Fabre, Cécile - Cosmopolitan Peace-Oxford University Press (2016)

OUP CORRECTED PROOF – FINAL, 28/05/16, SPi
C O S M O P O L I TA N P E A C E
Cosmopolitan Peace
C É C I L E FA B R E
1
3
Great Clarendon Street, Oxford, OX2 6DP,
United Kingdom
Oxford University Press is a department of the University of Oxford.
It furthers the University’s objective of excellence in research, scholarship,
and education by publishing worldwide. Oxford is a registered trade mark of
Oxford University Press in the UK and in certain other countries
© Cécile Fabre 2016
The moral rights of the author have been asserted
First Edition published in 2016
Impression: 1
All rights reserved. No part of this publication may be reproduced, stored in
a retrieval system, or transmitted, in any form or by any means, without the
prior permission in writing of Oxford University Press, or as expressly permitted
by law, by licence or under terms agreed with the appropriate reprographics
rights organization. Enquiries concerning reproduction outside the scope of the
above should be sent to the Rights Department, Oxford University Press, at the
address above
You must not circulate this work in any other form
and you must impose this same condition on any acquirer
Published in the United States of America by Oxford University Press
198 Madison Avenue, New York, NY 10016, United States of America
British Library Cataloguing in Publication Data
Data available
Library of Congress Control Number: 2016931578
ISBN 978–0–19–878624–5
Printed in Great Britain by
Clays Ltd, St Ives plc
Links to third party websites are provided by Oxford in good faith and
for information only. Oxford disclaims any responsibility for the materials
contained in any third party website referenced in this work.
To David Miller,
Critic, Mentor, Friend
Acknowledgements
In a speech delivered in 1919, on Bastille Day, standing at the battlefield of Verdun,
French Premier George Clémenceau, who governed France between 1917 and
1920 and was one of the driving forces behind the Treaty of Versailles, averred that
making war is easier than making peace. I would not commit myself to the view
that writing about war is easier than writing about peace. Still, I can say without
hesitation that the latter task has proved arduous—more so than I anticipated
when embarking on that project four years ago. The relative lack of philosophical
works on peace partly accounts for that, as does the sheer range and complexity of
the issues at stakes. In any event, I would not have been able to complete this book
had it not been for the help and support I received from the many friends and
colleagues who commented in detail on draft chapters. In particular, I would like
to record my gratitude to Annalise Acorn, Saba Bazargan, Chris Bennett, David
Birks, Daniela Dover, Gideon Elford, Helen Frowe, Iason Gabriel, Guy Goodwin-
Gill, Nicola Lacey, Rae Langton, Seth Lazar, Meira Levinson, Myfanwy Lloyd,
Larry May, John McLaughlin, Theodor Meron, David Miller, Jonathan Parry,
Hanna Pickard, Joanna Rozpedowski, Amia Srinivasan, Victor Tadros, Patrick
Tomlin, George Tsai, Suzanne Uniacke, Peter Vallentyne (who sent written com-
ments on Cosmopolitan War as I was writing this book, and who helped me clarify
my thoughts on state legitimacy), and Gerald Vong.
For their oral comments, and stimulating discussions, I thank participants at
conferences and seminars at the following institutions: the Oxford Institute for the
Ethics and Law of Armed Conflict (ELAC) and the Oxford Changing Character
of War Programme, St Hilda’s College, Balliol College, the Universities of Stanford,
Princeton, Reading, Essex and Sheffield, Charles Sturt University, the Centre for
Applied Philosophy and Public Ethics in Melbourne, the Oxford Blavatnik School
of Government, and University College London.
The mechanics of book production are complex and involve a large number of
individuals. Sections of ch. 1 first appeared in ‘War’s aftermath and the ethics of
war’, in H. Frowe and S. Lazar (eds), The Oxford Handbook of Ethics and War
(Oxford: Oxford University Press, 2016). Sections of ch. 2 first appeared in ‘War
Exit’, Ethics 125 (2015): 631–52. I am grateful to my commentators and referees
for their useful feedback on those papers, and to Oxford University Press and the
University of Chicago Press for permitting me to reprint the relevant sections.
Likewise, I thank the Estate of George Sassoon for kindly permitted me to reprint
eight lines of Siegfried Sassoon’s poem, On Passing the Menin Gate, in ch. 10.
Finally, I was spared from many typographical and stylistic mistakes by Phil Dines’
expert copy-editing, while Clare Kennedy efficiently steered the book through the
process.
I have deeper and more comprehensive debts of gratitude to (the individual
members of ) a number of groups/institutions. At Oxford, I have had the
viii Acknowledgements
e xtraordinary luck of belonging to the War Workshop, the Nuffield Political Theory
Workshop and the CSSJ Seminar, whose participants have critically read and com-
mented on several chapters of the book. In the spring of 2014, Allen Buchanan
organized a two-day session on the typescript at King’s College London. My
respondents on that occasion were Alejandro Chehtman (who read ch. 7 twice,
thereby going far beyond the call of duty), Chris Finlay, Alex Leveringhaus, Laura
Valentini, David Whetham, and Lea Ypi. Their penetrating observations and crit-
icisms, together with the audience’s wonderfully responsive feedback, proved
invaluable as I began the long process of knocking the book into shape. I am
deeply thankful to them all, as well as to Allen Buchanan and Mervyn Frost, for
their unstinting intellectual generosity. A month-long stay in the Department of
Philosophy at the Australian National University in the summer of 2014 provided
me with further opportunities to improve the typescript. In Seth Lazar, Christian
Barry, and Geoff Brennan, I could not have wished for better hosts. Finally, David
Edmonds and Nigel Wharburton, co-founders of Philosophy Bites, together and
separately interviewed me about various sections of the book in the last stages of
the writing process, which I found extremely helpful.
Talking about war and peace to academics is (relatively) easy. As a civilian, and
a woman to boot, trying out this material on soldiers and cadets is harder. I have
been very fortunate to do so at the Australian Defence Academy, the US Air Force
Academy, the Ecole Supérieure Militaire de St Cyr, and the US Naval Postgraduate
College. I owe special thanks to my hosts at those various institutions, in particular
Lindsay Kirchhoff at USAFA and B. J. Strawser at the USNPC.
Those trips were partly funded by the Michael Zilkha Trust and the British
Academy; the Academy also awarded me a Mid-Career Fellowship for one year,
thereby relieving me of all teaching and administrative obligations. I am grateful to
both institutions for their financial support.
I began writing this book at Lincoln College and finished it at All Souls College.
At Lincoln, the Rector and Fellows generously allowed me to go on leave for a year.
At All Souls, itself a war memorial to English soldiers who died in the Hundred
Years War, the Warden and Fellows have given me unparalleled freedom. At both
Colleges, I have been able to engage in wide-ranging discussions with historians,
lawyers, political scientists, and philosophers on many of the issues I tackle here.
At Oxford University Press, the two anonymous readers who provided comments
on the penultimate draft have provided a raft of helpful comments, which I have
done my very best to accommodate. If the book is more accessible to non-specialists
than the penultimate draft was, it is largely thanks to their stylistic, expository, and
structural suggestions. Cheyney Ryan, and Henry Shue deserve a special mention.
Over the course of many lunches, at Merton College, or at the Chiang Mai Kitchen
in Oxford, Cheyney, Henry, and I have exchanged ideas, bounced suggestions off
each other, and despaired at the state of the world. They too have pushed me to be
less relentlessly analytical and more willing to write about the affective valence of
our moral decisions. The book is the better for it. It goes without saying that if it
is still unduly hard to read, still unduly dry, the fault is entirely mine.
Acknowledgements ix
I began thinking seriously about the ethics of war and peace about eight years
ago. As I am bringing this project to an end, I can’t help feeling nostalgic, as well
as relieved. Peter Momtchiloff, at Oxford University Press, heard my plea to turn
what at first ought to have been a single monograph into two volumes. In the
course of those eight years, he let me get on with it, yet was patient with me when
I succumbed to bouts of anxiety about the overall structure and shape of the pro-
ject. I am deeply, deeply thankful for this support throughout.
Finally, on a number of occasions, I have acknowledged in print my gratitude to
David Miller. He is not a cosmopolitan, and he does not write on war and peace.
And yet, his influence on my work, though diffuse, has been enormous—not least
because accommodating his constructive criticisms always forces me to think
harder and to be clearer. More importantly, he has been a mentor and a friend for
twenty years. I owe him far more for his constant, warm, and generous support
(both professional and personal) than I can express in words, let alone repay. This
book is dedicated to him.
All Souls College, Oxford
Remembrance Day, 2015
Contents
1. Cosmopolitanism and War 1

1.1. Introduction 1
1.2. Cosmopolitan Justice 2
1.3. Cosmopolitan War 5
1.4. Jus Post Bellum in the Just War Tradition 13
1.5. Thinking about Peace after War 17
1.5.1. Just Peace v. Justifiedatc Peace 17
1.5.2. Jus Ad Bellum, Jus In Bello, Jus Ex Bello, and Jus Post
Bellum: Mapping the Relationship 20
1.6. Overview of the Book 22
2. Ending Wars 26
2.1. Introduction 26
2.2. Ending Wars and the Question of Authority 28
2.3. Ending Just Wars 34
2.3.1. Just Cause and Reasonable Chance of Success 35
2.3.2. Proportionality 37
2.3.3. Discrimination 40
2.3.4. How to End an Initially Just War: Suing for Peace,
or Surrendering? 41
2.3.5. To Whom is the Duty Owed? 44
2.4. Ending Unjust Wars 45
2.4.1. Ending an Unjust War: Conditional v. Unconditional Surrender 45
2.4.2. Acquiring Justifications for Continuing with
an Initially Unjust War 47
2.5. Conclusion 51
3. Peacekeeping and Military Occupation 53

3.2. Peacekeeping 54
3.3. Military Occupation 63
3.3.1. Just and Unjust Wars, Just and Unjust Occupations 64
3.3.2. Who Can Occupy? The Problem of Competent Authority 74
3.3.3. Rights, Competences, and Duties of Occupying Forces 77
3.3.4. Rights and Duties of Occupied Populations 79
3.4. Conclusion 88
4. Peace Agreements 89
4.2. On the Duty to Comply with Peace Agreements 90
xii Contents
4.2.1. What Peace Agreements Do: Declarative, Specifying,
and Creative Functions 90
4.2.2. Complying with Declarative Clauses 95
4.2.3. Complying with Creative Clauses 96
4.2.4. Complying with Specifying Clauses 100
4.3. Peace Agreements and Procedural Justice 102
4.3.1. The Question of Authority 103
4.3.2. Fraud, Corruption, and Coercion 112
4.4. Conclusion 115
5. Restitution 117
5.2. Ownership, Occupancy, and Sovereignty Rights 119
5.2.1. Justifying Ownership, Occupancy, and Sovereignty Rights 119
5.2.2. Restitution: The Basics 123
5.3. Returning Moveable Property 127
5.4. Returning Territory 130
5.5. Population Displacements: Expulsions and Returns 135
5.6. Conclusion 143
6. Reparations, Distribution, and Reconstruction 145

6.2. Reparations 146
6.2.1. Reparations for What, and How Much? 147
6.2.2. Reparations to Whom? 149
6.2.3. Who Should Pay and Why? 152
6.3. The Reconstruction Principle: Filling the Gaps of Reparations,
Bringing about Distributive Justice 163
6.3.1. Setting out the Reconstruction Principle 163
6.3.2. Reconstruction and Conditional Aid 165
6.3.3. Who Owes Reconstruction Duties? 166
6.4. Conclusion 171
7. Punishment 172
7.2. Justifying Punishment 174
7.2.1. Expressivist Punishment: A First Cut 174
7.2.2. Qualifying the Theory: A Second Cut 177
7.2.3. Who May Punish? 178
7.3. War-Related Crimes 180
7.4. War Criminals 185
7.4.1. Officers, Combatants, Leaders, Citizens: The Problem
of Participation in War-Related Crimes 185
7.4.2. Three Objections 189
Contents xiii
7.5. Which Punishment? Imprisonment, Lustration, and
Punitive Sanctions 192
7.5.1. Imprisonment 192
7.5.2. Lustration 195
7.5.3. Financial Penalties: Punitive Sanctions and Confiscation 197
7.6. Who Should Punish? Defending Universal Jurisdiction 198
7.6.1. Universal Jurisdiction: A Cosmopolitan Archipelago 198
7.6.2. An Objection to Universal Jurisdiction: The Problem
of Domestic Criminal Codes 206
7.7. From Amnesties to Amnesia: Sacrificing Punitive Justice at
the Altar of Peace? 209
7.7.1. Defending Amnesties: A First Cut 209
7.8. Conclusion 217
8. Transitional Foreign Administrations 218

8.2. Mandates, Trusteeship, and Transitional Foreign Administrations 221
8.3. Justifying Transitional Foreign Administration 227
8.3.1. TFAs as a Form of Occupation 227
8.3.2. The Difference a Treaty Makes 230
8.3.3. TFAs as Morally Mandatory 233
8.3.4. Constraining TFAs 234
8.3.5. Some Loose Ends 235
8.4. The Colonialism Objection 237
8.5. Conclusion 244
9. Reconciliation 246
9.2. Reconciliation and Cosmopolitan Jus Post Bellum 247
9.3. Emotions, Political Morality, and Reconciliation 253
9.3.1. Forgiveness after War 253
9.3.2. Trust and Reconciliation 257
9.4. Restorative Justice after War: The Case of Gacaca 260
9.5. Truth and Reconciliation Commissions 264
9.6. Saying Sorry: Apologies and Regret 268
9.7. Conclusion 280
10. Remembrance 281

10.2. Memory and War Remembrance 284
10.3. Past and Present: Two Problematic Arguments 286
10.3.1. Giving the Past its Due 286
10.3.2. Strengthening Community Bonds in the Present 291
xiv Contents
10.4. Remembrance and the Duty to Promote Peace 295
10.4.1. Justifying Remembrance 295
10.4.2. Three Objections 303
10.5. Conclusion 306
Conclusion 307
The Dependence v. Independence Question 307
The Individual v. Collective Question 309
The Just Peace v. JustifiedATC Peace Question 312
The Cosmopolitanism Question 314
Works Cited 317

Index 333
1
Cosmopolitanism and War
1 . 1 . I n t ro d u c t i o n
Apparently, 43 per cent of conflicts resume within five years of their ending. Call
me a pessimist, but the fact that over half either do not resume at all, or resume
only after five years, is not a cause for rejoicing, in the light of the enormous suffer
ing which they occasion.1 Yet, the fact that making and sustaining peace tout court
is so hard does not obviate the need for trying to work out what it takes to make a
just peace. Such is my aim in this book. More precisely, I build on my book on war,
Cosmopolitan War (henceforth, CW ), to offer a cosmopolitan account of war end
ings and of justice after war (jus ex bello and jus post bellum respectively.2)
War endings and war’s aftermath are the poor relative of contemporary just war
theory. I say ‘contemporary’ because both the natural law tradition as articulated
by, inter alia, Vitoria and Grotius, and positivist international law scholars of the
eighteenth century such as Vattel and Wolff, place great emphasis on what bellig
erents may do to one another once the war is over. Kant himself, of course, writing
a century after Grotius and at the same time as Vattel, put the jus post bellum firmly
on the table with his sketch of a Perpetual Peace. That said, until the 2003 invasion
of Iraq by the US-led coalition, and notwithstanding Walzer’s remarks on jus post
bellum in his seminal Just and Unjust Wars, war ethicists have devoted far more
energy discussing the grounds upon which resorting to war is just (jus ad bellum),
and the ways in which war is fought (jus in bello) than they have given to the nor
mative issues which arise once the guns have fallen silent.
Since 2003, however, there has been a steady trickle of works on both jus ex bello
and jus post bellum—though by no means as many as one might have expected. To
be sure, there also has been an enormous literature on transitional justice, to which
I will refer in due course. Transitional justice, however, is not the same as the jus
tice of suing for peace or the aftermath of war. The norms by which central
European countries transitioned into post-Communist rule are not norms of jus
post bellum, for those countries were not at war internally or externally with the
Soviet Union. Conversely, the 1815 Treaty of Vienna is not an apt candidate for
treatment at the bar of transitional justice, for the treaty essentially carved up
1 D. Philpott, Just and Unjust Peace: An Ethic of Political Reconciliation (Oxford: Oxford University
Press, 2012), 1.
2 C. Fabre, Cosmopolitan War (Oxford: Oxford University Press, 2012). The term ‘jus ex bello’ was
coined by Darrell Moellendorf. See D. Moellendorf, ‘Jus ex Bello’, Journal of Political Philosophy 16
(2008): 123–36.
2 Cosmopolitanism and War
Europe after the Napoleonic Wars, with no consideration of reconstruction, repa

rations, punishment of this or that agent as a war criminal, and so on.3
My concern, in other words, is with just—or rather, as we shall see, all-things-
considered justified—transition from war to peace, and peace after war. Although
disagreements abound as to when and how belligerents must end hostilities and
what constitutes a just post-war state of affairs, there is some degree of consensus
in contemporary war ethics on the following points: belligerents must sue for
peace not merely once they have achieved their just war aims, but sometimes even
though they have not won the war; demands for unconditional surrender are mor
ally impermissible; victorious belligerents must aim to restore the political sover
eignty and territorial integrity of their defeated enemy; some form of compensation
for wartime wrongdoings should be paid to victims; assistance should be given to
the defeated enemy and its civilian population towards the reconstruction of their
country; wrongdoers should be put on trial; and, crucially, a stable and durable
peace should as far as possible be the overarching aim of erstwhile belligerents
when dealing with one another.4
I shall discuss and defend variants of those principles throughout the book. In
this introductory chapter, I sketch out a normative framework for my cosmopoli
tan account of post-war justice. I first offer a brief statement of what I mean by
‘cosmopolitan justice’ (s.1.2), followed by a somewhat longer summary of my cos
mopolitan account of the just war (s.1.3). With those normative foundations in
place, I outline the central tenets of the jus post bellum in the history of just war
theory (s.1.4), and explain how thinking about transitional justice and justice after
war fits within just war theory in general (s.1.5). I conclude with an overview of
the chapters (s.1.6).
1 . 2 . C o sm o p o l i ta n J u s t ice
Across its many variants, cosmopolitanism is the view that human beings are the
primary loci for moral concern and respect and have equal moral worth. It is indi
vidualist, egalitarian, and universal, and insists that the mere fact that individuals
3 For a thoughtful discussion of the relationship between transitional justice and jus post bellum, see
J. Iverson, ‘Contrasting the Normative and Historical Foundations of Transitional Justice and Jus Post
Bellum’, in J. S. Easterday, J. Iverson and C. Stahn (ed.), Jus Post Bellum (Oxford: Oxford University
Press, 2014). For an illuminating treatment of the way in which the law can be used to secure individ
uals’ rights and obligations in transitional contexts, see R. Teitel, Transitional Justice (Oxford: Oxford
University Press, 2000).
4 G. J. Bass, ‘Jus Post Bellum’, Philosophy & Public Affairs 32 (2004): 384–412; M. Evans, ‘Moral
Responsibilities and the Conflicting Demands of Jus Post Bellum’, Ethics & International Affairs 23
(2009): 147–64; M. Evans, ‘Balancing Peace, Justice and Sovereignty in Jus Post Bellum: The Case of
“Just Occupation”’, Millenium—Journal of International Studies 36 (2008): 533–54; L. May, After War
Ends—A Philosophical Perspective (Cambridge: Cambridge University Press, 2012); B. Orend, The
Morality of War (Peterborough, On.: Broadview Press, 2006); B. Orend, ‘Justice after War’, Ethics &
International Affairs 16 (2002): 43–56; B. Orend, ‘Jus Post Bellum’, Journal of Social Philosophy 31
(2000): 117–37; R. E. Williams and D. Caldwell, ‘Jus Post Bellum: Just War Theory and the Principles
of Just Peace’, International Studies Perspectives 7 (2006): 309–20.
Cosmopolitanism and War 3
belong to a particular group rather than another has no bearing on what they are
owed as a matter of justice. To be clear: my claim is not that group membership is
irrelevant to all of individuals’ rights. In fact, in some cases it is highly relevant: if
I join an association such as a running club, a church, or a trade union, I acquire
some rights vis-à-vis other members and outsiders. Rather, my point is that mem
bership in certain groups alone does not affect individuals’ human rights—rights,
that is, to the freedoms and resources they need to lead a flourishing life.5 Not only
is group membership irrelevant to the conferral of those rights; it is also irrelevant
to the imposition of the corresponding duties—so that all human beings, wherever
they are, have those rights against all other human beings, irrespective of borders.
My account of justice, in turn, is sufficientist, and holds that individuals have
rights to the resources and freedoms which they need in order to lead a flourishing
life. This implies that they should be able to frame, revise, and pursue a conception
of the good with which they identify. Correlatively, they are under duties to respect
one another’s freedoms and to provide one another with the needed resources. This
may sometimes require that they sacrifice their own interests for the sake of others,
although there are limits to the sacrifices which they can be expected so to incur:
in particular, they cannot be expected to provide assistance to others if they would
have a less than flourishing life as a result.
Note that there are two kinds of human rights. Adopting Henry Shue’s termi
nology, basic human rights protect those goods and freedoms without which we
cannot lead a life worthy of a human being, or the denial of which clearly under
mines our status as moral and rational agents to be treated on a footing of equality.6
Torture, murder, enslavement, untimely death through avoidable lack of basic
healthcare, shelter, food or clean water, constitute violations of those basic rights.
So does the denial of the political right to have a formal say over the ways in which
our political, economic, and social institutions are arranged. There are several rea
sons for conceiving of that right as basic. First, as a matter of principle, as moral
and rational agents, we have the right (I submit) not to be subject to the coercive
power of laws which affect our fundamental interests without having a formal say
in how those laws will be made. Second, as a contingent matter, having such a right
may well afford us better protection than not having it against violations of other
basic rights. This is not to occlude the fact that societies which are governed along
somewhat democratic lines have an impeccable record in that respect. Nor is it to
overlook the fact that some non-democratic societies have managed to secure basic
material provision at least as well as their democratic counterparts. It stands to
reason, however, that given a choice between being given or denied a formal say in
5 In Cosmopolitan War, I claim that a just world is one in which individuals have opportunities for
a minimally decent life. I now think that this is not demanding enough, and that justice requires that
individuals have opportunities for a flourishing life—where the word ‘flourishing’ denotes some
thing better than ‘decent’. I was prompted to revise my views by Anna Stilz’s excellent critique. See
A. Stilz, ‘Authority, Self-Determination, and Community in Cosmopolitan War’, Law and Philosophy
(2013): 1–27 and, for my reply, C. Fabre, ‘Rights, Justice and War’, Law and Philosophy 33 (2014):
391–425.
6 H. Shue, Basic Rights: Subsistance, Affluence, and U.S. Foreign Policy (Princeton, N.J.: Princeton
the design of institutions whose coercive power does affect (for the better of worse)
our fundamental interests and their protective rights, we are entitled to demand
such a say, in the form of the vote. To say ‘we will not allow you to vote, but we will
listen to your grievances and then decide what we think is in your interests’ simply
is not reassurance enough.
Non-basic human rights, by contrast, protect those goods and freedoms without
which we can live a life worthy of a human being—though not a flourishing life.
The distinction will prove important at various junctures. At this point, we can at
least say that even if scarcity of material resources makes it impossible in a given
context to do more than respect individuals’ basic human rights, justice still
requires that steps be taken towards the realization of non-basic rights. (From now
on, when I speak of ‘human rights’, I shall mean both basic and non-basic human
rights, unless otherwise stated.)
Note also that there are two ways of construing cosmopolitan duties of justice.
On what one may call the institutionalist approach, those duties are discharged by
or through political institutions such as the state, and individuals are simply under
a duty of justice to support those institutions. On what one may call the individu
alist approach, duties of justice are discharged not just by institutions but also by
individuals in their private capacity. Now, to the extent that duties not to harm
others without warrant are duties of justice, most (if not all) will agree that they are
held by all agents, whether they occupy institutional roles or act in a private capac
ity. The dispute between the institutionalist and the individualist approaches
focuses on duties of distributive justice. On the former approach, I discharge my
duty of justice by voting for a political party which will set up a just taxation sys
tem, and by paying my taxes once that party is in power. On the latter approach,
I discharge my duty by doing all of that and by making choices, in my private life,
which will mitigate inequalities or alleviate poverty.7 In this book and in line with
Cosmopolitan War (s.1.3.1), I shall take the following claims for granted. In keep
ing with the institutionalist approach, to the extent that institutions can best
achieve both relevant knowledge of which policies will best bring about justice and
the high degree of cooperation between duty-holders that the discharge of those
duties requires, individuals’ primary duty of justice is to support those institutions.
At the same time, in keeping with the individualist approach, justice is not a prop
erty only of institutions: it also governs some of the choices of individuals in their
daily life, such as the choice of whether to help someone who is clearly and imme
diately in need, the choice to take part in an unjust non-political collective ven
ture, and the choice not to take advantage of manifestly unjust policies.
Once all have the resources required for a flourishing life, I also argued, the well-
off have as a matter of right the personal prerogative to confer greater weight to
their own goals and life-projects at the expense of the less well-off. This includes a
jointly held right to form political associations, whose standing as a source of
7 This overly simplified account of a complex debate will do for my purposes here. See J. Rawls, A
Theory of Justice (Harvard: Harvard University Press, 1971); G. A. Cohen, Rescuing Justice and Equality
(Cambridge, Mass.: Harvard University Press, 2008).
special obligations depends on the degree to which they (and officials acting on
their behalf ) respect and promote the human rights of their members as well as of
outsiders. Such rights as groups and institutional actors have ultimately flow from
and supervene on the rights of individual members of those groups.
Within that framework, cosmopolitan sufficientism is compatible with some
degree of patriotic preference for the resident members of one’s political commu
nity—precisely because a just world is one in which individuals have the freedom
to associate with one another and, provided sufficiency is met, to allocate amongst
themselves the benefits which result from their mutual cooperation. To exhibit
patriotic partiality, on that view, is to act on one’s personal prerogative to give
greater weight to one’s communal goals and projects than to those of others. On
my account of cosmopolitanism, thus, membership in this or that political com
munity—for short, borders—has limited ethical relevance. It is ethically relevant,
in so far as residents of the same political community are morally permitted to give
priority to one another once they have fulfilled their distributive obligations to
outsiders; but it is relevant only in a limited way, in that it does not generate obli
gations between fellow residents which outweigh obligations to distant strangers.
The foregoing account of cosmopolitan justice goes hand in hand with the fol
lowing accounts of sovereignty rights and state legitimacy. On the one hand, sov
ereignty rights are an instrumentally valuable way to discharge general, cosmopolitan
obligations of justice and an intrinsically valuable instantiation of individuals’ per
sonal prerogative to confer greater weight on their communal interests than on
outsiders’ similar interests. Thus, the relationship between human rights and sov
ereignty is not uni-directional, whereby the former provide preconditions for the
legitimate exercise of the latter. Rather, strengthening the exercise of power by state
institutions can (under certain conditions) promote human rights.8 On the other
hand, states are legitimate—that is to say, state officials have the morally justified
Hohfeldian power to govern—if and only if, through the laws which they vote for
and enforce and the executive decisions which they make on the basis of those
laws, they respect the human rights of both state members and outsiders. Moreover,
it is precisely because those directives provide agents with better conditions for the
protection of their moral rights and the fulfilment of their moral duties than they
would have in the absence of any state or under some other feasible state that
agents have a reason for regarding those directives as binding.
1 . 3 . C o sm o p o l i ta n Wa r
So much for cosmopolitan justice. What about cosmopolitan war? In CW, I argued
that the right to wage and fight in war is a right to kill and maim, or to authorize
such acts, in defence of one’s and third-parties’ individual and jointly held human
rights. I did not make it sufficiently clear in that book that the right can be held
8 See, e.g., K. Flikschuh, ‘On the Cogency of Human Rights’, Jurisprudence 2 (2011): 17–36. My
account of state legitimacy here is reproduced almost verbatim from CW, 46–7.
either by individuals taken singly (for example, by combatants who shoot at the
enemy in defence of their life) or by individuals who each participate in a collective
venture (for example, by each member of a platoon which embarks on a coordi
nated mission.) This is not to say, however, that the right to wage and fight in war
should be regarded as irreducibly collective. Rather, it is to say that its participants’
acts of killing and maiming must be seen, not just or not only as individual acts of
self- or other-defence, but also (at least often) as contributions to that enterprise.
On that account, the fact that war is a collective enterprise remains firmly in view, but
the right to wage and fight in war is still rooted in individualistic morality, since it
protects values and interests which are constitutive of individuals’ flourishing and
is held, not by groups whose moral status is irreducible to that of their members
but by those individuals themselves.
This calls for a further remark. I include in participating agents those who, in fact,
do not (e.g.) fire a single shot, or deliberately misfire. Although they do not directly
cause enemy deaths, their participation in the war makes the latter, and its concom
itant deaths, possible. It is in that sense that they can be deemed to contribute to the
war, and it is why they are liable to defensive harm. The further question is whether
they are liable to be killed for making that kind of contribution. I am inclined to
think not, but also to say that they are liable to deliberate non-lethal harm, or to
being exposed to a risk of a foreseen though unintentional lethal harm. In other
words, they are to be treated in the same way as non-combatants whose contribu
tions to an unjust war or phases thereof warrant harming, not killing, them.
Thus, the justification for the right to kill in war rests on causal contribution to
an unjust war. If so, anyone who significantly contributes to an unjust war is liable
to be killed—not just combatants, but non-combatants as well who significantly
participate in the war effort, for example by working in munitions factories, or by
taking the community to war in their capacity as its (civilian) leaders, or by directly
funding pro-war media. Contrastingly, non-combatants who make a significant
contribution to a just war are not liable to be killed. Admittedly, the notion of
causal contribution, and the thought that causal contributions vary such as to issue
in liability to different degrees of harm, are very problematic. Significance, coun
terfactual dependence, magnitude, and causal distance will in some cases give us an
intuitively plausible account of ‘vary’, but in some cases not. The alternative is to
abandon the view that causal contribution is decisive for establishing liability. But
this seems too costly—for we do need a way to distinguish the bystander from the
threat, and it is hard to see what else other than contribution (to a harm) can do
the work. It seems, thus, that we have no choice but to endorse a plurality of inter
pretations of ‘causal contributions’. It is unsatisfactorily ad hoc, but there really is
no other way, as far as I can tell.9
9 I develop the point about non-combatants at some length in C. Fabre, ‘Guns, Food, and Liability
to Attack’, Ethics 120 (2009): 36–63; see also J. McMahan, Killing in War (Oxford: Oxford University
Press, 2009), esp. ch. 5. For a recent attack on this kind of position, see S. Lazar, Sparing Civilians
(Oxford: Oxford University Press, 2015). I am grateful to Victor Tadros for help in clarifying this
(though he would, I think, reject my ‘plural interpretation’ move).
So much, then, for the individualist framework of this book. Admittedly,

I shall use throughout phrases such as ‘A wages an unjust war against B’, ‘A does
not have the resources to do x’, etc. Those phrasings should not be taken to sug
gest that I am, in fact and willy-nilly, conceiving of war and its constitutive rights
and obligations in collectivist terms. By ‘A’, in such phrases, I do not mean ‘com
munity A as a collective whose status is not reducible to that of its members and
who, has such, wages an unjust war.’ Rather, I mean ‘those members of A who
contribute in such and such ways to the war.’ (Incidentally, on this note, I shall
refer to those members as citizens. In some cases, it would be more precise to say
‘residents’ or ‘taxpayers within A’. But given that the overwhelming majority of
residents and taxpayers are in fact also citizens, I use the latter word for the sake
of terminological simplicity.)
With those points in hand, we are in a better position to see whether the ethics
of self- and other-defence in contexts other than war differs from the ethics of self-
and other-defence in war. It is my contention (though I shall not defend it here)
that the imposition of defensive harm is governed by the same norms in both
contexts. However, one’s conception of those norms should appropriately reflect
the specific features of those different contexts. Thus, what counts as a just cause
for defensive harm, whether a particular kind of defensive harm is necessary, or
indeed proportionate, is likely to vary across those contexts—not least because war
involves high numbers of agents and thus high numbers of both simultaneous and
diachronic threats.
Let me elaborate by reference to the act of killing—though what I say here
applies to harm imposition in general. Much if not most of recent just war theory
revolves around this very question—whether war violence is sui generis norma
tively speaking. Extant debates tend to conflate two questions: (a) is war violence
sui generis? (b) is combatants’ right to kill in war justified in the same way as an
individual’s right to kill outside war? If one answers ‘yes’ to the second question,
one is thereby committed to answering ‘no’ to the former. However, one may
consistently answer ‘yes’ to the first, and ‘no’ to the second. For, in the latter case,
one may hold that combatants, qua combatants, take part in a collective enter
prise (war), and that justifying their individual acts of killing requires that one
should take into account the fact that they act jointly with other combatants. At
the same time, one may hold that the norms governing combatants’ acts of killing
are different from the norms governing acts of killing carried out by collective
agents outside war, such as the Mafia. Conversely, one may also consistently hold
that acts of killing in war are justified on the same grounds and constrained by the
same conditions as acts of killing outside war (thus, ‘no’ to the first question), and
that when justifying such acts (and thus specifying those norms) one must con
sider whether the agent acted on his own when faced with a lone threat, as part of
a group, etc. The latter position is the one I adopt throughout the book. It is an
instance of what is now standardly called the reductive strategy for thinking about
war. The reductive strategy construes the right to kill in war as a right held by
individuals and not collectives; but it does not reduce all acts of killing in war to
acts carried about by individuals each facing a lone threat and each acting on their
own.10
In both contexts, however, an agent’s group membership (into a family, a tribe,
a nation) is irrelevant to his or her designation as a legitimate target of defensive
harm. This is of a piece with the central tenets of cosmopolitan justice: just as
group membership per se does not affect the conferring of human rights and duties
at the bar of justice, it does not affect the loss of those rights either. More precisely,
in the context of war, combatants lose their right not to be harmed, not in virtue
of their membership in a community which happens to be at war, but in virtue of
unjustifiably subjecting third parties to a wrongful threat of harm. On that view,
the moral status of the war ad bellum as just or unjust, together with the signifi
cance and relevance of individuals’ contributions to the war, does have a bearing
on whether they have the right to kill enemy combatants or, conversely, whether
they have lost their right not to be killed by them. Combatants who participate
and fight in an unjust war act wrongly and lose their right not to be killed—or, as
I shall sometimes say, are liable to be killed—whereas combatants who participate
and fight in a just war do retain that right. In other words, and pace the so-called
orthodox account of war as deployed by, e.g., Michael Walzer, the acts of killing
committed by combatants in prosecution of an unjust war are prima facie wrongful
whilst those committed by combatants in prosecution of a just war are prima facie
justified.11 Finally, combatants have the right to kill enemy combatants only if the
latter unjustifiably pose or contribute to posing a wrongful threat of grievous harm
to third parties (otherwise put, unjustifiably use lethal force in violation of third
parties’ human rights.) To say that they have the right so to act implies that the
targets of their defensive force have lost their right not to be killed and are under a
duty not to resist them. As we saw above, however, enemy combatants sometimes
do not meet the requisite degree of contributory responsibility for rights violations
and thus have not lost their right not to be killed, but it is nevertheless permissible
either deliberately to inflict on them harm lesser than death, or foreseeably though
unintentionally to kill them in prosecution of one’s just war. By the same token, it
is sometimes permissible foreseeably and unintentionally to kill enemy innocent
non-combatants—though not (crucially) on the mere grounds that they belong to
10 This paragraph expands on CW, 55. In this respect, my variant of reductivism differs from,
perhaps, McMahan’s. See, especially, J. McMahan, ‘War as Self-Defense’, Ethics & International Affairs
18 (2004): 75–80; and McMahan, Killing in War, 156.
11 See M. Walzer, Just and Unjust Wars—A Moral Argument with Historical Illustrations, 4th edn.
(New York: Basic Books, 2006). I make that point in CW s.2.3.2. For classic statements of that view
(modulo some differences which need not detain us here), see C. A. J. Coady, Morality and Political
Violence (Cambridge: Cambridge University Press, 2008); McMahan, Killing in War; D. Rodin, War
and Self-Defense (Oxford: Clarendon Press, 2002). For an excellent collection of essays on the debate
between the orthodox and the neoclassical accounts, see D. Rodin and H. Shue (ed.), Just and Unjust
Warriors (Oxford: Oxford University Press, 2008). The neoclassical account is rooted in the theory of
the just war which was articulated in the Renaissance by Spanish scholastic thinkers, notably Francisco
de Vitoria and Francisco Suarez. See F. Vitoria, ‘On the Law of War’, in A. Pagden and J. Lawrance
(ed.), Political Writings (Cambridge: Cambridge University Press, 1991 [1539]); F. Suarez, ‘De Bello’,
in J. B. Scott (ed.), Selections from three works of Francisco Suarez, S.J: De Triplici Virtute Theologica,
Fide, Spe, et Charitate (Oxford: Clarendon Press, 1944 [1621]).
the enemy but, rather, in justified defence of one’s personal prerogative to confer
greater weight on one’s communal projects. If there is a justification for harming
individuals who have not lost their right not to be harmed, one can say, not that
their right has been violated (for there is a justification for so acting) but, rather,
that it has been justifiably infringed.
As stated so far, my account of the right and permission to kill in war is neutral
as between three different accounts of right and wrong to which Derek Parfit draws
our attention. On the fact-relative account, an agent B is permitted to do x—in
this instance, to kill another A—just if, were we to know all the relevant facts,
killing A would be permissible. On the belief-relative account, B is permitted to
kill A just if killing A would be permitted if her relevant beliefs were true. On the
evidence-relative account, B is permitted to kill A just if killing A would be permit
ted in the light of the evidence available to B.12 Likewise, mutatis mutandis, with
the claim that B has the right to kill A; and likewise, invertatis invertandis, with the
claim that B commits a wrongdoing by killing A. Sometimes, those accounts yield
the same conclusion; often, they diverge. Suppose that A unjustifiably poses a
wrongful lethal threat to B. On a fact-relative account, B may kill A even if she
neither believes nor has evidence that A is threatening her. But on both the belief-
and the evidence-relative accounts, B may not kill A. Suppose now, by contrast and
against the facts, that B mistakenly believes, or has false evidence, that A is trying
to kill her. On the fact-relative account, B may not kill A; on the belief-relative and
evidence-relative accounts, she may. As Parfit notes, which account we should
endorse depends on whether we are asking what is the criterion for a right or
wrong act, whether we are asking whether the agent is blameworthy for so acting,
or whether we are asking what the agent may or ought to do if she does not know
the relevant facts. Tackling those issues in full is obviously beyond the scope of this
book, let alone this chapter. I shall simply posit the following theses. First, whether
an agent is permitted or has the right to harm another solely depends on the facts
of the matter, irrespective of both the agent’s beliefs and the evidence at her dis
posal. Second, whether an agent is blameworthy for so acting, when so acting is in
fact wrong, depends not on her beliefs alone but on the extent to which her beliefs
are informed by the best available evidence. Third, absent knowledge of the rele
vant facts, an agent ought to do that which would be permissible if the best avail
able evidence fitted the facts. If the best available evidence does not fit the facts, the
agent could be deemed to have acted objectively (or fact-relatively) wrongly though
she would have been subjectively justified in so acting. In this case, if her objective
wrongdoing consists in violating some other party’s right, she is liable to defensive
harm; but to the extent that she is not blameworthy for so acting, she would also
be excused (though not justified) for retaliating in her own defence. Unless other
wise stated, when I speak of agents’ post bellum right and duties, or of the rightness
or wrongness of post bellum acts, I employ a fact-relative account thereof.
12 See D. Parfit, On What Matters, vol. 1 (Oxford: Oxford University Press, 2011), 150–62. For a
clear application of the differences between those three accounts to liability to defensive harm, see
J. Quong, ‘Liability to Defensive Harm’, Philosophy & Public Affairs 40 (2012): 45–77. See also
McMahan, Killing in War, 43; 173–5.
So much, then, for the normative basis for the right and permission to kill in
war. Central to the claim that, in a given case, an agent does have that right or
permission, is the view that the end which he thereby serves is just. Whether this
is so often depends on the degree to which that end is part of a just war. By a just
war, I mean the following: (a) it has a just cause, where a just cause consists in the
violation, backed by the threat of lethal force, of some party’s human rights; (b) it
is a proportionate response to the injustice which the belligerent has suffered; (c) it
is not fought and won through the deliberate and indiscriminate targeting of inno
cent non-combatants; (d) it stands a reasonable chance of succeeding by military
means which do not breach the requirements of proportionality and discrimina
tion; and (e) it is the only way to pursue the just cause.
Those principles call for six clarificatory points. First, I use the word ‘war’ to
denote a violent conflict for political ends, where the number of victims is at least
1000 per year over the duration of the conflict, where casualties occur over several
months, and where both parties are genuinely in a position to inflict harm on each
other. This definition (which I adopt drawing on empirical studies of civil wars,
notably the Correlates of War Project13) covers large-scale interstate wars, dec
ades-long civil conflicts, as well as attritional struggles between insurgency groups.
Many will undoubtedly object that interstate wars are a different kind of conflict
altogether, subject as they are to the international laws of war and opposing as they
do well-defined and easily identifiable belligerents and combatants. As I noted
above however, on my account, wars are subject to the same norms as (collective
and individual) interpersonal violence—hence my definition. Terminology-wise,
the word ‘belligerent’ will refer to any community (whether politically sovereign or
not) which is at war, while the word ‘combatant’ will refer to any individual who
fights on behalf or at the behest of a belligerent. Throughout the book I shall gen
erally refer to soldiers as combatants, largely because the word ‘soldier’ usually
denotes someone who wears the uniform of a regular state army, to the exclusion
of irregular fighters. The word ‘combatant’ encompasses both and, as my argu
ments apply to both, works better for my purposes.
Second, although the set of principles which are thought to govern war are
standardly classified under the labels jus ad bellum, in bello, jus ex bello, and jus post
bellum, those labels themselves are simply a convenient way of demarcating different
phases in a war.14 They do not each constitute a set of substantially different and
mutually independent normative principles. This of course is not how the tradition
has tended to conceives of those categories. Thus, at various points in the history
of the just war tradition, it has been held that the moral status of combatants’
13 See http://www.correlatesofwar.org/ (accessed on 13/01/2016).

