 Notes to pages 104–8

duties. Human beings are competent to punish transgressions of perfect

duties, knowing that such transgressions are always morally culpable,
whereas imperfect duties need not be violated just because our actions do
not appear to conform to them.
17 See Kant, Critique of Pure Reason, A 551–2/B 579–80 (475).
18 I must therefore disagree with Herbert’s and with Fleischacker’s attempts
to interpret Kant’s use of “desert” as signifying something that does not
relate to moral motivations. See Herbert, “Immanuel Kant,” 67–72;
Fleischacker, “Kant’s Theory of Punishment,” 202–5.
19 As Hegel would later argue, some amount of uncertainty and contingency
must be accepted in all action, and therefore (perhaps somewhat paradoxi-
cally) we can be held responsible for consequences we did not intend or
anticipate. See The Philosophy of Right s. 118. See also Herbert, “Immanuel
Kant,” 67–8.
20 Cf. CJ 5:429–31 (317–19) with MM 6:246–7 (40–1).
21 MM 6:230–32 (24–6).
22 See again MM 6:307 ff. (86 ff.).
23 MM 6:219–20 (21).
24 Hill, “Kant on Wrongdoing,” 409.
25 Hill, “Kant on Wrongdoing,” 429.
26 Hill, “Kant on Wrongdoing,” 428–31. In his interpretation, Hill follows
Byrd’s “Kant’s Theory of Punishment” and Scheid’s “Kant’s
Retributivism.” Byrd reads Kant’s basic position as virtually identical with
that of H.L.A. Hart (see Byrd, “Kant’s Theory of Punishment,” 183).
27 Cf. Hill, “Kant on Wrongdoing,” 433–4, 438.
28 Tunick, “Is Kant a Retributivist?,” 60–78.
29 Cf. MM 6:235–6 (28).
30 Tunick, “Is Kant a Retributivist?,” 64.
31 Tunick, “Is Kant a Retributivist?,” 65–6. Cf. MM 6:336–7 (108–9).
32 MM 6:336 (109).
33 See MM 6:334 (107–8), 6:337 (109–10). Cf. Tunick, “Is Kant a
Retributivist?,” 63–4.
34 There is another passage that is often cited by Kant scholars as showing
him to be espousing what is essentially a deterrence theory of punish-
ment. In his posthumously published Lectures on Ethics, Kant says that “all
punishments imposed by sovereigns and governments are pragmatic; they
are designed either to correct or to make an example” (LE, “Reward and
Punishment,” 55). It may be that Kant changed his mind about punish-
ment from the time his lectures were composed to the time of the writing
of his Rechtslehre. In any case, if in the final analysis there remains a real
 Notes to pages 109–14 165

discrepancy between the two texts, more weight should be given to the
Metaphysics of Morals, which was published in Kant’s lifetime and there-
fore represents his mature view, than to his lectures, which were published
without his supervision.
35 Cf. MM 6:379–80 (145–6).
36 As an illustration of what Kant may have in mind in this passage, consider
the example of the titular character in Thomas Hardy’s The Mayor of
Casterbridge.
37 CPr 5:110–14 (114–18).
38 CPr 5:110 ff. (114 ff.).
39 CPr 5:114–32 (118–36).
40 See also MM 6:488–90 (230–2); and “On the Miscarriage of all Philosophical
Theodicies” 8:260n (28n), in Allen Wood and George di Giovanni, eds.,
Religion and Rational Theology. Cf. Byrne, Kant on God, 110–17; and Beck,
Commentary, 270–1.
41 The idea that divine retribution and human punishment follow the same
principle is suggested by Fleischacker, “Kant’s Theory of Punishment,”
203–6.
42 See also CPr 5:61 (63): “When … someone who delights in annoying and
vexing peace­-loving folk receives at last a right good beating, it is certainly
an ill, but everyone approves of it and considers it as good in itself even if nothing
further results from it; nay, even he who gets the beating must acknowledge,
in his reason, that justice has been done to him, because he sees the pro-
portion between welfare and well­-doing, which reason inevitably holds
before him, here put into practice” (my emphasis).
43 See Shell, The Rights of Reason, 94.
44 Similarly, Beck has argued that Kant does not adequately show how the
highest good might be derived from the categorical imperative, consider-
ing that the latter omits any reference to happiness. Beck speculates that
Kant’s synthesis of virtue and happiness might perhaps be defended on
the ground that “it is important for the architectonic purpose of reason in
uniting under one Idea the two legislations of reason, the theoretical and
the practical … [since r]eason cannot tolerate a chaos of ends.” Yet Beck
himself finds no such argument in Kant’s writings. See Beck, Commentary,
242–5. Taylor raises a similar doubt in “Kant’s Political Religion,” 11–12.
45 Fackenheim attempts to save Kant from inconsistency by arguing that the
key to understanding his many contradictions lies in grasping his new
kind of metaphysics. “Kant destroys the metaphysics which is based on
speculation and replaces it with a metaphysics which is based on moral
consciousness.” Kant “seeks to prove, not immortality and God, but that
166 Notes to pages 117–21