14 In fact, some people argue that on the reductivist strategy, those labels simply do not make sense,
for the only objects of evaluation are acts of killing (or planning thereof.) See the debate on Peasoup
on my paper ‘War Exit’, which forms the basis for ch. 2 of this book. (See http://peasoup.typepad.
com/peasoup/2015/05/ethics-discussions-at-pea-soup-cecile-fabres-war-exit-with-critical-precis-by-
helen-frowe.html accessed on 13/01/2016.) I still think that it makes some sense to speak of phases of
war. Those phases do not always follow on from each other in a temporally linear fashion: a war may
often stop and be followed by a fragile peace, only to break out again. This is particularly the case in
asymmetrical conflicts marked by long, attritional periods of violence.
acts of killing in bello in no way depends on the moral status of their leaders’
decision to resort to war ad bellum, which implies that those acts and decisions are
governed by different moral norms. As we have just seen, however, this is not the
case. Likewise, the principles of proportionality and necessity rule over both
the decision to go to war and operational decisions once the war has started. And
as we shall see throughout the book, requirements of proportionality, discrimina
tion, and success apply with suitable modifications to the ending of war and to
its aftermath.
Third, the just cause requirement, once unpacked, sets out justifications for
resorting to war, whereas the other requirements delineate the conditions under
which it may be waged and fought. Other things equal, a war which is unjust for
lacking a just cause is more grievously unjust than a war which has a just cause but
fails to meet other ad bellum requirements, for the latter at least has some justifica
tion (to wit, violations of human rights backed by the threat of lethal force) even if
it does not fulfil constraining conditions on its resort. Thus, it is more unjust for A
to go to war against B as a means to seize natural resources over which it has a claim
though not one sufficiently strong to provide it with a just cause for war, even
though there are alternative diplomatic routes to solve its conflict with B, than it is
for it to go to war, in breach of necessity, when it has a sufficiently strong claim
over those natural resources.15
Fourth, by stipulating that the war should stand a reasonable chance of success,
I do not mean to suggest that a war which only stands a minimal chance is ipso
facto unjust. Suppose that there is a way of winning the war which would be pro
portionate, in the sense that the goods achieved are worth the harms caused in
their pursuit; however, there is no certainty that the war will be won in that way;
and if it does not succeed, those harms will be caused in vain. The question, then,
is whether the mere fact that there is some chance of winning renders the war jus
tified. The kind of goods which the war would bring about if successful makes a
difference to how much of a chance of success the war should stand: a rather low
threshold might be enough if the war seeks to avert a genocide, but not quite so if
the war seeks to replace a foreign dictator with a properly authorized and indige
nous regime.
Fifth, the requirement of proportionality needs closer attention in this context.
In his seminal work on killing in war, Jeff McMahan distinguishes between wide
and narrow dis/proportionality.16 Take a domestic example. Suppose that unless I
kill you, you will tread on my toes for five minutes, causing me considerable pain. If
I do kill you, your relatives will launch a vendetta against my family. The claim that
killing you is a disproportionate response might mean either that the harm I inflict
15 I defend the view that we ought to distinguish between a justification for a particular course of
action and the conditions under which that course of action may be taken in Fabre, ‘Rights, Justice
and War’. For the view that wars to recover natural resources to which one has a claim are sometimes
justified, see CW, ch. 3. That view is heavily criticized in D. Rodin, ‘The Reciprocity Theory of
Rights’, Law and Philosophy 33 (2014): 281–308; D. Statman, ‘Fabre’s Crusade for Justice: Why We
Should Not Join’, Law and Philosophy 33 (2014): 337–60.
16 McMahan, Killing in War, esp. 20–32.
on you is disproportionate in relation to the harm which you inflict on me, or that
the bad consequences of my killing you are far worse (your death and that of sev
eral of my relatives) than the good it brings about (no treading on my toes). On
the former count, my act is narrowly disproportionate; on the latter count, it is
widely disproportionate. In the context of war, acts of killing are narrowly dispro
portionate to the extent that the good they bring about (achieving the just cause)
is outweighed by the bad they occasion for those agents who are responsible for
providing us with a just cause; they are widely disproportionate to the extent that
the good they bring about is outweighed by the bad they occasion for, typically,
non-participants. To illustrate: killing combatants who (e.g.) are not even firing at
us a minute before the armistice comes into effect is narrowly disproportionate;
holding position for a week only to retreat, at the cost of destroying a whole city’s
water and electricity infrastructure, is widely disproportionate.
Sixth, it is sometimes said that war must aim to bring about peace, or (which is
slightly different) cannot be just if its occurrence and/or the way it is fought under
mine prospects for peace. In Rawls’ words, ‘the aim of a just war waged by a just
well-ordered people is a just and lasting peace among peoples, and especially with
the people’s present enemy.’17 Peace, in this context, is not merely what peace
studies scholars call negative peace, which is characterized by the mere absence of
violence. Rather, it closely resembles what those scholars call positive peace,
characterized by cooperation between erstwhile belligerents. More expansively put,
peace as I understand it here is a state of affairs where all individuals actually enjoy
their human rights to the freedoms and resources they need to lead a flourishing
life. By implication, a just peace does not merely obtain between belligerents. It
concerns all other parties who may have been wrongfully harmed by the war
though were not themselves participants in it (e.g., bystander communities and
their members whose infrastructure was damaged by mis-targeted bombs), as well
as other parties who were neither harmed by the war nor contributed to it but
whose prospects for a flourishing life partly depend on the peace settlement. For
there to be a just peace, thus, it is not enough that belligerents no longer have
a justified grievance against each other: it must also be the case that the rights
of all human beings, wherever they are, are not violated in and by the aftermath of
the war.18
Yet I have not added securing and not undermining prospects for peace as a
sixth requirement for a just war. Securing such prospects folds into the just cause
requirement, while not threatening them folds into the wide proportionality
requirement. As they constitute a just cause for war, one may say of the victims of
those violations that they are justified in resorting to war as a means to redress
17 J. Rawls, The Law of Peoples (Cambridge, Mass.: Harvard University Press, 1999), 94.
18 The founding father of peace studies, to whom the distinction between negative and positive
peace is owed, is Johan Galtung. See, e.g., J. Galtung, Essays in Peace Research (Copenhagen: Ejlers,
1975). Some claim that the just peace requirement is dangerous, on the grounds that in striving to
fulfil it belligerents are likely unnecessarily to prolong the war (see, e.g., Y. Beilin, ‘Just Peace: A
Dangerous Objective’, in P. Allan and A. Keller (ed.), What is a Just Peace? (Oxford: Oxford University
Press, 2008).
those violations—in other words, to secure peace. As for the claim that a war can
not be just if it fails to bring about a just peace, it is in fact a point about wide
proportionality: on that view, a war is unjust if its occurrence and/or the way it is
fought threatens the establishment of a just peace, precisely because the harms it
thereby causes are widely disproportionate to the good it seeks to bring about.
Suppose that A’s bellicose leaders mount an unwarranted invasion into B. B’s army
repels the invasion successfully, but its leaders decide to overthrow A’s regime, on
the grounds that only then will B be safe from future invasions at A’s hands.
Although they are correct on this particular count, the fall of that regime leads to
a murderously violent and long civil war within A. Although citizensB are safe,
citizensA are not. In this case, B’s leadership may well have succeeded in securing
peace between their compatriots and citizensA, but at a (widely) disproportionate
cost.
In any event, the requirement that war not threaten prospects for a just peace is
overly strict. For as we shall see throughout this book, it is not always possible or
desirable for belligerents to seek a just peace: sometimes, all that they can and should
hope for is a justified peace all things considered (henceforth, justifiedATC peace). A
justifiedATC peace is a compromise between ensuring that individuals’ human rights
are secure and acknowledging that realizing a just and peaceful world is simply not
possible. I shall return to this distinction in s.1.5. Suffice it to say here that a state of
peace between two erstwhile belligerents is not justifiedATC if the basic human rights
of those belligerents’ individual members are compromised. Accordingly, although
compromising on a just peace is sometimes morally justified, indeed sometimes
morally mandatory, compromising on a justifiedATC peace is not morally mandatory,
for one cannot expect of anyone to sacrifice their basic human rights—in other
words, to allow themselves to be treated in de-humanizing ways.
So much, then, for war and its normative requirements. Our concern, though,
is with peace after war. I describe my overall framework for the book in s.1.6.
Beforehand, however, I offer a brief account of the just post bellum in the just war
tradition.
1.4. J us P ost B ellum i n t he J u s t Wa r T r a d i t i o n
The oldest known surviving peace treaty was concluded in c.1259 bc between the
King of the Hittites and Pharaoh Ramses II, following the battle of Kadesh
(though there are mentions of an older treaty, between two rival Sumerian cities,
in c.2550bc.) With neither side being able to claim a convincing victory, both
agreed to set their grievances aside, to renounce aggression as a means to settle
future disputes, to assist each other in the event of an aggression by a third party,
and to return refugees.19 The text is suffused by a commitment to peace and sets
19 The text of the treaty, two versions of which remain, can be found at http://www.languageandlaw.
org/TEXTS/CONST/KADESH.HTM For discussion of this and other ancient peace agreements,
see D. J. Bederman, International Law in Antiquity (Cambridge: Cambridge University Press, 2001),
ch. 5.
the stage for millennia of at time inchoate, at others systematic, thinking about
peace after war.
In the Western canon, it pays to go back to Cicero and, with respect to the jus
post bellum, to his admonishment in On Duties: ‘Something else that must very
much be preserved in public affairs is the justice of warfare . . . Wars, then, ought to
be undertaken for this purpose, that we may live in peace, without injustice; and
once victory has been secured, those who were not cruel or savage in warfare should
be spared.’20 If war must have peace as its aim, what belligerents do to each other
once the war is over is constrained by its consequences for peace.
Cicero’s plea for mercy for ordinary agents (those not cruel or savage) recurs in
different guises through the tradition, yet is matched by consistent support for the
punishment of the grievously guilty subject to considerations of proportionality.
In the Middle Ages, Augustine, who is widely described at the founding father of
Western just war theory, places peace at the heart of his account of war. While he
accepts that ‘the peace of the unjust’ is not worthy of the name, he also insists that
not being at war constitutes a kind of peace which we all, in fact, strive for.21
Throughout that millennium, a number of intellectual and religious movements
sought to set out rules for the recourse to war: the Peace and Truce of God move
ments in the tenth and eleventh centuries, whose values were given expression in
the Second Lateran Council of 1139; Bologna-based civil lawyers who revived
Roman jurisprudence based on their discovery of Justinian’s sixth-century Corpus
iuri civilis (twelfth century); Decretists and Decretalists who formulated a canon
law of war based on, respectively, the works of Gratian and decrees issued by Pope
Gregory IX (late twelfth–early thirteenth century); scholastic Paris-based theolo
gians (twelfth–thirteenth century); and, finally, the chivalric code. Post bellum,
property matters are central to those thinkers, and understandably so: in the
absence of regular armies, knights were paid from the spoils of war, were granted
the right to cover their losses and to get compensation for services rendered, and
could sue if none was forthcoming.22
But it is Vitoria who begins to develop an account of jus post bellum, the central
tenets of which apply to those whose war is just, or who have no reasons to believe
that it unjust. They are the following: (1) Punishment. Belligerents may punitively
execute enemy combatants once the war is over. Vitoria distinguishes between wars
against the infidels and wars against Christians: the wholesale destruction of enemy
combatants in the former case is permissible since the purpose of war is peace and
since peace cannot be hoped for with Infidels. In the latter case, proportionality is
20 Cicero, On Duties. ed. E. M. Atkins (Cambridge: Cambridge University Press, 1991 [44 BC]),
I-34, III-107.
21 See Augustine, The City of God against the Pagans. ed. R.W. Dyson (Cambridge: Cambridge
University Press, 1998 [426]), ch. 12. For an interesting discussion of Augustine’s view, see Coady,
Morality and Political Violence, 263–70.
22 Classic accounts of medieval just war theory are F. H. Russell, The Just War in the Middle Ages
(Cambridge: Cambridge University Press, 1975); J. T. Johnson, Ideology, Reason, and the Limitation of
War: Religious and Secular Concepts, 1200–1740 (Princeton, N.J.: Princeton University Press, 1975);
J. T. Johnson, Just War Tradition and the Restraint of War: A Moral and Historical Inquiry (Princeton,
N.J.: Princeton University Press, 1981).
key. (2) Restitution. The victorious party, if his war was just, is not under an obliga
tion to give restitution for plundered property: he can regard the latter as compen
sation for the costs he had to incur when waging the war. Moreover, he can extract
a tribute from the vanquished, as compensation for costs and as punishment. In
fact, he can even occupy his enemy’s land, forts, and cities, ‘to the extent necessary
for the compensation of losses’. Contrastingly, unjust belligerents are under a duty
to return what they have seized during the war. (3) Regime change. The victorious
party may sometimes remove the enemy regime, ‘when security and peace cannot
otherwise be ensured, when failure to do so would cause a dangerous threat to the
commonwealth’.23
Vitoria offers a useful analytical framework for thinking about peace after war,
distinguishing as he does between principles for just punishment, just entitle
ments, and just governance. Grotius, one of his successors in the natural law tradi
tion, defends similar principles. He too focuses firmly on what the victorious (and
just) belligerent may do to its enemy once the war is over—an emphasis which has
led some commentators to describe the emergent jus post bellum as, in fact, jus
victoriae. In this respect, early classical writers assume that one belligerent has deci
sive victory over the other and sharply distinguish between the entitlements of the
just belligerent from those (in fact, nugatory) of the unjust party. Moreover, in so
far as war seeks to redress the grievance which led to the war, the aforementioned
three principles aim at restoring, and avenging violations of, the just belligerent’s
ante bellum rights: peacemaking, in this context, does not involve the conferral of
new rights on belligerents.24 Finally, if peace requires in a given instance that the
victorious belligerent should show mercy and not insist on the full restoration of
his ante bellum rights, he must do so. As Grotius insists again and again, notably in
book III of De Jure Belli ac Pacis and echoing Vitoria, what the law of nations per
mits is one thing; what the natural law commands, which directs us towards mod
eration, is another.25
In contrast with early natural law theorists, however, Grotius places a strong
emphasis on peace treaties as an instrument for ending wars. From the claim that
faith between belligerents during negotiations is crucial to prospects for peace, to
the view that only those who have the authority to declare a public war have the
authority to end it, we find in De Jure Belli ac Pacis some of the founding blocks of
23 Vitoria, ‘On the Law of War’, Question 3, arts 5–9.

24 See S. C. Neff, ‘Conflict Termination and Peace-making in the Law of Nations: A Historical
Perspective’, in C. Stahn and J. K. Kleffner (ed.), Jus Post Bellum—Towards a Law of Transition from
Conflict to Peace (The Hague: T.M.C. Asser Press, 2008); A. Keller, ‘Justice, Peace and History: A
Reappraisal’, in P. Allan and A. Keller (ed.), What is a Just Peace? (Oxford: Oxford University Press,
2008).
25 H. Grotius, The Rights of War and Peace. ed. R. Tuck (Indianapolis, Ind.: Liberty Fund, 2005
[1625]), Bk III. For discussion see, e.g., L. May, ‘Jus Post Bellum, Grotius, and Meionexia’, in J. S.
Easterday, J. Iverson and C. Stahn (ed.), Jus Post Bellum (Oxford: Oxford University Press, 2014).
Grotius’ account of confiscation during and restitution after war was first set out in his De Jure Praede.
See H. Grotius, Commentary on the Law of Prize and Booty. ed. M. J. van Ittersum (Indianapolis:
Liberty Fund, 2006 [1606]). There is some dispute as to whether Grotius is heir to the scholastic tra
dition of Vitoria and Suarez. See, e.g., R. Tuck, The Rights of War and Peace: Political Thought and the
International Order from Grotius to Kant ((New York: Oxford University Press, 1999), ch. 3.
the modern law of treaties. Samuel Pufendorf, one of the last theorists of the nat
ural law in the modern era, and Grotius’ successors in the positivist law tradition,
notably Christian Wolff and Emerich de Vattel, further develop both points whilst
also committing themselves to Vitoria’s post bellum principle of restitution. By
implication, and as Wolff unambiguously asserts, an agreement made with a
non-sovereign power is not a treaty. Moreover, neither Vattel nor Wolff accepts
Vitoria’s punishment principle (so much so that Vattel goes as far as to counsel a
blanket amnesty for all parties). Wolff, however, allows for forcible regime change
wherever kingship is patrimonial, on the grounds that the victorious party is sov
ereign over the vanquished party. He also notes, in a somewhat deflationary man
ner, that as determining exactly who owes what to whom after the war is impossible,
a peace treaty is necessarily a compromise on what the just belligerent is owed.
Still, however imperfect the treaty, it is binding, not just on those on whose behalf
it is made, but on their successors as well.26
The territorially bounded sovereign states is to positivist international lawyers of
the late eighteenth century and onwards what the individual is to contemporary
lawyers and philosophers: the fundamental unit of analysis and main bearer of
rights and duties. In a somewhat radical departure from his contemporaries, Kant
avers in Perpetual Peace and in Metaphysics of Morals that although peace requires a
formal agreement between belligerents, a treaty between two such states simply
cannot on its own deliver peace, since peace necessarily requires that those states
unite in a worldwide federation. Outside such a federation, states are to one
another what individuals are to one another in the state of nature: namely in a state
of war, or constant threat thereof. Furthermore, each state, as a member of this
federation, must endow itself with a republican constitution, for unless those sub
ject to the law are also citizens, the head of state will regard the state as his patri
mony for him to dispose of as he wishes, and will embark on costly wars without
regard for the welfare of his subjects.
Interestingly, the claim that absent a federation states are in a state of war relative
to each other does not lead Kant to adopt the view that war is unconstrained by
moral norms. True, he grants states more extensive a right to wage war than most
of us would accept. In particular, given that absent an overarching authority states
are in a relationship of equality to one another, they may each exercise their judge
ment as to whether or not they have a just cause for war—which in effect licences
them to go to war whenever they believe that they are under threat, and irrespec
tive of evidence. That said, Kant unambiguously asserts that a state once at war is
26 See Pufendorf, On the Duty of Man and Citizen. ed. J. Tully (Cambridge: Cambridge
University Press, 1991 [1673]), Bk. II, chs. 16–17; E. d. Vattel, Le droit des gens, ou, Principes de
la loi naturelle appliques à la conduite et aux affaires des nations et des souverains. ed. A. Geouffre de
Lapradelle (Washington, D.C.: Carnegie Institute, 1916 [1758]), esp. Bk IV, ch. 2 for his discus
sion of peace treaties in general and amnesties in particular. See also C. Wolff, Jus gentium methodo
scientifica pertractatum. ed. J. H. Drake (Oxford: Clarendon Press, 1934 [1750]), esp. ch. IV and
VIII for treaties, ch. VII for restitution, regime change, and punishment. Ch. VIII, in addition,
contains a sketchy discussion of the right of just belligerents to continue with the war and of
unjust belligerents’ duty to end it, as well as of the conditions under which a just belligerent can
become unjust.
under an obligation not to use means or carry out acts which would jeopardize
prospects for peace. An unjust belligerent, on his view, is one which so conducts
itself. Moreover, once the war is over, there are limits on the conditions which the
victorious party is entitled to impose on the vanquished: no slavery, no conquest,
the return of prisoners without ransom and (interestingly) no compensation for
costs incurred in waging the war; and no punishment for acts committed during
the war (which implies that, as Kant explicitly notes, a peace treaty necessarily pro
vides for amnesties). Kant’s rationale for those provisions is straightforwardly sim
ple: in imposing those conditions, the victor would thereby imply that its
opponent’s war was unjust, and would deem itself to have waged a punitive war.
But independent, equal states simply cannot wage a punitive war against each
other, since punishment can only be meted out by a superior against an inferior. In
so acting, the victor would commit a contradiction in conception and act
immorally.27
Discussing each of those claims critically is beyond the scope of this section. In
summary and with inevitable oversimplification, the tradition construes peace as
war’s regulative ideal and addresses the following questions: whether the status of
one’s war as just or unjust has decisive bearing on one’s post bellum entitlements
and obligations; what those entitlements and obligations are; and whether they
must be couched in a peace treaty. The contemporary consensus I described in
s.1.1 draws on, e.g., Vitoria in its insistence that war crimes be punished, wrong
fully taken property be restituted, and reparations be paid. With Vattel, Wolff, and
Kant, it confers on peace agreements a legitimizing role in the search for peace. We
shall revisit these claims throughout the next nine chapters. In the remainder of
this Introduction, I offer a more precise account of what I mean by peace and
cognate concepts, and provide an overview of the book.
1.5 . Thi n ki n g A b o u t P eace A f t e r Wa r
1.5.1. Just Peace v. JustifiedATC Peace

My account of peace after war rests on the assumption that all human beings,
wherever they reside, have human rights to the freedoms and resources which they
need in order to lead a flourishing life. In s.1.3, I defined a just peace as a state of
affairs in which individuals are secure in their enjoyment of those rights. Prima
facie, widespread and sustained lethal attacks on those rights constitute a just cause
for war. As I also noted, however, it is not always possible or desirable for belligerents
to seek a just peace: sometimes, all that they can and should hope for is a justified
peace all things considered.28 Whether we demand just peace or a justifiedATC
27 I. Kant, ‘Perpetual Peace: A Philosophical Sketch’, in H. Reiss (ed.), Kant—Political Writings
(Cambridge: Cambridge University Press, 1991 [1795]); I. Kant, ‘Metaphysics of Morals’, in H. Reiss
(ed.), Kant—Political Writings (Cambridge: Cambridge University Press, 1991 [1797]), esp. §§
55–61.
28 Larry May holds the view that a just peace is the outcome of a compromise between what one is
owed in the aftermath of war stricto censu, and what may reasonably be hoped for given (e.g.) feasibility
peace is critically important. Thus, Brian Orend seemingly implies that a belliger
ent may not wage a war of intervention against a genocidal tyrant if it knows that
it will not be able to achieve punitive justice after war. He also seemingly endorses
the view that the victim of an unjust aggression may not defend itself, however
vicious the victorious invader would be, if it knows that it would not be able to
compensate enemy civilian victims for the poverty which they suffer as a result of
the destruction which it would wreak on their country.29 As F. M. Kamm shows,
both views are implausible, because they (respectively) overlook alternatives to
criminal justice and assume that one ought to sacrifice one’s basic human rights for
the sake of a third party’s not falling into poverty. It seems, on the contrary, that
belligerents may go to war even if they will not be able to achieve a just peace,
provided that they can achieve a justified peace all things considered. What they
may not do is wage a war in such a way as to jeopardize prospects for such a peace.
The key questions, then, are, first, what count as reasons for and against com
promising a just peace so long as we can get a justifiedATC peace; second, what a
justifiedATC peace minimally requires. As we shall see throughout this book, con
ceding that a justifiedATC peace, as opposed to a just peace, is (sometimes? always?)
all that we can hope for, often seems dictated by considerations of feasibility. But
the notion of feasibility itself need disambiguating. Let P be a moral principle of
the form: ‘individuals ought to do x’. What does it mean to say, or indeed to deny,
that P is feasible? In the growing literature on feasibility in political theory, two
rival accounts present themselves. Some (notably Kant) take the view that P is
feasible just if there are no practical obstacles to having or implementing P. To use
and adapt an example given in another context by G. A. Cohen, let P be the view
that a just society is one in which equality of (hedonistic) welfare obtains. P is not
feasible because implementing it requires knowledge of individuals’ relative levels
of happiness which we simply do not have the means to acquire. On this account
of feasibility, whether having or implementing P is morally desirable is an alto
gether different matter from the question of its feasibility.30
On other views, by contrast, such as defended by John Rawls, Thomas Nagel,
and David Miller, considerations of feasibility include considerations of moral
constraints, the importance of honouring values other than as embodied in the jus post bellum, etc. I
do not think that this is right. Justice—whether post bellum, ante bellum, extra bellum, and so on—pre
cisely consists in getting what one is due, irrespective of those constraints, period. I thus agree with
Mark Evans’ view that there are risks in congratulating ourselves that we have achieved justice even
though we in fact caved in in the face of those difficulties (albeit justifiably). See May, After War Ends;
May, ‘Jus Post Bellum, Grotius, and Meionexia’; M. Evans, ‘At War’s End’, in J. S. Easterday, J. Iverson
and C. Stahn (ed.), Just Post Bellum (Oxford: Oxford University Press, 2014).
29 See Orend, ‘Justice after War’, and F. M. Kamm, The Moral Target: Aiming at Right Conduct in
War and Other Conflicts (Oxford: Oxford University Press, 2012), 156–7, for the critical points adduced
here. Still, if enemy civilians will suffer grievous harm as a result of our waging war, our inability to
compensate them for such harms might well vitiate (morally speaking) our decision to resort to war.
30 I. Kant, ‘On the Common Saying: That May Be Correct in Theory but It Is No Use in Practice’,
in M. J. McGregor and A. W. Wood (ed.), Practical Philosophy (Cambridge: Cambridge University
Press, 1996 [1793]); G. A. Cohen, ‘On the Currency of Egalitarian Justice’, Ethics 99 (1989): 916–44.
For a contemporary defence of the Kantian view, see D. Estlund, ‘Utopophobia’, Philosophy & Public
Affairs 42 (2014): 113–34.
desirability. Even if we had the means to acquire knowledge of all individuals’ rela
tive levels of happiness at all times, to do so would be (in Cohen’s own words)
intolerably intrusive of their privacy: the moral costs of implementing P are so high
that P is simply not feasible. Likewise, and to borrow an example from David Miller,
assuming for the sake of argument that doubling the price of petrol is a requirement
of environmental and intergenerational justice, there is currently so little support
for such a policy that implementing it is simply not feasible, in the sense that its
implementation would be undemocratic.31 Note that constraints which are often
presented as practical often are moral as well. For example, a practical constraint
may lead us to desist from implementing P on the moral grounds that implement
ing P is morally too costly. In any event, whichever account of feasibility one
endorses, it is important to bear in mind the distinction between direct moral con
straints—of the form, ‘implementing P undermines the value of privacy’ (in Cohen’s
equality of welfare example) or ‘implementing P is undemocratic’ (in Miller’s fuel
prices example)—and morally inflected costly practical constraints.
The distinction just drawn between moralized and non-moralized accounts of
feasibility is crucial to properly understanding (un)feasibility claims. But it must
be supplemented by another distinction, between so-called hard and soft con
straints.32 Hard constraints include logical consistency, consistency with the his
tory of the world to date, and compatibility with physical laws. Soft constraints,
typically, are facts about human psychology and in particular agents’ motivation to
comply (or not) with moral demands, cultural contexts, agents’ extant beliefs and
socio-economic circumstances, the availability of financial resources, and lack of
knowledge.
One might wonder what hangs on the adoption of a moralized or non-moral
ized conception of feasibility. For what matters, one might think, is whether nor
mative principles should be implemented all things considered, and what
considerations underpin our conclusion to that effect or to the contrary. Some of
those considerations are purely practical, others moral—whether they all fall under
the heading of ‘feasible’ is irrelevant. I have some sympathy with that view. All the
same, the notion of feasibility has become so pervasive, ambiguous as it is, that
abandoning it altogether might not be helpful.
As we shall see throughout, obstacles to a just peace are epistemic constraints,
resource constraints (whether the resources in question are financial, time-related,
or human costs), and agents’ unwillingness to accept just terms combined with
their resolve to resume war if those terms are nevertheless imposed on them. With
those points in hand, we are in a better position to see what a justifiedATC peace
31 D. Miller, ‘Political Philosophy for Earthlings’, in D. Leopold and M. Stears (ed.), Political
Theory: Methods and Approaches (Oxford: Oxford University Press, 2008); T. Nagel, ‘What Makes a
Political Theory Utopian?’, Social Research 56 (1989): 903–20; J. Rawls, Justice as Fairness: A
Restatement. ed. E. Kelly (Cambridge, Mass.: Harvard University Press, 2001). See also J. Räikkä, ‘The
Feasibility Condition in Political Theory’, Journal of Political Philosophy 6 (1998): 27–40.
32 For recent discussions, see, e.g., P. Gilabert and H. Lawford-Smith, ‘Political Feasibility: A
Conceptual Exploration’, Political Studies 60 (2012): 809–25; M. Jensen, ‘The Limits of Practical
Possibility’, Journal of Political Philosophy 17 (2009): 168–84; H. Lawford-Smith, ‘Understanding
Political Feasibility’, Journal of Political Philosophy 21 (2013): 243–59.
consists in. It consists, at a minimum, in a state of affairs where individuals do not

enjoy all of their non-basic rights as a result of either one or more of those con
straints (which is why it is not just tout court), but where they nevertheless enjoy
their basic human rights (failing which it would not be justified ). A peace settle
ment or state of affairs whereby basic rights are not secure and/or non-basic rights
are under imminent threat of war is not a justifiedATC peace. Thus, when I say that
the adoption of this or that principle should not jeopardize prospects for a justifie
dATC peace, I mean that it should not bring about a state of affairs which falls below
the justifiedATC peace threshold. Above that minimum threshold, and below the
just peace threshold (as it were), what counts as a justifiedATC peace—more specif
ically, which non-basic rights may legitimately be sacrificed—will depend on
which constraints operate and how. Working this out is, in effect, one of the main
aims of the book.
1.5.2. Jus Ad Bellum, Jus In Bello, Jus Ex Bello, and

Jus Post Bellum: Mapping the Relationship
What constitutes a justifiedATC peace, and how one gets to it, are usually thought
to pertain to the jus ex bello and the jus post bellum. To set the stage for the argu
ments to follow, it is worth examining the relationship between jus post bellum and
jus ex bello on the one hand, and the conditions for a just war on the other hand.
As a starting point, let us turn to contemporary debates about the relationship
between jus ad bellum and in bello and see whether, and if so how, different con
ceptions of that relationship can help us understand the ways in which belligerents’
resort to and conduct in war have a bearing on their rights and duties ex bello and
post bellum. As we saw in s.1.3, on the so-called orthodox account of the morality
of war, the moral status of the war ad bellum has no bearing on combatants’ rights,
duties, and permissions once the war has started: combatants on either side of the
ad bellum divide (for example, the divide between unjustified aggression and justi
fied self-defence) have exactly the same rights and permissions, notably the permis
sion to kill enemy combatants; they are also similarly legitimate targets for one
another. As applied to the relationship between jus post bellum and the other two
jura, the orthodox account so construed holds that belligerents’ rights, permis
sions, and obligations vis-à-vis one another once the war has stopped are entirely
independent of the moral status of their war ad bellum and/or in bello. On the
so-called neoclassical account of war, by contrast, acts of killing pursuant to an ad
bellum unjust war are wrongful, unlike acts of killing pursuant to a just war. As
applied to the relationship between jus post bellum and jus ex bello and their coun
terparts, the neoclassical account would hold that failing to meet the requirements
of the latter two entails a failure to meet the requirements of the former two.
In order to assess those positions, we must draw a crucial distinction (generally
overlooked in this context) between procedural and substantive justice. In the
present context, procedural justice pertains to the conditions under which bellig
erents are deemed competent to sue for peace, and to negotiate, endorse, and
enforce a peace settlement. Substantive justice pertains to the content of that
settlement. This is a deeply important issue, for as I shall argue later on in the
book, a justifiedATC post-war world is characterized by a peace process issuing in a
peace settlement involving all parties, as opposed to the unilateral and coercive
imposition by the victor on the vanquished of a given state of affairs. A theory of
the jus ex bello and jus post bellum must therefore account for both just substance
and just process, whereas neither the jus ad bellum nor the jus in bello need do so,
since both are exclusively concerned with justifying the right or at least permission
to impose harm on others. Thus, when delineating the relationship between jus ex
bello and jus post bellum on the one hand, and jus ad bellum and in bello on the
other hand, in the light of the orthodox and neoclassical accounts of war, we must
distinguish between the following views:
Independence with respect to substance: Considerations other than as derived from
our moral assessment of resort to and conduct in war wholly determine whether
belligerents may or must sue for peace, and/or the terms which victorious belliger
ents may impose on their defeated enemy in the peace settlement.
Dependence with respect to substance: Whether belligerents may or must sue for
peace, and/or whether the terms of a peace settlement are just or unjust to them,
entirely depends on the moral status of those parties’ war ad bellum and/or in
bello.
Independence with respect to process: Belligerents are competent to sue for peace,
and/or to negotiate, endorse, and enforce peace agreements irrespective of the
moral status of their resort to and/or conduct in war.
Dependence with respect to process: Whether belligerents are competent to sue for
peace, and/or to negotiate, endorse, and enforce a peace settlement entirely
depends on the moral status of those parties’ war ad bellum and/or in bello.
Throughout this book, I shall argue that the rights and duties of belligerents (by
which I mean, here, individual members of communities at war) both ex bello and
post bellum, and at the bar of both procedural and substantive justice, are partly
determined, and in different ways, by the moral status of their resort to and con
duct in war. I should say at the outset that not all of those views are in fact held by
just war theorists—if only because, to repeat, the latter do not generally distinguish
between just substance and just process.33 That said, and as I aim to show through
out the book, keeping in view those four understandings of the relationships
between the different components of just war theory yields a rich and nuanced
account of belligerents’ rights and duties ex bello and ad bellum; it also enables us
better to see that bringing verdicts on the justice, or lack thereof, of a war ad bellum
and/or in bello to bear on ex bello and post bellum normative assessments is infinitely
more complex than bringing ad bellum verdicts to bear on in bello conclusions.
33 Orend is best known as an advocate of Dependence though he does not distinguish between
substance and process. As he famously puts it, ‘failure to meet jus ad bellum results in the automatic
failure to meet in bello and jus post bellum. Once you’re an aggressor in war, everything is lost to you
morally.’ (Orend, The Morality of War, 162.) See also Williams and Caldwell, ‘Jus Post Bellum’.
1 . 6 . Ove rview o f t he B o o k
In the remaining nine chapters, I offer a cosmopolitan account of just transitions

from war to peace, and of just and justifiedATC peace after war. In ch. 2, I examine
the conditions under which belligerents may, or indeed must, sue for peace. I argue
that this question cannot be settled solely by a verdict on the justness or unjustness
of their decision to go to war in the first instance. In particular, a belligerent which
embarks on a just war at time t1 is sometimes under a duty to sue for peace at t2
before it has achieved its ex hypothesi just war aims; conversely, a belligerent which
embarks on an unjust war at t1 sometimes acquires a justification for continuing
at t2. Moreover, when assessing whether a belligerent must exit or continue with
its war, one must pay attention to the various ways in which it might do so, and to
the goods and bads which a particular exit strategy would bring about relative to
other strategies.
In ch. 3, I assume that although hostilities have momentarily stopped, a peace
settlement has not yet been reached, and the situation remains so volatile that
the deployment of peacekeeping forces, or indeed a wholesale military occupa
tion, seems the best way to help bring belligerents to the negotiating table.
I mount a cosmopolitan argument in support of both peacekeeping and military
occupations, as follows: if belligerents, though willing to stop fighting in princi
ple, are nevertheless unable on their own durably to end violence as a precondi
tion for reaching a peace settlement, outside intervention in the form of the
deployment of peacekeeping forces is morally justified. Moreover, if a belligerent
has been so comprehensively defeated on the battlefield that its regime is no
longer able to govern, and if there is no available local actors yet with the requi
site legitimacy to succeed it, the victorious party or some properly authorized
institution may take on tasks of government and turn itself, in effect, into a
military occupier.
Suing for peace, peacekeeping, and military occupation (as I conceive of the
latter here) take place before a settlement has been reached. With those issues out
of the way, I begin my inquiry into the conditions under which a peace settlement
can be deemed just or, at least, justified all things considered. This is a difficult
question for cosmopolitans, for in so far as settlements generate special rights and
duties between their signatories, compliance might stand in the way of fulfilling
general obligations to humankind at large. A peace settlement is binding on its
parties (or so I argue) if, and only if, it meets the following two conditions: (1) the
procedural justice condition, whereby (1a) the agreement is not secured by fraud,
deception, or unjustified coercion, and (1b) parties who negotiate the agreement
are competent to do so; (2) the substantive justice condition, whereby the agree
ment’s clauses (2a) are justifiedATC vis-à-vis its parties and (2b) do not prevent the
latter from meeting their overriding obligations of justice to outsiders. In ch. 4, I
provide a justification for the claim that parties to a peace treaty are under an obli
gation to comply with its provisions, and I articulate and defend the procedural
justice condition. I argue that a belligerent’s standing as a peace negotiator is partly
shaped by the moral status of its war, which in turn has a bearing on the procedural
justness, or lack thereof, of the resulting peace settlement and, by implication, on

its bindingness.
As belligerents turn their mind to ending the war, attention must be paid to
restituting what has been wrongfully taken away, compensating its victims for
injuries to life and limbs, and reconstructing what has been destroyed. The princi
ples which dictate what is owed to whom by whom as a matter of right once the
war is over come under the headings of restitutive, reparative, and distributive
justice. At first sight, one might think there can be no such thing as a cosmopolitan
theory of restitution and reparations, for cosmopolitan justice pertains to individ
uals’ rights and duties vis-à-vis one another qua human beings. By contrast, resti
tutive and reparative rights and their correlative obligations are grounded in the
special relationship which unites soon-to-be-erstwhile belligerents. However,
I argue that cosmopolitan considerations shape what is owed in both restitution
and reparation, constrain the ways in which those duties must be discharged, and
must be brought to bear on the ways in which conflicts between the demands of
those different strands of justice must be solved.
I tackle restitutions in ch. 5. There, I argue that restitution of what was wrong
fully taken, though often an important component of a just peace settlement
between belligerents, not only is not always a necessary condition for it but also is,
at least sometimes, unjustified.34 I consider three cases: the restitution of publicly
and privately owned property, the return of territory, and refugees’ putative right
to return to their homeland. In all three cases, I offer an account of justifiedATC
restitution. In ch. 6, I turn to the issue of reparations for war wrongdoings. I argue
that they are a fundamental component of a just peace settlement between bellig
erents but that, in the light of resources and epistemic constraints, they have to be
partly replaced with non-reparative principles for the disbursement of reconstruc
tion help. In addition, outsiders, who did not take part in the war, may have to
assist reconstruction efforts. Taken together, those principles constitute a compro
mise on a just peace, towards a justifiedATC peace.
To claim that agents who commit wartime wrongdoings are liable to reparative
burdens vis-à-vis their victims is one thing. Whether they are liable to punitive
burdens is another matter. That some of those agents are thought to be liable to
punishment for war-related crimes and crimes against humanity committed in the
course of a war is generally beyond dispute, as witnessed by, e.g., the Nuremberg
and Tokyo trials, and the establishment of the International Criminal Tribunal for
the former Yugoslavia (ICTY) and of the International Criminal Court (ICC). In
ch. 7, I offer a cosmopolitan account of punishment for war-related crimes. I iden
tify war-wrongdoings which warrant punishment, and argue that citizens and indi
vidual combatants can sometimes be held accountable for those crimes—though
the punishment which they are liable to receive must fit both their individual act
of participation in war-related crimes and the fact that what they are participating
34 The maxim found its best-known articulation in US State Secretary Stimson’s warning to Japan
that the US would simply not recognize the latter’s 1931 forcible annexation of Manchuria. For an
exposition of the doctrine, see S. C. Neff, War and the Law of Nations: A General History (Cambridge:
Cambridge University Press, 2005), 295–6.
in is, precisely, a war which, by its very nature, is a collective enterprise. I also
defend a revised version of the principle of universal jurisdiction whereby human
ity at large, via international institutions such as the international criminal tribu
nal, has primary jurisdiction over war-related crimes, and whereby third-party
states sometimes have the right to punish war criminals even though they are not
connected by nationality or territory to those crimes. As will be clear throughout
that chapter, the principle that war criminals should be punished is subject to the
requirement that meting out punitive justice should not jeopardize prospects for a
justifiedATC peace. In many cases, it may well do so—particularly when the threat
of punishment would lead combatants and their leaders to derail the peace process.
Under such circumstances, I shall argue, amnesties are sometimes morally justified
albeit at the cost of punitive justice.
Those four chapters offer an account of the key requirements of a just peace and,
when it cannot be achieved, of the key features of a justifiedATC peace—both pro
cedural and substantive. But unless those requirements are properly enforced and
adhered to, prospects for a justifiedATC peace are rather poor. In chs. 8 and 9, I turn
to two different kinds of processes and institutions which, under some circum
stances, help enemies work towards peace. The first pertains to the mechanisms
through which punitive, reparative, and reconstructive measures can be imple
mented and enforced. Although belligerents are sometimes able to do so even if
they have been defeated on the battlefield (viz. France in the aftermath of the
Franco-Prussian war or Germany following its defeat in 1918), often they are not,
particularly if the war was a long and brutal civil conflict, or if pre-war institutions
have collapsed as a result of the conflict. So-called transitional foreign administra
tions, of the kind established in Bosnia-Herzegovina following the 1995 Dayton
Agreement, are one way—indeed the most intrusive way—in which external actors
take it upon themselves to strengthen prospects for peace by effectively governing
war-torn communities. In ch. 8, I apply my cosmopolitan argument for peace
keeping and military occupation to this particular kind of post-war governance
and defend it from the charge of neocolonialism. I conclude that if transitional
foreign administrations are morally justified, then a fortiori so are less invasive
modes of foreign involvement in post-conflict institutional rebuilding whereby
foreign actors assist, rather than lead, local agents’ reconstruction efforts—pro
vided of course that the latter tend towards a justifiedATC peace.
Whether or not transitional foreign administrations are in order, bringing about
a justifiedATC peace as the horrors of war are fresh in everyone’s mind requires that
erstwhile enemies should be able to conceive of each other as partners in peace.
This in turn requires that they not let their grievances entirely colour their percep
tion of each other. The task is particularly arduous as many wrongdoings will go
unpunished, and many reparative and reconstructive claims will go unmet. Under
those circumstances, processes of reconciliatory justice, for which I provide a nor
mative defence in ch. 9, come into play. In that chapter, I scrutinize two c ognitive-
cum-affective attitudes which are often thought to be key to, indeed necessary
for, reconciliation—to wit, forgiveness on the one hand, and trust on the other
hand. I argue that forgiveness is not central to political reconciliation, while trust
is. I then consider some of the institutional mechanisms which have been used for
reconciliation—such as nonjudicial fora (which are often construed as mechanisms
for non-punitive, restorative justice), truth commissions, and public apologies: I
show that they can facilitate the emergence of political trust after war and, thereby,
buttress reconciliation.
Finally, in ch. 10, I end by addressing an issue on which just war theorists have
been silent, and yet which seems more important than ever as we are in full First-
World-War-centenary flood—to wit, war remembrance. Arguments in support of
the view that there are good moral reasons for remembering wars of the past usu
ally appeal to the special relationship which binds together members of the same
political community, or of political communities which were enemies and who
must now learn to live together. It would seem that a cosmopolitan will have very
little to say, qua cosmopolitan, about war remembrance. Yet my aim in that chap
ter is to offer a justification for war remembrance which transcends national and
political borders and yet is appropriately sensitive to the specific historical and
personal importance which the war has for those who remember it.
As should be clear, the book covers a rather expansive territory. But it does not
claim to be fully exhaustive. In particular, I do not address at any length the
justifiability of imposing disarmament measures on a defeated belligerent; nor
do I tackle in depth the question of which kind of institutions—national,
regional, and/or international—are best suited, morally speaking, to foster peace.
One could for example write an entire book on the ethics of the establishment of
the EU as a means to ensure that France and Germany would never again go to
war against each other; one could conduct a relevantly similar exercise with
respect to the United Nations. There are several reasons as to why I do not engage
with those issues: lack of space for a start, but also the fact that this book is a
philosophical inquiry into the moral norms which regulate, or at least ought to
regulate, the transition from war to peace and the establishment of as close to a
just post-war world as we can possibly get. Although I address matters of legal
and institutional implementation, my primary concern is with norms. To the
extent that I discuss institutions and legal regimes, I do so mainly with a view to
discerning whether they are morally justified. A normative account of the EU, or
the UN, or of the implementation of coercively imposed disarmament measures
would take me far down an institutionalist road which I am not well equipped
to explore in depth.
Let me close with a remark on methodology. Although my arguments are una
shamedly analytical and my conclusions resolutely normative, I make liberal use of
historical and/or contemporary cases—in keeping with the approach taken in
Cosmopolitan War. As I did therein, I avail myself of relevant sources in political
science, international relations, history, and law. This partly explains why the book
is rather long. I ask you, my reader, to forgive me for perhaps taxing your patience.
I can only entertain the hope that my excursions in, e.g., the details of the 1919
Versailles Treaty, the post-First Gulf War UN Compensation Commission, the
Nuremberg trials, and collective apologies as issued by various heads of states, liven
up what would otherwise be an overly abstract piece of writing.
2
Ending Wars
2 . 1 . I N T RO D U C T I O N
One of the most often heard criticisms levelled at belligerents in policy circles and
the press is that they should have ended the war earlier, or later, than they did.
Thus, it is sometimes said of the US that they should have withdrawn their troops
from Vietnam as soon as they realized in the course of 1968 that they would never
succeed in overthrowing the Communists, instead of waiting until 1973. Likewise,
the American administration was criticized by many for not withdrawing US
troops from Iraq as soon as Saddam Hussein’s regime had collapsed; at the same
time it was urged by others to maintain a military presence in Iraq and fight with
insurgent forces until prospects for lasting peace and order were more secure. In a
related vein, senior French military staff objected to their government’s decision, in
1961, to end the Franco-Algerian conflict, on the grounds that the French army,
far from facing defeat, was in fact winning the war. And of course, whether or not
sovereign political communities and terrorist groups at war with one another
should stop fighting and instead engage in peace negotiations is an enduring ques-
tion which many such actors currently face.1
In contrast with other areas of just war theory, there is hardly any work on the
transition from war to peace, and more specifically on the ethics of war termina-
tion.2 My aim, in this chapter, is to start filling that gap. To be clear: my point is
1 For commentaries which reflect the two opposite views regarding Iraq described in this para-
graph, see, e.g., articles by George Galloway and John Bolton in The Guardian (15 December 2011).
For the view, articulated in 2011, that the US should withdraw from Afghanistan (at a time when they
still have thousands of ground troops there), see, e.g., R. W. Miller, ‘The Ethics of America’s Afghan
War’, Ethics & International Affairs 25 (2011): 103–32. For a brilliant review of some of the issues
which leaders face when considering whether to end their war, see F. C. Iklé, Every War Must End, 2nd
revised edn. (New York: Columbia University Press, 2005). For a thorough account of how terrorism
campaigns end, see A. K. Cronin, How Terrorism Ends—Understanding the Decline and Demise of
Terrorist Campaigns (Princeton: Princeton University Press, 2009).
2 With the following notable exceptions: Moellendorf, ‘Jus ex Bello’; D. Rodin, ‘Two Emerging
Issues of Jus Post Bellum: War Termination and the Liability of Soldiers for Crimes of Aggression’, in
C. Stahn and J. K. Kleffner (ed.), Jus Post Bellum—Towards a Law of Transition from Conflict to Peace
(The Hague: T.M.C. Asser Press, 2008); D. McCready, ‘Ending the War Right: Jus Post Bellum and
the Just War Tradition’, Journal of Military Ethics 8 (2009): 66–78. See also the special issue of Ethics
on war termination (vol. 125 (2015)). There is a tendency in some of the literature to merge the ques-
tion of war termination with the question of post-war rules of conduct—of which McCready’s article
is a good example. I believe that this is a mistake, and thus agree with Moellendorf and Rodin that a
different category (war exit) is needed. See also May, After War Ends, esp. 2–3 for a thoughtful discus-
sion of the difficulties inherent in establishing when a war has ended. For a concise discussion of the
Ending Wars 27
not that we need to add yet another, substantial set of principles to the other jura.
As I noted in s.1.3, those labels are simply a convenient way to demarcate various
phases in the initiation, conduct, and termination of a war, and do not have any
deep conceptual or normative significance. Rather, my point is that the ethical
issues raised by the termination of war warrant scrutiny—at the bar of the princi-
ples which govern the resort to, and conduct in, war.
Strictly speaking, a war is not legally ended until a peace treaty has been signed.
Thus, the Second World War did not end on 8 May 1945 in the West and 15
August 1945 in the East—the days Germany and Japan respectively surrendered.
Rather, the war between Germany and the Allies ended in 1990, when the Treaty
on a Final Settlement with respect to Germany was signed by the relevant parties.
The war between Japan and the US on the one hand and Japan and the USSR on
the other hand ended with the conclusions of (respectively) the Treaty of San
Francisco in 1952 and the Soviet-Japanese Joint Declaration of 1956. I shall scru-
tinize the ethics of peace agreements in ch. 4. In this chapter, meanwhile, I focus
on the decision to seek peace terms and stop fighting or what I call ‘war exit’—
though for stylistic convenience I shall sometimes use the phrase ‘to end the war’.
One might think that the normative issues raised by the cessation of hostilities
are not significantly different from those raised by their beginning. For the claim
that a belligerent is morally forbidden to go to war at t1 might seem to imply,
without further ado, that it must end its (ex hypothesi) unjust war at t2. Likewise,
the claim that a belligerent is morally permitted to go to war at t1 might seem to
imply, without further ado, that it may continue to prosecute its (ex hypothesi) just
war at t2 if it has not achieved victory yet. Finally, the claim that a belligerent is
morally obliged to go to war at t1 might seem to imply, without further ado, that it
must continue at t2. To quote from General Douglas MacArthur’s Farewell Address
to the US Congress on 19 March 1951, ‘War’s very object is victory, not prolonged
indecision. In war there is no substitute for victory’.3 If the war is just, if the victory
is noble, then of course (it might seem) just belligerents may end it by continuing
to fight until they are victorious. Often however, it does seem as if there is a substi-
tute for victory, and a morally acceptable one to boot—to wit, ceasing hostilities
under terms which fall short of obtaining redress for the wrongdoings which the
just belligerent suffered, but which nevertheless minimize suffering on all sides and
give the best hope for a lasting peace. As I shall argue throughout this chapter, the
question of when belligerents may, or indeed must, end war cannot be settled
solely by a verdict on the justness or unjustness of their war at t1.
The chapter proceeds as follows. In s.2.2, I examine the question of who has the
authority to end the war and distinguish between different ways of doing so. In
law on war termination, see Y. Dinstein, War, Aggression and Self-Defence, 3rd edn. (Cambridge:
Cambridge University Press, 2001), 33–5.
On a point of Latin terminology: Moellendorf speaks of jus ex bello, Rodin of jus terminatio. If Latin
must be used, the former strikes me as more apposite, for it captures the notion of exit, which is only
one aspect of the ending of war.
3 There is a good recording and transcription of the speech at http://www.americanrhetoric.com
(accessed on 13/01/2016).
28 Ending Wars
s.2.3, I argue that a belligerent whose war is just at t1 is nevertheless sometimes

under a duty to end it at t2 even though it has not yet met its ends. In s.2.4, I
argue, conversely, that a belligerent whose war is unjust at t1 may sometimes con-
tinue to fight.
2 . 2 . E N D I N G WA R S A N D T H E Q U E S T I O N
O F AU T H O R I T Y
My account of the ethics of killing in war partakes of the ethics of using lethal
force in other contexts. And yet, war and interpersonal violence are standardly
thought to differ in many respects, not least on the question of who is deemed
competent to resort to violence as a means to settle disputes. According to tradi-
tional just war theory as it developed in the modern era, a war is just only if it is
waged by a competent authority—namely a state, coalition of states, a national
liberation movement which can plausibly claim to aspire to statehood, or indeed
a whole people rising up against a tyrant (in a so-called levée en masse). Although
the requirement of competent authority is normally thought to determine who
counts as a lawful belligerent and thus applies to the decision to resort to war, it
plays a crucial part in determining who may kill once the war has started. On the
traditional view, a lawful belligerent is a competent authority whose public decla-
ration of war or bellicose acts formally mark the beginning of hostilities; only
combatants who act on behalf of and at the behest of a lawful belligerent enjoy
the privileges of combatantcy—notably the privilege of being able to kill the
enemy without being deemed guilty of murder, and the privilege of receiving the
status of prisoner of war rather than of common criminal if captured. If, then, a
war must be waged by a competent authority and fought by combatants acting at
the behest of such an authority for it to be just, hostilities can cease justly only if
a competent authority so decides.
However, in CW (s.4.3), I argue that the requirement of competent authority
ought to be jettisoned on the following grounds. All human beings, whichever
group they belong to, have human rights to the goods and freedoms which they
need in order to lead a flourishing life. The right to kill in defence of those rights
when the latter are violated through threats of lethal force is also as a human right,
the conferral of which does not depend on its holders’ membership in this or that
group. If so, to restrict the conferral of that right to states, quasi-states, or national
liberation movements is tantamount to discriminating against those individuals on
whose behalf no such actor is able or willing to fight. In addition, in so far as states’
rights are grounded in the rights of those states’ individual members, the right to
wage war, even if it held by states, reverts back to individuals if the state is unable
or unwilling to exercise it.
Thus, on my account of the just resort to war, it is not necessary for a war to be
just that it be initiated by a competent authority. By implication, an individual
acting on his own may wage war on another party without his war being thereby
unjust. Crucially, however, individuals acting in an individual capacity do not have
Ending Wars 29
the (Hohfeldian) power, on their own, to declare the community for whom they
fight (be it their own community in the case of national defensive war or another
community in the case of a war of intervention) to be at war with the wrongdoer.
Although it may be permissible for them to commit acts of war, they do not
thereby change their compatriots’ bundles of rights, duties, privileges, immunities,
and liabilities in respect of that war—unlike a (e.g.) duly mandated head of state
who, when declaring war against another party, can be deemed to act not just on
behalf, but also at the behest, of her compatriots.
The claim that an individual may wage a just war on his own is less controversial
than it seems. To the extent that they claim to act on behalf and in defence of the
victims of rights violations, those individual belligerents must have objectively
good reasons to believe that they act with victims’ presumptive consent—particu-
larly in view of the harms which the latter are likely to incur during the war. More
precisely, the more grievous the rights violations which give rise to the just cause,
the more helpless the victims; the greater the further harms which those victims
would suffer were the war not to take place, the less important it is that putative
rescuers should be a competent authority. Conversely, the less grievous the rights
violations, the less helpless the victims; the lesser those further harms, the more
important it is that some form of authorization be secured.4
Now, if it is not a necessary condition for a war to be just that it be waged by a
competent authority so defined, does it follow that those who fight it may end it
even if they are not permitted to do so by a competent authority? Not necessarily,
for it depends on how one decides to stop fighting. Most obviously, individual
combatants might decide to mutiny against their officers or to flee the battlefield.
These are not cases which I address here: my concern is with deliberate decisions to
seek peace terms or to surrender. A surrender can itself take different forms. For
example, ordinary combatants often surrender by laying down their arms without
being ordered to do so by their leaders. Generals themselves might surrender whole
armies in one part of the front whilst their own leaders persist in thinking that the
war can be won after all. This is what happened in the closing stages of the Second
World War in Europe, when some German generals surrendered to the British and
4 Two additional remarks. First, this simple point should not belie the huge complexities raised by
the necessity of (in those cases) securing authorization for starting a war. Suppose for the sake of
argument that Britain had a just cause for war against Argentina when the latter invaded the Falklands
Islands. Suppose further, counterfactually, that Argentina would have responded with wholesale
attacks against mainland Britain in response to the latter’s self-defensive move. Suppose, finally, that
the inhabitants of the Islands would have consented to the war, but that British citizens residing in
mainland Britain would not have done so. What should the British government do in that case? We
might think that the majority principle should govern our decisions; but suppose that the inhabit-
ants of the invaded, far-flung territory are under threat of genocide. Should we go with the majority
in this particular case? I prescind from addressing those complications here (though am grateful to S.
Lazar for pressing the issue.) Second, the problem of competence is more complex in the case of a
war of humanitarian intervention than in the case of a war of self-defence. For there are two sets of
agents involved there: those for whose sake the war is waged, to wit, individuals whose human rights
are violated by their own regime, and those at whose behest it is waged, to wit, combatants’ citizenry
as a whole. To the extent that authorization is required, either to start or end the war, it must be
issued by both. This point does not affect my argument in this section but should be borne in mind
throughout.
30 Ending Wars
the Americans in the Western parts of Germany whilst others continued to fight
against the Red Army in the East in order to ensure that as little of German terri-
tory (and thus prisoners of wars and civilians) as possible fell into the latter’s
hands.5 More typically, the leaders of a community at war might decide to surren-
der formally, as a result of which the community itself ceases hostilities. Thus,
Marshall Jodl signed Germany’s act of surrender on 8 May 1945 on behalf of
Admiral Dönitz, Hitler’s successor as Head of State; Japan’s War Cabinet, led by
Emperor Hirohito, surrendered to the United States on 10 August of that year.
Finally, a choice might have to be made between an unconditional surrender and
a surrender with conditions.
If for whatever reasons combatants—whether rank-and-file combatants or
their commanding officers—lack a just cause for continuing with the war (more
on which below) there is no reason to regard their act of local surrender as unjust.
Pace the 1955 US Code of Military Conduct, which prescribes (in article II) that
American combatants should never surrender of their own free will, surrender
can in fact sometimes be just. Similarly, if an entire army lays down its arms or
withdraws from the war either without having been ordered to do so by its com-
munity’s leaders, or having been ordered to continue, its act of surrender is not
rendered unjust simply because it lacks de jure authority to do so. To claim oth-
erwise is to imply that it would have been unjust on the part of Germany’s mili-
tary leaders to surrender to the Allies without Hitler’s, or subsequently Dönitz’,
agreement—which is deeply counterintuitive. The reason why that is so is not
that Hitler’s rule and orders were illegitimate (though they clearly were). Rather,
this is because the cause pursued by Germany was such as to warrant putting an
end to the war irrespective of who actually took the decision to surrender (though
as we shall see in s.2.4.1 German generals did have a case for continuing to fight
in the East.)
In the light of those points, it might seem natural to assume that if the cause is
just, individual combatants ought not to stop the war without being authorized to
do so by those on whose behalf they fight, particularly if the latter are subject to
grievous rights violations. Put differently, it might seem that grievous rights-viola-
tions such as genocidal killing and mass rapes allow individual combatants to pre-
sume, without being formally authorized to act, that victims consent to the war,
but by that token forbid them to stop fighting unless formally authorized to do so.
But this is not so. For a start, the conditions under which grievous rights violations
take place—typically, generalized chaos, terror campaigns, institutional vacuum,
5 For vivid accounts of the chaotic circumstances under which the German army gradually surren-
dered, see, e.g., G. L. Weinberg, A World at Arms: A Global History of World War II (Cambridge:
Cambridge University Press, 1994); I. Kershaw, The End: Hitler’s Germany, 1944–45 (London: Allen
Lane, 2011), 826ff; A. Beevor, The Fall of Berlin 1945 (London: Penguin, 2002); P. Kecskemeti,
Strategic Surrender: The Politics of Victory and Defeat (Stanford: Standford University Press, 1958). For
illuminating essays on surrender throughout history, see H. Afflerbach and H. Strachan (ed.), How
Fighting Ends: A History of Surrender (Oxford: Oxford University Press, 2012). For a thoughtful dis-
cussion of the rights and duties of combatants who surrender and thus become prisoners of war, see
M. Walzer, ‘Prisoners of War: Does the Fight Continue after the Battle?’, The American Political Science
Review 63 (1969): 777–86.
Ending Wars 31
or at the very least institutional unresponsiveness—are not propitious to formal

authorization. This is partly why individual combatants ought not to be placed
under an obligation to seek such authorization—whether to start the war or to end
it. More importantly, the claim that individual combatants are under an obligation
to continue to fight unless authorized to desist implies that they are under a duty
to expose themselves to the harms of war, such as incapacitating injury, psycholog-
ical trauma, and death. As I claimed in ch. 1, however, individuals generally cannot
be expected to provide assistance to others if they would have a less than flourish-
ing life as a result. By implication, combatants, whether acting in an individual
capacity or formally authorized to fight by their government, cannot be expected
to continue to fight purely for the sake of others if they would thereby incur the
aforementioned harms.
It is true, of course, that by taking up arms in the first instance, combatants
might have raised expectations of rescue among those individuals or indeed have
exposed them to further and greater harms at the hands of the enemy. On the first
count, ending the war might turn out to be more costly to its victims than if com-
batants had not fought in the first instance, which in turn might place combatants
under a more stringent obligation to keep fighting.6 On the second count, com-
batants might be held under a rectificatory duty to incur greater costs, in helping
those for whom they fight, than they would normally have been under an obliga-
tion to incur. However, in either case, even if combatants are under a duty not to
exit the war as and when they want it, they are not under a duty to continue it
whatever the costs to their life and limbs—and thus are permitted to end it when
those costs are unacceptably high.
Two points are worth making at this juncture. First, I am not aiming to fix a
threshold above which the costs to combatants become so high that the latter are
permitted to end the war: I am not even sure how one would ago about fixing
such a threshold. Second, one might think that combatants are paid to incur very
high risks to their life and limbs, and that in refusing to continue with the war on
grounds of personal safety they are derelict in their contractual duties to the bene-
ficiaries of their protective undertaking—in just the same way as a bodyguard
who voluntarily contracted to protect another party would be derelict in his con-
tractual duties if he refused to take the bullet as it is fired, thereby letting his
charge die.7 I do see the force of both the point and the example. But I still think
that there are limits to the sacrifices which combatants (and indeed bodyguards)
can be reasonably expected to make even in the presence of a prior agreement,
and even if they can generally be expected to take higher risks in virtue of such
agreement.
That said, we must further distinguish the act by which armed forces, whether
as a whole or in part, surrender or withdraw from the battlefield, from the act by
6 For the view that extricating oneself from a course of action can be unacceptably costly to third
parties, see C. A. J. Coady, ‘Escaping from the Bomb: Immoral Deterrence and the Problem of
Extrication’, in H. Shue (ed.), Nuclear Deterrence and Moral Restraint (Cambridge: Cambridge
7 I owe this example to Gabriella Blum.
32 Ending Wars
which a whole community is deemed no longer to be engaged in hostilities.

Whilst armed forces may (at least in some cases) justly surrender or simply with-
draw without the permission of their leaders, only properly authorized actors may
surrender or sue for peace. This is because negotiations over the terms of the sur-
render or of the transition from war to peace are carried out on behalf of individ-
uals who will themselves be affected, possibly adversely, by those terms. Note that
this is compatible with the foregoing argument against the requirement of com-
petent authority. I, an ordinary French citizen, may sometimes justly go to fight
on my own initiative to defend my and my compatriots’ rights from some wrong-
doer, including a foreign state, or to defend distant strangers from their own
regime: that is to say, my use of lethal force in defence of those rights is not unjust
merely in virtue of the fact that I act without authorization (though it may of
course be unjust for other reasons). I may also decide, on my own, not to pursue
this fight. However, just as I do not have the (Hohfeldian) power on my own to
deem those for whom I fight to be at war against that wrongdoer, I do not have
the power on my own to deem them no longer to be at war. Nor can I decide, on
my own, that they will not surrender or sue for peace. The reason why I cannot on
my own make such decision is that I cannot claim to speak, as a lone individual,
on behalf of those individuals.
From now on, I set aside the normative issues raised by combatants’ decision to
stop fighting, whether those decisions are reached in an unstructured way by indi-
vidual combatants acting on their own initiative or by military leaders who no
longer trust in their political masters’ judgement. Instead, I focus on wholesale
military exits and ask whether and when leaders must or may cease hostilities on
behalf of those for whom they fight. But before turning to that question, we should
note, right away, that proponents of the view that combatants’ in bello rights and
permission are to a significant extent dependent on the moral status of their war
should refrain from applying the dependence thesis to the ending of war. That is to
say, they should reject the view (which I sketched out in s.1.4) that whether belli-
gerents are competent to sue for peace, and/or to negotiate, endorse, and enforce a
peace settlement is entirely dependent on the moral status of those parties’ war ad
bellum and/or in bello: for if there is one kind of belligerents of whom we want to
say that they can sue for peace, it is unjust belligerents. At the same time, one
should not rush to embrace the view that the moral status of a belligerent’s war
makes no difference to its standing when it comes to suing for peace: sometimes it
does, when the war it has fought, and/or the way it has fought it, are so grievously
wrong as to vitiate any trust its enemy may have in its ability and willingness to,
e.g., respect its commitment to stand its troops down and approach the beginning
of peace negotiations in good faith.
Further, it is also worth examining Paul Kecskemeti’s suggestion that a victori-
ous belligerent whose war aim is the complete political defeat of its enemy is seem-
ingly presented with a dilemma. On the one hand, such a belligerent wants an
orderly surrender of its enemy’s armed forces; on the other hand, it seeks to dis-
mantle its political authorities. Yet, those authorities have to be in place in order
for an orderly and mass surrender to occur; but once the surrender has occurred,
Ending Wars 33
and once the enemy has thereby renounced its status as a belligerent, it seems that
it is no longer liable to be removed by force.8
However, the dilemma is more apparent than real, once one sees that a regime
is liable to be removed by force even if it is not formally at war—for example on
the grounds that it is guilty of gross human rights violations. If that point is cor-
rect, a victorious belligerent can stop short of forcibly removing enemy officials
from office up until the point at which the latter have ordered their armed forces
to stop fighting and to surrender, and then proceed to dismantling that regime.
Note that the requirement that actors may not bring an end to the war on behalf
of those for whom they fight unless they have some authority to do so is not tan-
tamount to insisting that only legitimate regimes can end wars. Recall that a legit-
imate regime is one whose officials have the morally justified power to govern over
a given territory; that power, in turn, is morally justified only if the regime’s direc-
tives over time respect and promote the human rights of both the country’s citizens
and outsiders who are subject to those directives. On this admittedly very demand-
ing conception of legitimacy, very few regimes, if any at all, are legitimate, since
most if not all regularly breach those rights. Moreover, many wars are civil wars, or
oppose two distinct communities only one of which is a properly authorized
regime. In many cases, thus, the actors who surrender or sue for peace on behalf of
a community are not state actors and thus cannot be deemed to respect their fellow
community members’ political rights since they have not been elected (more pre-
cisely, since the conditions are not there yet for the conduct of elections). To insist
that only legitimate regimes are properly competent to put an end to mass rights
violations as occur through war by seeking peace terms undermines the very basis
(human rights protection) on which the value of legitimacy rests.
In any event, the claim that a regime is not legitimate—in other words, that its
officials do not have the morally justified power to govern—does not imply that
none of its directives are binding. As I argue in CW (s.1.4), it could be that some
of its directives in fact enable those who are subject to them de facto to do better at
the bar of their human rights and duties than they would otherwise. This explains
why an illegitimate regime such as Stalin’s did have the (Hohfeldian) power to
organize armed resistance against Nazi Germany (though the claim that it was
illegitimate, and to the degree that it was so, also implies that Soviet citizens would
not have wronged its officials or the latter’s supporters by overthrowing it after the
Second World War). In the context of war endings, thus, a regime might be illegit-
imate but nevertheless still be able and willing to issue and enforce directives relat-
ing to the ending of the war which, under the precise circumstances, would go
further towards the respect and promotion of human rights (though fall short of
full protection) than would directives issued by other actors. If and to the extent
that those directives meet this particular condition, this regime, though illegiti-
mate, is competent to sue for peace on behalf of their community. By implication,
a belligerent A is not under an obligation to be responsive to diplomatic overtures
8 See Kecskemeti, Strategic Surrender, 24–5. His case study for this putative dilemma is Germany’s
surrender to the Allies in 1945.
34 Ending Wars
just because the latter are made by its enemy’s de facto regime B (though as we shall
see below, it might nevertheless be under a duty to do so, for separate reasons.) It
must be the case that the latter is competent in the aforementioned sense to make
those overtures in the first instance. Moreover, if B is competent to sue for peace,
then A is under a prima facie obligation to negotiate with that particular actor. To
illustrate, the Allies refused to negotiate peace terms with Nazi leaders in the clos-
ing stages of the Second World War—on the grounds that the nature of the Nazi
regime was such that its officials could not possibly be characterized as able and
willing to issue better directives in relation to such terms than others might have
been. At the same time, they also refused to negotiate with resistance movements,
even in 1944–45 when opposition to the war in general and the regime in particu-
lar was mounting, particularly within the Wehrmacht, and when those movements
had already made peace overtures. Whilst they may well have had a strong justifi-
cation for doing so (a point to which I shall return in s.2.4), their adamant com-
mitment to the wholesale military occupation of Germany and reluctance to
endow those movements with any degree of legitimacy stands in need of further
justification (an issue which I shall consider in ch. 3).9
2 . 3 . E N D I N G J U S T WA R S
Suppose, then, that A is at war against B. On what grounds, and when, must
A either surrender or approach B with a view to ceasing hostilities? It might seem
obvious at first that whilst A may continue to fight until it has achieved its war aim,
it must stop as soon as it has done so since it seemingly no longer has a justification
for fighting. On that view, thus, assuming that the defeat of Nazism was a major
war aim of the Allies, the latter were under a duty to seek an armistice with
Germany as soon as the latter’s regime was vanquished, and precisely on those
grounds. Likewise, and bearing in mind that, as I argued in CW (ch. 4), the norms
which govern civil wars are the same as those which govern interstate wars, insur-
gents who fight against their own regime with the aim of toppling it may continue
to fight until they have achieved this aim, and must stop once they have. As
Emerich de Vattel puts it of sovereign states, ‘when the sovereign has had no choice
but to go to war, for a just and important cause, he may continue with military
operations until he has attained the legitimate aim, which is to obtain justice and
to put himself in a state of security’.10 This holds, if it does at all, whether A was
justified, or more strongly was morally obliged, to go to war in the first instance.
Conversely, suppose that A’s war is unjust—and in particular, lacks a just cause.
A, it seems, may not continue to fight until it has achieved its ends. Just as its
combatants ought to lay down their arms, its leaders ought to surrender. Hitler and
9 For an interesting discussion of the German case, see Coady, Morality and Political Violence,
271–3. For historical accounts of peace overtures made by Nazi leaders which the Allies rebuffed, see,
e.g., Kecskemeti, Strategic Surrender, esp. ch. 7; and Kershaw, The End, 284–5.
10 Vattel, Le droit des gens, ou, Principes de la loi naturelle appliques á la conduite et aux affaires des
nations et des souverains, Book IV, ch. 1, §6. My translation.
Ending Wars 35
his generals were under a duty to surrender to the British in 1940, without ques-
tion. For similar reasons, the French High Command’s insistence that the French
Army was winning the war in Algeria against the FLN and that the war must thus
continue rings hollow. The thought here is that the moral status of the war has
some bearing on the moral status of its ending, such that the fact that A lacks a just
cause for going to war against B puts its under a prima facie duty to end its war here
and now, whilst the fact that it has a just cause for starting the war gives it a prima
facie permission not to end it here and now.
As I shall argue, however, the unjust belligerent’s duty to surrender or seek an
armistice, and the just belligerent’s permission not to do the latter, are only prima
facie. In Darrel Moellendorf ’s seminal words, (1) ‘it could be morally required to
end a war that initially satisfied all four of the principles of jus ad bellum even
though a victory has not been obtained. (2) It could be right to continue a war that
initially failed to satisfy any one (or more) of the . . . principles of jus ad bellum’.11 I
would also add to Moellendorf ’s points the claim that under certain conditions, a
belligerent can be morally required to end a war which it was under a duty to wage.
Now, remember that a war is just if and only if it satisfies the following five
principles: it has a just cause, where a just cause consists in the violation, backed by
the threat of lethal force, of some party’s human rights; it is a proportionate
response to the injustice which the belligerent has suffered; it is not fought and
won through the deliberate and indiscriminate targeting of innocent non-combat-
ants; it stands a reasonable chance of succeeding by military means which do not
breach the requirements of proportionality and discrimination; it is the only way
to pursue the just cause. With those principles in hand, we must distinguish
between the case where at t1, A’s war meets all five conditions and is therefore just,
and the case where at t1 A’s war fails to meet one or more of those conditions. I
tackle the first case in this section, and the second case in s.2.4.
2.3.1. Just Cause and Reasonable Chance of Success

Suppose that A’s war meets all requirements at t1, but that new and accurate infor-
mation comes to light, or new weapons technology is available to B, such that the
war no longer satisfies the requirement that the war should stand a reasonable
chance of success. Under those conditions, the war becomes unjust, and A is mor-
ally required to end it. For war kills, and a war which has no chance of successfully
pursuing its just cause is one where the deaths which it occasions are utterly point-
less. If there is a duty to end it, then, that duty is owed to those who will die from
it—to A’s combatants of course, but also to A’s citizens or neutral parties who will
be caught in the crossfire, as well as innocent enemy civilians who, precisely in so
far as they are innocent of the wrongdoings which gave A its just cause, have not
lost their right not to be killed.
11 Moellendorf, ‘Jus ex Bello’, 124; and D. Moellendorf, ‘Two Doctrines of Jus Ex Bello’, Ethics
125 (2015): 653–73. See also Rodin, ‘Two Emerging Issues of Jus Post Bellum’; D. Rodin, ‘The War
Trap: Dilemmas of jus terminatio’, Ethics 125 (2015): 674–95. Kamm, The Moral Target, ch. 1. I shall
indicate where I agree and disagree with them as I go along.
36 Ending Wars
But perhaps this is too quick. For suppose that continuing to fight might enable
A to gain something, though less than what it is entitled to. Is A entitled to con-
tinue? It depends on whether A would have had a just cause for going to war
against B for such prospects in the first instance. Suppose that B mounts a wholes-
cale and unwarranted invasion of A’s territory at t1 with a view to annexe A—a just
cause if there is one. (In the remainder of this chapter, I shall assume for ease of
exposition that A’s war against B is a war of self-defence—that A is defending the
rights of its own members to the resources and freedoms necessary for a flourishing
life. Mutatis mutandis, what I say here applies to other wars, notably wars of
humanitarian intervention.) Suppose further that, at t1, A’s defensive war meets
the other conditions for a just war, but that circumstances change in such a way
that, at t2, A will only be able to recover part, though not all, of its territory: A will
have to settle for the loss of its territory, 50 kms inland, alongside its border with
B, which represents (say) 10 per cent of its territory.12 If A would have had a just
cause for going to war against B had B’s forces only sought to capture 90 per cent
of A’s territory, then the mere fact that A no longer stands a reasonable chance of
expelling the enemy from the whole territory does not imply that it is under a duty
to stop, given that it could still gain 90 per cent by continuing. If, on the other
hand, A would make only very minimal gains by continuing to fight—so minimal
in fact that it would not have had a just cause for going to war in defence of those
gains to begin with—then it is under a prima facie duty to stop. In other words,
much depends on what we mean by ‘success’. It need not be success in obtaining
redress for the precise wrongdoing which provided its victims with a just cause for
going to war: it can be success in obtaining redress for whatever wrongdoing B
commits in the course of its war which, had it been committed before the war
started, would have been serious enough so as to provide A with a just cause for
initiating the war.
This, incidentally, raises an interesting question. Some belligerents—particu-
larly insurgent groups or national liberation movements—know that they will not
succeed if they take up arms, yet do so as a means to improve their bargaining
position in future negotiations. Are they justified in so acting? Only if that which
they seek to bargain for would itself provide them with a just cause for war in the
first instance. By that token they are under a (prima facie) obligation to end their
war if and when the enemy offers them those particular concessions (subject to no
additional just cause emerging.) Note that the justificatory bar for continuing with
the war is high: for it requires of belligerents that they desist unless they would, by
continuing, obtain gains which are by themselves a just cause.
In her recent work, Frances Kamm argues that even if a particular cause y (for
example, the defence of free speech) does not justify Victim resorting to war against
Dominant at t1, it may justify continuing to fight at t2 once the just cause x (for
example, the end of slavery) has been achieved.13 Kamm offers two arguments in
defence of that view, both of which are problematic. First, by incurring considerable
12 I borrow and adapt this example from Rodin, ‘Two Emerging Issues of Jus Post Bellum’, 55.
13 See Kamm, The Moral Target, 14ff.
Ending Wars 37
costs for the sake of defeating slavery, Victim may well have earned the permission
to continue to fight for the sake of free speech. But it is not clear why incurring
those costs would render the losses which Victim would inflict on Dominant by
continuing the war proportionate to the achievement of a cause which was not
deemed weighty enough at t1 to warrant those very same losses.
Kamm’s second (admittedly tentative) argument is that by refusing to end slav-
ery without war, Dominant has forfeited its claim that Victim not wage war on it
for the sake of a good—free speech—which it was under a duty to provide anyway.
Again, this is problematic, particularly in the light of the fact that war involves
killing. For the fact that one is in a position to use the war justly started at t1 in
defence of x as a means to achieve y does not make killing in defence of the latter
justified—particularly the killing (even in collateral damage) of innocent civilians.
Ex hypothesi, those civilians have not forfeited their right not to be killed, whether
in defence of the abolition of slavery or for the sake of free speech. The claim that
the abolition of slavery provides Victim with a just cause for going to war implies
that it provides Victim with a justification for (collaterally) killing them. But if
killing them in defence of free speech is unwarranted at t1, then surely it is also
unwarranted at t2. That their government wrongfully rejected Victim’s demand for
the abolition of slavery at t1 is neither here nor there.
2.3.2. Proportionality
A, thus, must end the war if it turns out that it will not in fact succeed in obtaining
that for which it resorted to war in the first instance, unless it can gain something
new by continuing to fight which would have provided a just cause for going to
war from the outset.
Consider now the proportionality requirement, whereby a war is just only if the
goods it brings about outweigh the bads it occasions.14 Deaths—of combatants
and non-combatants—are one such bad, and the requirement of proportionality
thus acts as a fundamental moral constraint on harming in war. Suppose that at t1,
A would be able to repel B’s invasion at the proportionate and morally acceptable
cost of 15,000 casualties—beyond which the requirement would not be met. But
A’s self-defensive campaign goes badly, so that 10,000 lives are lost for negligible
gains. At t2, further estimates yield the conclusion that an additional 8000 lives
would be lost, though victory would be achieved, if A’s leaders decided to go on.
Given that they will breach proportionality if they go on, must they desist? I have
put the example in a highly stylized way. To bring it home, however, just think
about the First World War in this way. In the summer of 1914, all parties believe
that it will be over by Christmas of that year. Within six weeks, it becomes quite
clear that it will not be over by then. Dozens of thousands of young men have
already died. But—the generals and civilian leaders say—the next big push will do
it, surely! And so on, for four years, with devastating losses on all sides. How can
14 This is a simplified way to put it. See, e.g., T. Hurka, ‘Proportionality in the Morality of War’,
Philosophy & Public Affairs 33 (2005): 34–66.
38 Ending Wars
this possibly be justified, morally, even on the assumption that the French and the
Belgians had a just cause?15
Here are two arguments to the effect that belligerents may continue, both of
which strike me as problematic. First, some might argue that the loss itself of
10,000 lives provides A with a justification for continuing. One such justification
takes a punitive form, such that A has a justification for killing combatantsB puni-
tively for their wrongful killings of A’s members. However, there are very good
reasons for resisting punitive killings, not just in one-to-one domestic contexts, but
in war contexts too. As I and others have argued elsewhere, combatants may be
grievously harmed only in virtue of unjustifiably subjecting third parties to a
wrongful threat of grievous harm and provided that harming them would block
that threat. Killing, thus, is not justified as a backward-looking retribution for past
wrongdoings. Accordingly, A would not be justified, on those grounds, in contin-
uing with its war.16
On another construal, proposed by McMahan, the claim that lost lives provide A
with a justification for not ending the war takes a ‘redemptive’ form: to the extent
that prosecuting the just cause at least partially redeems those lives, so that they have
not been sacrificed in vain, A is justified (subject to war’s other requirements) in
fighting. This is a powerful argument—and one which was consistently deployed
during the First World War. Notwithstanding its strong intuitive pull, however, the
redemption thesis faces the following dilemma. Either it holds that redemption
alone furnishes A with a new just cause, in which case it is tantamount to claiming
that combatantsA are justified in killing combatantsB even though, ex hypothesi, their
acts of killing clearly cannot be construed as forestalling a lethal threat; or it claims
that combatantsA would be justified in killing combatantsB because, and only if,
they would thereby and at the same time prosecute their just cause, in which case
the original just cause alone provides combatantsA with a justification for killing
combatantsB and redemption does not do any justificatory work.17
The other argument in favour of the claim that A may continue its war at t2
even though it would breach the proportionality requirement holds that the
10,000 lost lives ought to be regarded as sunk costs. Given that the 8000 lives
which A’s victory would cost are well within the upper limit of 15,000 lives for that
very same end, it would seem odd to require of A that it relinquish its just cause—
even though the extra 3000 lives which will be lost would have been deemed an
15 In the plethora of books generated by the First World War publishing industry, David Stevenson’s
monograph is particularly helpful. See D. Stevenson, 1914–1918: The History of the First World War
(London: Allen Lane, 2004).
16 For a devastating critique of the view that punishment is a just cause for war, see D. Luban, ‘War
as Punishment’, Philosophy & Public Affairs 39 (2011): 299–330.
17 See J. McMahan, ‘Proportionality and Time’, Ethics 125 (2015): 696–719. An anonymous ref-
eree for Ethics (where a shorter version of this chapter first appeared) suggested that redemption might
play a partial justificatory role, by allowing A to include in its proportionality calculus the value of
ensuring that its deceased combatants did not die in vain—which in turns might license A to continue
with the war. However, as we saw in ch. 1 (s. 1.3), the requirement of proportionality is not a justifi-
cation for waging or continuing war: rather, it is one of its constraining conditions.
Ending Wars 39
unacceptable cost at t1. There are some difficulties with this claim.18 Let us first
consider the fate of the 10,000 agents who die between t1 and t2. At t1, those
deaths are regarded as a bad to be weighed relative to the good the war would bring
if we initiated it, and thus count as a reason against going to war—albeit not a
sufficiently decisive one. At t2, those deaths simply do not count as a bad any more
to be weighed relative to the good the war would bring if we continued it. On this
view, thus, it would become permissible for any belligerent to inflict however many
losses it thinks are required to win its war, so long as it would be in a position
plausibly to declare, at t2, that from now on, it will manage to achieve victory at a
cost, calculated between t2 and victory, which is lesser than the upper limit set at
t1. The problem is that proportionality seems to lose most of its bite as a constraint
on killing, for only notionally does that upper limit operate as an appropriately
constraining restriction on permissible war killings.
The sunk costs view has a similarly troubling implication with respect to the fate
of the additional 3000 agents who will die if A continues with its war. At t1, they
are protected, since their deaths would exceed the upper limit of 15,000 lives. At
t2, however, they lose such protection, since it now becomes permissible for A to
continue. But then it appears that no one is protected from death above that upper
limit, so long as, at the point at which the decision to continue with the war is
made, their death together with that of those who will also die does not exceed that
upper limit. On that account too, that upper limit does not seem to be constrain-
ing enough a restriction.
Some might think that my reading of the proportionality requirement is too
restrictive, for if A desists, thereby renouncing its control over part of its territory,
it has failed to bring about any of the good which the war would have produced
whilst at the same time caused 10,000 lives to be lost. This seems morally worse,
arguably, than killing 18,000 agents, 3000 of whom we admittedly were not per-
mitted to kill at t1, but thanks to which we win the war.
But now consider the following case. At t1, we can bring about some good by
inflicting a harm of magnitude 5 (h5) onto another agent P: any harm greater than
that would be disproportionate. We try, but fail. At t2, we realize that we can
achieve our desired good end by inflicting a further harm, h2, on P—taking the
overall harm he would suffer above the proportionality threshold. Can we say that
we may proceed, on the grounds that it is worse to inflict h5 for nothing than to
subject P to h7 thereby successfully achieving our end? Can we say in other words
18 For a defence of the sunk costs view, see McMahan, ‘Proportionality and Time’. I agree with him
that if war leaders discover after initiating the war that they had unavoidably been mistaken in assess-
ing the evidence in favour of starting it, they are not blameworthy for initiating the war. But this claim
is compatible with the view that they are not objectively justified in continuing with it under the cir-
cumstances. For a point similar to the one I am making in the in-text paragraph, see Kamm, The Moral
Target, 16. For worries about sunk costs, see also Rodin, ‘Two Emerging Issues of Jus Post Bellum’; D.
Rodin, ‘Ending Wars’, Ethics & International Affairs 25 (2011): 359–67; Rodin, ‘The War Trap’;
Moellendorf, ‘Two Doctrines of Jus Ex Bello’. Moellendorf and I independently reached the same
conclusion, namely that discounting sunk costs altogether in effect destroys the requirement of
proportionality.
40 Ending Wars
that the harm P sustained at t1 ought to be seen as a sunk cost?19 I do not think so:
P would have a grievance, it seems to me. He would be justified in complaining
that if it was impermissible to subject him to h7 in one go at t1, then it is imper-
missible to subject him to that same overall harm in two steps. Admittedly, the war
case is different, for there the question is whether we may permissibly kill add-
itional and different agents at t2 whom we were not permitted to kill at t1. However,
even if those agents themselves do not have a grievance in this case, the intraper-
sonal case teaches us two lessons. First, harms sustained by the same individuals
across different phases of the war ought not to be regarded merely as sunk costs.
Second, by killing combatants, A does more than harming them: it also harms
those on whom those combatants depend, and/or to whom they are closely associ-
ated—for example, their fellow citizens who, post war, will need as many surviving
combatants as possible for reconstruction efforts; more importantly, their relatives,
whose lives will be much the worse for the loss of several family members. Those
considerations might not be particularly relevant to wars as fought by contempo-
rary Western armies, since too few of those combatants die for it to make a differ-
ence post war, and since very few families have more than one relative serving in
the armed forces. But they are acutely relevant to conflicts in which whole commu-
nities are involved to such a degree that people lose several of their relatives. A
judgement to the effect that a military campaign is proportionate must take those
wider harms into account. Accordingly, even if A does not breach proportionality
when solely construed in terms of combatants’ lives, there still is room for the view
that once account is taken of the harms accruing to third parties as a result of those
lives being lost, A ought not to regard those losses as sunk costs. To be sure, some-
times, those harms will not be such as to outweigh the harm of killing or maiming
a number of combatants for nothing, and thus as to bar A from continuing with
the war; sometimes however they will act as precisely one such constraint.
2.3.3. Discrimination
To recapitulate, we have seen that the ad bellum requirements of proportionality
and reasonable chance of success sometimes yield the conclusion that a belligerent
must prematurely end a war which it was all-things considered justified to wage in
the first instance. So does the requirement of discrimination, which insists that
combatants not target innocent civilians. Suppose, for example, that A justly starts
a war at t1 and can win its war by targeting only military objectives. At t2, however,
the enemy has procured weapons which A’s forces cannot counter, and the only
19 The worry about restrictiveness is well put by Rodin, on the basis of which he argues in ‘Ending
Wars’ that A may (sometimes) continue to fight as the lesser of two evils. Rodin, ‘Ending Wars’;
Rodin, ‘Two Emerging Issues of Jus Post Bellum’. I am very grateful to Victor Tadros for his penetrat-
ing comments in Peasoup on the Ethics version of this chapter. I owe him the intrapersonal case, which
he offered in support of his criticism of my sunk costs argument. (See http://peasoup.typepad.com/
peasoup/2015/05/ethics-discussions-at-pea-soup-cecile-fabres-war-exit-with-critical-precis-by-helen-
frowe.html accessed on 26/10/2015.)
Ending Wars 41
way for A to win its war is by deliberately targeting civilian populations. Under
those circumstances, A is under a prima facie obligation to stop.
The obligation is only prima facie however because there is one, very restricted
case where the deliberate targeting of innocent civilian populations is morally jus-
tified: where it is a necessary means to adverting a grave evil such as mass extinc-
tion, genocide, and mass enslavement. (Or so I argue in CW, s.7.2.) If targeting
would be morally justified at t1 on such grounds, then it is morally justified at t2.
2.3.4. How to End an Initially Just War: Suing