the belief in immortality and God is implicit in finite moral conscious-

ness.” We should therefore attribute the tensions and ambiguities we find
in Kant’s arguments – such as the ambiguity of the summum bonum – not
to Kant’s incompetent philosophizing, but to finite moral consciousness.
“The philosopher too is a finite moral agent; and it is in his latter rather
than his former capacity that he is in touch with ultimate moral reality:
and as philosopher he recognizes this fact.” See Fackenheim, “Kant’s
Philosophy of Religion,” 9, 15–18. Thus, according to this line of argument,
Kant would urge us to embrace retribution, despite the unanswered ques-
tions we have raised regarding the relation between happiness and virtue,
because it is a product of our finite moral consciousness. This does not
seem persuasive to me. Kant’s fundamental intention is to show how mo-
rality is rational; he can do so by demonstrating its coherence to specula-
tive reason, but certainly not by allowing speculative reason to dissolve
into a mysterious metaphysics of moral consciousness.

Chapter 5

1 The passage is from Rousseau, On the Social Contract, 2.5.

2 See also White’s brief discussion of this point in Foucault, in “Le pouvoir
et la parole,” in Artières et al., Surveiller et Punir de Michel Foucault, 291–2.
3 See, again, the Leviathan, 27.41. Emphasis added.
4 Locke, Second Treatise of Government, 2.11.
5 Given Foucault’s philosophical commitments, it would no longer be possi-
ble for him to have recourse to the principle of retribution along Kantian
lines, as a way of solving the problem of the unlimited modern punishing
power. Foucault does not even entertain this as a possibility, since the
Kantian concept of punishment relies on a view of man as constituted by a
transcendental (“noumenal”) autonomous self – a view that, from Foucault’s
point of view, has been widely discredited as a historical artefact of the Age
of Enlightenment. See Foucault, The Order of Things, ch. 9; “Nietzsche,
Genealogy, History,” in The Foucault Reader, 76–100. Foucault’s historicism
appears to presuppose and to depend on the vast success (and subsequent
developments) of Rousseau’s critique of rationalism in the name of what
Rousseau called natural goodness. (See especially the Discourse on the Origins
and Foundation of Inequality among Men, in The First and Second Discourses, ed.
Masters and Masters.) According to that epoch­-making critique, rationality
is only an acquired faculty – one that, as such, cannot serve as the source of
natural right. On the contrary, it is the essentially non­-rational passions of
self­-love and love of liberty (which are themselves crucially shaped by
 Notes to pages 122–5 167

society in all of its forms) that express the “perfectibility” of our nature – or,
more accurately, its malleable and non­-teleological character. Modern society
has become more prosperous and secure, but at the cost of the self­-alienation
of the individual, whose reason and passions forever remain in contradic-
tion. One might say that, in a certain sense, Foucault’s Discipline and Punish
draws out the implications of this Rousseauan analysis for liberal criminal
justice. While liberal citizens have been sufficiently enlightened (as to the
causes of crime and the corrupting effects of society itself on the individual)
to feel pity for the criminal, they nevertheless cannot help constantly consid-
ering their own self­-interest, which teaches that the criminal is a dangerous
threat that needs to be incapacitated or destroyed altogether.
6 See also Boulant, Michel Foucault et les prisons, 106–24.
7 Cf. Beccaria, On Crimes and Punishments, ch. 19; Montesquieu, The Spirit
of the Laws, 12.4.
8 See also Adam J. Hirsch, The Rise of the Penitentiary.
9 In this respect, Foucault’s account of prison administration closely paral-
lels Weber’s theory of bureaucratic administration.
10 This fact about penitentiary science has been admitted even by Foucault’s
staunchest critics. See Pinatel, “Philosophie carcérale, technologie politique,
et criminologie clinique,” in Artières et al., Surveiller et Punir de Michel
Foucault, 115. See also, more generally, Saleilles, L’individualisation de la peine.
11 See also Foucault, Discipline and Punish, 129–30, 183, 245–7.
12 Cf. Foucault, Discipline and Punish, 199, 249. White, in Artières et al.,
Surveiller et Punir de Michel Foucault, 286–90, discusses Foucault’s penchant
for ascribing causality to obscure forces, and links this feature of Discipline
and Punish to the earlier Words and Things.
13 See Foucault, Discipline and Punish, 128–9, and contrast with our discussion
of Rush’s pamphlet on the penitentiary in chapter 3.
14 Pinatel observes that Foucault understates the influence of Catholicism on
the European penitentiary movement. See “Philosophie carcérale,” 103–4.
15 See Foucault, Discipline and Punish, 264–71. See also Rothman, The
Discovery of the Asylum; Hirsch, The Rise of the Penitentiary.
16 The criticism of the effectiveness of the carceral penal system is quite ex-
tensive. Those who claim that the penal system cannot be salvaged as an
instrument of correction or deterrence, and that it must be abandoned root
and branch, include Knopp et al., Instead of Prisons; Cohen, Against
Criminology; West and Morris, The Case for Penal Abolition; and Irwin and
Austin, It’s About Time. Others argue that the effectiveness of prisons in
their corrective and deterrent capacities is marginal, but believe this prob-
lem can be remedied with further reforms. See, for example, Dilulio,
168 Notes to pages 126–35

Governing Prisons. It should be noted that no one denies that prisons are
effective at incapacitating offenders.
17 Foucault concerns himself with manifestations of despotic disciplinary
structures in all aspects of modern life, and his focus on the modern penal
system in particular is explained by the fact that it represents the furthest
development of this phenomenon. The question of this broader claim in
Discipline and Punish is beyond the scope of the present study.
18 Polet, too, observes this inconsistency in the argument of Discipline and
Punish. See “Punishing Some, Disciplining All,” in Koritansky, The Philosophy
of Punishment, 214.
19 See the note on the restorative justice movement in our Introduction, 6–7
and n2.
20 Stuntz argues that in previous generations, local communities in the United
States had significantly more control over criminal justice than they do to-
day. See The Collapse of American Criminal Justice, chs. 3–4, 7.
21 Thus Stuntz, in The Collapse (283, 307–8), recommends “decentralization” and
“local democracy” as a way of fixing the broken criminal justice system. But
he does not regard these changes as probable, since they would depend on
a political majority that lives in relatively safe neighbourhoods and thus has
no incentive to support extensive reforms. See also the findings in Schneider,
Refocusing Crime Prevention, regarding the non­-participation of residents of a
Vancouver neighbourhood in local community crime prevention programs.

Conclusion

1 Dzur, “An Introduction,” 1.

2 For an overview of the issues, see Piehl and Useem, “Prisons,” in Wilson
and Petersilia, Crime and Public Policy, 532–58.
3 For an overview of the competing theories of this phenomenon, see for ex-
ample Siegel and McCormick, Criminology in Canada, ch. 3; Rosenfeld,
“Changing Crime Rates,” in Wilson and Petersilia, Crime and Public Policy,
559–88.
4 See, especially, Wilson’s Thinking About Crime.
5 Versions of this explanation have since been proposed by political scientist
Francis Fukuyama (in The Great Disruption) as well as, more recently, by
psychologist Steven Pinker. In The Better Angels of Our Nature: Why Violence
Has Declined, Pinker attributes the sharp spike in crime in the 1960s to a
“decivilization” of society, and the subsequent decline of crime in the
1990s to a “recivilization” brought about by a collective reaction against
the radicalism of preceding decades, a concern in urban and economically
 Notes to pages 136–40 169

disadvantaged communities about their declining conditions, and (in the

United States) the resurgence of social conservative movements such as
Christian evangelicalism. On the other hand, Stuntz, in The Collapse of
American Criminal Justice, attributes the late­-twentieth­-century crime boom
to “pendulum justice”: according to Stuntz, the crime boom was a conse-
quence of the excessive leniency of American criminal justice immediately
following the Second World War (see esp. ch. 9).
6 See the March 1982 Atlantic Monthly article, “Broken Windows,” authored
by Wilson and George L. Kelling.
7 Similar problems can be identified in another book popular at that time,
which also criticized the traditional approach to criminal justice: psychia-
trist Karl Menninger’s provocatively titled The Crime of Punishment.
8 See also Zimring, Perspectives on Deterrence (esp. 32–107) for an overview
of the wide range of factors that may affect the efficacy of deterrence and
thus infinitely complicate the search for an exhaustive theory of the funda-
mental causes of crime.
9 Thacher, “Order Maintenance Reconsidered,” 412–13.
10 As Koritansky also notes. See his “Introduction” in The Philosophy of
Punishment.
11 Rawls, “Two Concepts of Rules,” 3–32; Rawls, A Theory of Justice, 211–12,
276–7; Hart, “Prolegomenon,” 1–27; Feinberg, Doing and Deserving, ch. 5.
See again our discussion above, 10–12.
12 Brubaker, “Can Liberals Punish?” 821–36.
13 Brubaker, “Can Liberals Punish?” 821–2, 825.
14 Brubaker, “Can Liberals Punish?” 828–31. See also Corlett’s criticism of
Rawls in “Making Sense of Retributivism,” 81–3.
15 Brubaker, “Can Liberals Punish?” 833.
16 On the controversy in the scholarly literature over the extent of the break
between modern political thought and classical republicanism, compare
Pocock, The Machiavellian Moment, with criticism of Pocock in T. Pangle, The
Spirit of Modern Republicanism, and M. Zuckert, The Natural Rights Republic.
This page intentionally left blank
 Bibliography

Primary Sources

Aristotle. Nicomachean Ethics. Trans. Robert C. Bartlett and Susan D. Collins.