for Peace, or Surrendering?
So far, I have claimed that A, who has a just cause, is sometimes under a prima facie
obligation to exit a war which it cannot win justly—where winning means obtain-
ing full redress for the wrongdoings to which it was subject without breaching
other requirements of the just war. A further and crucially important question is
that of the means by which it must do so. Broadly speaking, A can either sue for
peace or surrender. Whether it should do the latter partly depends on the military
situation. Suppose that A, though unable to win the war, is nevertheless in a
stronger position than B: it has more combatants and military equipment at its
disposal, or it has the capacity to inflict far heavier losses on B than the latter is able
to inflict, or there is greater support for the war on its home front than on B’s, and
so on. To insist that A, who ex hypothesi is a just belligerent, ought to surrender to
B under those circumstances—in other words, to admit defeat—seems prima facie
unfair. Instead, A may offer terms for a justifiedATC peace—terms in other words
which on the one hand protect the basic human rights of its members (as well as
of innocent third parties) and on the other hand are as propitious to a just peace
over time as feasibility and desirability constraints allow. All the same, if A could
secure better terms at the bar of its members’ and third parties’ human rights by
suing for peace than by continuing to fight, it is under a duty to do so.20 If B
refuses those terms and continues to fight, thereby subjecting A’s members to the
violation, backed by lethal force, of their human rights, A may continue until B
desists, so long as its superior military strength (combined if necessary with other
legitimate means it might have at its disposal, such as economic sanctions, or par-
allel diplomatic negotiations) means that it has a reasonable chance of achieving
20 There is another reason, thus, why the French High Command’s position vis-à-vis the Algerian
war was morally dubious—namely that it implicitly relied on the traditional view that one wins or
loses a war on the battlefield and not in the corridors of power. See Iklé, Every War Must End, 52, for
a particularly effective way of putting the point. It is worth noting, in fairness to military and civilian
leaders, that discerning what constitutes a victory, or indeed a defeat, is much harder than we tend to
think—certainly harder than in the days (pre-1789 in Europe) where leaving the battlefield was ipso
facto an admission of defeat. See, e.g., D. P. Johnson and D. Tierney, Failing to Win: Perceptions of
Victory and Defeat in International Politics (Cambridge, Mass.: Harvard University Press, 2006); R.
Mandel, ‘Reassessing Victory in Warfare’, Armed Forces and Society 33 (2007): 461–96; J. Angstrom
and I. Duyvesteyn (ed.), Understanding Victory and Defeat in Contemporary War (London: Routledge,
2007). For a fascinating account of conceptions of defeat and victory pre-1789, see J. Q. Whitman,
The Verdict of Battle—The Law of Victory and the Making of Modern War (Cambridge, Mass.: Harvard
42 Ending Wars
this particular, limited objective within the constraints of the requirement of

proportionality.
Suppose now that A is losing the war from a strictly military point of view, and
that B demands a surrender. At this juncture, some commentators are likely to
draw a distinction between conditional and unconditional surrenders and claim
that A is not under a duty to accept the latter, on the grounds that a demand for
unconditional surrender is wrongful. I shall dissent from that view in s.2.4.1.
Meanwhile, let us suppose that B does not insist on unconditional surrender. To
surrender, even with conditions attached, is an admission of defeat. In so doing, A
would in effect weaken its bargaining position vis-à-vis B, which might in turn
undermine its ability to obtain some form of compensation for the wrongdoing
which it suffered at the latter’s hands, or more widely undermine its geostrategic
status on the world stage. Is A nevertheless under a duty to stop fighting?
Those considerations are routinely invoked by civilian and military leaders in
support of their decision to continue a war which they know they will not win—of
which the policy conducted by the Nixon Administration in Vietnam between
1968 and 1973 is a case in point.21 The view that A may continue to fight as a
means to avoid those costs is far less innocuous than it sounds: it implies that
avoiding termination costs provides just belligerents with a justification for delib-
erately killing enemy combatants and (more often than not) foreseeably killing
enemy civilians. I shall return to the problem of collateral damage very briefly at
the close of s.2.3.5. For now, let us focus on the killing of enemy combatants. As I
claimed in s.1.3, a war is just only if it is a response to violations backed by lethal
force of some party’s human rights to the freedoms and resources which they need
in order to lead a flourishing life. Furthermore, combatants are liable to be killed
only if they unjustifiably subject those parties to those rights violations—in other
words, only if they bear some significant degree of contributory responsibility for
those wrongdoings. Finally, killing non-liable combatants is permissible only as a
means to forestall a graver evil than their deaths themselves would be. By implica-
tion, then, A is justified in killing combatantsB as a means to avoid termination
costs only if being subject to those costs would wrongfully undermine the human
rights of those on whose behalf it fights, and if combatantsB bear a significant
degree of contributory responsibility for the latter harm. On the first count, it does
not follow from the fact that A cannot win its war anyway that termination costs
themselves will undermine citizensA’ prospects for a flourishing life. True, A might
not be able to get the compensation it deserves; it might also suffer a loss of status
on the world stage. But that alone does not mean that the fundamental rights of
A’s members would be undermined as a result. One need only think of the vastly
improved living conditions which post-war Germany witnessed, at least in the
West, soon after the end of the Second World War.
21 This view is defended in Rodin, ‘Two Emerging Issues of Jus Post Bellum’, 57. For the Vietnam
case, see, e.g., M. Atwood Lawrence, The Vietnam War—A Concise International History (Oxford:
Oxford University Press, 2008), ch. 7. Similar considerations are thought by some to inform President
Obama’s policy on Afghanistan. See, e.g., Miller, ‘The Ethics of America’s Afghan War’.
Ending Wars 43
Moreover, on the second count, even if the human rights of A’s members would
be undermined if the war were to stop, it remains to be seen whether A is justified
in killing combatantsB as a means to avoid paying termination costs. Suppose that
if A were to stop fighting, B’s leaders, and not its combatants, would impose on it
the costs of termination, via the negotiating process. On what grounds may A
decide to continue the war and in so doing kill combatantsB?
I can think of two plausible arguments to that effect. First, one might want to
say that killing enemy combatants for the sake of preserving one’s status on the
world stage is permissible even if those combatants themselves are not responsible
for the fact that one would have to incur early termination costs. Given that com-
batantsB are not causally responsible for A’s termination costs, they should be
treated in the same way as innocent enemy non-combatants. Now, in CW (s. 7.2),
I argued that one may deliberately kill innocent non-combatants only if their
death averts genocide, mass enslavement, or (wrongfully caused) starvation. If that
is correct, killing combatantsB is permissible only as a means to avert the aforemen-
tioned evils—of which losing one’s status on the world stage is not one, unless
(which I suspect is in fact unlikely) the loss for A of its international status or its
diminishing negotiating strength are overwhelmingly likely to lead to mass starva-
tion, genocide, or mass enslavement within its borders.
The second argument appeals to combatantsB’ initial wrongdoing—to wit, the
wrongdoing they committed by participating in their community’s unjust war
against A. By so acting—the argument goes—combatantsB have made themselves
liable to be killed by A, not in self-defence (since they are not at present imposing
on A the harms attendant on early termination) but, rather, in self-preservation.
Moreover, by participating now in that war, not only are they contributing to the
violation of citizensA’ human rights: they are also contributing to make A choose
between incurring the costs of an early termination and continuing to fight at the
expense of the lives and limbs of innocent persons. They, and not citizensA, should bear
the costs of exiting a war which they should not have started in the first instance:
by refusing to do so, they are making themselves liable to be killed by A.22
To recapitulate, if the costs of terminating the war are such as to impair citizensA’
prospects for a flourishing life, and if under the circumstances combatantsB either
are liable to be killed as a means for A not to incur those costs or may be killed as
the lesser of two evils, then A is not under a duty to surrender to B. Failing those
two conditions, however, A must do so.
22 For the distinction between self-defensive and self-preserving killing, see J. McMahan, ‘Self-
Defense and the Problem of the Innocent Attacker’, Ethics 104 (1994): 252–90, esp. at 256–8. For
criticisms of the forced choice argument for defensive killing, see S. Lazar, ‘Responsibility, Risk, and
Killing in Self-Defense’, Ethics 119 (2009): 699–728. One might object to my point here that A is not
forced into doing anything: it could have decided not to go to war in the first instance and submitted
to B’s demands at t1. I argue against this particular way of construing A’s predicament in C. Fabre,
‘Cosmopolitanism and Wars of Self-Defence’, in C. Fabre and S. Lazar (ed.), The Morality of Defensive
War (Oxford: Oxford University Press, 2014), esp. s.4.
44 Ending Wars
2.3.5. To Whom is the Duty Owed?

So far, I have assumed that A’s war meets all the requirements of a just war at t1.
Before addressing the case of unjust belligerents, some loose ends need tying up.
At various junctures in s.2.3, I have argued that a just belligerent is sometimes
under a duty to end its war even though it has not successfully prosecuted its just
cause. If the justification for the duty lies in the fact that it cannot continue with
the war other than by killing enemy combatants who ought not to be killed, then
the duty itself is owed to those combatants. In some cases, however, A’s duty to
exit the war is owed, not to those of combatants and citizens of B’s who contrib-
uted to its predicament in the first instance but rather to their own forces and
civilians, as well as to innocent civiliansB who would get caught in the crossfire if
the war were to continue. Vis-à-vis their fellow citizens, A’s leaders are under a
duty to choose the course of action which will minimize the harms they are incur-
ring as a result of B’s wrongdoing. However, its duty so to act is constrained by its
duty to spare enemy civilians if it can. The question, then, is whether the rights of
citizensA should count more heavily than, or as heavily as, those of innocent civil-
iansB in A’s decision to continue with the war, surrender, withdraw its troops, or
sue for peace.
At the bar of cosmopolitan justice, group membership is irrelevant to individu-
als’ basic entitlements. On that view, A cannot confer greater weight on the lives of
its own civilians than on the lives of B’s civilians just in virtue of the fact that the
latter belong to the enemy. However, they enjoy the personal prerogative not to
have to divest themselves of the resources and freedoms which they need in order
to lead a flourishing life even if, in exercising their prerogative, they would fail to
provide others with much needed assistance. Crucially, and as I argue in CW
(s.2.4), the personal prerogative also sometimes permits them to harm others fore-
seeably, though not intentionally, with a view to securing their own prospects for
leading such a life. By that token, the armed forces of a just belligerent are some-
times permitted to inflict collateral damage on enemy non-combatants—not in
virtue of the fact that the latter belong to the enemy community, nor indeed out
of a duty to one another and to their fellow citizens to protect their rights; rather,
they may so act in fulfilment of their personal prerogative not to lead a less than
flourishing life. Of all of A’s options, thus, A must choose that which minimizes
the loss to its members of prospects for such a life. Other things equal, if A’s exit
options leave citizensA equally badly off, then A must choose that which minimizes
harm to those of civiliansB who do not bear (significant) contributory responsibil-
ity for their community’s unjust war against A. Replace ‘fellow citizens’ with ‘fellow
community/group members’, and everything I have said in this section applies,
mutatis mutandis, to civil conflicts.
To conclude, the mere fact that A will not win their just war by continuing to
fight is not enough to show that they must desist. It matters too what their
options are for ending hostilities. The point remains, however, that the fact that
A’s war is just ad bellum does not entail that A may continue to fight that war
until they win it.
Ending Wars 45
2 . 4 . E N D I N G U N J U S T WA R S
Suppose now that A is waging an unjust war. If, as we have just seen, circumstances
can change in such a way as to turn A’s just war into an unjust one and impose on
A a duty to cease hostilities at the cost of fulfilling its just ends, it stands to reason
that they can also change in such a way as to turn its unjust war into a just war and
relieve A from its duty to stop fighting. Suppose for the sake of argument that the
US-led coalition lacked a justification for invading Iraq in 2003. Nevertheless, it
seems plausible to hold that it acquired a just cause for continuing to fight—this
time, insurgents—which just cause consisted in bringing stability to, and consoli-
dating, the new Iraqi regime in the face of armed rebellion by some factions within
the Iraqi population.23 It might also be that even if war is not a necessary means to
prosecute a just cause at time t1, diplomatic routes have now all been closed as a
result of the war, so that the latter has become necessary. Likewise, even if it stood
no chance of succeeding when it was first started, it might now stand a very good
chance (thanks to unforeseeable progress in the development of certain [morally
acceptable] military technologies, etc.) And so on.
2.4.1. Ending an Unjust War: Conditional v. Unconditional Surrender

Let us assume that A’s war is unjust at t1 and remains unjust at t2. Is A under a duty
to surrender to B? According to some war ethicists, B may not impose on A an
unconditional surrender—for (those ethicists say) it is wrong to beat your enemy
into abandoning its prerogative to protect its innocent members from wrongful
harm which your own troops might inflict on them, such as indiscriminate rape or
vengeful killing. If, then, B were to continue to fight their (unjust) war against A
to that end, they would act wrongly.24 Presumably, then, A would not be under a
duty to stop fighting and to accede to B’s demand.
But this is not quite right. To see why, consider the often-discussed case of the
Allies’ demand for unconditional surrender on the part of Germany during the
Second World War. The Allies’ demand had two parts: a ‘no-negotiation rule’,
which stipulated that they would not engage in any negotiation over armistice and
peace terms with German authorities but would issue the latter with instructions
for a mass and orderly capitulation; and a ‘no-recognition rule’, whereby the Nazi
regime would cease to exist as soon as the ceasefire came into force and no local,
indigenous entity would take its place. The ‘no-recognition’ rule set the stage for
the military occupation of Germany.
23 See Moellendorf, ‘Jus ex Bello’, 162. For an interesting exception to the academic philosophical
consensus on the 2003 Iraqi War (namely, that it was an unjust war), see N. Biggar, In Defence of War
(Oxford: Oxford University Press, 2013), ch. 7.
24 See, e.g., M. Walzer, Just and Unjust Wars, 111–17. See also Coady, Morality and Political
Violence, 270–8; J. McMahan, ‘Just Cause for War’, Ethics & International Affairs 19 (2005): 1–21, at
2. McMahan makes an exception in the case of Nazi Germany, on account of the nature of the regime.
See Kecskemeti, Strategic Surrender, ch. 7, for a good treatment of unconditional surrenders which
focuses on the cases of Italy, Germany, and Japan in 1943–45.
46 Ending Wars
I shall address the issue of occupation in ch. 3. Suffice it to say for now that if
enemy citizens are liable to a military occupation, or if there are lesser evil argu-
ments in favour of occupation, the no-recognition rule is not excessively demand-
ing. As to the no-negotiation rule, it issued in this particular case from the Allies’
refusal to deal with their German counterparts as anything but criminals and to
consider the possibility of having to offer them concessions.
In fact, we must distinguish between two senses in which a surrender is uncon-
ditional. In the sense used by aforementioned critics, a surrender is unconditional
in that it cannot but commit its addressee to refusing to defend the human rights
of its members. However, it is implausible to suppose that the word ‘uncondi-
tional’ cannot be taken to mean ‘unconditional within the constraints of morality’.
In a more plausible sense, thus, a surrender is unconditional in that the victorious
belligerent refuses to negotiate with its enemy—which is compatible with a com-
mitment to respecting and promoting the rights of the latter’s population.25
To demand an unconditional surrender in that second sense is not necessarily
wrongful. Sometimes, it is—when the pursuit of a justifiedATC peace requires some
concessions prior to the beginning of peace negotiation and as hostilities end.
However, as we saw in s.2.3, sometimes a regime does lose its claim to speak on
behalf of its community as it seeks to end the war, and it is not in itself morally
impermissible to refuse to engage with its officials.
Suppose that I am wrong—that when B imposes on A an unconditional surren-
der, it by definition imposes on A to waive their compatriots’ rights. B clearly acts
wrongly. Some might think that, by the same token, A is clearly not under a duty
to subject itself to B’s demand for an unconditional surrender, on the following
grounds: in agreeing not to make the respect of their compatriots’ human rights a
condition of their ceasing to fight, they would waive those rights on behalf of their
compatriots; yet, that is not something which they have any authority to do; given
that they lack that authority, they cannot be held under a duty so to act.
Again, this would be too quick. The surely correct point that A’s leaders lack the
authority to alienate their compatriots’ rights not to be raped, killed arbitrarily,
summarily detained, and so on, is entirely compatible with the claim that under
some circumstances A nevertheless is under an obligation to accept B’s demand.
Suppose that B will not stop the war unless A surrenders unconditionally, and that
B’s post-conflict terms will include neither systematic violations of citizensA’ basic
human rights nor imminent lethal threats to their non-basic rights—in other
words, that those terms will not bring about for A anything less than a justifiedATC
peace. By continuing with the war, A’s leadership would expose its own forces and
compatriots, as well as (possibly) innocent civilians in B who might be casualties
of A’s missions, to the harms attendant on a prolonged war which they could not
hope to win. Under circumstances such as those, A’s leadership are under a duty to
offer a surrender to B, even though they could not be deemed to have waived their
compatriots’ fundamental rights in so doing, and even though B would act wrongly
25 As Churchill in fact averred in a speech to the British Parliament in 22 February 1944. (Hansard,
HC Deb 22 February 1944 vol. 397 cc663–795.)
Ending Wars 47
by issuing those terms. Put differently, A’s leadership is under a duty to stop fight-
ing given that its acceptance of B’s ‘no condition’ demand will not subject its
members to those harms. But were B’s terms to fall short of what a justifiedATC
peace requires, A’s leadership would not be under a duty to accept them, since it
could not be deemed to have waived those rights. Likewise, if B’s ‘no condition’
demand for surrender is meant to provide B with a pretext for committing griev-
ous rights-violations against citizensA, A’s leadership not only is not under a duty
to accept that demand: it may justifiably refuse to surrender altogether.
2.4.2. Acquiring Justifications for Continuing with an

Initially Unjust War
My opening move, at the outset of s.2.4, was to note that a wrongdoer can some-
times acquire a justification for continuing with a course of action which it should
not have started. We should distinguish between two cases. In the first case, the
new justification is wholly unrelated to the initial unjust war. At this point a new
war itself begins, which implies that the proportionality clock is set back to zero
and the lives lost in the initial war can be wholly discounted.26
In the second case—my main focus here—the new justification is related to the
initial unjust war. Suppose that A unjustly aggresses B at t1, that the latter repels
the invasion and that A’s regime no longer poses a threat to its interests. However,
at t2, B takes advantage of its troops’ momentum to invade A with a view to occu-
pying it. Given that there is no justification for B’s occupation of A’s territory,
A prima facie may continue to fight its war against B, even though it lacked a just
cause for initiating it at t1.27 Or suppose that while fighting A, combatantsB com-
mit grievous rights violations against innocent members of A. Recall the example
of Germany in the last few days of the Second World War: some German military
leaders recognized that the war was lost but nevertheless persisted in fighting in
Eastern Prussia so as to enable as many civilians as possible to flee from the East to
the West and to ensure that as many combatants as possible could surrender to
American and British troops rather than the Red Army—on the grounds that the
latter were more likely than the former to kill, rape, and torture prisoners of war
and civilians. In the light of the exactions committed by Soviet combatants in their
campaign in Eastern Prussia, including mass summary executions of German pris-
oners of war and mass rapes of women (not all of whom were German), it seems
that they had a prima facie justification for so acting.28
In those examples, A lacks a just cause at t1, yet is justified in continuing with
the war if it acquires a new just cause. A fortiori, if A does have a just cause at t1 yet
26 See Moellendorf, ‘Two Doctrines of Jus Ex Bello’. For the view that there is no difference, from
the point of view of proportionality, between fighting an ongoing war and starting a new one, see
J. McMahan, ‘Proportionality and Time’.
27 See Rodin, ‘Two Emerging Issues of Jus Post Bellum’, 58, for a similar point.
28 See Weinberg, A World at Arms, 822–3. The claim that German combatants were permitted to
carry on fighting against the Soviet Army as a means to forestall imminent serious violations by the
former of the rights of German prisoners of war and women civilians presupposes that pre-emptive
killing is sometimes permissible. I take that assumption for granted here.
48 Ending Wars
goes to war in breach of one or more of the other requirements, it is justified in

continuing at t2 if circumstances have changed in such a way that it no longer
breaches those requirements: the reason why A’s war was initially unjust makes no
difference to the moral status of its decision to carry on.29
However, we must pay greater attention than I have done so far to the features
of the process whereby A acquires a justification for continuing to fight. In particu-
lar, it matters whether A is placed in that position as a result of its own wrongdo-
ings, or for extraneous reasons. In illustration of the latter case, suppose that in the
course of A’s unjust war against B, a natural disaster occurs in a part of B’s territory
whose minority population has hitherto proved rather hostile to B’s majority and
authoritarian regime. B’s regime takes advantage of the resulting chaos to move
some of its armed forces in the area and to mount a murderously violent campaign
of repression. This is the kind of wrongdoings which would have provided A with
a just cause for starting a war of intervention against B at t1, and A thus has a
strong justification for not ending its otherwise unjust war with B and for sending
some of its own troops to help the latter’s victims.
Suppose now that A acquires a just cause as a result of its own wrongdoings: this
time, B’s takes advantage of the chaos created by A’s wrongful war to murder its
opponents. Or suppose that factions within B take advantage of the collapse of B’s
institutions to try and seize power violently and illegitimately. In order to assess
whether A may continue, we must distinguish between cases where those who
would benefit from the continuation of the war either are responsible for the initial
wrongdoing or can be justifiably presumed to have consented to it, and cases where
they fall in neither category. In the former cases, A has two options. In continuing
with the war, it would protect from harm agents who bear some degree of respon-
sibility for the deaths wrongfully caused before t2 and would do so at the expense
of individuals who would die from t2 onwards and are innocent of that initial
wrongdoing. In exiting now, by contrast, it would protect those very same individ-
uals from harm though it would thereby expose initial wrongdoers to the harms of
military defeat. Given the choice (and other things roughly equal such as numbers
of individuals killed or spared on both sides), it should opt for the exit option, for
those wrongdoers surely have a lesser claim to protection than the innocent.
Compare with cases where A’s decision to continue with the war would benefit
neither agents responsible for wrongdoings nor individuals on whose behalf the
war was fought and who could justifiably be presumed to consent to it. In those
cases, the new cause—to wit, the protection of the innocent’s human rights—does
indeed provide A with a justification for not exiting just yet, subject of course to
meeting the other requirements of a just war. The same considerations apply, by
parity of reasoning, in those cases where A’s war meets the other four conditions as
a result of its wrongdoing. (For the sake of simplicity and to spare you from further
29 I used to think differently, and argued as much in my article ‘War Exit’. In her Peasoup response
to the article, however, Helen Frowe convincingly rebutted my argument. I am grateful to her for
proving me wrong. See http://peasoup.typepad.com/peasoup/2015/05/ethics-discussions-at-pea-
soup-cecile-fabres-war-exit-with-critical-precis-by-helen-frowe.html (accessed on 26/10/2015).
Ending Wars 49
distinctions I prescind from addressing cases where the group of beneficiaries

includes both kinds of agents.)30
To be sure, the claim that A may continue to fight raises the problem of moral
hazard: A is more likely to start an unjust war if it knows that it might subse-
quently acquire a justification for continuing what it unjustly initiated.31 In mit-
igation of those moral hazards, however, the fact that (as I shall argue in ch. 7) A’s
leaders should nevertheless still be regarded as liable to punishment and other
forms of sanctions for having started an unjust war at t1 might well give A’s emu-
lators potential pause for thought. Moreover, even if A’s decision gives other
wrongdoers strong enough an incentive to act unjustly, it does not follow that A
may not so act. A faces the following dilemma. In desisting here and now, it
would fail to assist members in B who may well have a very strong interest in,
indeed a rectificatory claim to, its intervention; in continuing with its war, A
would contribute to the wrongful killing of innocent people by those other
wrongdoers. The question then is whether an indirect contribution to future acts
of killing is morally worse than a direct failure to help victims here and now—
particularly victims whose predicament one has caused in the immediate past. In
answering that question, one must take into account (inter alia) the number of
lives saved versus the number of lives lost, the degree and strength of one’s reme-
dial responsibility for one’s victims, the time frame within which other wrongdo-
ers might act following one’s decision to continue to with the war, and so on. But
unless one endorses the highly implausible view32 that it is always morally forbid-
den to act in such a way as to provide third parties with an incentive to commit a
wrongdoing, it remains the case that A sometimes has a justification for continu-
ing a war which it started unjustly.
That claim, note, yields the unfortunate worry that, in so acting, A might end
up achieving its initially unjust cause. Suppose that although A has acquired a
justification for pursuing its advance through B’s territory, for example to destroy
military installations which B will use impermissibly to target civiliansA, it has not
acquired a justification for doing that which it intended to do in the first instance,
30 The following objection has been raised against my claim here, to wit: the moral status of A’s war
at t1 has a bearing on what other parties might do at t2; in particular, those parties might have a jus-
tification for preventing A at t2 from continuing with its war, as a means to deter other states from
carrying out unjust aggressions. The objection works, however, only if it is generally permissible to kill
as a deterrent combatants who are ex hypothesi justified in acting as they do, indeed, to kill as a deter-
rent innocent civilians in the target state—in fact, only if the deterrence of future wrongdoing is a just
cause for killing. I am sceptical, for reasons adduced by, e.g., D. Luban in his article on punitive war.
(See Luban, ‘War as Punishment’.) That said, the objection makes the interesting and correct point
that the justifiability of an act of war partly depends on the agent’s specific situation. For development
of that point, see S. Bazargan, ‘The Permissibility of Aiding and Abetting Unjust Wars’, Journal of
Moral Philosophy 8 (2011): 513–29.
31 I am grateful to Larry May for pressing me on this.
32 I regard the view as highly implausible because it implies by parity of reasoning that, for exam-
ple, police officers would act wrongly by killing a suicide bomber in defence of 500 innocent bystand-
ers in the following circumstances: if they kill him, they would provide a corrupt fellow officer with
an incentive to plant evidence accusing an innocent person of being a terrorist (as opposed to a low-life
criminal with evidence of corruption) and then using this as a pretext to kill him. Surely, however, it
is permissible for them to kill the bomber for the sake of saving 500 innocent lives.
50 Ending Wars
namely overthrowing B’s regime. Suppose further that, as a result of the continuing
war, B’s regime collapses. In pursuit of its new just cause, A’s forces kill agents who
have not acted in such a way as to lose their right not to be killed, such as innocent
civilians in B who are caught in the crossfire. They also kill members of B’s armed
forces who are not themselves committing the rights violations which provide A
with its new justification for continuing to fight. The difficulty, here, consists in
deciding what matters more morally speaking: that A should be able successfully
to prosecute its newly acquired just cause, at the risk of also successfully prosecut-
ing its unjust one, or that it should not be able successfully to prosecute its initial
unjust cause as would happen if it were to sue for peace, but at the risk of not
obtaining redress for its just grievance. Whether A ought to continue the war or to
end it depends on one’s assessment of what is the lesser of those two evils in the
particular circumstances of the case.
In all those cases, A acquires a justification for continuing the war which it did
not have when it started it, and must sue for peace as soon as it has achieved its new
end. In addition, it remains to be seen how it must end it. If its justification flows
from an additional just cause without affecting the moral status of its initial cause,
A must surrender once the new end is achieved, in an admission that it has indeed
lost their (unjust) war overall—as Germany’s leaders did on 8 May 1945. But
when A’s newly acquired just cause stems from its initial wrongdoing, whether it
must surrender here and now depends on whether by prolonging the war it signif-
icantly contributes in the end to improving the prospects of B’s citizens. If that is
the case, this might provide its leaders with a justification for refusing to surrender
and instead for seeking to end the war under more favourable terms than those to
which it would be entitled otherwise, and thus for continuing to fight with a view
to improving its negotiating position.
So far I have canvassed various scenarios where A acquires a justification for con-
tinuing with a war which it should not have started in the first instance. Before
concluding, it is worth revisiting the issue of termination costs. In s.2.3, I argued
that avoiding those costs provides a just belligerent with a just cause for continuing
a war which it knows it cannot win only if paying those costs would undermine its
members’ prospects for a flourishing life to such an extent as to provide them with
a just cause for war. I further observed that under some circumstances combatants
are liable to be killed as a means to avoiding those costs. The question here is whether
the costs of suing for peace—such as loss of prestige, loss of status on the interna-
tional stage, impairment to one’s ability to implement one’s foreign policy, and so
on—can sometimes tip the balance in favour of continuing the war though the
latter was unjustly started. They do so only if they themselves were imposed on an
unjust belligerent unjustly by its (ad bellum just) enemy, and only if the latter’s
combatants can be held liable for imposing them or may be killed as a means to
avert a grievous evil. In other cases, of which the following two are interesting exam-
ples, A may not invoke termination costs as a justification for not suing for peace.
In the first case, the prospects for a flourishing life of some of citizensA would be
undermined were A to end its war now simply because war and its economy provide
opportunities, jobs, and income to vast segments of the population. The example
might seem far-fetched; yet, one of the reasons why some conflicts are seemingly
Ending Wars 51
intractable is precisely that war, notwithstanding the destruction it causes, has

become an indispensable source of revenues to many of those who fight in it. This
is particularly so when belligerents are enmeshed in lucrative criminal activities
which both serve to fund the war and are caused by it. But in so far as A’s war is ex
hypothesi unjust, to prolong it at the cost of the lives of B’s members (be it combat-
antsB or citizensB killed as collateral damage) is wrongful: instead, the costs of ensur-
ing that A’s members do not suffer from the aforementioned harms should be borne
by wrongdoers within A who started the (unjust) war in the first instance.
The second kind of case I have in mind is this. Suppose that A goes to war
against B without just cause, realizes that it will not be able to win its war, but also
has very good reasons to believe that, if it were to stop now, its powerful neighbour
C would take advantage of its weakened loss of status to mount a wrongful aggres-
sion. To say that A has a justification for continuing its ad bellum unjust war against
B as a means to thwart C’s future aggression is to imply that its forces are morally
justified in deliberately killing combatantsB and foreseeably killing non-combat-
antsB as collateral damage, even though those agents are innocent of C’s wrongdo-
ing. In this case A would treat those innocent individuals as a means only to its
ends and thus act wrongly. Instead, A should target C’s wrongdoers, and not B’s
innocent members, in defence of its own members—and thus sue for peace against
B and repel C’s invasion when the latter occur.
This conclusion is subject to the following caveat. Suppose that C’s aggression of
A, following the latter’s loss of status on the international stage, would overwhelm-
ingly likely result in the mass enslavement, genocide, or starvation of A’s population
by C’s forces. Suppose further that the majority of A’s population is not responsible
for A’s unjust war against B. In this case, A has a choice between suing for peace with
B and allowing those grievous wrongdoings to occur, and blocking those wrongdo-
ings by continuing its war with B in the course of which it will kill a few thousands
of the latter’s members. Now, according to the doctrine of acts and omissions, ceteris
paribus it is morally worse to harm than to allow harm to happen—to kill than to
let die. I endorse the doctrine for reasons I articulate in CW (Introduction) but am
nevertheless not committed to holding A under a duty to sue for peace in this case.
This is because in this case, precisely, things are not roughly equal: the choice which
A faces is between letting vast numbers of individuals suffer the most grievous rights
violations imaginable, and killing considerably fewer other agents. It seems to me,
at least impressionistically at this stage, that despite the fact that A’s leaders and
some of its citizens are responsible for the predicament in which A’s population
would find itself were C to attack, they are morally permitted not to end their war
against B, as a means to avert a much greater evil. This of course is entirely compat-
ible with the further claim that A would owe additional compensation to B for so
acting—a claim to which I shall return in ch. 6.
2 . 5 . C O N C LU S I O N
In this chapter, I have argued that the moral status of a belligerent’s war sometimes
has a bearing on that belligerent’s authority to sue for peace on behalf of its
52 Ending Wars
i ndividual members. I have also claimed contra much of the literature that demands
for unconditional surrender are not always unjustified. Finally, I have argued that
we must to some extent sever the ethics of war termination from the ethics of war
initiation: a belligerent who embarks on a just war at t1 might be under a duty to
sue for peace at t2 before it has achieved its ex hypothesi just war aims; conversely,
a belligerent who embarks on an unjust war at t1 might acquire a justification for
continuing at t2. Moreover, when assessing whether a belligerent must exit or con-
tinue with its war, we must pay attention to the various ways in which it might do
so, and to the goods and bads which a particular exit strategy would bring about
relative to other strategies. To anticipate on subsequent chapters, however, although
we must disentangle the ethics of war exit from the ethics of war initiation, we
must bring the moral status of both to bear on post-war settlements.
3
Peacekeeping and Military Occupation
The guns have fallen silent. Perhaps belligerents are finally ready to sue for peace.
Or perhaps one has defeated the other comprehensively enough as to undermine
its rule but not so comprehensively that one can say with any confidence that war
will not resume under the auspices of a new regime. In that fragile, delicate phase
of the transition from war to peace, what may and must soon-to-be erstwhile bel-
ligerents do? If continuing to fight would be wrong, they are under a moral obli-
gation to work towards the establishment of a just peace. Yet the path to peace is
fraught when overshadowed by the continuing threat of war and the tumultuous
political, economic, and social conditions in which warring parties often find
themselves—so tumultuous, in fact, that they might not be able to do it on their
own. Hence the questions which I address in this chapter. First, if belligerents,
though willing to stop fighting in principle, are nevertheless unable on their own
durably to end violence as a precondition for reaching a peace settlement, is out-
side intervention in the form of the deployment of peacekeeping forces morally
justified? If so, how may or must those forces conduct themselves vis-à-vis belliger-
ents? Second, if a belligerent has been so comprehensively defeated on the battle-
field that its regime is no longer able to govern, and if there are no available local
actors yet with the requisite legitimacy to succeed it, may the victorious party turn
itself into a military occupier and take on the tasks of government?
The distinction between peacekeeping and military occupation is not as sharply
drawn in practice as I have sketched it out here, for a peacekeeping operation can
morph into an occupation if state institutions collapse. That said, it is easier to
show, as I do in s.3.2, that the deployment of peacekeeping forces whose mandate
is ‘merely’ to prevent the resumption of war is morally justified, than to show, as I
do in s.3.3, that institutional actors who are not members of a defeated belligerents
sometimes have the right, indeed are under a duty, to exercise temporary powers of
government over that belligerent’s community. This, incidentally, largely explains
why I have much more to say about occupation than about peacekeeping.
A definitional remark before I begin. To claim that belligerents are no longer
fully at war supposes that one can differentiate a state of war from a state of non-
war, which includes the transitional state which concerns us here. As indicated in
s.1.3, I adopt the definition of war adopted by the Correlates of War project,
whereby a war is a sustained armed conflict involving relatively organized armed
forces which fight for control over territory, and marked by at least 1000 battle
54 Peacekeeping and Military Occupation
deaths per annum. Fewer deaths, and either clear signs from at least one of the bel-
ligerents that they are willing to negotiate or clear signs that they are no longer able
to fight let alone to control the territory over which they claim jurisdiction—and
we can talk about approaching a state of non-war.
3 . 2 . P E A C E K E E P I N G
The (vast) literature on peacekeeping standardly distinguishes between peacekeep-

ing missions, whose mandate consists in helping secure the conditions under
which belligerents can reach a ceasefire and negotiate peace, from peacebuilding
operations whose mandate consists in overseeing or implementing comprehensive,
institution-shaping peace settlements. Another distinction is sometimes drawn
between so-called first-generation and second-generation peacekeeping. First-
generation missions were deployed as interposition forces between consenting bel-
ligerents, usually states, on their way to peace. The first of such missions, which
was to pave the way for three decades of UN involvement in transitions from war
to peace, was set up in 1956 in the immediate aftermath of the invasion of Egypt
by Israel, France, and the United Kingdom (the so-called Suez Crisis.) Under the
terms of resolution 1001, which the UN’s General Assembly adopted in November
of that year, the UN established a United Nations Emergency Force, whose main
tasks were to secure and monitor a ceasefire, to oversee the withdrawal of French,
British, and Israeli troops from Egypt, and to act as a buffer between Israel and
Egypt by monitoring cross-border movements. Subsequent first-generations mis-
sions have included, inter alia, a mission in Congo in the early 1960s, another
mission to act as a buffer between Egypt and Israel in the Sinai in 1974–79, the
deployment of a UN monitoring force to monitor the ceasefire between India and
Pakistan in 1965, and the establishment of a UN mission in Lebanon in 1978.
Those earlier missions—staffed mostly by military personnel with distinctive blue
helmets—were constrained by three key principles: the explicit consent of all belli-
gerents to the deployment of peacekeepers; peacekeepers’ neutrality between belli-
gerents; and a prohibition on the use of force by peacekeepers for any purpose other
than the defence of the mission’s own staff and the successful prosecution of the
mission.
The end of the Cold War and the outbreak of a number of civil wars, with con-
comitant humanitarian disasters, institutional collapses, and regional instability,
led the United Nations, some of their member states, and regional organizations
such as the African Union to deploy and support more complex missions involving
all of the above as well as the provision of humanitarian relief and close involve-
ment in state-building. Those latter missions, of which good examples are the UN
mission in Bosnia before the signing of the Dayton Agreement in 1995 and the
UN mission in Rwanda before the 1994 genocide, also relied on the parties’ con-
sent, aimed to remain neutral between them, and operated on the aforementioned
and highly constraining rules of engagement. The genocidal murders of civilians in
those conflicts have since led the United Nations significantly to revise the tryad
Peacekeeping and Military Occupation 55
consent—neutrality—limited force in favour of more robust—some would say,

interventionist—forms of peacekeeping.1
There are other ways to label and distinguish those various kind of operations,
which need not detain us here. My concern in this chapter is with outside assist-
ance with, or intervention in, the tortuous march towards a peace agreement—
rather than with post-settlement peacebuilding and enforcement missions. (I will
deal with the latter in ch. 8.) Accordingly, I shall understand peacekeeping as a way
to help warring factions reach the negotiating table. Peacekeepers so construed do
not themselves oversee and shape those negotiations; nor do they enforce them.
Their role is limited to enforcing ceasefires, demilitarizing belligerents, providing
humanitarian aid under conditions of acute emergency and (if absolutely neces-
sary) resorting to force—albeit force short of war—against those who unjustifiably
pose or contribute to posing an ongoing or imminent threat to human rights. To
frame their mandate in the language of the UN Charter, they operate under a
combination of Chapter VI and Chapter VII of the Charter. As per Chapter VI,
their aim is to facilitate the peaceful settlement of disputes when belligerents are
ready to stop fighting; as per Chapter VII, they may employ force as a necessary
means to protect international peace from the consequences of belligerents’ unwill-
ingness to respect ceasefires and lay down their arms. As UN Secretary General
Dag Hammarskjöld, who served between 1953 and 1961, apparently put it, peace-
keeping falls within the remit of ‘Chapter Six and a Half.’2
On what grounds, then, are peacekeeping missions morally justified? In so far as
they are meant, not to intervene militarily in an ongoing conflict, but to get belli-
gerents who are ready to stop fighting to the point where they can negotiate peace,
one might think that the set of conditions and principles which regulate the resort
to war will not apply here at all. But that would be a mistake. Those principles, you
recall, are the following: a war is just if, and only if, (a) it has a just cause, where a
just cause consists in the violation, backed by the threat of lethal force, of some
party’s human rights; (b) it is a proportionate response to the injustice which the
belligerent has suffered; (c) it is not fought and won through the deliberate and
indiscriminate targeting of innocent non-combatants; (d) it stands a reasonable
chance of succeeding by military means which do not breach the requirements of
proportionality and discrimination; (e) it is the only way to pursue the just cause.
1 For good summaries and analyses of the operations and missions mentioned here, see, e.g., A. J.
Bellamy, S. Griffin and P. Williams, Understanding Peacekeeping (Cambridge: Polity, 2003); N.
MacQueen, Peacekeeping and the International System (London: Routledge, 2006). For nuanced
defences of interventionist peacekeeping (on the grounds, mostly, that they do actually work when
properly designed), see M. W. Doyle, and Sambanis, Nicholas, Making War and Building Peace: United
Nations Peace Operations (Princeton: Princeton University Press, 2006); V. P. Fortna, Does Peacekeeping
Work? (Princeton, N.J.: Princeton University Press, 2008); M. L. Howard, UN Peacekeeping in Civil
Wars (Cambridge: Cambridge University Press, 2008). The United Nations’ Peacekeeping department
has a comprehensive website detailing past and current missions, at http://www.un.org/en/peacekeep-
ing/operations/.
2 Almost every book and article I have read on this topic attributes those words to him, but I have
not been able to trace the original source. In a twist of tragic irony, he died in a plane crash while on
his way to negotiate a ceasefire in Congo.
A peacekeeping mission takes place as belligerents are (seemingly) willing to

stop fighting. In addition, as I noted above, their mandate usually involves moni-
toring and implementing ceasefires—which in turn might require monitoring the
withdrawal of troops from certain areas and the surrender of weapons—as well as
the delivery of emergency humanitarian assistance. While this may require that
they use force in defence of their mission (a point to which I shall return later on
in this section) it is not a defining feature of their mission that they should fight
and kill in its prosecution—unlike parties at war. Taking those differences into
account, we can adapt the requirements for a just war into requirements for just
peacekeeping, whereby a peacekeeping mission is just if, and only if: (a*) it has a
just cause, where a just cause consists in an imminent risk that the war will resume;
(b*) it is a proportionate response to this imminent risk; (c*) peacekeepers do not
deliberately cause wrongful harms to innocent parties as a means to secure the
success of their mission; (d*) it stands a reasonable chance of success by financial,
human, and military means the deployment of which violates neither (b*) nor (c*),
where success consists in getting the belligerents to the point where they can nego-
tiate at the very least a justifiedATC, and if possible just, peace settlement; (e*) it is
the only way to get belligerents to move to a justifiedATC peace.
As with war, the key condition is the just cause requirement, which provides
actors with a justification for (in this instance) intervening between belligerents. In
CW (s.5.2), I argued in favour of a right to wage a war of humanitarian interven-
tion along the following lines: individuals whose human rights are violated by their
own regime and who have the right to kill in defence of those rights also have the
Hohfeldian power to transfer that right to potential rescuers, either wholly if they
themselves do not fight, or in part if they do. Furthermore, a similar argument can
be made to the effect that members of a politically sovereign community locked in
a just war against another similar community also have the power to enlist assist-
ance from a third party, by way of a military alliance. At first sight, the best case for
peacekeeping proceeds along similar lines: peacekeepers are justified in enforcing
ceasefires, demilitarizing belligerents, and providing humanitarian relief, on the
grounds that individuals whose human rights would be threatened by the resump-
tion of war have a right to take precisely those steps so as to prevent such resump-
tion and thus to secure peace, or at any rate, a justifiedATC peace, by the
aforementioned means, and the power to transfer that right to outsiders if neces-
sary. If this is correct, the consent of those individuals to the peacekeeping opera-
tion is a necessary condition for the latter to be just. In this respect, it would seem
that my case for peacekeeping does not radically depart from standard arguments
as deployed by e.g., the UN in favour of traditional peacekeeping.
However—and this is crucial—although consent plays an important part in my
justification for peacekeeping, the latter differs from those standard arguments. On
my account, the explicit consent of all belligerents is not a necessary condition for
the justness of all and any peacekeeping operations. In some cases belligerents are
simply not organized and structured enough to be in a position durably to consent
on behalf of those whom they claim to represent. The UN missions in Somalia
(1992–95) are good examples of the difficulties faced with peacekeepers in those
fragmented situations. Following the downfall of its President Siad Barre in January
1991, civil war erupted in that country as various factions took up arms in a bid
for territorial control. In the face of a worsening humanitarian crisis and the col-
lapse of state institutions, the UN and nongovernmental and regional organiza-
tions embarked on extensive relief operations and pressed for a political solution to
the crisis. But whereas all faction leaders (bar one) agreed to a ceasefire overseen by
the UN in January 1992, armed militia men—some under the putative control of
those leaders, some not—made the task of peacekeepers impossible by looting and
attacking humanitarian convoys.3 Under those conditions, to say that the UN
were not entitled to act on the grounds that there were no ‘stable’ leaders whose
consent, on behalf of their constituents, could be sought, is to leave the victims of
the resumption of the war to their own devices, which, given that they would be
likely to suffer grievous harm, is too high a price to pay for holding on to the value
of consent.
Suppose that there are in fact such leaders. Is their explicit consent necessary?
On some views, which may well have shaped first-generation UN missions, out-
side intervention by peacekeepers in the last stages of a civil war is an interference
in that country’s internal affairs; likewise, outside intervention by peacekeepers in
the last stages of a transnational war between, e.g., two political sovereign commu-
nities is an interference in those communities’ internal affairs. In both cases, peace-
keepers would wrong the belligerents were they to intervene without the latter’s
explicit consent.
Yet, this is not quite right. As we saw in ch. 1, individuals have a jointly held
right to political self-determination only if they fulfil their fundamental obliga-
tions of justice to one another. As we saw in ch. 2, one such obligation is the obli-
gation to stop fighting if one’s war is or has become unjust. If those points are
correct, the claim that a peacekeeping operation to which belligerents have not
consented wrongs them is not particularly strong in this context. For consider.
Suppose that A and B are at war and that A is not under a duty to sue for peace, on
the grounds that, were it to do so now, the resulting settlement would not secure a
justifiedATC peace. In so far as A is not under a duty to sue for peace, its consent to
the deployment of peacekeeping forces is a necessary condition for the latter to be
just. By implication, then, B’s consent is not necessary: since its agents are waging
an unjust war against A (ex hypothesi), they would not be wronged were peacekeep-
ers to deploy. Furthermore, and again by implication, if it would be unjust for both
A and B to continue fighting, and if they are thus both under a duty to sue for
peace, the forcible deployment of a peacekeeping mission would not wrong them
either. In fact, they are under a moral duty to accept the presence of the mission.
Note that whether A and B are politically sovereign communities or factions locked
3 See, e.g., R. Murphy, UN Peacekeeping in Lebanon, Somalia and Kosovo—Operational and Legal
Issues in Practice (Cambridge: Cambridge University Press, 2007); T. Findlay, The Use of Force in UN
Peace Operations (Oxford: Oxford University Press, 2002), ch. 6. For an interesting discussion of when
warring factions are more or less likely to consent to the deployment of peacekeepers, see Fortna, Does
Peacekeeping Work?, esp. ch. 2.
into a civil war is irrelevant: on cosmopolitan grounds, as I argued in CW (s.4.2),

the moral norms which regulate those two kinds of conflict are the same.
Is it true to say that there are no circumstances under which B’s consent to the
deployment of a peacekeeping mission is necessary to render it just? Not so. If
the mission will fail absent such consent, forcible deployment would violate the
requirement that it should stand a reasonable chance of success, and would there-
fore be unjust. Importantly, the point is not merely pragmatic: peacekeeping mis-
sions are extraordinarily costly, and the material, financial, and human resources
which they require could be easily deployed elsewhere instead—for example to
alleviate non-war related global poverty. Even if the populations of war-torn com-
munities have a stronger claim to peacekeeping resources than the global poor have
a claim to the resources they need to lead a flourishing life (by no means a foregone
conclusion), it is a moral imperative vis-à-vis the latter that the deployment of
those resources should stand a reasonable chance of being successful.4 By implica-
tion, though, if the mission can succeed without the explicit consent of all belliger-
ents (for example, because some belligerents are simply too weak to provide much
opposition to peacekeepers), the deployment of peacekeepers can be just absent
such consent provided that individuals whose human rights depend for their pro-
tection on the presence of those peacekeepers can plausibly be presumed to con-
sent. Claims by peacekeepers that those individuals would consent if given a
chance ought not to be taken at face value.
The explicit consent of all belligerents, thus, is not a necessary condition for a
just peacekeeping mission. Nor is it a sufficient condition for it—at least so stated.
This might seem surprising, for suppose that all parties do in fact consent, even
though some of those belligerents have a justification for continuing with the war.
Why is it not enough to deem the mission justified? Quite simply, because A’s
regime has to have some standing in order to agree to a ceasefire and to its being
monitored by peacekeepers on behalf of their members. It is subject to this particu-
lar condition that the regime’s consent to the deployment of a peacekeeping force
is sufficient to deem the latter just.
So far, I have argued that there sometimes is a just cause for the deployment of
a peacekeeping mission. That however is not enough to deem the mission justified
overall, for it also matters what the peacekeepers’ mandate is and how they will
carry it out—considerations which are captured by requirements (b*) to (e*) above.
I have already said a few words about the reasonable chance of success requirement
((d*)). The other requirements warrant explaining. Requirement (b*), whereby the
mission must be a proportionate response to the imminent risk of war resumption,
is relatively straightforward: as we saw above, those missions are materially costly;
moreover, as I shall argue below, their members have the right to kill in self- and
4 In fact, there is an ongoing debate in the relevant policy circles on the importance of funding
peacekeeping missions relative to the importance of focusing on development in developing coun-
tries—a tension which is sometimes described as ‘the Northern peacekeeping agenda v. the Southern
poverty alleviation agenda’. See Findlay, The Use of Force in UN Peace Operations, 343. For various
accounts of what is required to increase a mission’s chances of success, see, e.g., Doyle, Making War
and Building Peace; Fortna, Does Peacekeeping Work; Howard, UN Peacekeeping in Civil Wars.
other-defence. It stands to reason, in the light of the harms which would accrue to
those individuals (for example, the global poor who will not get help, or the inno-
cent who will be killed by peacekeepers or by belligerents acting in hostile reaction
to the latter’s presence5) that the mission be a proportionate response to the unfold-
ing crisis, in two senses. In the narrow sense of proportionality, the harms imposed
by peacekeepers on those who are liable to the imposition of a mission should be
proportionate to those agents’ wrongful contributions to the just cause for that
mission. In the wide sense, the mission itself may not be conducted in such a way
as to jeopardize prospects for a justifiedATC peace.
By the same token, it also stands to reason that, as per requirement (e*), it
should be the only way to get belligerents to the point where they can negotiate a
justifiedATC peace settlement: if they can do so themselves, then they ought to do
so, so that the resources which the mission would require could be deployed else-
where where they are actually needed. Requirement (c*), whereby peacekeepers
should not deliberately cause wrongful harms to innocent parties as a means to
secure the success of their mission, is similarly straightforward. But it should be
supplemented by a subtly different principle, to the effect that peacekeepers must
not deliberately cause wrongful harm to innocent parties whilst on deployment.
Unfortunately, that principle has been honoured as much in the breach as in the
observance, as peacekeepers have often taken part or been complicitous in serious
human rights violations such as rape, looting, and human trafficking.6 Those acts
are grievously wrong per se. But they also taint the peacekeeping operation by
undermining both its chances of success (measured in getting the belligerents to
the negotiation table) and prospects for a justifiedATC peace. Adding this principle
is thus far from trivial.
Finally, as I noted at the outset, in practice, peacekeeping missions used to be
constrained by two related requirements: a requirement that peacekeepers remain
neutral between belligerents, and a requirement that they use force only in defence
of their life and of the mission (since if they were to use to force for other ends,
they could no longer be regarded as outsiders to the conflict whose endings they
were meant to oversee—or so the thinking went).7 This has led peacekeepers to
stand by while one party carried out massacres against another—as in Srebrenica
in 1995, a supposedly safe enclave in Bosnia, where thousands of Muslim civilians
were killed by Bosnian Serb forces under the eyes of powerless Dutch troops. Since
then, the United Nations have revised their commitment to neutrality and limited
force. In particular, peacekeepers are now allowed to use force in defence of their
own mission broadly understood—and thus not just in defence of their or their
fellow peacekeepers’ lives but also in defence of civilians whose lives are under
5 I thus include harms caused by other, intervening agents in the proportionality calculus which
those who authorize peacekeeping missions must undertake, on the grounds that the latter, and not
just intervening agents, are morally responsible for those harms. See, e.g., M. J. Zimmerman,
‘Intervening Agents and Moral Responsibility’, The Philosophical Quarterly 35 (1985): 347–58.
6 See, e.g., Bellamy, Griffin and Williams, Understanding Peacekeeping, ch. 16; Murphy, UN
Peacekeeping in Lebanon, Somalia and Kosovo, 230–8.
7 See Findlay, The Use of Force in UN Peace Operations.
threat at the hands of armed forces hostile to the nascent peace process.8 These are
welcome developments, for on the view of defensive killing which underpins my
account of the just war, agents have the right to kill anyone who unjustifiably vio-
lates their or some other party’s human rights by way of a lethal threat. More
strongly still, they are sometimes under a duty to do so when acting in defence of
others.9 A peacekeeping mission which treats all parties on a par irrespective of
their willingness to abide by the terms of a justified ceasefire and of the threats
which they pose to the innocent is not just. Rather, a just mission must protect the
innocent (whether civilians or combatants who are willing to work towards peace).
This specific claim is contained in the requirement of proportionality widely
understood, which stipulates that neither the deployment of a peacekeeping mis-
sion nor the way it is conducted must unnecessarily threaten, once it is over, the
establishment of a justifiedATC peace. In so far as ongoing grievous human rights
violations poses this kind of threat, for peacekeepers to be neutral as between the
parties irrespective of what the latter do during the ceasefire violates that
condition.
So much, then, for the claim that peacekeeping missions are morally justified
under the aforementioned conditions—even though their forcible deployment
and presence on the territory of one of the belligerents might be regarded by some
as a breach of sovereignty. Still, some loose ends need tying. First, if peacekeepers
justifiably kill opponents to the (justified) peace process, those opponents are not
permitted to retaliate in their own defence, since ex hypothesi they are liable to be
killed. In that sense, peacekeepers are immune from being killed by factions which
oppose the peace process by unjustifiably posing wrongful threats of lethal harm.
But their immunity flows from their or third parties being so threatened: it is not
conferred on them in virtue of the fact that they are peacekeepers. Were they them-
selves unjustifiably to pose a wrongful threat of lethal harm in the course of their
mission, they would become liable to be killed. In s.1.3, I argued that the status of
combatant does not automatically confer on its holders the privilege to kill in self-
and other-defence, and that the status of non-combatant does not automatically
confer on its holders immunity from being killed. By parity of reasoning, the status
of peacekeepers does not automatically confer on its holders either the right to kill
or the immunity from being killed. Thus, the norms regulating the use of force
against and by peacekeepers are the same as the norms which regulate the use of
force in domestic conflicts and in war.10
Second, I suggested a few paragraphs ago that belligerents who are under a duty
to sue for peace are under a duty to accept the presence of a justified peacekeeping
mission. Conversely, putative peacekeepers are sometimes under a duty to deploy.
8 Current UN doctrine is set out in (the so-called) Brahimi Report. For comprehensive accounts of
the use of force in UN peacekeeping missions, see, e.g., Findlay, The Use of Force in UN Peace
Operations; Murphy, UN Peacekeeping in Lebanon, Somalia and Kosovo, ch. 4.
9 See CW, ss.2.2 and 5.3.
10 For a discussion of peacekeepers’ legal status in the laws of war, see D. Blocq, ‘The Fog of UN
Peacekeeping: Ethical Issues regarding the use of Force to protect Civilians in UN Operations’, Journal
of Military Ethics 5 (2006): 201–14.
For as I argued in CW (s.5.3), the general duty to provide assistance to the needy
generates a more specific duty to help individuals whose regime carries out sys-
tematic human rights violations, by authorizing and financing a war of humani-
tarian intervention, as well as a duty to send some of their (professional)
combatants to fight that war. That duty, moreover, is conditional upon the satis-
faction of the no-undue sacrifice proviso, that is to say, on it being the case that
the costs of waging the war do not prevent residents of intervening powers from
enjoying opportunities for a flourishing life. On similar grounds, and subject to
the same proviso, there is a duty to authorize, finance, and deploy peacekeeping
forces as a means to facilitate the conditions under which belligerents can embark
on peace negotiations.
The question, then, is who can and ought to do it? When defending the right
and duty to wage a war of humanitarian intervention in CW (s.5.4), I claimed that
victims of human rights violations at the hands of their own regime are (usually)
better off entrusting multinational institutions with the task of authorizing the
quick deployment of an international army, since those institutions are more likely
to be impartial and to reach their decision to use force in a transparent and account-
able way than lone, unmandated interveners. Impartiality, here, does not mean not
favouring this or that belligerent on the grounds that such a policy would further
those interests; rather, it means treating all parties in the nascent peace process
strictly as they are entitled to be treated. Unlike impartiality as construed by the
UN as neutrality (that is, even-handedness, between belligerents irrespective of
their conduct), it is a fundamental moral requirement of peacekeeping. Accordingly,
those institutions and their armed forces, both of which should be funded by all
member communities in proportion to the latter’s GDP, constitute the primary
institutional rights-holders and duty-bearers when humanitarian intervention is at
issue.
This approach may be dubbed ‘heavily qualified multilateralism’: it clearly rec-
ommends multilateral action, but subject to its effectiveness. Just as it applies to
military humanitarian intervention, it also applies to peacekeeping—not just when
the peacekeeping mission follows a war of humanitarian intervention but also as a
response to a war of aggression between A and B—whether the latter are belliger-
ents in a civil war or political sovereign communities. In fact, the case for multilat-
eralism is even stronger in the transition from war to peace than it is in war itself.
For in that situation, where belligerents are willing to negotiate but unable to trust
one another sufficiently so as to mutually provide and enjoy secure conditions for
negotiations, it is imperative that peacekeepers not be suspected of favouring a
former ally over the latter’s enemy when discharging their tasks. Multilateral insti-
tutions are in a better position than single parties to do that—unless the latter have
a proven reputation for impartiality, or do not have a special, partisan stake in
ensuring that one faction wins over the other in ensuing peace negotiations.
Incidentally, the requirement of impartiality and the conclusion which it sup-
ports to the effect that multilateral operations are prima facie better suited to the
tasks at hand, shed some light on the normative relationship between the moral
status of belligerents on the one hand, as having waged a just and unjust war, and
the moral status of peacekeepers as carrying a just or unjust operation on the other
hand. Let me explain. In the next section, I shall argue that a just belligerent some-
times has a justification for occupying a defeated enemy. I shall also argue that an
unjust belligerent does not have at the outset, but may come to acquire, a justifica-
tion for so doing. Thus, in the case of occupation, we may draw some conclusions
about the moral status of occupiers by referring to their moral status as belligerents
in the war that led to the occupation. Not so here. For peacekeeping operations
must be impartial as between belligerents in virtue of the tasks which they carry
out. Accordingly, a belligerent, whether just or unjust, simply cannot oversee the
transition from war to peace as a peacekeeper, since it is a party itself in the peace
process (with its preliminary steps such as ceasefire, etc.) which the mission is
meant to oversee. It is qua belligerent, and irrespective of the status of its war, that
it is thus disqualified.
To be sure, the multilateral approach is not without difficulties. For a start, a
multilateral force by definition comprises national forces deployed under the
authority of an overarching institution—typically, though not necessarily, the
United Nations. It is crucial therefore that national military codes should not con-
flict with the mission’s imperatives. In particular, rules of engagement must be
uniform and understood to be so by all peacekeepers, as must expectations arising
from peacekeepers’ place in the chain of command.11 Those operational require-
ments add grist to the cosmopolitan mill, since they hold, in effect, that peacekeep-
ers’ membership in this or that national army has no bearing on their rights, duties,
and liabilities qua peacekeepers.
Moreover, allocating the costs of financing and staffing peacekeeping fairly
between its parties is likely to prove extremely difficult. As a matter of fact, it is a
standard complaint of many developing countries that they provide a dispropor-
tionately high number of combatants relative to their population size, the size of
their army, and/or the size of their economy compared with wealthy nations.12
Assume (plausibly) that a fair allocation must take those factors into account. Still,
the risks of jeopardizing the peace process and thereby of exposing its beneficiaries
to the dangers of war by entrusting the task of overseeing it to parties who will be
seen as partisans outweigh (when roughly equal) the risks of erring on the side of
unfairness to some of the providers of peacekeepers by adopting a multilateral
approach.
To conclude, peacekeeping missions are morally justified—indeed, are some-
times a duty. As we saw, however, the tasks of peacekeepers, however complex on
11 See, e.g., Murphy, UN Peacekeeping in Lebanon, Somalia and Kosovo, ch. 3.
12 See, e.g., Blocq, ‘The Fog of UN Peacekeeping’, 206. As of October 2015, the top ten contrib-
utors of on-field manpower to peacekeeping operations were Bangladesh, Ethiopia, India, Pakistan,
Rwanda, Nepal, Senegal, Ghana, and China. The first contributor who is also a member of the OECD
(whose members are the most developed, economically speaking, of the world’s countries) is Italy—
whose contribution is the 25th most important. For what it is worth, as of January 13 2016, the UK
stands at 51, and the US at 73. See http://www.un.org/en/peacekeeping/resources/statistics/contribu-
tors.shtml (accessed on 13 /01/2016). For a very good discussion of the difficulties involved in multi-
lateral, or multiagency, approaches to peacekeeping, see MacQueen, Peacekeeping and the International
System, esp. ch. 11.
the grounds, are nevertheless rather easy to justify, since they consist in the main in
preventing belligerents from resuming war. The longer the negotiation process
and/or the less able local actors are to exercise functions of government, the more
likely it is that peacekeeping missions might have to take on that particular
mantle—the dreaded ‘mission creep’—dire forewarnings of which abound

whenever peacekeepers are deployed. Peacekeeping, in fact, might easily turn into
a military occupation.
3 . 3 . M I L I TA RY O C C U PAT I O N
Military occupation involves more than ensuring that belligerents do not keep
fighting: it consists in taking on some limited powers of government over the ter-
ritory of one or several of the belligerents for a limited time, on the understanding
that the occupying party qua Occupier will not annex the territory and will leave
once an indigenous government has been instated. It is not part of the remit of an
occupying force as I shall consider it here actively to engage in state-building.13 At
the same time, occupying forces carry out a greater range of tasks than peacekeep-
ing forces, and thus are in greater contact with local populations. This, perhaps, is
the most significant feature of military occupation for those who experience it.
And yet, notwithstanding the revival of the ethics of war in the last three or four
decades, Israel’s long occupation of the West Bank, and the occupations of Iraq
and Afghanistan by the US-led coalition, there is a dearth of works on this particu-
lar topic.14 My aim in this section is to provide a normative account of the rights
and wrong of occupation. In s.3.3.1 I argue that a just belligerent sometimes has a
right qua just belligerent to conduct a military occupation, and that unjust belli-
gerents should be presumed to lack that right unless circumstances change in such
a way as to provide them with a just cause for occupying. In s.3.3.2, I bring to bear
my defence of qualified multilateralism in the context of peacekeeping on the case
of military occupation. In ss.3.3.3 and 3.3.4, I set out the rights and duties which
occupying forces and occupied populations have vis-à-vis one another.
13 For a good definition, see D. M. Edelstein, Occupational Hazards—Success and Failure in

Military Occupation (Ithaca: Cornell University Press, 2008), 3. For an important typology of military
occupation, see A. Roberts, ‘What Is a Military Occupation?’, British Yearbook of International Law 55
(1984): 249–305. For recent discussions of the law of occupation, see the articles in The International
Review of the Red Cross, No 885 (2012).
14 Michael Walzer’s Just and Unjust Wars only has a few pages on the permissibility of guerrilla
warfare against occupiers. See Walzer, Just and Unjust Wars, 176–9. He has since written (a little bit)
about the US occupation of Iraq. See M. Walzer, ‘Just and Unjust Occupations’, Dissent, Winter 2004.
Two philosophical articles which are entirely given over to occupation are J. McMahan, ‘The Morality
of Military Occupation’, Loyola International and Comparative Law Review 31 (2009): 101–23 and
J. Rocheleau, ‘From Aggression to Just Occupation? The Temporal Application of Jus Ad Bellum
Principles and the Case of Iraq’, Journal of Military Ethics 9 (2010): 123–38. See also Evans, ‘Moral
Responsibilities and the Conflicting Demands of Jus Post Bellum’; Orend, ‘Justice after War’. The
latter two articles however address the issue of non-belligerent occupation. For a set of fascinating
witness accounts, by Israeli combatants, of life as a member of occupying forces in the West Bank, see
http://www.breakingthesilence.org.il. I am grateful to Ruvi Ziegler for drawing my attention to those
sources.
3.3.1. Just and Unjust Wars, Just and Unjust Occupations

Under Chapter VII of the UN Charter, the Security Council may respond to an
act of aggression against a member state or to any threat to, or breach of, peace, by
authorizing measures short of military force such as economic sanctions and,
should those measures fail, by taking military action (arts 39–42 esp.) Although
Chapter VII is standardly read as allowing the United Nations to resort to war, it
can and has been read as allowing them to mount a military occupation, notably
following civil conflicts, as a means to restore and maintain peace.15 Once occupy-
ing troops are on the grounds, the wide-ranging legal provisions which govern
their conduct are set out in the 1907 Hague Regulations (s. III, arts 42–56), as well
as in the Fourth Geneva Convention adopted in 1949 (s. III, arts 47–78.)16 Taken
together, those documents are the basis for the law of military occupation. Thus,
the UN Charter can plausibly be read as applying the jus ad bellum to the case of
occupation, whilst the Hague Regulations and the Geneva Convention pertain to
the jus in bello. In so far as the law of military occupation is deemed to partake of
jus in bello, it applies to all occupiers, irrespective of the legal status of the war
which they waged and which led them to occupy enemy territory. By implication,
the liabilities and rights of occupied populations under those rules are the same
irrespective of the legal status of the war which their country waged against
Occupier. On that view, even if (as many think) the war waged by the Allies against
Germany in the Second World War was lawful whilst the 2003 invasion of Iraq by
the US-led coalition was not, the Allied Occupation of Germany between 1946
and 1949 was subject to exactly the same legal regime as the occupation of Iraq by
that coalition between 2003 and 2004.17
15 For such an interpretation of Ch. VII, see, e.g., G. H. Fox, Humanitarian Occupation
(Cambridge: Cambridge University Press, 2008).
16 Strictly speaking, those regulations are annexed to the Fourth Hague Convention. All legal docu-
ments mentioned here can be found in A. Roberts and R. Guelff (ed.), Documents on the Laws of War,
3rd edn. (Oxford: Oxford University Press, 2000). I restrict my account to non-customary international
law. Note, moreover, that several provisions of both the Hague Regulations and the Geneva Conventions
concern belligerent occupation (or occupatio bellica, in the juridical Latin of scholars)—cases, in other
words, where Occupier is still formally at war with Occupied and the latter’s allies and where Occupied’s
regime has been weakened or displaced rather than vanquished. For a historical account of the law and
practice of military occupation in the nineteenth and early twentieth century, see K. Nabulsi, Traditions
of War: Occupation, Resistance, and the Law (Oxford: Oxford University Press, 1999); Fox, Humanitarian
Occupation, chs. 1 and 2. Two classic contemporary books on the law of belligerent occupation are
E. Benvenisti, The International Law of Occupation (Princeton, N.J.: Princeton University Press,
2004); Y. Dinstein, The International Law of Belligerent Occupation (Cambridge: Cambridge University
Press, 2009). Throughout this chapter, my account of those legal provisions which apply to cases of
non-belligerent occupation follows both books, unless otherwise stated.
17 The view set out in this paragraph is accepted by more or less all legal scholars. That said, some
commentators note that, according to a declaration passed without a vote by the UN General
Assembly, ‘the territory of the state shall not be the object of military occupation resulting from the
use of force in contravention of the provisions of the Charter’. See Declaration on principles of interna-
tional law concerning the friendly relations and co-operation among states in accordance with the Charter
of the United Nations. GA Res. 2625, UN GAOR, 25th Sess., Supp. No. 28, at 121, UN Doc. A/8028
(1970). This principle was reasserted by the International Court of Justice in a 2004 landmark opinion
on the legality of the Israeli occupation of the West Bank. See Legal Consequences of the Construction of
a Wall in the Occupied Palestinian Territory, paras 75 and 117.
However, it is a key tenet of this book that, contra orthodoxy, the jus in bello is
not independent of the jus ad bellum. As we shall see in the next two subsections,
by parity of reasoning, whether a belligerent who comes to occupy a defeated com-
munity de facto has the right to do so does not depend on the fact of its occupying
presence, but at least in part on the moral status of the war which it has waged to
get there.
From a Just War To a Just Occupation

Suppose that B attacks A without just cause. Or suppose that B’s regime carries out
a genocidal campaign against an ethnic minority within its borders, or that it is
engaged in a murderous and large-scale campaign of repression against any and all
who dare resist its rule. In the course of its justified war against B, A makes consid-
erable inroads into B’s territory, which leads to the collapse of B’s regime. A’s lead-
ers have objective reasons to believe that if they withdraw their troops from B’s
territory, factions within B—either faithful to the defeated regime or wanting to
seize powers for themselves—would then attempt to resume the war—be it a war
of aggression against A or yet another campaign of violence and repression against
a subset of B’s population. Are A’s leaders justified in maintaining their army in B?
Are international agents justified in sending occupying forces to B?
I believe so. Consider the case of an occupation which follows a just war of
national defence. Suppose that if there is no ongoing military presence in B follow-
ing the latter’s defeat, some agents within B will resume hostilities. The military
occupation of B can be justified on two grounds. First, it might help block the
threat which those agents continue to pose to A even though the fighting has
stopped: when issuing and enforcing directives against B’s occupied civilians, or
when requisitioning or using B’s resources, A, in his capacity as Occupier, serves its
continuing needs as a belligerent which has not yet fully thwarted the threat which
those factions pose. Or, following a war of humanitarian intervention, the occupa-
tion might serve as a bulwark against perpetrators of atrocities which have not yet
been caught and who might reoffend as soon as occupying forces depart. In both
cases, those combatants and civilians who would make a significant contribution
to B’s unjust resumption of hostilities have no claim against incurring the harms of
occupation, such as the harm attendant on not being able collectively to exercise
full de facto power over the entirety of B’s territory, or the harms attendant on their
freedoms being restricted (such as their freedom to move around without a permit,
or their freedom to be out past a certain time in the evening, etc.), or indeed the
harms attendant on not being able fully to control the resources which happen to
be on, or under, the region which occupying forces occupy. Put simply, they have
lost their right to govern over that territory in a way which is inimical to bringing
about a justifiedATC peace. In so far as A has objectively true reasons to believe, in
the light of B’s past actions, that B cannot be trusted to respect and promote that
value, it may temporarily claim B’s sovereignty rights for itself.
In those cases, military occupation is justified by appeal to imminent rights
violations—as in instance of self- or other-defence. But it can also be justified by
appeal to past violations—as an instance of self- or other-preservation. Suppose

that A, or some party on whose behalf it acts, are subject to imminent rights viola-
tions which can be parried only if A subjects to the harm of occupation members
of B. Suppose further that although those members are not the agents of those
imminent threats, they nevertheless bear a significant degree of responsibility for
the unjust war which B waged against A. Those wrongdoers’ past contributions to
A’s predicament render them liable to the harms of occupation as a means to rectify
a current and unjust state of affairs.18
Whether the occupation is justified overall does not merely depend on whether
it has a just cause: it also depends on whether it meets the other requirements for
justified force which apply to war and from which there is no reason to exempt
occupation any more than there was to exempt peacekeeping forces: the occupa-
tion is a proportionate response to the injustice which victims of B’s liable agents
have suffered; it is not conducted through the deliberate and indiscriminate killing
of innocent agents; it stands a reasonable chance of succeeding by means which do
not breach the requirements of proportionality and discrimination; it is the only
way to pursue the cause of a justifiedATC peace. Finally, and as I shall argue in
s.3.3.2, a just occupation, unlike a just war, must meet the requirement of compe-
tent authority.
The foregoing remarks do not imply that A becomes fully sovereign over B’s
territory, for to reiterate, it is a defining feature of a military occupation that its
officials only temporarily exercise limited powers of government, and aim to return
full powers to an indigenous regime as soon as they can do so without jeopardizing
prospects for a justifiedATC peace. What they do imply is that those of citizensB who
have provided A with a just cause for occupation have temporarily forfeited their
sovereignty rights (widely understood here to include rights over public property
and resources) over their territory. When citizensB forfeit their sovereignty rights
for the aforementioned reasons, by implication, citizensA temporarily acquire those
rights in respect of B. This in turn grounds A’s power, as Occupier, to issue direc-
tives to which B’s occupied civilians are subject, and places the latter under an
obligation to comply with those directives.
The Problem of Innocent Civilians

But now a serious difficulty arises, pertaining to innocent civilians. When issuing
its directives, levying taxes, and requisitioning resources, Occupier cannot discrim-
inate between those amongst occupied civilians whose contribution to the war
warrants their subjection to occupation, and those who did not contribute at all to
their regime’s (unjust) war, or who did do so but whose contribution falls below
the required threshold. The difficulty is not specific to occupation: likewise, a just
belligerent is usually unable to make fine-grained distinctions between those civil-
18 Some agents are doubly liable, as it were: not only are they currently posing a threat to occupy-
ing forces by, e.g., funding guerilla groups; in addition, they are also responsible for the war in the first
instance. This case does not impugn my arguments in this section. I am grateful to Alejandro
Chehtman for pressing me on this.
ians who may be deemed legitimate targets and those who may not. Proponents of
the neoclassical account standardly seek to solve that difficulty, in the case of tar-
geting decisions, by invoking the doctrine of double-effect, and by arguing that
killing non-liable civilians is permissible if their death is a collateral harmful con-
sequence of those decisions. As McMahan argues in his important article on mili-
tary occupation, one can deploy the doctrine in support of the claim that subjecting
non-liable civilians to the harms of occupation is permissible as an unintentional
though foreseen consequence of imposing those very same harms on liable agents.19
That said, the doctrine of double-effect has more limited reach in this case than
appears at first sight. I noted earlier that Occupier exercises limited governmental
powers over the occupied territory (OT). In so doing, it deprives occupied civilians
of the opportunity to exercise sovereignty rights over OT. Moreover, Occupier
typically uses those powers to restrict the freedoms of occupied civilians. Finally,
and this is crucial, Occupier also extracts material resources from the occupied
population by levying taxes, requisitioning materials, using military installations
and public buildings, and so on. Non-liable civilians suffer from the following
harms: the harm of not being able to exercise their sovereignty rights, the harms of
not being able fully to exercise their civil rights (for example, the rights to freedom
of movement, freedom of speech, etc.), and the harms resulting from income and
resource loss. The question is whether the doctrine can support the claim that
Occupier may justifiably hamper the exercise by occupied citizens of their sover-
eignty rights, as well infringe their civil and property rights.
For the argument to work, it must be the case that the group of civilians on
whom Occupier imposes a harm comprises both individuals who are still, or were,
a threat to Occupier and thus are liable to that harm, and individuals who are not
liable. It must also be the case that Occupier does not infringe non-liable civilians’
rights as a means to its ends. However, only some of Occupier’s acts can be charac-
terized in this way, to wit, those acts which infringe occupied civilians’ individual
rights. The imposition of a curfew, which McMahan himself briefly mentions, is a
good example. Only some occupied civilians are liable to suffer this restriction on
their freedom of movement; others are not, but as it is not possible to discern who
is liable and who is not, Occupier has no choice but to impose a blanket restric-
tion. Likewise with individual rights to material resources. Occupier cannot iden-
tify who, among those civilians, is liable to be subject to the burdens of rationing
and who is not—which burdens Occupier admittedly need not intend to impose
on the non-liable. Infringing the rights of non-liable civilians can plausibly be
construed as a merely foreseen side effect of Occupier’s self-protective ends rather
than a means to those ends.
However, and this is crucial, the doctrine of double-effect cannot justify the
infringement of the rights which non-liable civilians jointly hold with their liable
compatriots, such as jointly held rights over publicly owned resources and, cru-
cially, jointly held sovereignty rights. In the case of individual rights, such as the
right not to be killed, the right to freedom of movement, or the right to own
19 McMahan, ‘The Morality of Military Occupation’, 105–6.

resources for one’s daily needs, those who are not liable to suffer the burdens
attendant on rights infringement could enjoy those rights if it were possible to
identify them as non-liable and to distribute the burden accordingly. Put differ-
ently, it is not necessary to non-liable civilianB1’s survival or freedom of movement
that liable civilianB2 also enjoy or even hold those rights. But jointly held rights,
particularly those rights which citizens must exercise together, are very different.
Suppose that as a French citizen, I have pre-institutional moral rights over the
territory known as France. Even so, I can exercise those rights only if enough other
individuals, similarly situated in relation to the community known as ‘France’ both
hold and exercise those very same rights with me. Likewise, qua French citizen I
hold collective ownership rights over the natural resources to be found under that
territory and the public buildings, installations, and infrastructures which it con-
tains only if enough other individuals both hold and exercise those very same
rights. In so far as military occupation necessarily consists in the temporary suspen-
sion of all occupied citizens’ jointly held sovereignty rights, the loss of those rights
for the non-liable is not a collateral damage of Occupier’s policy: it is its main,
constitutive feature.20
Although the doctrine of double-effect does not provide support for some of
Occupier’s main rights qua Occupier, this should not lead us to reject all and any
kind of occupation—even occupation proceeding from a just war—as unjustified.
Most obviously, an occupation proceeding from a war that is unjustly undertaken
by B’s erstwhile regime might be welcome to those civilians as a means to provide
them with better opportunities for a flourishing life, and the temporary loss of
their sovereignty rights might be a small price to pay for this. Moreover, even if the
doctrine does not work in the case of jointly held rights, other arguments might,
and in particular, lesser-evil arguments—of a kind deployed to justify the deliber-
ate killing of innocent civilians in situations of supreme emergency. Still, one
should be wary of appeals to lesser evils. For such appeals in turn call for an account
of what constitutes an evil—as distinct from what is morally bad simpliciter—and,
more precisely, what constitutes an evil of a kind as to warrant subjecting civilians
to occupation-related harms. In CW (s.7.2), I argued that the evil of large numbers
of non-liable people suffering the loss of all their rights can justify the deliberate
and lethal targeting of innocent civilians. In so far as the harms which a military
occupation inflicts on those civilians are typically lesser than the loss of all their
rights, they can be justified by appealing to the importance of Occupier, not just
of forestalling the loss by its own members of all their rights, but of protecting
lesser rights, such as the sovereignty rights the violation of which gave them a just
cause for war in the first instance. Crucially, and to reiterate, in so far as Occupier
has a justification for so acting, it cannot be said to violate the rights of those civil-
ians. At the same time, however, it deprives the latter of freedoms and resources to
which they still have rights. The point is worth stressing, because according to the
20 For an analysis of the notion of suspended sovereignty in public international law, see A. Yannis,
‘The Concept of Suspended Sovereignty in International Law and Its Implications in International
Politics’, European Journal of International Law 13 (2002): 1037–52.
laws of occupation, compensation for the harms of occupation is in order only if