Chicago: University of Chicago Press, 2011. http://dx.doi.org/10.7208/
chicago/9780226026763.001.0001.
– Politics. Trans. Carnes Lord. Chicago: University of Chicago Press, 1984.
http://dx.doi.org/10.7208/chicago/9780226026701.001.0001.
Aquinas, Thomas. On Law, Morality, and Politics. Ed. William P. Baumgarth
and Richard J. Regan. Indianapolis: Hackett, 1988.
Augustine. The City of God against the Pagans. Trans. Robert Dyson. Cambridge:
Cambridge University Press, 1998.
Beaumont, Gustave de, and Alexis de Tocqueville. On the Penitentiary System
in the United States and Its Application in France. Trans. Francis Lieber.
Carbondale: Southern Illinois University Press, 1979.
Beccaria, Cesare. On Crimes and Punishments. Trans. David Young. Indianapolis:
Hackett, 1984.
Benn, S.I. “An Approach to the Problems of Punishment.” Philosophy 33,
no. 127 (1958): 325–41. http://dx.doi.org/10.1017/S0031819100055017.
Bentham, Jeremy. The Rationale of Punishment. Ed. James McHugh. Amherst:
Prometheus Books, 2009.
Blackstone, William. Commentaries on the Laws of England. Facsimile of the
1st ed. in 4 vols. Chicago: University of Chicago Press, 1979.
Boonin, David. The Problem of Punishment. New York: Cambridge University
Press, 2008. http://dx.doi.org/10.1017/CBO9780511819254.
Bradford, William. “An Enquiry How Far the Punishment of Death Is
Necessary […]” In Pennsylvania. Philadelphia: Dobson 1793, 1–73.
172 Bibliography

Braithwaite, John. Restorative Justice and Responsive Regulation. New York:

Oxford University Press, 2002.
Braitwaite, John, and Pettit Philip. Not Just Deserts: A Republican Theory of
Criminal Justice. New York: Oxford University Press, 1990.
Brubaker, Stanley C. “Can Liberals Punish?” American Political Science Review
82, no. 3 (1988): 821–36. http://dx.doi.org/10.2307/1962493.
Brubaker, Stanley C., and Ronald J. Terchek. “Punishing Liberals or Rehabilitating
Liberalism.” American Political Science Review 83, no. 4 (1989): 1309–16. http://
dx.doi.org/10.2307/1961671.
Calvin, John. Institutes of the Christian Religion. Grand Rapids, MI: W.B.
Eerdmans, 1995.
Christie, Nils. “Conflicts as Property.” British Journal of Criminology 17 (1977):
1–26.
Clark, Ramsey. Crime in America. New York: Simon and Schuster, 1970.
Cohen, S. Against Criminology. Englewood Cliffs: Transaction, 1993.
Corlett, J. Angelo. “Making Sense of Retributivism.” Philosophy 76, no. 295
(2001): 77–110.
Dagger, Richard. “Playing Fair with Punishment.” Ethics 103, no. 3 (1993):
473–88. http://dx.doi.org/10.1086/293522.
Dilulio, John J. Governing Prisons: A Comparative Study of Correctional Management.
New York: The Free Press, 1987.
Duff, R.A. Intention, Agency, and Criminal Liability: Philosophy of Action and the
Criminal Law. Oxford: Basil Blackwell, 1990.
Durkheim, Emile. Sociology and Philosophy. Trans. D.F. Pocock. Glencoe: The
Free Press, 1953.
Dzur, Albert. “An Introduction: Penal Democracy: Good Society Symposium on
‘Democratic Theory and Mass Incarceration.’” Good Society 23, no. 1 (2014):
1–5. http://dx.doi.org/10.5325/goodsociety.23.1.0001.
Feinberg, Joel. Doing and Deserving. Princeton: Princeton University Press,
1970.
– Harm to Others. Oxford: Oxford University Press, 1984.
– Harmless Wrongdoing. Oxford: Oxford University Press, 1988.
Foucault, Michel. Discipline and Punish. Trans. Alan Sheridan. New York:
Vintage Books, 1995.
– The Foucault Reader. Ed. Paul Rabinow. New York: Vintage Books, 2010.
– The Order of Things. [Trans. anon.] New York: Vintage Books, 1994.
Fukuyama, Francis. The Great Disruption: Human Nature and the Reconstitution
of Social Order. New York: Simon and Schuster, 2000.
Furman v. Georgia. 408 U.S. 238 (1972).
Gregg v. Georgia. 428 U.S. 153 (1976).
 Bibliography 173

Habermas, Jürgen. The Structural Transformation of the Public Sphere: An Inquiry

into a Category of Bourgeois Society. Trans. Thomas Burger. Boston: MIT Press,
1989.
Hahn, Paul. Emerging Criminal Justice: Three Pillars for a Proactive Justice System.
Thousand Oaks: Sage Publications, 1998.
Hart, H.L.A. “Prolegomenon to the Principles of Punishment.” In Punishment
and Responsibility. Oxford: Clarendon Press, 1968, 1–27.
Hegel, G.W.F. The Philosophy of Right. Trans. Alan White. Newburyport: Focus
Publishing, 2002.
Hesiod and Theognis. Theogony, Works and Days, and Elegies. New York:
Penguin Books, 1973.
Hirsch, Adam J. The Rise of the Penitentiary: Prisons and Punishment in Early
America. New Haven: Yale University Press, 1992.
Hobbes, Thomas. De Cive. Trans. Richard Tuck and Michael Silverthorne.
Cambridge: Cambridge University Press, 1998.
– Elements of Law. Ed. J.C.A. Gaskin. Oxford: Oxford University Press, 1994.
– The English Works of Thomas Hobbes. Ed. Sir William Molesworth. Elibron
Classics, 2005.
– Leviathan. Ed. Edwin Curley. Indianapolis: Hackett, 1994.
The Holy Bible: King James Version. New York: Ivy Books, 1991.
Irwin, John, and James Austin. It’s About Time: America’s Imprisonment Binge.
Belmont: Wadsworth, 1997.
Jefferson, Thomas. The Complete Jefferson. Ed. Saul Padover. New York: Duell,
Sloan, and Pierce, 1943.
Kant, Immanuel. Critique of Judgment. Trans. Werner S. Pluhar. Indianapolis:
Hackett, 1987.
– Critique of Practical Reason. Trans. Lewis White Beck. New York: Liberal Arts
Press, 1956.
– Critique of Pure Reason. Trans. Norman Kemp Smith. New York: St Martin’s
Press, 1965.
– Groundwork for the Metaphysics of Morals. Trans. H.J. Paton. New York:
Harper & Row, 1964.
– Lectures on Ethics. Trans. Louis Infield. Indianapolis: Hackett, 1980.
– Political Writings. Ed. H.S. Reiss. New York: Cambridge University Press,
1991.
– Metaphysics of Morals. Trans. Mary Gregor. Cambridge: Cambridge
University Press, 1996. http://dx.doi.org/10.1017/CBO9780511809644.
– Religion within the Bounds of Reason Alone. Trans. Allen Wood and George
di Giovanni. In Religion and Rational Theology. New York: Cambridge
University Press, 1996.
174 Bibliography

Knopp, F.H., B. Boward, and M.O. Morris. Instead of Prisons: A Handbook for
Abolitionists. Syracuse: Prison Research Education Project, 1976.
Lewis, C.S. “The Humanitarian Theory of Punishment.” In God in the Dock:
Essays on Theology and Ethics. Grand Rapids: W.B. Eerdmans, 1970, 287–300.
Livy. The Rise of Rome. Ed. T.J. Luce. New York: Oxford University Press, 1998.
Locke, John. A Letter Concerning Toleration. Ed. James Tully. Indianapolis:
Hackett, 1983.
– Second Treatise of Government. Ed. Richard Cox. Wheeling: Harlan Davidson,
1982.
Machiavelli, Niccolo. 1996. Discourses on Livy. Trans. Harvey Mansfield and
Nathan Tarcov. Chicago: University of Chicago Press, 1996. http://dx.doi
.org/10.7208/chicago/9780226500331.001.0001.
– The Prince. Trans. Harvey Mansfield. Chicago: University of Chicago Press,
1985.
Menninger, Karl. The Crime of Punishment. New York: Viking Press, 1996.
Mill, John Stuart. “April 1866 Speech on Capital Punishment.” In
Utilitarianism, 2nd ed. Ed. George Sher. Indianapolis: Hackett, 2001.
– On Liberty. Indianapolis: Hackett, 1978.
Montesquieu, Charles Louis de Secondat, Baron de. The Spirit of the Laws.
Trans. Anne Cohler. New York: Cambridge University Press, 1989.
– Oeuvres complètes de Montesquieu. Ed. André Masson. Paris: Nagel, 1950–5.
Morris, Herbert. “Persons and Punishment.” Monist 52, no. 4 (1968): 475–501.
http://dx.doi.org/10.5840/monist196852436.
Murphy, Jeffrie G. “Marxism and Retribution.” Philosophy and Public Affairs 2,
no. 3 (1973): 217–43.
Pangle, Thomas L., ed. The Roots of Political Philosophy: Ten Forgotten Socratic
Dialogues. Ithaca: Cornell University Press, 1987.
Pestritto, Ronald J. Founding the Criminal Law: Punishment and Political Thought
in the Origins of America. DeKalb: Northern Illinois University Press, 2000.
Pinker, Steven. The Better Angels of Our Nature: Why Violence Has Declined. New
York: Penguin Books, 2011.
Plato. Alcibiades I. Trans. Carnes Lord. In T. Pangle, The Roots of Political
Philosophy, 175­–221.
– “Apology of Socrates.” In West and West, Four Texts on Socrates, 63–97.
– Euthyphro. In West and West, Four Texts on Socrates, 41–61.
– Gorgias. Trans. James H. Nichols, Jr. Ithaca: Cornell University Press, 1998.
– Laws. Ed. Jeffrey Henderson. Cambridge, MA: Harvard University Press,
1926. http://dx.doi.org/10.4159/DLCL.plato_philosopher­-laws.1926.
– Laws of Plato. 2 vols. Ed. E.B. England. Manchester: University of
Manchester Press, 1921.
 Bibliography 175