Occupier requisitions privately owned resources from occupied civilian popula-
tions, irrespective of the latter’s contribution to the war and of the status of that
war. On my account, by contrast, it is owed only if civilians have not acted in such
a way as to warrant losing their resources.
Let us take stock. I have argued that, following a just war, the military occupa-
tion of a community a significant part of whose population poses an ongoing
threat to peace is sometimes justified—particularly if it can thereby bring about a
justifiedATC peace—notwithstanding the fact that the occupation necessarily
infringes the rights of innocent civilians.
From an Unjust War to an Unjust Occupation

So far, however, I have assumed that A becomes an occupier as a result of waging a
just war against B. Suppose now that at time t1, A’s regime orders A’s army to
invade B’s territory without just cause. At t2 B’s regime surrenders and A’s armed
forces occupy B’s territory. Ex hypothesi, in so far as combatantsA were fighting an
unjust war against B at t1, they lacked the right to kill combatantsB in prosecution
of that war. By the same token, they lack the right to stay in B once the fighting
has ceased. Consider a domestic analogy. Andrew wrongfully breaks into Ben and
Charlotte’s house and orders them at gunpoint to hand over all their belongings.
Andrew decides that he might as well use the house as a base for his criminal oper-
ations. He is willing to have Ben and Charlotte stay there provided they make
themselves scarce and supply him with the resources he needs to keep his business
going. He makes it clear that he will not hesitate to use lethal force to ensure their
compliance. They for their part justifiably believe that it is safer for them and their
children to stay in the house: it is not as if they have anywhere else to go, and at
least, that way, the children will remain warm and fed. Clearly Andrew lacks the
right to remain in the house: neither Charlotte nor Ben are under a duty to him to
provide him with any assistance at all, or a duty not to try to kick him out. In fact,
they would do him no wrong by killing him should the opportunity arise and
should this be the only way to get rid of him.
The same applies to Occupier. Ex hypothesi, it does not have a just cause for
invading B’s territory, from which it follows that it lacks a just cause for remaining
there (unless—as we shall see below—circumstances change in such a way that it
comes to acquire a justification for staying). To the extent that B’s members have
not threatened the rights of A’s members to territorial integrity and political sover-
eignty, and to the extent that they are not posing an ongoing threat either to those
rights or to the rights of fellow members, they are not liable to lose their own
sovereignty rights over their territory. More precisely, they are not liable to lose the
right to govern themselves, to having their freedoms curtailed, and to the appro-
priation by occupying forces of their privately and publicly owned resources.
Some might object that B’s citizens are liable to incur those burdens if their
regime has explicitly transferred its powers over B to A’s forces following its mili-
tary defeat (as happened in some European countries whose regimes acceded to
German occupation during the Second World War). In this case (my opponent
might press), the aforementioned domestic analogy does not work since B’s regime
has authority over B’s population in a way that Ben does not have over Charlotte.
However, the objection strikes me as implausible. Consent given under the
duress of impending military defeat at the hands of an unjust aggressor cannot
bind B’s officials and citizens to comply with the directives enacted by A’s o fficials—
any more than powerful insurgent factions within a democratic community can
elicit valid consent to the imposition of a dictatorship following an irrepressible
coup d'état.
That very last point is worth stressing: on my account, there is no morally rele-
vant difference between a foreign unjust occupier and an indigenous dictatorial
regime. This is of a piece with the key tenets of my cosmopolitan account of jus-
tice. For on that account, you recall, individuals have sovereignty rights on two
grounds: as a means to ensure that they enjoy the freedoms, and have access to the
resources, which are constitutive of a flourishing life, and in fulfilment of their
personal prerogative to constitute political associations once they have met their
obligations of justice to distant strangers. The duty to respect those rights, in turn,
is held by all and sundry, irrespective of borders. In so far as individuals’ member-
ship in this or that political community is morally irrelevant to the conferral of
those rights, rights-violators’ membership in this or that political community is
also irrelevant to the moral status of their dereliction of duty. Violations of (in this
instance) sovereignty rights are not the more or less grievous for being committed
by foreigners as opposed to compatriots.
From an Unjust to a Justified Occupation

Sometimes, however, circumstances might change, such that A comes to acquire a
justification for occupying B even though the war which lead to the occupation
was unjust. In particular, the outbreak of a civil war and concomitant rights viola-
tions between different factions in the aftermath of the invasion might be thought
to impose on third parties a duty to take appropriate measures if, first, those who
started the civil war did so unwarrantedly and, second, the unjust belligerent’s
troop withdrawal would lead to the collapse or continuous absence of all major
social and economic institutions and facilities. In transforming itself into an occu-
pying force, Occupier exercises de facto power over both victims and initiators of
the civil conflict. If the factions who started the civil war did so without just cause,
or had a just cause but are fighting that war unjustly (for example by committing
atrocities against their opponents), then they lack a claim not to be occupied. As
for the victims, if the unjust war waged against their community is partly caused
by the civil war, they might be said to have a claim to be occupied. The 2003 war
against Iraq and the subsequent occupation of that country by the US-led coalition
aptly illustrate the point that a belligerent may acquire, if not an all-things-considered
justification, at least a just cause, for militarily occupying a country against
which it should not have gone to war in the first instance. The occupation of Iraq
was sanctioned by the UN Security Council in Resolution 1483 and the law of
occupation was deemed to apply to it as soon as coalition forces started exercising

control over Iraqi territory. As a matter of fact, a just cause is all that the coalition
had, for it was conducted in flagrant violation both of the law of occupation and
of basic principles of morality—as attested by the number of Iraqi laws which the
CPA overturned without warrant, by the Bush administration’s failure properly to
prepare for it, and so on. But had the occupation been conducted differently, it
might have been justified all things considered notwithstanding its roots in an
unjust war.21
To some, the foregoing points will seem too hasty, vulnerable as they might
seem to the familiar problem of moral hazard. For if an unjust war can issue in a
justified occupation as a result of the chaos created by the war, an unjust belligerent
would have every incentive to prosecute its war in such a way as to bring about civil
strife, thanks to which it would then acquire a justification for occupying the
country and appropriating its resources. By that token, citizens of the unjustly
aggressed community (the objection might continue) ought not to resist the
aggression, precisely for fear of triggering a civil war and thereby providing their
unjust aggressor with a justification for its occupation.
We encountered a similar objection in s.2.4.2, when discussing the view that a
belligerent which started an unjust war might come to acquire a justification for
continuing it. To apply here the reply I gave there, in so far as Occupier would
need to devote enormous resources of its own in order to maintain order, and as it
would need to maintain order to further its own ends in the occupied territory, I
am not sure that moral hazard is as serious a problem as it appears. In any event,
here we are concerned with delineating whether an occupation is justified. That an
unjust belligerent would have an incentive to trigger a civil strife for the aforemen-
tioned reason does not render its conduct just. Even if its subsequent occupation
of that territory would be justified as a means to bring about peace, it would still
originate in a serious wrongdoing, which in turn would have some bearing on that
belligerent’s rights, powers, duties, and liabilities. In particular, as we shall see in
s.3.3.3, occupying powers are under a duty to promote the welfare of occupied
civilians. This requires that they sometimes expose their own forces to considerable
risks for the sake of those whom they occupy. One of the most difficult questions
in war is that of the balance to be achieved between the interests of combatants in
surviving the war more or less safe, and the interests of the civilian populations
which are caught in the crossfire. It would seem appropriate to hold occupying
forces under a duty to incur greater risks for the sake of occupied civilians when the
21 See Rocheleau, ‘From Aggression to Just Occupation?’ and McMahan, ‘The Morality of Military
Occupation’. See also D. J. Scheffer, ‘Beyond Occupation Law’, The American Journal of International
Law 97 (2003): 842–60; P. Cockburn, The Occupation—War and Resistance in Iraq (London: Verso,
2007); L. Diamond, ‘What Went Wrong in Iraq’, Foreign Affairs 83 (2004): 34–56; G. H. Fox, ‘The
Occupation of Iraq’, Georgetown Journal of International Law 36 (2005): 195–297; A. A. Allawi, The
Occupation of Iraq—Winning the War, Losing the Peace (New Haven, Conn.: Yale University Press,
2007). The occupation of Iraq is sometimes described as transformative, so comprehensive was the
Coalition Provisional Authority’s programme of institution-building. I will scrutinize this kind of
occupation in ch. 8.
occupation, though justified for the aforementioned reasons, nevertheless proceeds

from an unjust war, than when it is fully just.22
The claim that occupying forces are under a duty to occupied civilians under
those circumstances might elicit the following objection:23 it implies that Occupier
wronged occupied civilians by waging a war one outcome of which is the outbreak
of a civil war. This in turn seemingly implies that those civilians had a right that the
pre-war conditions under which the civil war did not occur should have been
maintained. Quite often, however, they experienced pre-war civil peace thanks to
their erstwhile regime’s highly repressive measures (routine use of torture, extraju-
dicial imprisonment or killing of opponents, etc.) In so far as no one has a right to
be protected from harm at that particular cost, Occupier did not wrong those
civilians, at least on that specific count—a point which, incidentally, is compatible
with the claim that its war was unjust.
Let us accept that civilians did not have a claim to be protected from civil
war-related harms at that cost, and that Occupier therefore did not violate, or
indeed justifiably infringe, that particular claim. Nevertheless, on any plausible
morally neutral definition of harm, those civilians are harmed by the outbreak of
the civil war; furthermore, the civil war itself is partly caused by Occupier’s con-
tribution to the fall of the enemy regime. We must then distinguish between two
cases: the case where Occupier could not have conducted its war against that
regime in such a way as to minimize or lessen the risk that a civil war would ensue,
and the case where it could have done so. In the latter case, Occupier did act
wrongfully towards civilians by not taking the required steps, and can be held
under a remedial duty to stay in that country. In the former case, and subject
to the aforementioned effectiveness caveat, Occupier can be held under a duty to
stay, not on remedial grounds, but at the bar of a general duty of assistance—here,
humanitarian assistance—to those in need, if only because the fact of its presence
in the country might make it particularly apposite for its forces to carry out the
relevant task.
Note that in the case I have just discussed, the justification for the occupation
lies in the importance of promoting the human rights of occupied civilians from
threats imposed by their own compatriots. In some instances, however, it is con-
ceivable that an unjust belligerent might be justified in occupying its enemy out of
legitimate concern for its own members. Suppose that once B’s regime has surren-
dered, some factions within B decide to carry on fighting but use wrongful
means—such as terroristic attacks against A’s citizens back home. Suppose further
that were A to withdraw from B’s territory, this would not stop those attacks—but
that the contrary were true. This, I think, is another instance where, absent a cred-
ible indigenous regime taking on the task of governing it in such a way as to block
further threats to its own members, A has acquired a justification for remaining in
22 See also McMahan, ‘The Morality of Military Occupation’, 118, for a similar point.
23 The objection was put to me by a participant at the Princeton PEPA seminar in February 2013,
where an earlier draft of this chapter was discussed.
B. But here too the fact that it lacked a justification for going to war in the first
instance has a bearing on its rights, duties, powers, and liabilities.
When To Stay, When To Leave

In just the same way as an account of legitimate resort to war must attend to the
question of war exit, an account of legitimate military occupation must attend to
the question of when Occupier may, or must, withdraw its forces. Earlier I argued
(in line with the law of occupation) that occupying a territory is not (descriptively
and normatively) tantamount to annexing it, and that an occupation is deemed
successful if it has brought about genuine prospects for a justifiedATC peace. I also
noted that a party whose occupation is unjust at t1 may come to acquire a justifi-
cation for remaining in situ at t2, which justification lies in the importance of
bringing a justifiedATC peace within the occupied community, between the occu-
pied community and Occupier, or both. Those claims imply that Occupier is
under a duty to withdraw as soon as the occupation has achieved its goal, unless
the now-effectively-sovereign community has already agreed to Occupier’s contin-
uing military and/or civilian presence (without concomitant rights of govern-
ment)—as was the case in Germany after 1949.
However, the claim that Occupier is under a duty to withdraw under those cir-
cumstances does not imply that it is under a duty to stay until those circumstances
obtain. Suppose that its presence triggers waves of sectarian violence within B
which it will not be able to stem by remaining in place as an occupying force. Or
suppose that its ongoing presence in B triggers waves of attacks against its own
troops and/or its own citizens back home. On the first count, Occupier faces a
dilemma: the worse the violence, the harder it is to envisage withdrawing; yet, by
remaining, it contributes to the escalation in the violence.24 Under those circum-
stances, in so far as the longer it remains in place the less likely it is that the occu-
pation will succeed, it may, in fact must, withdraw. Yet, this does not imply that B
may justifiably be left to its own devices. For it could be that a party other than
Occupier might be in a better position to bring about prospects for a justifiedATC
peace (as, for example, when an occupying power which has not managed to gain
enough trust amongst occupied civilians is replaced by an international force
which is perceived to be more impartial.) It could also be that although the occu-
pation will not succeed at bringing about such a peace, a full-scale war could,
which provides Occupier with a justification for turning itself into a belligerent
against B. At that point, though, our inquiry would partake of the ethics of the
resort to war, and no longer of the ethics of the transition from war to peace.
On the second count, whether Occupier may leave (though it would not have
achieved its just end by so doing) on the grounds that the occupation is too costly
24 See Edelstein, Occupational Hazards, 88ff. For a discussion of Occupier’s duty to leave, see
Evans, ‘Moral Responsibilities and the Conflicting Demands of Jus Post Bellum’, 160–2. For a philo-
sophical discussion of some of the complex issues raised by violence resulting from occupation, see
S. Bazargan, ‘Proportionality, Territorial Occupation, and Enabled Terrorism’, Law and Philosophy 32
(2013): 435–57.
to its troops and/or citizens, depends on the relative weight it may attach to the
interests of its own members relative to the interests of occupied populations. As I
argued in s.3.2 in connection with the ethics of peacekeeping, agents who are in a
position to protect third parties from severe rights violations at the hands of their
compatriots are under a duty both to incur the material costs of the war and to
expose themselves to the risks inherent in waging it. If so, there is such a thing as a
duty to carry out a military occupation with its concomitants risks and costs, at
least when the occupation is justified as a means to bring about a justifiedATC peace,
but provided that interveners and occupiers’ own prospects for a flourishing life
would not be jeopardized by the military occupation.25
One final point on the ending of occupation. Suppose that Occupier is under a
duty to leave at t2 on the grounds that its presence is a factor for dangerous insta-
bility. Occupier may still have moral obligations vis-à-vis formerly occupied popu-
lations if the latter have been rendered so vulnerable by the occupation that they
are not yet in a position independently to exercise their political, civil, and eco-
nomic rights. Thus, in 2008, the Israeli Supreme Court held that, although Israeli
forces had left Gaza, the fact that, as a result of the occupation of that territory, the
population of Gaza was entirely dependent on Israel for the provision of electricity
placed the latter under an obligation to continue to do so—an obligation which,
it also held, Israel was in fact meeting at the time.26 The general point holds
whether or not the occupation was just to begin with, but particularly so if it was
unjust in the sense that it had no justification in the first instance and/or that it was
conducted unjustly.
3.3.2. Who Can Occupy? The Problem of Competent Authority

I have argued that belligerents sometimes have or acquire a just cause for occupy-
ing and exercising governmental powers over their vanquished enemy, subject to
the requirements of proportionality, necessity, reasonable chance of success, and
discrimination. It remains to be seen whether it is a necessary condition of a just
occupation that it be conducted by a competent authority.
In s.2.2, we saw that a decision to seek peace terms on behalf of a political com-
munity must be taken by actors who have a claim to authority over that commu-
nity—where it has that authority in virtue of its ability and willingness ‘to issue
and enforce directives relating to the ending of the war which, under the precise
circumstances, would go further towards the respect and promotion of human
rights than would directives issued by other actors’. This is because that decision is
a fundamentally political act—bearing as it does on the ways in which the commu-
nity will be governed after the war, to which ends, and by which means. Similar
25 For an extensive account of the duties owed by the international community to communities
torn by civil war, see C. Hermanson, Duties in the Wake of Atrocities: A Normative Analysis of Post-
Atrocity Peacebuilding, D. Phil thesis, Oxford, 2012.
26 See http://elyon1.court.gov.il/files_eng/07/320/091/n25/07091320.n25.pdf and, for discus-
sion, Y. Ronen, ‘Post Occupation Law’, in J. S. Easterday, J. Iverson and C. Stahn (ed.), Jus Post Bellum
(Oxford: Oxford University Press, 2014).
considerations underlie the issue of competent authority and military occupation.

In s.3.3.1, I described two scenarios in which a belligerent A could justifiably
occupy part or whole of B’s territory: when, following a war of aggression by B
against A, factions within B would attempt to resume the war following their
regime’s defeat; when, following A’s war of humanitarian intervention against B,
factions within B would mount another campaign of violence against the interven-
tion’s beneficiaries. In both cases, A as an occupying power exercises justified pow-
ers of government over B’s territory: it enacts laws, punishes wrongdoers, and levies
taxes; and it does so sometimes against citizensB who are liable to be so treated in
virtue of their responsibility for providing A with a just cause, sometimes against
citizensB who are not liable but who may nevertheless be subject to the harm of
occupation as the lesser of two evils.
In the first case, A, as an occupying power, justly occupies B only if its own
members, on whose behalf the occupation of B is carried out, consent to it, either
presumptively or, as the case may be if conditions are propitious, explicitly. One
might think that this claim contradicts the view that a war in defence of rights
need not be waged by a competent authority in order to be just: for if that claim is
true, then it seems unclear why a military occupation, which also derives its justi-
fication from the importance of rights, must be carried out by a competent author-
ity in order to be just. In fact, there is no contradiction. For the claim that fulfilment
of the competent authority requirement is not a necessary condition for a just war
is grounded in the importance of ensuring that the victims of rights violations have
the right to defend one another when the conditions for authorization are not
propitious—when, that is, they are subject to ongoing lethal threats. But our con-
cern here is with occupation post-conflict, when there is a high risk that, absent
occupying forces, some agents in B will resume hostilities. Under those circum-
stances, A’s officials are in a position to seek an occupation mandate from their
fellow citizens, and are under an obligation to do so on pain of the occupation
being unjust.
In the second case, consent is also required from those civilians in B on whose
behalf the intervention was initially carried out, who stand to benefit from it and
yet are at risk of being harmed by it. In both cases, it is precisely because the fight-
ing has ceased, lives are not immediately under threat, and A, as an occupying
force, exercises powers and carries out tasks of government, that it makes sense to
accept the requirement of competent authority in the particular context of an
occupation: the decisions to be made in the aftermath of a life-threatening emer-
gency are complex, require careful handling and garnering of all relevant informa-
tion, call for accountability, and more generally are of inherently public, political
nature.
There are important differences, thus, between the ethics of defensive killing and
the ethics of occupation, such that jettisoning the requirement of competent author-
ity as a necessary condition for a just war does not entail that one should reject it as
a necessary condition for a just occupation. However, those points alone do not tell
us which kinds of political actors have the requisite competent authority. As we saw
in ch. 1, the rights of institutional actors flow from and supervene on the rights of
individuals; furthermore, to say that those actors have political authority over those
who are subject to their directives is to say that they enable the latter to do better, at
the bar of their fundamental rights and duties, than they would do under available
and feasible alternatives. The right to carry out an occupation is a right held by
individuals which they can transfer to institutional actors if doing so is a more effec-
tive way of protecting their human rights. On that view, whether an institutional
actor has the right to carry out a military occupation on behalf of the latter’s bene-
ficiaries is entirely dependent on its ability to carry out the task at issue. Standard
reasons as to why it is better, all things considered, to transfer rights of government
to institutional actors are that the public authorization of the use of coercion is more
likely to be transparent, fair, and impartial, and that the public use of coercion as a
means to enforce those directives is more likely to be accountable.
In line with my arguments for (qualified) multilateral peacekeeping in s.3.2,
those points tell in favour of (qualified) multilateral occupation. Yet some might
perhaps object that granting occupiers the right directly to govern over political
communities undergoing a difficult transition from war to peace is an unaccept-
able breach of the latter’s sovereignty rights, in a way that granting those institu-
tions the right to oversee a ceasefire and provide humanitarian relief is not: the risk
that the war will resume, together with the humanitarian crises which both result
from and fuel the threat of imminent violence, permit breaches of sovereignty,
whereas the relatively peaceful, or at any rate less violent, conditions of a post-conflict
situation do not. However, in so far as B is ex hypothesi unable and/or unwilling to
govern itself in a way which is conducive to a justifiedATC peace, and in so far as its
inability and unwillingness so to govern itself will cause some of its members and/
or outsiders to suffer severe rights violations, its members lack a claim against
interference by outsiders. Under those circumstances, the fact that those outsiders
are international institutions is entirely irrelevant, so long as they are capable of
discharging the tasks of government. In fact, when the occupation follows a trans-
national war between two sovereign political communities, the case for having it
carried out by those institutions is all the stronger, since members of the unjustly
aggressed party might not be able to quell their desire for revenge and thus might
be more tempted to authorize or commit exactions against occupied civilians, to
appropriate more resources from their erstwhile enemy than it morally permitted,
etc. The occupation of what was to become the Democratic Republic of Germany
by the Soviet Army in the immediate aftermath of the Second World War provides
a good example of such abuses.27
Granted, institutions, whether national or international, are not always trans-
parent, fair, and impartial—far from it. With respect to the resort to war, when
they are derelict in their duty to protect the fundamental rights of their citizens, or
are simply incapable of doing so, the right to authorize and wage war reverts to the
latter. By that token, when multinational institutions are similarly derelict in their
duty to conduct a military occupation, the right to occupy reverts to whichever
27 See, e.g., G. MacDonogh, After the Reich—From the Liberation of Vienna to the Berlin Airlift
(London: John Murray Publishers, 2007).
institutional actor is in the best position to fulfil the tasks of an occupier and, in so
doing, to bring about a justified peace. In other words, just as there is nothing
intrinsically valuable in entrusting states with the task of resorting to, and using,
violence, there is nothing intrinsically valuable in entrusting multinational institu-
tions with the task of occupying a defeated belligerent.
3.3.3. Rights, Competences, and Duties of Occupying Forces

To recapitulate, I have argued that the moral status of the war which B conducted
and which led to the occupation of B’s territory has a normative bearing on the
moral status of that occupation. While defending that claim, I sketched out some
of occupying forces’ and occupied populations’ rights and duties. My aim in the
next two subsections is to develop the latter point.
According to 1907 Hague Regulations and the 1949 4th Geneva Convention,
occupying forces must respect the laws in force in the occupied territory at the
time the occupation starts, except in the following three cases: those laws violate
the laws of war; the preservation of law and order in the territory requires that it
should pass new laws; Occupier’s war effort outside that territory requires new
laws. Occupier may enact security measures as well as measures aimed at meeting
the acute needs of civilians (via rationing, for example). Occupier may also decide
to raise taxes via legislation, provided it does so for the purpose of defraying the
costs of the occupation and meeting the needs of the army in place (as opposed to
enriching itself ). Finally, not only is Occupier legally competent to legislate and in
so doing to confer new rights and duties on Occupied; it may also enforce both its
laws (typically via military tribunals) and indigenous laws in force at the start of
the occupation should local courts be unavailable. Whilst the latter have jurisdic-
tion over occupied populations, they have no power over members of Occupier’s
armed forces and associates. This proved to be a point of contention (and tension)
in the occupation of Iraq: shortly before leaving his post as Head of the Iraqi pro-
visional government, US Consul Paul Bremner exempted private security contrac-
tors who worked for the US armed forces from the jurisdiction of local Iraqi courts.
That decision elicited considerable anger in Iraq given that, at the time, American
courts were slow in bringing those contractors to trial despite evidence of serious
wrongdoings on their part.28
Furthermore, Occupier is under a duty to ‘restore and ensure, as far as possible,
order and civic life while respecting the laws in force in the country unless absolute
necessity dictates otherwise’.29 A duty to restore and ensure civic life can include a
28 The relevant order, CPA 17, is available at http://www.iraqcoalition.org (accessed on

13/01/2016).
29 1907 Hague Regulations, art. 43. (My translation.) The standard English translation (as found
in, e.g., Roberts and Gueff (eds.), Documents on the Laws of War), reads more restrictively as ‘to restore,
and ensure, as far as possible, public order and safety [as distinct from civic life—l’ordre et la vie pub-
lics], while respecting….’ But the Hague Conventions were adopted in French, and it is the French
version, therefore, which is authoritative. (See, e.g., Dinstein, The International Law of Belligerent
Occupation, 89–90, for discussion of this particular clause.) My description of the Convention accord-
ingly refers to the French text.
duty to provide the conditions under which occupied civilians are able to engage
in a wide range of economic, social, cultural, indeed political activities. A duty to
restore and ensure order, on the other hand, may easily be read as a duty to protect
the lives and limbs of civilians—not merely from attacks by their fellow citizens
(when local courts are unwilling or unable to step in) but also from threats posed
by its own forces in violation of the laws of war (which the Occupier is under a
duty to respect, as any belligerent party is.) At the same time, Occupier is also
under a duty to take any measure necessary to quash unrest as triggered by resist-
ance activity on the part of non-uniformed civilians who, in so far as they are not
incorporated in regular forces, do not enjoy the privileges of combatantcy. It thus
behoves Occupier to determine through legislation which acts of civic life consti-
tute an unacceptable breach of order—sabotage being the most obvious example.
Finally, Occupier also has some rights over Occupied’s resources. For example, it
may levy taxes on the occupied population; it may also seize privately owned mate-
rial resources and use them to meet the needs of both its forces and occupied
populations, or in prosecution of a war it is still waging against a third party (irre-
spective of that war’s moral status). Crucially, such requisitions as it orders cannot
serve to meet the domestic needs of its own citizens; finally, it cannot conscript
occupied civilians into its own army or workforce.
The law so understood applies to all occupying forces, irrespective of the moral
status of the war which led to the occupation. In the light of the neo-classical
account of the relationship between jus ad bellum and jus in bello, the morality of
occupation differs from the law thereof in important respects. Most obviously, an
Occupier, whether just or unjust, lacks the right to use occupied civilians’ resources
to fund its unjust wars, as well as the right to enforce their (wrongful) directives by
killing those who oppose them. Moreover, an unjust Occupier lacks the moral
power to issue directives aimed at safeguarding its forces from civilians’ justified
attacks; it also and no less clearly lacks the moral power to raise taxes, requisition
movable resources and use immovable property as a means to meet their occupa-
tion-related needs. Those points apply whether the Occupier is unjust in virtue of
being in situ as a result of waging an unjust war and not having acquired a justifi-
cation for staying, or whether it is unjust because, notwithstanding the fact that it
did wage a just war, it lacks such a justification all things considered. They also
apply whether Occupier lacks a just cause for its occupation, or whether it has a
just cause but fails to meet the other requirements for a just occupation. By con-
trast, a just occupier is morally permitted and has the moral power to carry out all
those tasks. Finally, an unjust Occupier does not have the right to kill insurgents
who justifiably resist its rule by resorting to lethal force (a point to which I shall
return in s.3.3.4), whereas a just Occupant has the right to kill unjust insurgents
who so act (subject to the usual conditions of proportionality, necessity, etc.).
That said, the laws of military occupation stipulate that Occupier is competent
to govern not merely to those ends but also to protect occupied civilians—indeed,
more widely, to ‘restore and ensure public order and civic life’. If Occupier occu-
pies B unjustly, does it have the moral power to legislate to those protective ends?
Suppose, first, that Resistants blow up an enemy military convoy—which (let us
assume) serves no purpose at all in this context. Some citizens are foreseeably
though unintentionally killed in the explosion (e.g. they live by the railway track.)
If the account of state legitimacy which I briefly sketched out in s.1.2 is sound, it
seems that Occupier may not punish Resistants. For on that account, a state is
legitimate only if it enables those who are subject to its directives better to abide by
their duties and to enjoy their rights than under available feasible alternatives.
Given that Occupier ex hypothesi lacks a justification for occupying B and ought to
withdraw, by implication it is not the case that there is no available feasible alter-
native, and Occupier therefore lacks the power to issue the relevant directives
against Resistants.
Note that this is compatible with the point that, if Occupier does not leave, and
if its forces find themselves in a position to protect some occupied civilian from
having her rights violated, here and now, by some of her compatriots (who are
intent on, e.g., raping or killing her), then they may so act. Likewise, should Ben
act in such a way as to violate a fundamental human right of his son’s whilst
Andrew is still unjustly occupying the house, Andrew may forcibly prevent him
from doing so. That either Occupier or Andrew should be placed in a position so
to act as a result of their own wrongdoing is irrelevant to their being permitted to
provide basic, life-saving assistance—which is not tantamount to exercising powers
of government—to innocent third parties: whilst Ben’s son and B’s innocent citi-
zens are wronged by (respectively) Andrew and Occupier’s invasion and subse-
quent occupation of their home and country, surely they are not wronged by those
wrongdoers’ attempt to rescue them from another grievous wrongdoing.30
3.3.4. Rights and Duties of Occupied Populations

So far, I have focused on what Occupier may or may not do to occupied popula-
tions. In this section, I address the issue of the grounds upon which, if any, the
latter owe an obligation to the former to comply with its directives. If Occupier has
a justification for maintaining a military presence in B, and if its officials’ directives
satisfy the requirement that they should enable occupied civilians better to enjoy
their rights and fulfil their duties, then those directives are binding and occupied
populations are under a duty to comply. Note that they owe that duty not just to
one another (as grounded in a prior moral duty to respect those moral rights which
officials’ directives protect in law), but also to those officials themselves, who (as I
argue in CW, s.1.4) have a right not to be interfered with when carrying out the
tasks of government.
30 The question I address here is subtly different from another question, which I do not tackle,
namely: what occupying forces may do when faced with insurgents who do not abide by moral
norms—for example, who deliberately target innocent non-combatants or unjustifiably use them as
human shields or who deceive occupying combatants into believing that they are civilians, in order the
more easily to kill them. In this case (unlike the case raised in the main text), insurgents commit those
rights-violations as part of their fight against occupying forces. That question arises in war, and not just
in the context of occupation. I make a stab at it in CW, ch. 7.
On the Duty To Comply with an Unjust Occupier’s Directives

Suppose, however, that the occupation is unjust. May civilians resist? Not accord-
ing to the law, particularly if their defeated regime has surrendered and concluded
an armistice with Occupier, following which it has dissolved itself. Whether or not
the defeated regime was morally legitimate and thus had proper authority to con-
clude the armistice is irrelevant. Nor, according to the law, may civilians act in such
a way as to prevent Occupier from exercising its legal rights and fulfilling its legal
duties qua occupying power. For example, they may not commit acts of sabotage
or use their protected status as civilians to approach occupying combatants and kill
them. If they do, Occupier’s officials may punish them as if they were common
criminals, indeed its combatants may kill them in self-defence. In a well-known
passage on guerrilla warfare, Michael Walzer discusses the example of French par-
tisans who, during the occupation of France by German in 1940–44 disguised
themselves as peasants and were thus able to come close to and kill German com-
batants. On standard interpretations of the laws of occupation, they were deemed
to have betrayed the armistice agreement passed between the Vichy regime and
Germany and thus to commit a grievous breach of faith. At the same time, Walzer
notes, to the extent that individuals are not subsumed under the decisions of their
government, and to the extent that the German Occupation was unlawfully con-
ducted, those partisans could be deemed to have acted legitimately. Yet, those
German combatants who attempted to kill them in self-defence, and those officials
who punished them for murder, also had a justification for so acting, since they
could not hope to ensure and restore public life under the constant threat of attack.
As Walzer put it, ‘resistance is legitimate, and the punishment of resistance is legit-
imate. That may seem like a simple stand-off, and an abdication of ethical judg-
ment. It is actually a precise reflection of the moral realities of military defeat’.31
Insurgency is an unavoidable feature of occupation. On Walzer’s account, it seems,
resistants, or insurgents, and occupiers are morally on a par, even if the occupation is
unjust. As I suggested in s.3.3.3, I do not think that this is right. For a start, the fact
that Occupier is unjust means that occupied civilians are not under a duty to its
officials not to resist their rule by force or stealth, nor are they under a duty to its
forces to provide them with the resources they need to function as an occupying
force.
To be sure, the point that occupied civilians do not owe a duty of allegiance to
Occupier is compatible with the claim that they can sometimes be under a duty to
obey its directives. And this claim in turn might provide support for Walzer’s view
that unjust occupying forces act legitimately when punishing resistants who kill
their combatants.
Here is an interesting, though in the end unsuccessful, argument in support of
that claim. Suppose that a group of us pool some of our resources in common and
set out on a hiking trip. We all agree that, in the event that we will be attacked by
highway robbers, we will do what John decides, for he is an experienced hiker who
31 Walzer, Just and Unjust Wars, 176–9.

knows the area well and we trust his judgement: put differently, we confer on John
the authority to decide how best to preserve our commonly owned resources.32 Lo
and behold, two days into our hike, we are set upon by robbers who threaten to kill
us unless we hand over our money. John decides to comply. In the light of our
unanimous decision to follow his lead, I would wrong both John and all other
members of the groups if I were singlehandedly to override John’s decision by
resisting the robbers. If that is correct, B’s occupied civilians owe it both to one
another and to their officials not to resist the decision made by B’s regime to hand
over power to A’s officials. Or so the argument claims.
I remain unconvinced. Political regimes never secure unanimous consent for
their decision and are hardly ever in a position to secure consent from their citizens
whilst trying at the same time to repel an invasion. However, suppose that they do
secure such consent. Even so, the argument remains limited in scope—though
understanding why that is so provides us with the right kind of justification for the
view that occupied civilians must sometimes comply with Occupier’s directives.
The rights which B’s officials surrender to A’s officials on behalf of occupied civil-
ians are of two kinds: jointly held rights over communal goods such as collectively
owned property or sovereignty rights; and individual rights, such as the right to
freedom of movement (which might be threatened by curfew), rights over pri-
vately owned resources (which might be threatened by requisitions), or rights not
to be tortured, arbitrarily imprisoned, summarily killed, and so on. Even if civil-
ians consent to have their individual rights protected under A’s rule, the mere fact
that they have consented does not put them under a duty to B’s officials or to their
fellow citizens to comply by those of A’s directives which are relevant to those
rights. Should they change their mind as to the appropriateness, for the protection
of those rights, of B’s decision to surrender, then prima facie they may do so.
Similarly, even if I undertook at time t1to respect John’s decision, when faced with
the robbers, as between giving them my money or letting them shoot me, surely I
am not thereby under a duty to John or indeed to the rest of you not to revoke that
commitment at t2 when faced with the robbers.
Surprisingly perhaps, the point holds even when jointly held rights are at issue.
For it is a central assumption of this book that individuals are not under a duty to
divest themselves of the resources and freedoms which they need to lead a flourish-
ing life for the sake of protecting the rights of others—even when those rights are
jointly held. By implication, the fact that civiliansB entrust their regime with the
task of government does not bind them to its decisions if their prospects for a
flourishing life would be undermined by their regime’s decision to surrender power
over B’s territory to A’s forces. Likewise with the robbers’ case. That I undertook at
t1 to entrust John with the task of deciding to surrender wealth over which I have
some right is not enough to place me under an obligation at t2 to abide by his
decision if, in so doing, I shall suffer human-rights violations.
32 I owe this example to Patrick Tomlin. Both he and Will Kymlicka pressed me hard on this par-
ticular issue, for which I am grateful.
That being said, and as Walzer rightly notes, in some cases a decision by some
occupied civilians to resist occupying forces’ violations of their rights will result in
severe harm to their compatriots. In other cases, those directives might in fact help
protect those rights. Failure to comply under those circumstances would be mor-
ally wrong—not vis-à-vis Occupier, but vis-à-vis other occupied civilians. The fact
that they consented to their regime’s making such decisions on their behalf in the
first instance is irrelevant to the duty to comply.33
Where does this leave Walzer’s claim that occupiers and resistants both act legit-
imately when exercising force against each other and that the former are neverthe-
less permitted to punish the latter for so acting? Note, first, that resistants may well
be under a duty to their compatriots not to kill the occupiers. However, second,
even if they act impermissibly, it does not follow that occupiers are justified in
punishing them. True, resistants’ acts of killing make it very difficult for occupiers
to secure order and civic life. Ex hypothesi, however, the occupation is unjust. The
right response, for occupiers, is not to inflict further harm on civilians, but simply
to leave.
But now here is a twist. Suppose that A occupies B unjustly but wages a just war
elsewhere, against C. Consider the following example. In June 1940, under the
terms of a secret protocol which formed part of the Molotov–Ribbentrop Pact of
Non-Aggression signed by the USSR and Germany in August 1939, Soviet forces
invaded all three Baltic states, eliminated their officials, and installed Soviet cadres
in their place. In June 1941, Germany breached the Pact and invaded the USSR.
The latter ended up occupying and annexing the Baltic states (as well as Romania,
Bessarabia, and Eastern Poland) whilst at the same time fighting a war of self-de-
fence against German forces. The Soviet occupation of the Baltic states was clearly
unjust (as well as illegal), whilst its war of self-defence against Germany was just—
or at any rate, had a just cause. Was the Soviet regime morally justified in requisi-
tioning resources from the Balts towards its war against Germany? I do not think
so, precisely because its occupation was unjust. In taking from occupied civilians
the resources (military and otherwise) which the latter need to resist the occupa-
tion and in so doing restore their fundamental civil and political rights, A com-
pounds its initial wrongdoing of unwarranted aggression with the wrongdoing of
a strengthened unjust occupation.
And here is a further twist. In 1941, the German Army, totalling about 3.5 mil-
lion men, invaded the USSR, including the Baltic States. Populations of those
states, on any account of the just war, had a just cause for resisting that invasion.
33 See Walzer, Just and Unjust Wars, 178. One might think that it would be appropriate for those
civilians to bear the costs of partisans’ war of liberation, from which they would benefit, in the form
of greater exposure to reprisals. I assume in this paragraph that the harms endured by those civilians
exceed those which they can be legitimately asked to incur for the sake of liberation. For the view that
exposing the beneficiaries of a military action to a risk of death is sometimes permissible, see
J. McMahan, ‘The Just Distribution of Harm Between Combatants and Noncombatants’, Philosophy
& Public Affairs 38 (2010): 342–79; Fabre, Cosmopolitan War, ss.2.4, 4.4, and 5.5.2. In CW (s.4.4.2)
I address the question of whom partisans may kill should they choose rightfully to resist Occupier—
their own, collaborationist compatriots, or members of occupying forces. In this section however, my
focus is on the conflict between insurgents and occupiers.
Did the Soviet occupying authorities have the right to fight the Germans and, to
that end, to requisition resources from local populations? Were the latter under a
duty to comply with those requisition orders?34 It might be tempting to answer the
first question in the negative, on the grounds precisely that the Soviet were unjust
occupiers. Tempting, but erroneous: as I noted in s.2.2, an illegitimate regime
might in fact better enable those who are de facto subject to its directives to exercise
their human rights and fulfil their correlative duties. This is why (I also noted) an
illegitimate regime such as Stalin’s did have the (Hohfeldian) power to organize
armed resistance against the Nazi’s invasion. Their point applied to Soviet citizens,
but it also applies here to occupied Baltic populations. It is of course contingent on
it being the case that resisting Germany’s invasion in compliance with the Soviet
occupying authorities’ directives did enable the populations of the Baltic states to
do better at the bar of their human rights and duties than they would have done
under a German occupation. From what we now know, the Nazis had planned to
work and starve to death all populations of the USSR, including populations
annexed by the latter after 1939, with a view to have those territories settled by
Germany; by contrast, Stalin had ‘merely’ aimed to subject them to his totalitarian
rule, which (in this case) did not include wholesale extermination. Furthermore, it
is also pretty clear that the Baltic citizens would simply not have been able to resist
the German invasion on their own. On both counts, the Soviets did have the
power to issue the relevant directives, and Baltic citizens were under duties to one
another to comply.
Prima facie, then, unless A’s requisition orders enable occupied citizensB better
to exercise their human rights and fulfil their correlative duties, the latter are justi-
fied in failing to comply with those orders, so long as A’s retaliatory measures in the
face of those acts of disobedience are not disproportionate to what rebellion would
achieve. To be sure, failure to comply with requisition orders in the light of A’s
refusal to withdraw its occupying troops might leave those for whom A fights
(whether citizensA in a war of self-defence, or third parties in a war of intervention)
exposed to C’s unjust threat. The question then is whether occupied citizensB are
under a duty of assistance to those individuals to comply with A’s policy of requi-
sition even though the policy is unjust. I do not believe that they are. Most obvi-
ously, when A wages a self-defensive war against C, citizensB are not under a duty
to help those citizens of A who significantly contribute to A’s unjust occupation of
their territory. But nor are they under an obligation to help those who, though not
supporting A’s occupation, are vulnerable to C. For by diverting towards the latter
resources such as munitions, means of transport, food, oil, etc., occupied civilians
are depriving themselves of the wherewithal to renew their (just) war of resistance
against their (unjust) occupier.
To be sure, occupying regimes can be more or less harsh towards those over
whom they govern. However, and returning to our example, even if one takes the
34 I owe these penetrating questions to an anonymous reader for Oxford University Press. For the
historical background to my discussion of this case, see Weinberg, A World at Arms, ch. 7. For a dis-
cussion of the law of occupation as was breached in the Baltic States, see Benvenisti, The International
Law of Occupation, 67–8.
view that the indigenous populations of the Baltic states had a less horrendous
time at the point of Soviet guns than Ukrainians did under German boots, the
former were not under a duty to the latter to help the USSR repel the German
invasion. As we saw in s.1.2, individuals have a personal prerogative to confer
greater weight on their own prospects for a flourishing life, for which the enjoy-
ment of fundamental civil and political rights is a necessary condition, than on
other agents’ similar prospects. By implication, they are not under a duty to con-
tinue to live under a rights-violating dictatorial Occupier, or (if their war of resist-
ance against Occupier would be unsuccessful anyway) to incur even more
hardship, for the sake of ensuring that the latter’s own citizens do not endure
totalitarianism.35
My overall conclusion in this case, thus, is that Baltic citizens were under a duty
to comply with the Soviet’s directives to fight German troops in the Baltic states,
but were not under a duty to comply with those directives as pertained to the
Soviet war effort in Russia itself. At first sight, this seems to contradict the cosmo-
politan principle that individuals are not permitted to confer greater weight on
their compatriots’ fundamental rights than on strangers’ similar rights: on cosmo-
politan grounds, a citizen of Estonia was not justified let alone obliged to give
priority to a fellow Estonian over a Russian (assuming that the latter was innocent
of the Soviet occupation of Estonia). However, there is a crucial difference between
the two cases. In the other-defensive case, where Estonians are asked to give
resources to the Soviets so that the latter may defend Russians, Estonians are enti-
tled as per the no-undue-sacrifice proviso to give priority to their own fundamental
human rights over the Russians’ similar rights, and thus to hold on to their
resources in defiance of Soviet’s directives. In the communal self-defensive case, by
contrast, the resources which the Soviet are seeking to extract from Estonians serve
to defend the latter’s human rights from a (ex hypothesi) worse violator. Estonians,
who would themselves benefit from the Soviets’ defensive war against Germany in
Estonia thus cannot refuse to help their compatriots (by refusing to comply with
requisition orders) on the grounds that those resources would be better used to
fight the unjust occupier: the point of requisitions is precisely to wage that particu-
lar fight. That—and not patriotic preference—is why Estonians in this case are
under a duty to one another to comply.
Engaging with an Unjust Occupier: Cooperation and Collaboration