– The Laws of Plato. Trans. Thomas L. Pangle. Chicago: University of Chicago

Press, 1980.
– Lesser Hippias. Trans. James Leake. In T. Pangle, The Roots of Political
Philosophy, 281–99.
– Lovers. Trans. James Leake. In T. Pangle, The Roots of Political Philosophy,
80–90.
– Minos. Trans. Thomas L. Pangle. In T. Pangle, The Roots of Political
Philosophy, 53–66.
– The Republic of Plato. Trans. Allan Bloom. New York: Basic Books, 1980.
– Works, vol. 5. Ed. John Burnet. Oxford: Oxford University Press, 1907.
Primoratz, Igor. “Punishment as Language.” Philosophy 64, no. 248 (1989):
187–205. http://dx.doi.org/10.1017/S0031819100044478.
Rawls, John. A Theory of Justice, rev. ed. Cambridge, MA: Harvard University
Press, 1999.
– “Two Concepts of Rules.” Philosophical Review 64, no. 1 (1955): 3–32. http://
dx.doi.org/10.2307/2182230.
Rothman, David J. The Discovery of the Asylum. Toronto: Little, Brown, 1971.
Rousseau, Jean­-Jacques. Collected Writings of Rousseau. 11 vols. Ed. Roger Masters
and Christopher Kelly. Hanover: University Press of New England, 1990.
– The First and Second Discourses. Trans. Roger D. Masters and Judith R.
Masters. New York: St Martin’s Press, 1964.
– On the Social Contract. Trans. Roger D. Masters and Judith R. Masters. New
York: St Martin’s Press, 1978.
Rush, Benjamin. A Plan for the Punishment of Crime: Two Essays. Ed. Negley
Teeters. Philadelphia: Pennsylvania Prison Society, 1954.
Saleilles, Raymond. L’individualisation de la peine, 3rd ed. Paris: F. Alcan, 1927.
Schneider, Stephen. Refocusing Crime Prevention: Collective Action and the Quest
for Community. Toronto: University of Toronto Press, 2007.
Siegel, Larry J., and Chris McCormick, eds. Criminology in Canada: Theories,
Patterns, and Typologies, 4th ed. Toronto: Nelson, 2010.
Smart, J.C.C., and Bernard Williams. Utilitarianism, For and Against.
Cambridge: Cambridge University Press, 2007.
Spinoza, Benedict. Theologico­-Political Treatise. Trans. Martin D. Yaffe.
Newburyport: Focus Publishing, 2004.
Stuntz, William J. The Collapse of American Criminal Justice. Cambridge, MA:
Harvard University Press, 2011. http://dx.doi.org/10.4159/harvard
.9780674062603.
Sutton, Adam, Adrian Cherney, and Rob White. Crime Prevention: Principles,
Perspectives, and Practices. Port Melbourne: Cambridge University Press,
2008. http://dx.doi.org/10.1017/CBO9780511804601.
176 Bibliography

Templeton, Laura J., and Timothy F. Hartnagel. “Casual Attributions of Crime

and the Public’s Sentencing Goals.” Canadian Journal of Criminology and
Criminal Justice 54, no. 1 (2012): 45–65. http://dx.doi.org/10.3138/cjccj.2010
.E.29.
Thacher, David. “Order Maintenance Reconsidered: Moving beyond Strong
Causal Reasoning.” Journal of Criminal Law and Criminology 94, no. 2 (2004):
381–414. http://dx.doi.org/10.2307/3491374.
West, Gordon, and Ruth Morris, eds. The Case for Penal Abolition. Toronto:
Canadian Scholars’ Press, 2000.
West, Thomas G., and Grace Starry West. Four Texts on Socrates. Ithaca: Cornell
University Press, 1984.
Wilson, James Q. Thinking About Crime, rev. ed. New York: Basic Books, 2013.
Wilson, James Q., and George L. Kelling. “Broken Windows.” The Atlantic,
March 1982.
Wilson, James Q., and Joan Petersilia, eds. Crime and Public Policy, 2nd ed.
New York: Oxford University Press, 2011.
Zimring, Franklin E. Perspectives on Deterrence. Washington, D.C.: National
Institute of Mental Health, Center for Studies of Crime and Delinquency, 1971.

Secondary Sources

Adkins, Arthur. Merit and Responsibility. Oxford: Clarendon Press, 1960.