To recapitulate briefly, occupied civilians are not under an obligation to unjust
Occupier, and to one another, to comply with all of Occupier’s directives. May
they engage with its forces? I use the term ‘engage’ to refer both to the mere act of
entering into transactions and relationships (widely understood) with members of
the occupying forces in their private capacity, the act of actively cooperating with
them in their public capacity, and the act of engaging with them in a role-based,
35 However, and again by implication, individuals who are subject to a just occupation are under a
duty to help A prosecute just wars abroad, subject to the no-undue-sacrifice proviso.
though not public, capacity. Examples of the former include selling them goods
such as food and tobacco, serving them in cafés, having sex with them (whether
prostitutional or not), and so on. Examples of the latter include cooperating with
them when they act in such a way as to protect and enforce just laws in force or
when they enact legislation which helps promote conditions for civic life. Examples
of role-based interaction include the case of a doctor from the occupied commu-
nity who has to decide whether or not to treat a badly wounded occupying
combatant.
It might be that, in the long term, cooperating with an unjust Occupier and
thereby helping to entrench its rule would be morally worse, at the bar of the fun-
damental rights of occupied civilians, than resisting it. On the other hand, cooper-
ating now might help further some just end. The point applies to ordinary civilians
(in so far as they are expected to respect Occupier’s directives) but also, and in fact
perhaps more so, to officials from the displaced regime who have remained in post.
Suppose that some occupying combatants have committed exactions against civil-
ians before going AWOL, and that Occupier, though unjustly in situ, is deter-
mined to bring them to trial. Suppose further that it needs the help of indigenous
police officers to track them down. Obdurate refusal to collaborate would seem
wrong in this case. Must public officials resign in order not to be tainted with the
shameful mark of collaboration? I do not think so. Quite apart from the fact that
they might jeopardize their and their family’s main source of income in so doing,
not all aspects of public life within the occupied territory can and will fall within
the remit of Occupier’s concerns. Laws in place still need to be enforced (subject
to their being just), common criminals still warrant arresting, and local taxes still
need to be raised. Discharging those tasks requires some degree of cooperation
with Occupier—cooperation which might well be permissible.36
I say ‘might’ for this really is a genuinely complex issue—the sharp end, perhaps,
of the ethics of occupation. I lack the space in this chapter to offer a full account
of permissible and impermissible cooperation with the enemy. As a first, rough cut,
however, it seems to me that the following considerations have a bearing on our
moral assessment of what B’s public officials may or may not do. For a start, there
is a difference between actively participating in Occupier’s wrongful ends (as when
36 For a useful typology and illuminating normative account of different kinds of collaboration,
cooperation, connivance, and collusion with wrongdoers, see R. E. Goodin and C. Lepora, On
Complicity and Compromise (Oxford: Oxford University Press, 2013). For an interesting historical
account of resistance and collaboration in the Second World War, which draws on a typology that is
similar to though pre-dates Goodin’s and Lepora’s, see W. Rings, Living with the Enemy—Collaboration
and Resistance in Hitler’s Europe 1939–1945 (London: Weidenfeld and Nicolson, 1982). For the tor-
turously difficult case of Jewish community leaders who cooperated with the Nazis by selecting fellow
Jews for deportation (on the grounds that they would cause less harm overall than if the Nazis oper-
ated such selection themselves), see Kamm, The Moral Target, ch. 5. For a study of the different ways
in which French officials reached accommodation with the Germans during the Second World War
(from simple cooperation over logistical matters such as road safety to collaboration with racist poli-
cies), see R. Gildea, Marianne in Chains—Daily Life in the Heart of France during the German
Occupation (New York: Picador, 2002). It should not be forgotten, of course, that the Channel Islands
were occupied by German troops, and that islanders too had to find ways of living with the enemy.
Their predicament is vividly described in M. Bunting, The Model Occupation—The Channel Islands
under German Rule, 2nd edn. (London: Pimlico, 2004).
French, Belgian, or Dutch police officers actively and willingly assisted the local
Gestapo in rounding up Jews), and cooperation with Occupier’s officials towards
rightful ends without participating in its wrongful ends (such as ensuring that com-
mon murderers be apprehended and punished). The latter course of action, even if
it were deemed impermissible, clearly would not be as wrong as the former, and this
even though one knows that in so cooperating, one facilitates Occupier’s wrongful
ends. In addition, whether cooperation of that kind is morally permitted partly
depends on the extent to which, if at all, one may give priority to the short-term
interests of one’s compatriots, whom one helps by remaining in post, over their
long-term interest in Occupier’s military defeat. When Occupier is concurrently
engaged in a conflict elsewhere, the permissibility of cooperation also depends on
the extent to which, if at all, one may give priority to the interests of one’s compa-
triots over the interests of C’s members in not being subject to Occupier’s ongoing
unjust war against their country. Finally, it also partly depends on the relative moral
weight one should attach to refraining from helping one’s compatriots by resigning
one’s post and thereby not facilitating Occupier’s wrongful attacks on third parties,
as against helping one’s fellow residents and thereby contributing to those third
parties’ predicament. Even if harming (or contributing to harming) is morally worse
other things roughly equal than refraining from helping, the reverse might well be
true when other things are considerably unequal—for example, when one’s contri-
bution to a harm is exceedingly marginal in absolute terms and exceedingly smaller
than one’s contribution to a benefit, or when one stands in a special relationship
with the beneficiaries of one’s willingness to remain in post.
Discerning morally permissible cooperation from morally prohibited collabora-
tion with an unjust Occupier is no easy task—and certainly harder, in fact, than
those who have not lived under a military occupation have generally supposed.
That said, the horrors of Nazi occupation of Europe and of Japan’s occupation of
China prior to and during the Second World War, ought not to blind us to the fact
that occupiers, however unjust their occupation, sometimes also and at the same
time act rightly, and in a way which warrants formal assistance. Nor should it
blind us to the thought that, however condemnatory we might be of formal collab-
oration with occupying forces, it remains nevertheless morally permissible, at least
sometimes, to engage with them in a private capacity. In fact, unless occupying
forces are rigidly kept apart from occupied civilians, the latter have no choice but
to engage with them (down the street, in shops, etc.) and it would not be apposite
to condemn them for doing something which they cannot avoid doing. Nor would
it be apposite always to condemn them for giving in either to the threat of coer-
cion, or to the simple need to earn a living, or both. It is one thing for a publican
to refuse to serve an uncouth customer, but quite another to turn away a gun-yielding
combatant. It is one thing to shut down one’s business for fear of having to transact
with the enemy when one has other sources of income, but quite another to do so
when that is the only way to feed one’s family. It is one thing for a doctor not to
make oneself actively available to occupying forces; it is quite another for her
actively to refuse to treat a wounded enemy occupier who presents himself, in great
pain, to her surgery.
Those sketchy points raise two further issues. First, occupied civilians engage
with occupying forces through both commercial transactions and altruistic
exchanges. In some contexts, a commercial transaction is more appropriate than an
altruistic exchange, in so far as it puts occupied civilians and occupying combat-
ants on an equal footing; in other contexts, the reverse might be true. Suppose that
you have a choice between selling or giving some good to occupying combatants—
say a turkey at Christmas time. Selling implies transacting on a footing of equality
but still getting something from the enemy. Giving enables you, perhaps, not to
place yourself on their level, and in so doing to reverse, albeit for a very brief time,
the unequal relationship of occupier/occupied. That said, if your business is to sell
turkeys, you probably should sell, not give, for in doing the latter you would be
doing those combatants a favour which turkey-sellers do not normally do. But if
you simply happen to keep turkeys for your own consumption, perhaps you should
give, not sell—as you would indeed normally do for people who have been staying
in your house. Generally put, whether you may give or sell a particular good to
someone who is, ex hypothesi a wrongdoer (not merely vis-à-vis you but also vis-à-vis
others) partly depends on the nature of your relationship with that person. This
might well explain why the French by and large did not frown upon prostitutes
who took on German combatants as clients, but strongly objected to non-commercial
sexual relationships with the enemy—so strongly in fact as to inflict, post-war,
harsh, humiliating punishments on French women suspected of having taken
German lovers.
Second, for all that occupied civilians and occupying combatants stand in a
broadly unequal relationship, they coexist under the shadow of death, alive to the
very real possibility that, sooner or later, they might be locked again in a lethal
conflict. That too brings with it its unfair share of dilemmatic moments. Suppose
that the enemy soldier who has been billetted in your house breaks down in your
courtyard at 10 a.m. having heard news of his parents’ death in aerial bombings.
Should you show him compassion? Well, it depends. If you saw him the day before
harrass your neighbours at checkpoint or take a suspected Resistant to the local
torture centre—maybe not. But if he has so far behaved impeccably and
courteously—in your kitchen, on the street, at the local shops—perhaps you should
show him compassion. At any rate, it would not be morally inappropriate to do so.
All the same, you would not wrong him by detonating a mine, several hours later,
at the exact moment the military convoy on which you know him to be is passing
by. To be sure, if you know at 10 a.m. that you will attempt to kill him later that
night, you might think it hypocritically wrong to show him compassion—and yet
struggle to hold on to your resolve in the face of his obvious distress. But then
again, you might not give in, and you might be justified in not feeling so torn apart
if you have good reasons to believe that for all his polite reserve towards you and
your family he would not hesitate to kill you upon ordered to do so by his superiors.
Those considerations are shaped partly by the moral status of the Occupation
itself, but also by the highly textured context within which interpersonal rela-
tionships between occupying individual combatants and occupied individual
civilians unfold. The judgements which occupied civilians must make when engaging
with occupying forces in their daily life bring to light the fact that occupying com-
batants are not simply enemies who (in the case at hand) have no claim to issue
directives on us, consume our resources and make use of our infrastructure—
indeed, whom we may sometimes justifiably kill: they are fellow human beings
whom it is sometimes appropriate to treat as such. The point might seem naively
obvious, yet it does bear stressing, if only as a reminder of the emotional resonance
of our moral decisions and the moral valence of our emotional reactions.
3 . 4 . C O N C LU S I O N
In this chapter, I have mounted a justification for two kinds of intervention in the
internal affairs of belligerents during the transition from war to peace: the deploy-
ment of peacekeeping forces on the one hand, and military occupation on other
hand. Both are sometimes justified, or so I argued, not just as a matter of right, but
also as a matter of duty—the latter as an instance of a general duty to provide
assistance to those in need, irrespective of borders. Both are subject to moral norms
which are relevantly similar to and as stringent as the resort to and conduct in war.
But while the tasks of peacekeepers are relatively restricted, military occupiers, who
take on the reins of government, actually form relationships with occupied civil-
ians. Negotiating those relationships—literally living with the enemy—does not
always require grand decisions and solemn gestures. More often than not, it is to
be exposed to the daily necessity of acts of moral compromise, many of which
might seem insignificant when taken on their own but whose cumulative effect can
corrode one’s self-respect and moral integrity. That is the harsh reality of military
occupation, and it is partly for this reason that it must end as soon as possible. To
this issue, and more generally to the issue of properly building peace, I now turn.
4
Peace Agreements
As I noted in s.1.4, the oldest surviving peace treaty was concluded in c.1259 bc.
Since then, peace treaties, or agreements, have become a staple of international
diplomacy. They are also increasingly used, not merely to carve out territory
amongst belligerents but rather to provide a multipronged framework for rebuild-
ing communities shattered by war. Inter alia, they act as constitutional settlements,
set out conditions for reparations and reconstruction, provide for the prosecution
of major war criminals and the amnesties of small (and not so small) fry, and set
out guidelines for the return of refugees.1
By their very nature, treaties generate special rights and duties between signato-
ries the exercise and fulfilment of which is inherently exclusionary. Cosmopolitans
are wary of special relationships based on communal membership in so far as those
relationships are exclusionary vis-à-vis outsiders, and they should thus be particu-
larly concerned with treaties precisely for that reason. And yet, they have not paid
attention to those kinds of special relationships. Nor have war ethicists spent much
time on this issue.2 A cosmopolitan theory of the jus post bellum must provide an
account of the grounds upon and conditions under which such treaties can be
binding on their parties, and of the degree to which treaty obligations supersede,
or are outweighed by, those parties’ general obligations to outsiders.
In the course of this and the next five chapters, I argue that an agreement, or
treaty, is binding on its parties if and only if it meets the following two conditions:
(1) a procedural condition, whereby (1a) the agreement is not secured by fraud,
deception, or unjustified coercion, and (1b) parties who negotiate the agreement
are properly empowered to do so; (2) a substantive condition, whereby the agree-
ment’s clauses (2a) are at the very least justifiedATC vis-á-vis its parties and (2b) do
not prevent the latter from meeting their overriding obligations of justice to out-
siders. My aim in this chapter is twofold: to provide a justification for the claim
1 The UN maintains a database of over 800 peace agreements–related documents at http://peace-

maker.un.org/ For a useful and concise study of contemporary peace agreements, see J. S. Easterday,
‘Peace Agreements as a Framework for Jus Post Bellum’, in J. S. Easterday, J. Iverson and C. Stahn (ed.),
Jus Post Bellum (Oxford: Oxford University Press, 2014). For a comprehensive treatment, see C. Bell,
On the Law of Peace: Peace Agreements and the Lex Pacificatoria (Oxford: Oxford University Press,
2008). I shall use the words ‘treaty’ and ‘agreement’ interchangeably.
2 For a notable exception, see D. Schwartz, ‘The Justice of Peace Treaties’, The Journal of Political
Philosophy 20 (2012): 273–92.
90 Peace Agreements
that parties to a peace treaty are under an obligation to comply with its provisions,
and to articulate and defend the procedural condition. In subsequent chapters, I
offer a detailed account of the substantive condition.
I proceed as follows. In s.4.2, I claim that peace agreements fulfil a number of
different functions and I provide a justification for the duty to comply which takes
those functions into account. Having thus shown that, and why, parties are under
a duty to meet their treaty obligations, I argue in s.4.3 that the moral status of the
war has a bearing on the procedural justness, or lack thereof, of a peace treaty, and
by implication on its bindingness.
Before I start, however, a definitional caveat is in order. As Christine Bell notes
in her comprehensive study of contemporary practice, as a matter of fact, peace
treaties, or peace agreements (I shall use the terms interchangeably), range from
pre-negotiations undertakings (for example, to agree on a ceasefire), to substantive
frameworks for institution-building and reconstruction. They also comprise agree-
ments to implement the latter. Moreover, again as a matter of fact, they have been
used to put an end to relatively small-scale conflicts in which a few thousand peo-
ple have died, as well as to large-scale conflicts characterized by genocidal cam-
paigns. They have been concluded between sovereign states, between rulers of
empires and kingdoms vastly different in form and status from modern nation
states, between factions within states, between states and so-called indigenous
actors, and/or in intra-state cases under the auspices of international organizations
or other states. Faced with this bewildering diversity of practices and documentary
and legal diversity spanning several thousand years, it is crucial to adopt a working
definition which makes sense of the notion of peace agreement in general. I draw
on and revise Christine Bell’s definition, to the effect that ‘peace agreements are
documents produced after discussion with some or all of the conflict’s protago-
nists, that address militarily violent conflict with a view to ending it’, and whose
constitutive clauses solely set out parties’ rights and duties.3
4 . 2 . O N T H E D U T Y TO C O M P LY W I T H
PEACE AGREEMENTS
4.2.1. What Peace Agreements Do: Declarative, Specifying, and

Creative Functions
The maxim pacta sunt servanda is a cornerstone of international law. It is also taken
entirely for granted. Yet, in so far as compliance might not work to the advantage
of signatories, and absent robust mechanisms for enforcement, it pays to inquire into
the normative foundations of the obligation to abide by the terms of international
3 Bell, On the Law of Peace, 53. Chs. 2 and 3 of that book offer an illuminating account of the
diversity of peace agreements and a fascinating recounting of their history. As an anonymous referee
for Oxford University Press pointed out to me, Bell’s definition as it stands will not do, since it counts
a formal document containing a threat to resume war as a peace agreement, so long as the threat is
intended to stop war. Hence my adding the proposition ‘and whose constitutive clauses . . .’
Peace Agreements 91
treaties in general and peace agreements in particular. At first sight, the answer seems
obvious: we all are under a duty to promote peace; in so far as violating peace trea-
ties threatens peace, we all are under an obligation to comply with their provisions.
But there are two difficulties with this move. First, in so far as we all are under a
general obligation to promote peace, the view under scrutiny cannot on its own
explain what is specific to the duty to comply with peace agreements—as distinct,
for example, from the duty not to invade another country without just cause.
Second, and relatedly, it does not account for the intuition that the victims of
treaty violations are themselves wronged, and that their being wronged (as distinct
from peace being undermined) provides at least part of the justification for the
obligation to comply (and the right to seek redress for noncompliance). Germany’s
invasion of Belgium and France in the Spring of 1940, in violation of the Versailles
Treaty, may well have been a crime against peace: but it was also and foremost a
crime against French and Belgian citizens, who did have a special claim against
their 1914 aggressor that it comply with the Treaty—a specific claim which other
signatories to the Treaty, for example the United Kingdom and the United States,
did not have.
How, then, can we best justify pacta sunt servanda? One can think of a peace
agreement as a tool whereby a victorious belligerent imposes its will on its defeated
enemy (for example, the 1919 Versailles Treaty), as a contract between two equally
situated belligerents seeking to put an end to their dispute (a vision which seems
encapsulated in the 1969 Vienna Convention on the Law of Treaties), or as a joint
undertaking between two moral equals who agree to set their past grievances aside
and to cooperate with one another so as to bring about peace (for example, the
1998 Good Friday Agreements between the UK, the Republic of Ireland, and eight
political parties from Northern Ireland). On the first model, treaties are a tool of
statecraft suited to the demands of realpolitik. On the second and third of those
models, treaty obligations are voluntary and thus relevantly similar to promissory
and contractual duties.
In so far as the first model is incompatible with cosmopolitan morality, it is to
the literature on promissory and contractual obligations that we should turn,
which provides a useful set of lenses for unearthing the moral foundations of treaty
obligations. That literature is too vast for me fully to account for it here. In particu-
lar, its preoccupation with the similarities and differences between promissory and
contractual obligations need not be ours.4 Its other preoccupation—why we are
under a duty to comply with promises and contracts—is relevant since we must
defend the view that signatories are under an obligation to comply with just peace
agreements. As a preliminary step, however, we need to elucidate what it is that
4 For a classic defence of the view that contractual obligations are a species of promissory obliga-
tions, see C. Fried, Contract as Promise: A Theory of Contractual Obligation (Cambridge, Mass.:
Harvard University Press, 1981). For a illuminating critique, see D. Kimmel, From Promise to
Contract—Towards a Liberal Theory of Contract (Oxford: Hart Publishing, 2003). The view that a
treaty is a promise by definition is crucial to Vattel’s and Wolff’s accounts of treaty obligations. See
Wolff, Jus gentium methodo scientifica pertractatum, ch. IV; and Vattel, Le droit des gens, ou, Principes de
la loi naturelle appliques à la conduite et aux affaires des nations et des souverains, Bk IV, ch. 3.
92 Peace Agreements
those agreements do. At first sight, we might think that they confer on their signa-
tories special, as distinct from general, rights and obligations. The distinction
between general and special rights is usually drawn as follows.5 General rights are
rights which we have in virtue of (depending on one’s account of rights) being
human, or being a person. Special rights, by contrast, are standardly bestowed on
their holders either as a result of some past event or deed, or through transactions
such as contracts, agreement, and promises. However, while peace agreements to a
large extent confer rights and obligations, the jural relations so established are not
always special in that sense. For peace agreements have three functions as far as
jural relations between signatories are concerned: a declarative function, whereby
they state independently justified rights and obligations which can themselves be
either special or general; a specifying function, whereby they particularize and spec-
ify independently justified rights and obligations which (again) can be either spe-
cial or general; and a creative function, whereby they create rights and obligations
whose sole justification lies in the agreement itself.
To illustrate the declarative function, consider the 1973 Paris Peace Accords,
which put an end to the Vietnam War and whose signatories were the United
States, the Democratic Republic of Vietnam, and the South Vietnamese Provisional
Revolutionary Government. Article 1 states that ‘The United States and all other
countries respect the independence, sovereignty, unity and territorial integrity of
Vietnam as recognised by the 1954 Geneva Agreements on Vietnam’. Assuming
for the sake of argument that the Vietnamese have a jointly held right to political
self-determination, article 1 does nothing but affirm that (independently justified)
right—which is a general right held by human beings in virtue of being moral and
rational agents. Similarly, consider the 1919 Versailles Treaty—and in particular
Part VIII, which sets out Germany’s reparative obligations to the Allies. Let us
assume for the sake of argument that the Germans did have reparative obligations
on account of their unlawful and wrongful invasion of Belgium and France.
Reparative obligations and their correlative rights are clearly special—not gen-
eral—in so far as they arise as a result of a past event, and are not held in virtue of
being human or of some feature of agency. However, here too, the relevant clauses
of the Treaty (articles 231–47) merely state a set of independently justified rights
and obligations—justified, that is, at the bar of reparative justice.
Often, of course, peace agreements do more than declare those rights and obli-
gations: they specify them in a way which creates further rights and obligations.
Here is a simple domestic analogy. Suppose that while out on a walk the two of us
stumble upon a child who has got separated from her parents and is in considera-
ble distress, having broken her arm and fallen into a pond. Let us assume (plausi-
bly) that this child has a right against us that we assist her, in virtue of a general
right to help which we all have against everyone else. We quickly agree that I will
get her out of the pond and that you will gather material for a makeshift splint. As
a result of our agreement, we each bear a specific obligation which contributes to
5 See H. L. A. Hart, ‘Are There Any Natural Rights?’, The Philosophical Review 64 (1955): 175–91;
J. Waldron, The Right to Private Property (Oxford: Oxford University Press, 1988), ch. 4.
Peace Agreements 93
the fulfilment of our general obligation to help. Note, crucially, that there is no
prior, independent justification for allocating to me, rather than you, the obliga-
tion to get the child out of the pond, and to you, rather than me, the obligation to
prepare the splint: I am not a lifeguard, and you are not a nurse. If we each respec-
tively occupied those roles, our agreement would be declarative, not specifying,
since we each would have an independently justified obligation to help the child in
those particular ways—justified, that is, on the grounds that the best qualified
person for a particular task should perform that task. Note, further, that agree-
ments of the kind just described can specify independently justified special rights
and obligations. Suppose that you and I accidentally damage another person’s
property. We clearly are under a special obligation to provide her with assistance (at
a greater cost to us than would be the case if we were not responsible for her pre-
dicament) and to pay compensation for the harm we caused her. However, there
are many morally equivalent ways in which we can discharge those reparative
obligations, and the two of us might thus come to an agreement as to what you and
I should do precisely.
Let me now bring those considerations to bear on the issue of international
treaties in general and peace agreements in particular. Recall that at the bar of
cosmopolitan justice, all individuals wherever they reside have rights to the free-
doms and resources which they need in order to lead a flourishing life. Those
rights in turn impose duties on everyone else. But while there is little difficulty in
assigning on everyone in the world a general negative obligation not to, e.g., kill,
torture, or lie, positive duties to provide assistance must be assigned to specific
parties. Simply to say that those in a position to help must do so will not work,
since vast numbers of agents may well be in a position to help, coordination
might be required to deliver assistance, and so on. The point is familiar, of course:
less familiar is the thought that treaties can function as a mechanism for the
assignment of such duties. Consider the right not to be left to die of starvation. It
correlates into an obligation on the part of everyone to provide food to anyone
who is starving. But we still need to know who among those who have more than
enough must provide food to whom among the starving. Or consider the right to
political self-determination. It correlates into a negative duty on the part of all
others not to mount an unwarranted military attack on politically sovereign com-
munities and a positive duty to provide military assistance to those who are
attacked (subject to costs). But we still need to know who must provide which
kind of assistance for which threatened community. In the case of food, it is often
said that domestic institutions, or states, are the best way to allocate the general
obligation to help. Suppose however that a political community and its domestic
institutions are unable on their own to discharge that obligation to their own
members. A further layer of duty-holders is needed. Likewise, in the case of help
to thwart a military aggression, some agent other than the state who is attacked
must be identified.6 In both cases, bilateral and multilateral agreements or treaties
6 See, e.g., R. E. Goodin, ‘What Is So Special about Our Fellow Countrymen?’, Ethics 98 (1988):
663–86; H. Shue, ‘Mediating Duties’, Ethics 98 (1988): 687–704.
94 Peace Agreements
can serve to identify who is under the relevant obligations to whom. They can also
serve to specify how those obligations will be met.
How does this apply to a peace agreement? Whether one conceives of it as a
contract, an instrument for the imposition of the victor’s will on the vanquished,
or a joint undertaking between moral equals, a peace agreement is passed between
parties which in virtue of the conflict which opposes them stand in a relationship
characterized by the commission of wrongdoings and attendant grievances on
(usually) both sides (even if one side is in a position to claim that its war was just
in toto). The rights and obligations set out in the agreement are special in that they
are justified by past events and are constitutive of erstwhile belligerents’ relationship
with one another qua erstwhile belligerents. Moreover, as we saw a few paragraphs
ago, in so far as many of its clauses pertain to compensation, restitution, and
punishment for wrongdoings, the content of the terms it imposes is determined
by prior principles of compensatory, restitutive, and retributive justice. However,
to the extent that discharging those obligations necessitates joint action by multi-
ple parties and admits of morally equivalent decisions, a peace agreement will
cash out those independently justified obligations into specific obligations.
Examples include a schedule of reparation payments, as set out (e.g.) in article 235
of the Versailles Treaty, or a timeframe for organizing a handover of legislative
and administrative competences from Occupier to Occupied and the precise
modalities of such transfers, as set out (e.g.) in the 1993 Oslo Accords between
Israel and the Palestinians. Moreover, a peace agreement might also specify exist-
ing cosmopolitan and thus general rights and duties. Suppose that one of the
parties, B, has been devastated by the war, and that its population is on the brink
of starvation. To anticipate on ch. 6, at the bar of cosmopolitan morality, those
individuals have a right to assistance which imposes obligations of assistance not
just on the belligerent responsible for those harms but on outsiders as well. The
peace agreement might serve as a tool to allocate those obligations if those outsid-
ers (for example, some subsections of the international community) are willing to
oversee the peace process.
Finally, treaties often create rights and duties de novo, which are necessarily spe-
cial since they derive their justificatory force solely from treaties. Suppose that it is
in X’s interest to get Z but that Y is not under an obligation to provide it. Under
some circumstances, Y might be morally permitted, and have the moral power, to
bind himself to help X. If you and I agree via a contract of sale that you will sell me
your car for 5000 pounds, I acquire a right against you to that car, and you acquire
a right against me to my money: we both will it to be the case that the content of
both rights is as stipulated by the contract. Likewise, a peace agreement creates
rights and obligations de novo when, for example, its parties use the peace process
and resulting settlement as an opportunity to set up a military or trade alliance.
Thus, treaties concluded by Britain and China in the middle of the nineteenth
century following the two Opium Wars (1839–42 and 1856–60) stipulated terms
(highly unequal to the Chinese) under which Britain and other Western merchants
could trade with and within China. Under those terms, Britain was granted the
status of so-called Most Preferred Nation (for trade) with no obligation on its part
Peace Agreements 95
to reciprocate. On no plausible grounds did Britain have a prior, independent right

to be granted such status, and it is appropriate therefore to regard the relevant
treaty provisions as creative.7
Importantly, the creation and conferral of de novo special rights and obligations
are constrained by moral norms, for there are limits to the extent to which others
can justifiably take advantage of a superior negotiating position to get us to agree
to terms which we would otherwise reject. As we shall see in s.4.3, the prohibition
on defrauding, deceiving, and exploiting the other side delineates such limits on
the making of peace agreements, whereas the imposition of otherwise warranted
coercion does not.
4.2.2. Complying with Declarative Clauses

In the next few chapters, I shall provide an account of the rights and obligations
which erstwhile belligerents are justified in imposing on one another via such
agreements. For now, let us assume that a given peace agreement is both (at the
very least) substantively justifiedATC and procedurally just. On what grounds are its
parties under an obligation to comply with its terms? The source of the obligation
depends on whether the treaty clauses are declarative, specifying, or de novo.
Consider declarative agreements first. In so far as they merely state belligerents’
independently justified rights and duties vis-à-vis one another, belligerents are
under a duty to comply in virtue of a general duty to do what justice requires,
which in turn is grounded in the principle of fundamental equality. The fact that
belligerents agreed to confer those rights and impose those obligations on one
another does not place them under an additional duty to comply with the agree-
ment qua agreement. By analogy, Andrew does not acquire a special obligation to
keep his promise to Barbara, whom he has no justification whatsoever for killing,
not to kill her: his obligation is not to kill Barbara, period.
Note that this is so even if Andrew promises not to kill Barbara as a means to
secure her compliance with grievously unjust demands. Suppose that Andrew says
to Barbara whom he is sexually assaulting at gunpoint, ‘if you stop resisting I will
not kill you, I promise’. And suppose that, after he has finished raping Barbara—a
rape made easier by Barbara’s non-resistance—he does kill her. It would be odd to
say that Andrew did not just wrong Barbara by raping and killing her but also by
defaulting on his promise. This is because although he uttered the words ‘I prom-
ise’, he did not in fact make a promise. At the very least, he did not make a binding
promise—one by which he voluntarily imposed on himself an obligation which he
did not previously have. Since one can (definitionally) default only on binding
promises, Andrew cannot be described as defaulting. By parity of reasoning, a
belligerent who undertakes via a peace treaty not to invade its erstwhile enemy in
exchange for the latter’s willingness to pay extortionate (and thus unjust) repara-
tions, and who then defaults on its undertaking as soon as payment is received, is
7 See, e.g., D. Wang, China’s Unequal Treaties: Narrating National History (Lanham, Md.: Rowman
& Littlefield, 2005), ch. 1. I am grateful to Vivienne Shue for bringing this case to my attention.
96 Peace Agreements
guilty of the wrongdoings of, first, making that demand in the first instance and,
second, carrying out an unwarranted aggression. It is not guilty (in my view) of the
additional wrongdoing of defaulting on the agreement.
As this point, it has often been objected to me that enshrining independently
justified obligations in a peace agreement gives additional moral weight to those
obligations, in two different ways. First, it provides signatories with an opportunity
to signal to the world at large that they enjoy the status of contracting parties with
whom one can make peace. By way of reply, however, it does not follow from the
claim that an agreement or clause thereof has a particular function that it is more
binding on duty-bearers in virtue of having that function.
Second, and contrary to what I argued above, some might think that undertak-
ing to do φ when one is in any case under a duty to do it does raise expectations in
such a way that defaulting on the undertaking renders the failure to do φ morally
worse than if one had not so bound oneself. Suppose that A undertakes by treaty
not to invade B without just cause. A’s willingness to endorse the treaty induces B
to spend less on its defence capabilities than it would have done had A not made
that commitment. Five years later, A invades B. Its act of invasion seems morally
worse than if it had not committed itself by treaty not to invade B, precisely in so
far as it made it harder for B to defend itself effectively.
But we need to know more about A to assess this case. Either its leaders knew
when signing and endorsing the treaty at t1 that they would default five years later
at t5, or they had not formed that intention then but subsequently changed their
mind. In the first case, they are guilty of manipulating B into making their invasion
at t5 more successful. In the second case, they are guilty of violating B’s sovereignty
rights, period. Either way, A is not under an obligation not to invade B which it did
not have before making the agreement. In so far as promissory and relevantly simi-
lar obligations are obligations of that kind, we cannot account for A’s wrongdoing
by appealing to the wrongfulness of defaulting on promises. To see this, suppose
that A enlists C’s help to invade B at time t5: theirs is a joint invasion to which they
contribute equally. C was not a signatory to the treaty at t1. Yet, it does not commit
a less grievous wrongdoing by invading B than A does. To claim otherwise is to
hold, inter alia, that combatantsA are liable to greater defensive harm than combat-
antsC, and that C’s leaders are liable to greater punishment post war than A’s leaders,
for the crime of invasion. But this does not seem right. To be sure, they are liable to
some degree of punishment for having deceitfully weakened B’s defensive capacity.
But they would also be liable to punishment for having weakened B ante bellum by,
e.g., hacking into its IT infrastructure and scrambled communication systems
between units. What justifies the imposition of that kind of punishment is the act
of wrongfully increasing someone’s vulnerability to one’s wrongful attacks—not the
invasion itself, and thus not the breach of the non-aggression clause.
4.2.3. Complying with Creative Clauses

So much, then, for the duty to comply with declarative clauses to the effect that
φ—which, I have argued, is not grounded in the agreement itself but rather in
Peace Agreements 97
whatever considerations justify the duty to do φ per se. By contrast, defending the
duty to comply with de novo treaty clauses is as elusive a task as defending con-
tractual and promissory obligations has proved to be. On Hume’s and Rawls’
influential accounts, promises and contracts are useful social conventions whose
function is to provide a framework for cooperation and coordination between
self-interested agents. On Hume’s account, contracting parties are bound by their
own interest, notably their interest in being trusted in the future, to comply with
the terms of the contract or promise. On Rawls’ account, defaulting on one’s
contractual obligations is morally wrong in so far as it is tantamount to free-riding
on other people’s willingness to comply with those agreements. As critics of the
conventionalist view have long noted, however, a contract entered into by two
strangers who are ignorant of that social convention, or who are willing to bind
themselves prior to its emergence, would not be binding—which is counterintu-
itive. Moreover, the conventionalist view seems unable to account for the surely
plausible view that, when I fail to fulfil my contractual or promissory obligations
to you, I wrong you primarily, and not other participants in the general conven-
tions of contracting and promising.8
In my view, the most plausible account of the wrongfulness of noncompliance
with voluntary undertakings such as promises and contracts is owed to T. M.
Scanlon, and goes like this. When I explicitly and voluntarily undertake to do
something for you, I raise your expectation that I will act as I said I would. In so
far as I thereby lead you to make certain decisions on the basis of those expecta-
tions which you would not normally have taken and as making those decisions will
prove costly if I default, I ought not to default. Moreover, when I issue a promise
with the declared aim to get you to reciprocate by doing something for me which
you are not under an independently justified duty to do and which you would not
do but for my undertaking, I derive a benefit from having raised your expectation
and from your performance of your part of the bargain. To raise your expectations
in that way whilst not intending to do my part is wrong, just as not reciprocating
is wrong, for reasons which also account for the wrongfulness of lying to and
manipulating agents in general. Namely, in so acting, I use you merely as a means
to my end and in so doing fail to treat you with the respect that I owe you as a
moral and rational agent who is, in all relevant respects, my moral equal. We can
thus explain the wrongfulness of defaulting without referring to the fact (however
incontrovertible) that promises fulfil the crucially important social and relational
function identified by Hume.9
8 D. Hume, A Treatise of Human Nature. ed. P. H. Nidditch (Oxford: Clarendon Press, 1976
[1740]), Bk III, part II, s. V; Rawls, A Theory of Justice, 344–8. Hume focuses on promises, not con-
tracts, but what he says in the Treatise applies to the latter. For criticisms of the conventionalist
account, e.g., T. Scanlon, What We Owe to Each Other (Cambridge, Mass.: Harvard University Press,
1998), ch. 7; J. J. Thomson, The Realm of Rights (Cambridge, Mass.: Harvard University Press, 1990),
ch. 12.
9 Scanlon, What We Owe to Each Other, esp. 296–309. For discussion, see, e.g., John Deigh,
‘Promises under Fire’, Ethics 112 (2002): 483–506; N. Kolodny and J. Wallace, ‘Promises and
Practices Revisited’, Philosophy & Public Affairs 31 (2003): 120–54. One difficulty with the account is
that someone who is known for repeatedly breaking his promises and who thus does not raise
98 Peace Agreements
Although Scanlon focuses on promises, his argument applies mutatis mutandis

to contracts understood as voluntary and mutual enforceable undertakings. For
whilst the fear of coercive legal sanctions might motivate contracting parties not to
default, it does not explain why they ought not to default, both in cases where they
would incur sanctions and in cases when they would not (because, for example,
enforcement mechanisms are unreliable, inexistent, etc.) And it seems to me that
the best explanation for why contracting parties should comply is precisely that,
were they to default, they would fail to treat one another with the respect they owe
to each other.
It might be objected that in appealing to the principle of fundamental equality
as the source of contractual obligations, I am occluding a fundamental difference
between an obligation to do what justice requires of us (such as not kill without
just cause) irrespective of whether we consent to be so obliged, and an obligation
which we voluntarily impose on ourselves. For if the same principle justifies both
kinds of obligations, then (one might think) it is not at all clear what normative
weight is added by our consenting to be so bound.
However, remember that the rights and obligations thus set out are created de
novo: prior to A explicitly undertaking to B that he will do φ, B had no claim
against A that the latter do φ. The reason why A is now under an obligation to B
to do φ is that, by having made that undertaking, A raised B’s expectations to that
effect. It is in virtue of A’s undertaking that he is under an obligation [to do φ]—
where the brackets denote the justificandum—and not some other act to which B
has no claim. But it is in virtue of the principle of fundamental equality that A is
under an obligation [not to disappoint B’s deliberately raised expectations with
respect to φ], whatever those expectations are. Put differently, we all have a claim
not to be deceived and manipulated into benefiting others without receiving what
we were led to expect in return: what those benefits are is determined by the under-
taking (promise, contract, agreement).
So much, then, for contractual obligations between two individual parties. As it
stands, the foregoing account cannot be brought to bear without further ado on
creative peace agreements. For on that account, A’s contractual obligations are jus-
tified both by the principle of fundamental equality and by his having induced B
to rely on his intention to do what he undertook to do. If parties in a peace agree-
ment, which are typically states, or relatively cohesive political actors aspiring to
statehood, are to be regarded as relevantly analogous to individuals, then one can
justify state (or community) A’s agreement-based obligations to state or commu-
nity B by invoking both the requirement that A (qua state or community) treat B
(qua state or community) with equal concern and respect and the requirement that
romisees’ expectations does not wrong the promisee by, once again, breaking his promise.
p
The Scanlonian account must bite this particular bullet. Redeemingly, however, the account can also
hold that the recurrently derelict promisor commits the wrongdoing of attempting falsely to raise
expectations; or, even and more bluntly put, the wrongdoing of treating the promisee like an idiot and
thus disrespecting him. Either way, the wrongdoing does not lie in breaking the promise: it lies in
making it in the first instance. I am grateful to Jonathan Parry and an anonymous reader for Oxford
University Press for pressing me on this.
Peace Agreements 99
A not disappoint B’s deliberately raised expectations that A comply with the treaty.
Unsurprisingly, this is how public international law tends to think of treaty agree-
ments, notwithstanding the human-rights-centred developments it has witnessed
of late.10 However, on the moral individualist view of the relationship between
states and individuals which underpins my account of the ethics of war and peace,
states’ or communities’ rights flow from and supervene on the rights of their indi-
vidual members. By that token, belligerents’ leaders negotiate peace agreements on
behalf of those individuals. The Scanlonian expectation-based defence of contrac-
tual obligations will work, in this context, only if the following two claims are true:
first, those leaders can and do believe that their counterparts at the negotiating
table will do what they undertake to do; second, those on whose behalf they nego-
tiate will accept the terms of the agreement and not, for example, knowingly return
in subsequent elections a political party which will renege on those terms. By
implication, state A’s obligation to comply is nothing more than citizensA’ obliga-
tion to treat citizensB with equal concern and respect: the content of that obliga-
tion is settled by the undertaking which leadersA made on behalf of citizensA to
leadersB—an undertaking which the latter accept on behalf of citizensB.
Unfortunately, construing the obligation to comply with treaties along individ-
ualistic lines will only take us so far. It will not solve yet another difficulty, namely
that peace agreements are usually meant to bind their parties not just at the time
at which they are concluded, but over many years. Suppose, then, that leadersA
negotiate a peace agreement with leadersB at time t1, some clauses of which are
meant to be binding for three decades. Why should citizensA comply with those
terms at t28? On a conventionalist account of agreement-based obligations, one
could argue that there is a general, peace-promoting expectation that the descend-
ants of contracting parties should deem themselves bound by the convention pacta
sunt servanda. As we saw, however, the conventionalist account is wanting. The
challenge, then, consists in showing how citizensA at t28 can be bound by their
predecessors’ voluntary undertaking given that they themselves did not raise
citizensB’ expectations at t1, and given, moreover, that the current generation
of citizensB did not itself have its expectations raised then.
10 That said, as public international lawyers know all too well, the legal difficulties raised by changes
in statehood over a given territory are enormous. Here are but two. First, can a treaty be made by a
state which was not in existence before the war? The case of Germany following the Second World War
comes to mind here. As international law scholar Josef Kunz noted in 1952, in law the conclusion of
a peace treaty between belligerents supposes that each party is the same state before and after the war.
But given that the Nazi state had been dismantled in 1945 and (more importantly still) that Germany
had been partitioned into two independent sovereign states, no such treaty could be made between
Germany and her erstwhile foes until the former had been reconstituted as a whole sovereign entity.
(See J. L. Kunz, ‘Ending the War with Germany’, The American Journal of International Law 46
(1952): 114–19.) Second, should a new state, such as the Austrian and the Hungarian Republics in
1919, be held liable for the public debts incurred by its altogether different predecessor, in this
instance, the Austrian-Hungarian Empire? As it happened, they were held liable by the victorious
belligerents. See, e.g., I. Brownlie, Principles of Public International Law, 6th edn. (Oxford: Oxford
University Press, 2003), ch. 29, for some of the difficulties raised by such changes. For the specific case
of Austria and Hungary, see M. Macmillan, Peacemakers—Six Months that Changed the World (London:
John Murray, 2001), chs. 19 and 20. We will return to those difficulties at the close of s.6.2.3.
100 Peace Agreements
To be sure, if breaking the treaty would jeopardize peace, they ought to comply.
However, breaking a treaty need not lead to war, particularly when the treaty is
genuinely creative (as opposed to declarative or specifying). This is because creative
clauses confer on the parties non-fundamental rights only, and it is consequently
hard to imagine that signatories would go to war over breaches of the correlative
obligations. But we are still left, then, with the task of justifying compliance with
de novo agreements.
If the Scanlonian account of promissory obligations is sound, it yields the con-
clusion that, in such cases, citizensA are under an obligation to continue to comply
with the terms of the treaty even though they may not have been born or of age at
t, only if citizensB, who themselves might not have been born or of age at t1, formed
the expectation that they would comply. We thus need to know how, if at all, those
expectations can be raised and formed over time. The peace agreement is reached at
t1 by citizensA via their leaders. To the extent that citizensB have objectively sound
reasons to believe that those leaders are properly authorized to act on citizensA’
behalf, it is appropriate of them to form the expectation that the latter will comply.
At t2, A’s citizenry comprises individuals who were (again, via their leaders) signato-
ries of the agreement, as well as newly adult citizens. So long as A’s citizenry so
constituted does not in its majority signal to B’s contemporary citizenry that they
do not consider themselves so bound by the treaty, they can be deemed to induce
the latter to rely on their performance and be under an obligation so to act. At t3,
citizenry A might not longer include original signatories, but it does include mem-
bers of the cohort which, at t2, continued to nurtured citizensB’ expectations; so
long as they, together with their contemporaries, do not signal an intention to with-
draw, they too are under an obligation so to act—and so on.
Note that the clause ‘so long as…’ is crucial. It implies that, should one of the
parties wish to renegotiate, or indeed withdraw, from the peace agreement, it may
do so—provided of course that it does so in a procedurally fair manner, and that it
compensates other parties to the agreement for the burdens attendant on its with-
drawal. A creative peace agreement, in other words, is not one to which its parties
are bound ad infinitum, precisely because it does not impose obligations and con-
fer rights which are independently justified and which there are thus independent
reasons to respect.
4.2.4. Complying with Specifying Clauses