Artières, Philippe, et al. Surveiller et Punir de Michel Foucault: Regards critiques
1975–1979. Maquette: Presses universitaires de Caen, 2010.
Beck, Lewis White. Commentary on Kant’s Critique of Practical Reason. Chicago:
University of Chicago Press, 1960.
Boulant, François. Michel Foucault et les prisons. Vendôme: Vendôme Impressions,
Groupe Landais, 2003. http://dx.doi.org/10.3917/puf.boull.2003.01.
Brown, Keith C., ed. Hobbes Studies. Oxford: Basil Blackwell, 1965.
Bruell, Christopher. “On Plato’s Political Philosophy.” Review of Politics 56,
no. 2 (1994): 261–82. http://dx.doi.org/10.1017/S003467050001843X.
– On the Socratic Education. Lanham, MD: Rowman and Littlefield, 1999.
Byrd, B. Sharon. “Kant’s Theory of Punishment: Deterrence in Its Threat,
Retribution in Its Execution.” Law and Philosophy 8, no. 2 (1989): 151–200.
http://dx.doi.org/10.1007/BF00160010.
Byrne, Peter. Kant on God. Burlington: Ashgate Publishing, 2007.
Carrese, Paul O. The Cloaking of Power: Montesquieu, Blackstone, and the Rise of
Judicial Activism. Chicago: University of Chicago Press, 2003. http://dx.doi
.org/10.7208/chicago/9780226094830.001.0001.
 Bibliography 177

Carrithers, David W., Michael Mosher, and Paul A. Rahe, eds. Montesquieu’s
Science of Politics: Essays on “The Spirit of the Laws.” Lanham: Rowman and
Littlefield, 2001.
Cattaneo, Mario A. “Hobbes’s Theory of Punishment.” In Brown, Hobbes
Studies, 275–97.
Fackenheim, Emil. “Kant’s Philosophy of Religion.” In The God Within: Kant,
Schelling, and Historicity. Toronto: University of Toronto Press, 1996, 3–19.
Fleischacker, Samuel. “Kant’s Theory of Punishment.” In Williams, Essays on
Kant’s Political Philosophy, 191–212.
Galston, William A. Kant and the Problem of History. Chicago: University of
Chicago Press, 1975.
Gauthier, David. The Logic of Leviathan. Oxford: Clarendon Press, 1969.
Goldsmith, M.M. Hobbes’s Science of Politics. New York: Columbia University
Press, 1966.
Hampton, Jean. Hobbes and the Social Contract Tradition. Cambridge: Cambridge
University Press, 1986.
Herbert, Gary. “Immanuel Kant: Punishment and the Political Preconditions
of Moral Existence.” Interpretation 23, no. 1 (1995): 61–76.
Heyd, David. “Hobbes on Capital Punishment.” History of Philosophy Quarterly
8, no. 2 (1991): 119–34.
Hill, Thomas E., Jr. “Kant on Wrongdoing, Desert, and Punishment.” Law and
Philosophy 18, no. 4 (1999): 407–41.
Hostettler, John. Cesare Beccaria: The Genius of “On Crimes and Punishments.”
Hampshire: Waterside Press, 2010.
Johnston, David. The Rhetoric of “Leviathan.” Princeton: Princeton University
Press, 1986.
Koritansky, Peter, ed. The Philosophy of Punishment and the History of Political
Thought. Columbia: University of Missouri Press, 2011.
Krause, Sharon. “Despotism in The Spirit of the Laws.” In Carrithers, Mosher, and
Rahe, Montesquieu’s Science of Politics: Essays on “The Spirit of the Laws,” 231–71.
– “The Politics of Distinction and Disobedience: Honor and the Defense of
Liberty in Montesquieu.” Polity 31, no. 3 (1999): 469–99. http://dx.doi.
org/10.2307/3235250.
Mackenzie, Mary. Plato on Punishment. Berkeley: University of California
Press, 1981.
Macpherson, C.B. The Political Theory of Possessive Individualism: Hobbes to
Locke. Oxford: Clarendon Press, 1962.
Maestro, Marcello. Cesare Beccaria and the Origins of Penal Reform. Philadelphia:
Temple University Press, 1973.
178 Bibliography

Martinich, Aloysius. Hobbes. New York: Routledge, 2005.

Merle, Jean­-Christophe. German Idealism and the Concept of Punishment. New
York: Cambridge University Press, 2009. http://dx.doi.org/10.1017/
CBO9780511770425.
Moore, Michael. Placing Blame: A General Theory of the Criminal Law. Oxford:
Oxford University Press, 1997.
Morrow, Glenn. Plato’s Cretan City: A Historical Interpretation of the “Laws.”
Princeton: Princeton University Press, 1960.
Murphy, Jeffrie G. “Does Kant Have a Theory of Punishment?” Columbia Law
Review 87, no. 3 (1987): 509–32. http://dx.doi.org/10.2307/1122669.
Norrie, Alan. “Thomas Hobbes and the Philosophy of Punishment.” Law and
Philosophy 3, no. 2 (1984): 299–320. http://dx.doi.org/10.1007/BF00144330.
O’Brien, Michael. “Plato and the ‘Good Conscience’: Laws 863E5–864B7.”
Transactions and Proceedings of the American Philological Association 88 (1957):
81–7.
– The Socratic Paradoxes and the Greek Mind. Chapel Hill: University of North
Carolina Press, 1967.
Orwin, Clifford. “On the Sovereign Authorization.” Political Theory 3, no. 1
(1975): 26–44.
Pangle, Lorraine. “Moral and Criminal Responsibility in Plato’s Laws.”
American Political Science Review 103, no. 3 (2009): 456–73. http://dx.doi
.org/10.1017/S0003055409990074.
Pangle, Thomas. Montesquieu’s Philosophy of Liberalism. Chicago: University
of Chicago Press, 1973.
– “The Political Psychology of Religion in Plato’s Laws.” American Political
Science Review 70, no. 4 (1976): 1059–77. http://dx.doi.org/10.2307/1959374.
– The Roots of Political Philosophy. Ithaca: Cornell University Press, 1987.
– The Spirit of Modern Republicanism. Chicago: University of Chicago Press, 1988.
– The Theological Basis of Liberal Modernity in Montesquieu’s Spirit of the Laws.
Chicago: University of Chicago Press, 2010. http://dx.doi.org/10.7208/
chicago/9780226645520.001.0001.
Pitkin, Hanna. “(a). “Hobbes’s Concept of Representation – I.” American Political
Science Review 58, no. 2 (1964): 328–40. http://dx.doi.org/10.2307/1952865.
– “(b). “Hobbes’s Concept of Representation – II.” American Political Science
Review 58, no. 4 (1964): 902–18. http://dx.doi.org/10.2307/1953289.
Pocock, J.G.A. The Machiavellian Moment. Princeton: Princeton University
Press, 1975.
Saunders, Trevor. “The Socratic Paradoxes in Plato’s Laws: A Commentary
on 859c–864b.” Hermes 96, no. 3 (1968): 421–34.
– Plato’s Penal Code. New York: Oxford University Press, 1991.
 Bibliography 179

Schaub, Diana J. Erotic Liberalism: Women and Revolution in Montesquieu’s

“Persian Letters.” Lanham: Rowman and Littlefield, 1995.
Scheid, Don E. “Kant’s Retributivism.” Ethics 93, no. 2 (1983): 262–82. http://
dx.doi.org/10.1086/292433.
Schrock, Thomas. “The Right to Punish and Resist Punishment in Hobbes’s
‘Leviathan.’” Western Political Quarterly 44, no. 4 (1991): 853–90. http://
dx.doi.org/10.2307/448798.
Shell, Susan M. The Rights of Reason: A Study of Kant’s Philosophy and Politics.
Toronto: University of Toronto Press, 1980.
Shorey, Paul. What Plato Said. Chicago: University of Chicago Press, 1965.
Shuchman, Philip. “Comments on the Criminal Code of Plato’s Laws.” Journal
of the History of Ideas 24, no. 1 (1963): 25–40. http://dx.doi.org/10.2307/
2707857.
– “Kant on Punishment.” Kantian Review 1 (1997): 116–24.
Stauffer, Devin. The Unity of Plato’s Gorgias. Cambridge: Cambridge University
Press, 2006. http://dx.doi.org/10.1017/CBO9780511499005.
Strauss, Leo. Natural Right and History. Chicago: University of Chicago Press,
1950.
– The Political Philosophy of Hobbes: Its Basis and Its Genesis. Trans. Elsa Sinclair.
Chicago: University of Chicago Press, 1952.
– The Argument and Action of Plato’s “Laws.” Chicago: University of Chicago
Press, 1975.
Sussman, David. “Shame and Punishment in Kant’s Doctrine of Right.” Philo-
sophical Quarterly 58, no. 231 (2008): 299–317. http://dx.doi.org/10.1111/j.1467­-
9213.2007.530.x.
Taylor, A.E. “The Ethical Doctrine of Hobbes.” In Brown, Hobbes Studies, 35–55.
Taylor, Robert. “Kant’s Political Religion: The Transparency of Perpetual Peace
and the Highest Good.” Review of Politics 72, no. 1 (2010): 1–24.
Tuck, Richard. Natural Rights Theories. New York: Cambridge University Press,
1979. http://dx.doi.org/10.1017/CBO9781139163569.
Tunick, Mark. “Is Kant a Retributivist?” History of Political Thought 17, no. 1
(1996): 60–78.
Uleman, Jennifer. “External Freedom in Kant’s ‘Rechtslehre’: Political,
Metaphysical.” Philosophy and Phenomenological Research 68, no. 3 (2004):
578–601. http://dx.doi.org/10.1111/j.1933­-1592.2004.tb00367.x.
Warrender, Howard. The Political Philosophy of Hobbes: His Theory of Obligation.
Oxford: Oxford University Press, 1957.
Watkins, J.W.N. Hobbes’s System of Ideas, 2nd ed. London: Hutchinson, 1973.
Weinrib, Ernest. “Law as Idea of Reason.” In Williams, Essays on Kant’s
Political Philosophy, 15–49.
180 Bibliography

White Beck, Lewis. Commentary on Kant’s Critique of Practical Reason. Chicago:

University of Chicago Press, 1960.
Williams, Howard, ed. Essays on Kant’s Political Philosophy. Cardiff: University
of Wales Press, 1992.
Zuckert, Catherine. Plato’s Philosophers: The Coherence of the Dialogues. Chicago:
University of Chicago Press, 2012.
Zuckert, Michael. The Natural Rights Republic. Notre Dame: University of
Notre Dame Press, 1996.