Let us take stock. On the one hand the duty to comply with declarative clauses to
the effect that φ is simply the duty to do φ and is justified as such; on the other
hand, the duty to comply with a creative clause to the effect that φ is grounded in
a Scanlonian account of promissory obligations. The case of what I have called
specifying agreements, or clauses, is more complex than either declarative or crea-
tive agreements. Such agreements, you recall, render more specific rights and obli-
gations which are themselves independently justified, when the precise ways in which
those rights and obligations are made more specific are not themselves generated
by independent principles (such as the ‘best qualified should do it’ principle.) The
Peace Agreements 101
question, thus, is whether signatories’ duty to comply with the relevant provisions
is simply a natural, independently justified obligation to do what justice requires
(as per the duty to comply with declarative clauses), whether it has its source in the
signatories’ voluntary undertaking and the mutual expectations thus raised (as per
the duty to comply with creative clauses), or both. Consider the 1993 Oslo
Accords. Following months of initially secret negotiations between the Israeli gov-
ernment and the PLO under the auspices of the Norwegian authorities, Prime
Minister Rabin and PLO Chairman Yasser Arafat signed the Accords in September
1993, thereby undertaking to bring an end to the decades-long conflict between
Israel and Palestine. Suppose, for the sake of argument, that Palestinians have a
jointly held right to exercise territorial sovereignty over the Gaza Strip and the
West Bank, and that the Israelis thus have an obligation to withdraw their troops
from those areas and hand over all judicial, legislative, and administrative compe-
tences to the relevant Palestinian authorities. The Accords delineate a set of princi-
ples to which both parties commit themselves as they begin peace negotiations. In
particular, they provide for the establishment of a so-called Palestinian Interim
Self-Government Authority for a transitional period of five years, elections to
which are to be held no later than nine months after the Accords come into force
(article II-2). They also stipulate that negotiations for a permanent peace settle-
ment should begin as soon as possible, and by the beginning of the third year of
the interim period at the latest (article V.2). Ex hypothesi, Palestinians have an
independently justified political right to take part in decisions which affect their
community; they also each have independently justified rights and obligations to
enter into negotiations for a long-lasting peace. However, the time frames for the
implementation of those rights and obligations are not themselves independently
justified: time frames rarely are, and are set with some unavoidable degree of arbi-
trariness. It thus makes sense to think of those two articles as specifying rather than
declarative clauses. The duty to comply with them can be couched as, respectively,
‘a duty to ensure that free elections are held within nine months’ and ‘a duty to
begin permanent peace negotiations within two years’. Each of those duties can in
turn be decomposed into two duties, where the bracketed clause identifies the
justificandum, as follows:
a. a duty [to ensure that free elections are held];
b. a duty to ensure that free elections are held [within nine months];
c. a duty [to begin permanent peace negotiations];
d. a duty to begin permanent peace negotiations [within two years].
The justification for duties (a) and (c) flows from a general duty, incumbent on all
human beings wherever they reside, to do as justice requires, whereas the justifica-
tion for duties (b) and (d) lies in the Scanlonian account of the obligation to com-
ply with creative agreements. Prior to the Israelis and the Palestinians explicitly
undertaking to do what justice requires within the indicated time frames, neither
had a claim against the other that they should do exactly that. By making that
undertaking voluntarily, the Israelis and Palestinians, via their leaders, raised
amongst one another the expectation that they would do as stated. To the extent
that they each benefited from having done so, they were in turn under a duty to
each other to do their part. The point applies not just to those individuals on
whose behalf the agreement is made at time t1: it also applies to their successors,
on the basis of the same considerations as were adduced earlier in support of the
view that generations who have not themselves signed a creative treaty are never-
theless under a prima facie obligation to abide by its terms unless they signal to the
other party that they are no longer willing to be so bound.
We will return to the problem of diachronic obligations in chs. 5 and 6. For
now, it is worth raising—briefly—the case of treaties which are made by two sig-
natories or more affecting another party not itself a signatory, and which must be
fulfilled decades later. For example, between 1842 and 1898, the United Kingdom
and China (then governed by the Qing dynasty) signed a series of treaties under
the terms of which China would grant the UK a ninety-nine-year-long lease (start-
ing in 1898) over Hong Kong. Full sovereignty over Hong Kong was duly trans-
ferred back to China in 1997. The inhabitants of Hong Kong were not consulted
in the nineteenth century; nor were their successors formally asked, in 1997,
whether they wished to remain part of Britain, re-join China, or become inde-
pendent. On orthodox interpretations of international law, Britain clearly was
under an obligation, which it honoured, to cede the territory. On a human rights–
based, individualist understanding of justice in international relations, those vari-
ous treaties were the unjust product and legacy of a colonial mindset whereby great
powers would dispose of territories and their inhabitants at will. To my mind,
those treaties were not morally binding, and Britain was not under a moral obliga-
tion to China to comply as it did—but nor, equally, was it morally entitled to hold
on to Hong Kong without consulting its inhabitants. To be sure, there might well
have been very good all-things-considered moral reasons as to why Britain’s leaders
were right not to antagonize China—as the islanders themselves might have
argued, unwilling as they presumably were to be subject to a threat of military
invasion. But that is not the same as to say that the source of Britain’s obligation
lied in its having passed some treaties with China in the second half of the nine-
teenth century: that fact was neither here nor there.
To conclude, I have defended parties’ duty to comply with a justified peace
agreement by appealing to the general obligation to do what justice requires as well
as the general obligation not to take advantage, by noncomplying, with the other
party’s willingness to do their own part based on the expectations one has volun-
tarily raised. As I noted at the outset of this section, however, defending the obli-
gation to comply is not the same as establishing the conditions under which a
peace agreement is justified, and therefore binding. To the latter task I now turn.
4.3. P E A C E A G R E E M E N T S A N D P RO C E D U R A L J U S T I C E
I suggested in s.4.1 that an agreement is binding on its parties if and only if it

meets the following two conditions: (1) the procedural justice condition, whereby
(1a) the agreement is not secured by fraud, deception, or unjustified coercion, and
(1b) parties who negotiate the agreement are properly empowered to do so; (2) the
substantive justice condition, whereby the agreement’s clauses (2a) are just vis-à-vis
its parties and (2b) do not prevent the latter from meeting their overriding obliga-
tions of justice to outsiders. In this section, my focus is on the procedural justice
condition.11 An agreement which fails to meet the procedural condition so under-
stood can be deemed invalid, in the sense that it fails to confer new rights and
impose new obligations on its signatories, and that the latter therefore are not
bound to one another to comply with it. As we shall now see, in accounting for
that condition, we must attend to two separate issues, namely the issue of which
actors have the authority to negotiate and ratify the agreement, and the issue of
which conditions the negotiation process must meet in order for the resulting
agreement to be binding.
4.3.1. The Question of Authority

To complicate matters even further, the issue of authority in turn yields three
questions. First, is it a necessary condition for actors to be empowered to negoti-
ate a peace agreement that they should act on behalf and at the behest of a sover-
eign political community or coalition thereof, and/or serve as officials of a
legitimate state? Second, who should endorse peace agreements? Third, does the
status of a belligerent’s war as just or unjust have a bearing on its authority as a
peace negotiator?
Beginning with the first question, recall that in s.2.2, I argued that actors may
not bring an end to the war on behalf of their community unless they have the
authority to do so; I also noted that this is not tantamount to insisting that only
legitimate regimes can end war. Both points apply to the act of negotiating peace.
Consider the authorization point. Suing for peace—declaring that one is willing to
stop fighting if an acceptable peace can be found—is in relevant ways less porten-
tous than peace negotiations themselves, which through the creation and conferral
of various rights and obligations change the jural and moral situation of those who
are subject to their terms. Thus, if suing for peace requires some form of authori-
zation, a fortiori so does negotiating it. As for the illegitimacy point, since direc-
tives issued by illegitimate states can nevertheless be binding on those subject to
them, directives as issued by an illegitimate regime in fulfilment of peace terms can
be similarly binding.
None of this implies that only actors who are authorized by a sovereign commu-
nity, or coalition thereof, can negotiate peace terms. To insist otherwise would
make it impossible for parties in a civil war between a state and insurgent groups
to negotiate with each other. On that view, no Palestinian leader would be empow-
ered to negotiate terms with Israel and any resulting agreement would be null and
11 A full normative account of treaty-making should also attend to the conditions under which a
signatory to a treaty which suffers neither from a vice in procedure nor from a vice in its terms can
become exempt from its obligations. Material incapacity is one such condition, but see, more gener-
ally, articles 61 and 62 of the 1969 Vienna Convention. Those difficulties, though real and interesting,
are somewhat tangential to my aim in this chapter.
void, as would the mid-1990s peace agreement concluded by warring factions in

Bosnia. One could perhaps weaken the statist construal of authorization and hold
that only those who are authorized to negotiate by a community aspiring to
become a sovereign state are properly competent. But this concessive move also has
counterintuitive implications. For example, it implies that the World Jewish
Congress, which negotiated reparation payments with Germany on behalf of
members of the diaspora (as distinct from the state of Israel), was not properly
empowered to do so, and thus that the relevant clauses of the resulting Luxembourg
Agreement were null and void.12 But to deem all of those agreements as null and
void simply on the grounds that some of the signatories were not representatives of
sovereign states, or of groups aspiring to statehood, is to fetishize the value of sov-
ereignty to the detriment of the value of peace. In so far as sovereignty has value
only to the extent that it does not threaten peace (as we saw in ch. 1), this is not a
tenable position.
That said, the more transformative and wide-ranging the clauses of a peace
agreement, the more reason there is to insist on formal authorization procedures.
Moreover, when one of the belligerents is not already a sovereign community,
those procedures might in turn help constitute it into one and, in so doing, bring
about a more just, or at least less unjust world. Consider again the 1993 Oslo
Accords. They did not set out solutions to any of the disagreements opposing the
two parties—such as the Palestinians’ right to return, the status of Jerusalem, and
the status of Jewish settlements in the West Bank. Had the negotiators attempted
to tackle those, there would have been no Accords. Rather, they enshrined a prin-
ciple of mutual recognition, an agreement to make territorial concessions, provi-
sions for gradual transfers of legislative and judicial competence from Israel to the
Palestinians in some of the areas subject to the former’s military occupation, and
an undertaking to start negotiations for a permanent peace settlement. Crucially,
and as we saw above, the Accords also stipulated that elections should be held
within the Palestinian community to authorize leaders who would then embark on
the next, and infinitely more controversial, stage of the peace process. In so doing,
the Accords helped Palestinians constitute themselves into a sovereign people.
Depressingly, the peace process has failed so far. But the requirement of elections
within Palestine, whose community was not (to say the least) unanimously sup-
portive of the PLO and its Chairman, illustrates the importance of ensuring that
the more complex a full-scale peace agreement, the more important it is that nego-
tiators be properly authorized to act, and crucially to make concessions, by those
who will be directly affected.13
12 See A. Colonomos and A. Armstrong, ‘German Reparations to the Jews after World War II: A
Turning Point in the History of Reparations’, in P. De Greiff (ed.), The Handbook of Reparations
13 For an accessible account of the Oslo Accords, see A. Shlaim, ‘The Rise and Fall of the Oslo
Peace Process’, in L. Fawcett (ed.), International Relations of the Middle East (Oxford: Oxford University
Press, 2013). The problem, of course, with leaving issues unresolved, is that it consolidates injustice
(in this case, the injustice that befalls Palestinian refugees) and makes it harder to remedy it as time
goes by. There are both advantages and disadvantages to lack of precision, and omissions, in the draft-
ing of agreements. For discussion, see Bell, On the Law of Peace, ch. 8.
So much, then, for the authority to negotiate. The question which it raises is
distinct from that of who has the authority to endorse a peace agreement. In some
cases, the same political actor is empowered to do both. In the United Kingdom,
for example, neither Parliament nor the British people need approve international
treaties which are negotiated and signed by the British government: treaty ratifica-
tion is a prerogative of the Crown—even though Parliament must approve of any
change in domestic law which such treaties might require for their implementa-
tion, which in effect imposes some democratic limit to what the government can
agree to.14 By contrast, article II-2 of the US Constitution stipulates that the
President may make international treaties subject to ‘the advice and consent’ (my
emphasis) of at least two thirds of Senators. In a similar vein, article 53 of the
French Constitution provides that peace treaties must be ratified by the legislative
power—and so not necessarily by referendum.
The issue, then, is whether it is a necessary condition for a peace agreement to
be binding on those on whose behalf it is negotiated and signed that a majority of
those individuals either directly (by referendum) or indirectly (via their elected
representatives) consent to it. I argued in CW (s.4.3.2) that democratic authoriza-
tion is not a necessary condition for a decision to go to war to be just: had Roosevelt
dispensed with Congress approval in the aftermath of the attack on Pearl Harbour,
his decision to enter the war would nevertheless have been all things considered
justified, precisely because the attack was grievously and obviously unjust. In the
context at hand here, we must (once more) distinguish between declarative, speci-
fying, and creative agreements or clauses thereof. In the case of a declarative agree-
ment, whose provisions are thus independently justified, democratic approval is
not required since all human beings are under a duty to comply with those provi-
sions whether or not they consent. To be sure, it might be judicious to seek such
approval, as a means to endow the resulting agreement with a kind of legitimacy it
would not otherwise have in the eyes of those subject to it. To the extent that
implementing the agreement requires some degree of cooperation on the part of
those subject to its terms, not seeking their approval in such instances would be
morally undesirable and thus morally wrong all things considered (since it would
vitiate prospects for a justifiedATC peace). My point, though, is that individuals
subject to that agreement are not wronged if the agreement is negotiated and
endorsed without their consent—any more than I am wronged by a decision, to
which I do not consent, to enshrine in law a general obligation not to kill another
person without justification.15 Specifying and creative agreements, on the other
hand, do require such approval, since they impose on their signatories a range of
obligations and liabilities which they would not have otherwise. Whether such
approval should be obtained by referendum or through representative institutions
14 See A. V. Lowe, International Law (Oxford: Oxford University Press, 2007), 67.
15 The point, note, is compatible with the claim that my consent might be required to endow this
or that actor with the power to enforce that (now-legal) obligation: my concern here is with endorsing
peace agreements, not with enforcing them. I shall tackle the issue of enforcement in ch. 8 when
dealing with transitional foreign administrations.
need not detain us here at length—though a referendum confers greater legitimacy

(we may plausibly suppose) on the treaty itself.
The third question raised by the issue of authority is trickier, to wit: does the
moral status of a negotiating belligerent as just or unjust have a bearing on its
authority to negotiate and endorse the peace agreement? In s.1.5, I brought to bear
the view that jus in bello is independent of or, as the case may be, dependent on, jus
ad bellum, on the distinction between the substantive and the procedural justice of
war endings and war’s aftermath. I considered the two views which that analytical
discussion generated:
Independence with respect to process: Belligerents are competent to sue for peace
irrespective of the moral status of their resort to war, or conduct in war, or both.
Dependence with respect to process: Whether belligerents are competent to sue for
peace is entirely dependent on the moral status of those parties’ war ad bellum
and/or in bello.
As applied to the question of negotiating and endorsing a peace agreement, this
gives us:
Independence with respect to process: Belligerents are competent to negotiate and
endorse peace agreements irrespective of the moral status of their resort to and/
or conduct in war.
Dependence with respect to process: Whether a belligerent is competent to negoti-
ate and endorse a peace settlement is entirely dependent on the moral status of
those parties’ war ad bellum and/or in bello.
When discussing war endings in ch. 2, I argued that unjust belligerents ought not
to be deemed incompetent to sue for peace just in virtue of being unjust, on the
grounds that they, par excellence, ought to be able to end hostilities. However,
negotiating and endorsing peace settlements are different from ‘merely’ suing for
peace. To say that a belligerent has the standing to reach a peace agreement is to say
that it has the (Hohfeldian) power to change its jural bundles of claims, privileges,
immunities, and powers with respect to the content of the agreement’s clauses, and
that it is recognized as having that power by the other party. Suppose, for example,
that A invaded B as a result of a long-standing dispute over some territory which
had been in B’s possession but which A claimed it in fact owned. Suppose further
that B’s leadership instruct their pilots to carpet-bomb some of A’s cities, and that
A’s leadership, sensing both that they will not win the war and that they neverthe-
less are in a position to inflict considerable losses on B, sue for peace. It is in the
interest of both A and B to end the war. A and B agree that A will withdraw its
troops from the disputed territory and formally renounce its claim to it; they also
agree that B will pay part of A’s reconstruction costs. Setting aside the question of
whether those terms are just, to say that A and B have the standing to reach a set-
tlement with one another on those terms is to say that they both have the power to
change one another’ jural bundles over, respectively, the disputed territory (such
that citizensA no longer have claims over it, while citizensB acquire such claims),
and the resources needed for reconstructing A. In addition, the power to change
jural bundles must be protected by rights as held by the relevant officials not to be
interfered with when negotiating (for example, rights not to be kidnapped, assas-
sinated, unjustly coerced and defrauded, etc.) and by claims against those subject
to those agreements that the latter comply. Furthermore, this protected power is a
fiduciary power which belligerent leaders have and exercise on behalf of their citi-
zens. Finally—and this is crucial—neither has the standing to reach the settlement
if the other party has a justification for refusing to grant it the power to negotiate
and endorse it. By analogy, Andrew may well have the power to give his car to
Bernard; but if Bernard refuses to accept Andrew’s gift on the grounds that Andrew
bought the car with morally tainted money, Andrew does not in the end have the
standing to change his jural bundles over his car vis-à-vis Bernard.
Our task, then, is to assess whether, and if so, how, belligerents’ standing to
negotiate and endorse a peace settlement is affected by the moral status of their war
(in a way that their standing to sue for peace would not be). At this juncture, it
might be thought that if, as I suggested a few paragraphs ago, a specifying or crea-
tive treaty ought to be ratified democratically, we need not worry about the issue
of whether an unjust belligerent has the standing to ratify the treaty, since ratifica-
tion proceeds from the people. However, a people might have been wholly behind
an unjust war, and the question of its standing as a ratifier thus arises in those cases
where ratification takes place via referendum. Moreover, even in a referendum
conducted in a community only a section of which was in favour of the war, actors
such as political parties will campaign for or against approval. The question, at this
point, is whether some such actors lose their standing as ‘institutional’ advocates or
opponents to the referendum in virtue of their conduct during the war. Finally, it
is not always the case that an agreement should be ratified by referendum—as we
saw above with the example of the Oslo Accords and as we shall see in ch. 8 when
discussing the case of the Dayton Agreement pertaining to Bosnia-Herzegovina.
Those two agreements were negotiated and signed by the leaders of the different
parties. There too the question is whether those leaders lose their standing as
potential negotiators in virtue of their wartime deeds.
Let us begin, then, with the view that one’s standing to negotiate and/or endorse
a peace settlement is entirely dependent on the status of one’s war. So construed,
the dependence view is clearly too strong. For a start, it seemingly implies that it is
enough for a belligerent to have standing as a peace negotiator and endorser that it
should have waged a just war. Yet, this cannot be right: a regime might have waged
a just war and yet be so profoundly illegitimate with respect to its own members as
to not have the power to negotiate on their behalf.
Conversely, the dependence view also holds that a belligerent which has waged
an unjust war has thereby forfeited its power to negotiate and/or endorse a peace
settlement. As it stands, that view has two variants: on the one hand, it can be
taken to hold that a political community which has waged an unjust war has thereby
forfeited its power to negotiate and endorse such a settlement (and in so doing to
change its own members’ and other parties’ Hohfeldian bundles); on the other
hand, it can be taken to hold that a regime which has waged an unjust war has
forfeited its power to negotiate and/or endorse that settlement on behalf of the
political community which it claims to represent, even though (some) members of
that community themselves, innocent as they are of their regime’s wrongdoings, do
retain that power.
It is hard to see how a political community could in its entirety forfeit its power
to negotiate and endorse a peace settlement, unless all of its members are suffi-
ciently responsible for the unjust war—a fact which is not likely to obtain in
practice. And yet, such was the position taken by the United States, Britain,
France, and Italy vis-à-vis Germany’s standing as a peace negotiator at the conclu-
sion of the First World War. The Versailles Treaty (as well as all the other treaties
to emanate from the Paris Peace Conference) was the object of prolonged negoti-
ations between all four victorious powers: German leaders were summoned to
Versailles to be handed a copy of the proposed Treaty, had two weeks to issue
written comments but were given to understand that there was in fact no room
for negotiations. The high-handed, top-down, authoritarian ways in which the
winners treated their erstwhile enemy strike us today as astonishing, largely (I
surmise) because it disempowered all of German subjects irrespective of their
degree of contribution to the war.16
Would the Allies have been justified in denying standing, not to the German
people as a whole, but to their leaders? I do not think so, for although the invasion
of Belgium and France by Germany clearly was a war crime, it was not of such a
nature as to warrant wholly excluding the political community on whose behalf it
was carried out and its allies from any say in the post-war settlement, all the less so
as the regime which had authorized the invasion (the German Empire) was by then
defunct.
Note that in this particular example, the defeated and unjust belligerent lacked
a just cause for war in the first instance. Sometimes, though, a belligerent might
have had a just cause but nevertheless resorted to war unjustly—for example in
breach of the requirement of proportionality and/or necessity. I claimed in s.1.3
that other things equal, a war which is unjust for lacking a just cause is more griev-
ously unjust than a war which has a just cause but fails to meet other ad bellum
requirements, for the latter at least has some justification (to wit, violations of
human rights) even if it does not fulfil constraining conditions on its resort. If so,
and if (as we saw with the case of Germany in 1919), lacking a just cause does not
in itself cause a belligerent to lose its standing as a peace negotiator and endorser,
neither does (a fortiori) breaching the other requirements.
In the light of those points, proponents of the orthodox account of the just war,
whereby combatants’ permission to kill in war is independent of the moral status
of the cause which they thereby pursue, might feel drawn to the view that a bellig-
erent has the standing to negotiate and endorse a peace agreement irrespective of
16 The Allies held the dependence view with respect to treaty negotiations, but not with respect to
its endorsement: although the Allies refused to negotiate with Germany, they did not deem the
German National Assembly, which voted in favour of the treaty in June 1919, incompetent to endorse
it. For an engrossing account of the peace negotiations which took place in Paris from January 1919
onwards, see Macmillan, Peacemakers.
the moral status of its war. Notwithstanding their hostility to the orthodox account,
friends of the neoclassical view might do well to heed the siren calls of independence,
particularly if they accept my account of the power to govern in general. For on
that account, as set out in s.1.2, state officials have that power if and only if their direc-
tives better enable those who are subject to them to enjoy their pre-institutional
moral rights and to fulfil their pre-institutional moral duties than they would have
in the absence of any state or if they were included in any other feasible state. It
could well be, thus, that an unjust belligerent (whether ad bellum, in bello, or both)
would, through the peace settlement which it negotiates, provide its members as
well as outsiders with better conditions overall for the protection of their moral
rights and the fulfilment of their moral duties than they would have otherwise. To
claim that a peace agreement is invalid just in virtue of the fact that one or more of
its parties ended up at the negotiating table thanks to conducting an unjust war risks
adversely affecting individuals—typically, those who are already the weakest—for
whom the alternative, in the form of a long and destructive war, might be far
worse.17 When the moral status of the war does not affect belligerents’ de facto
ability to provide such protection for its citizens and outsiders, those belligerents
prima facie have the standing to negotiate and endorse the peace settlement. By
implication, under those conditions, proponents of the neoclassical account of the
morality of war, who hold that the moral status of war killings is dependent on
the moral status of the cause which those acts serve, ought to resist the view that
the moral status of a belligerent as a peace negotiator and endorser is wholly
dependent on the moral status of their war.
In some cases, however, the acts carried out by a belligerent are such as to vitiate
the latter’s standing. This seems particularly so of a belligerent which does not have
a just cause in the first instance and which commits atrocities in prosecution of
that unjust cause. As we saw in s.2.2, the Allies famously refused to negotiate with
the Nazi regime in the closing stages of the Second World War precisely because
the war the latter conducted was morally so abhorrent that its officials simply
could not be granted any moral authority to act on behalf of the German people
and to enter into a negotiating relationship with the Allies. Likewise, in the spring
of 1994, the Tutsi Rwandan Patriotic Front refused to negotiate peace terms with
Rwanda’s extremist Hutu-dominated government, whose officials had orchestrated
the genocide of c.800,000 moderate Hutus and Tutsis. Note though that even
belligerents which have a just cause may well commit atrocities of a kind as to
undermine their standing to negotiate and endorse a peace agreement. True, the
thought that Hitler, had he not killed himself, might have sat with F. D. Roosevelt
and Winston Churchill and hammered out a peace agreement as if on a footing of
equality, is repellent. Then again, so is the thought that Stalin did sit with Roosevelt
and Churchill at Yalta: the USSR may well have had a just cause in resisting the
German invasion, but its troops committed atrocities on a massive scale, with the
permission and encouragement of the leadership.
17 For a similar point, see Schwartz, ‘The Justice of Peace Treaties’.
The question, though, is why one might be justified in refusing to negotiate a

peace agreement with grievously unjust belligerents of that ilk. One possibility is
that committing certain acts is in itself disqualifying. It is hard not to have sympa-
thy for that view. However, it is vulnerable to the objection that, in some cases,
there might not be any alternative, and that, for all that they did, those institu-
tional actors nevertheless are the best available to represent the interests of their
citizens. Granted, the objection does not impugn the Allies’ decision not to confer
negotiating standing on the Nazis; but it should alert us to the dangers of depriv-
ing some actors of standing without considering what their absence from the nego-
tiating table might mean for those whom they would represent and who, however
grievous their leaders’ wrongdoings and their possible complicity in it, have not
forfeited their right to be treated humanely.
Another argument for the view that grievously unjust belligerents lack the
standing to reach a peace settlement is that their wartime deeds are such that they
simply cannot be trusted to do better than any other feasible alternative at the bar
of the rights and duties of those who are liable to the terms of the peace settlement.
I sketched that argument in s.2.2 in connection with the question of suing for
peace. To elaborate on it in the present context, a peace settlement binds two cat-
egories of agents whose trust in its justice is crucial for the realization of a durable
peace: those on whose behalf, and against whom, the war was waged. Amongst the
former, some may well have supported the war whilst others opposed it but had
little control over it. Those agents cannot reasonably be presumed to trust in a
leadership which authorized, indeed ordered, the commission of crimes as grievous
as the mass killings of innocent non-combatants or an unwarranted wholesale
aggression against another community. In fact, they can justifiably assume that on
the balance of probabilities, those wrongdoers will negotiate and endorse peace
terms which will either consist in or lead to their rights being violated, or render
them less able to fulfil their obligations to one another than feasible alternatives
would do. Note that it is not the commission of the wrongdoings itself which
deprives wrongdoers of standing; rather, the nature of the wrongdoings is such as
to suggest that trust would not be given. If those points are correct, both ad bellum
and in bello wrongdoings can deprive their perpetrators of standing so long as they
are sufficiently grievous.
To be sure, in cases where an evil belligerent will, as a matter of fact and despite
available clear evidence, be able to negotiate just terms, a refusal to negotiate is
wrong on a fact-relative account of wrongness. Crucially however the wrongdoing
is not done to those leaders themselves (for in general no one has a right against
those on whose behalf one seeks to act to be authorized by them to do so). Rather,
the wrongdoing is done to those other agents to whom victims and bystanders owe
obligations—obligations which the latter would be better enabled to carry out by
authorizing those leaders to negotiate on their behalf. But although withdrawing
trust from perpetrators and in so doing contributing either to an unjust peace
agreement (as negotiated and endorsed by worse agents than perpetrators) or to
the resumption of an unjust war would be wrong in the fact-relative sense, availa-
ble evidence is all that agents have to guide their actions under conditions of
e pistemic uncertainty. Thus, while they might in fact be acting wrongly they do
nevertheless have a (subjective) justification for withholding both trust and con-
sent to have those particular leaders speak in their name.18
Suppose, then, that B’s leadership is disqualified on the aforementioned grounds
from negotiating a peace agreement with A’s leadership. Does it follow that A may
set out and impose peace terms unilaterally on community B? In international law,
although a victorious belligerent is not under an obligation to negotiate peace
terms with its defeated enemy, a treaty qua treaty can confer rights and impose
obligations only with the consent of the relevant parties.19 At the bar of the moral-
ity of the transition from war to peace, however, and to reiterate a point made
earlier, whether such consent is required depends on the function fulfilled by the
treaty. To the extent that declarative measures set out independently justified rights
and duties, it is not necessary that B’s members should consent to their insertion
in the treaty for them to be so bound—in just the same way as I do not need your
agreement to insist that you abide by your obligation not to kill me or invade my
house. To put the point more strongly, with declarative measures, there really is
nothing to negotiate.
By contrast, specifying and, a fortiori, creative clauses/agreements do leave room
for negotiations, since in both cases they are binding, qua such clauses, in virtue of
parties’ consent. Recall that B’s leadership ex hypothesi lacks the standing to nego-
tiate with A. May A impose terms unilaterally in this case? I do not think so. For
even if B’s wartime leaders are morally barred from the negotiating table on account
of their wartime wrongdoings, there are very good moral and prudential reasons to
foster and encourage the emergence of a (relatively) untainted regime with which
negotiations will be possible. If measures must be taken in the meantime to stabi-
lize B’s situation, A can act as a military occupier, under the conditions set out in
ch. 3. The immediate aftermath of the 1870 Franco-Prussian war provides an
interesting example. On the 1st of September 1870, the French Army, led by
Emperor Napoleon III, suffered a crushing defeat at the hands of the Prussian
troops. Following the capture of the Emperor, leaders of the Republican opposi-
tion to the regime declared the advent of the Third Republic, formed a provisional
government, and continued with the war. In January 1871, they surrendered to
Prussia. Chancellor von Bismarck’s armistice terms included the payment of repa-
rations and the cession of the Eastern provinces of Alsace and Lorraine, as well as
the holding of legislative elections in France. The elections took place in February
1871 and returned a monarchist majority, which in turn designated a provisional
government (pending a new constitutional settlement). That government negotiated
18 The claim that war victims and bystanders in this case would act wrongly in a fact-relative sense
yields the conclusion (at least in some cases) that they are liable to incur some harm as a means to
redress unjust peace terms. Both the claim and the conclusion are compatible with the point that they
are nevertheless justified on the basis of the available evidence and under conditions of epistemic
uncertainty not to have perpetrators of the most grievous war crimes negotiate and endorse peace
terms on their behalf.
19 1969 Vienna Convention, art. 9. There are exceptions, for example when a treaty between two
countries bordering a waterway confers on non-parties’ ships rights of way through the waterway. See
Lowe, International Law, 81–2.
and concluded the Treaty of Frankfurt with the newly founded Germany in May
1871. Only an elected government (Bismarck held) could have the authority to
conclude the final peace agreement.20
So far, I have focused on cases where the victorious belligerent’s war is just in
toto. Suppose however that the war is won by a grievously unjust party—for exam-
ple, the Soviet Union over Germany in 1945. Does my argument so far imply that
its leaders lack the standing to negotiate and endorse a peace agreement? I believe
that it does: those leaders would not be wronged by a withdrawal of trust on the
part of those subject to the terms of the peace agreements. Analogously, assuming
that a given terrorist group lacks a just cause for taking hostages, the claim ‘we
should never negotiate with hostage-taking terrorists’ is misguided in so far as not
negotiating could result in the death of many innocent people; it is apt, however,
if all it implies is that we do not owe it to those agents and those whom they rep-
resent to negotiate with them.
To recapitulate, I have argued that the claim that a belligerent has waged an
unjust war is not sufficient to disqualify its leaders as a peace negotiator. Put differ-
ently, a peace agreement is procedurally valid even if one of the parties’ superior
bargaining position comes about as a result of violations of the laws of war—
subject to the qualifications just introduced. To many, notably victims of those
violations, this will often seem morally unpalatable to the point of being unaccept-
able. But given that the alternative is the resumption of war, they simply have no
choice morally speaking. There is, however, a way to mitigate their grievance and
at the same time buttress such agreements, which is wholly consonant with cosmo-
politan morality—namely to have outsiders to the conflict mediate between nego-
tiating parties. Thus, US Senator George Mitchell played a crucial part in the peace
process in Northern Ireland, and the United Nations oversaw the negotiations
which lead to the Dayton Agreement in Bosnia. Involvement of that kind, whether
by international institutions or state actors, indeed whether by political actors or
civil society organizations, may make it more likely that resulting agreements will
be procedurally just.21
4.3.2. Fraud, Corruption, and Coercion

As I noted at the outset, however, procedural justice in peace negotiations requires
not simply that competent parties with the requisite authority should negotiate
and endorse the agreement; it also requires that the process of negotiations itself (as
distinct from the process by which belligerents came to occupy their respective
positions at the negotiating table) should meet certain conditions. Articles 49–52
of the 1969 Vienna Convention on the Law of Treaties set out the following
invalidating features of peace negotiations: a treaty is void if one party secures the
20 See, e.g., W. Fortescue, The Third Republic in France, 1870–1940: Conflicts and Continuities
(London: Routledge, 2000); J.-M. Mayeur and M. Reberioux, The Third Republic in France, 1870–1940:
Conflicts and Continuities (Cambridge: Cambridge University Press, 1988).
21 See, e.g., Bell, On the Law of Peace, ch. 9.
other’s agreement through fraud, corruption, and the coercion of its representa-
tives, as well as the coercion of that other party through use of force in violation of
the UN Charter.
The claim that obtaining concessions from one’s enemy through fraud and/or
corruption of its negotiators is morally wrong and vitiates the resulting agree-
ment seems straightforwardly convincing at first sight. Suppose, however, that
those concessions are morally justified—such that defrauding or manipulating
the other side would secure a substantively just agreement. May we do so? Even
if the other side has by its conduct during war forfeited their right not to be
deceived, we might have very good reasons overall not to deceive them—on the
grounds, for example, that in iterative diplomatic relationships, one such act
would jeopardize future negotiations and, thereby, prospects for a justifiedATC
peace.
By contrast, the connection between coercion (or the threat thereof ) and
bindingness warrants further scrutiny. The thought that agreements made under
the threat of coercion can nonetheless be valid, and therefore binding, is at the
heart of (amongst others) Grotius’ and Hobbes’ accounts of promises in particu-
lar and political authority in general.22 On those accounts, voluntarily agreeing
to terms is a sufficient condition for being bound to comply: neither fear nor
coercion vitiate the agreement. Even if Andrew coerces Barbara into promising
to have sex with him by issuing a credible threat to kill her otherwise, we must
hold that Barbara is bound to fulfil her promise, for she has voluntarily acceded
to Andrew’s demands. (She could after all have opted to die instead of submit-
ting to sexual violence.) But this is utterly implausible. True, Barbara voluntarily
agrees to having sex with Andrew, but this in no way entails that she is under a
duty to submit to his attack without resisting. By implication, then, it would
seem that a belligerent who is coerced into acceding to unjust peace terms is not
bound by the agreement.
A more interesting question is whether coercion vitiates just agreements. We
need to distinguish between two plausible interpretations of that view. On the one
hand, it might mean that A’s leaders may not directly threaten B’s leaders with
lethal force unless the latter accede to peace terms—for example by putting a gun
to their head. On the other hand, it might mean that A’s leaders may not credibly
threaten B’s leaders with the resumption of hostilities, or with force short of war,
as a means to get B’s leaders to accede to their terms.
It is tempting to think that, in both cases, A’s leaders may so act and the result-
ing agreement is binding on B notwithstanding its leaders having been coerced,
just if and on the grounds that the terms are (declaratively) just. Not so, in fact, for
two reasons. First, one should resist the view that one may threaten another party
with lethal force as a means to get them to comply with a just request to do
22 Grotius, The Rights of War and Peace, Bk II, ch. 11; T. Hobbes, Leviathan. ed. N. Malcolm
(Oxford: Oxford University Press, 2012 [1651]), ch. XIV. For contemporary discussions with specific
mention of war, see, e.g., John Deigh, ‘Promises under Fire’; T. M. Scanlon, ‘Reasons, Responsibility,
and Reliance: Replies to Wallace, Dworkin, and Deigh’, Ethics 112 (2002): 507–28.
something.23 For threatening to do φ is morally permissible only if doing φ is

permissible under the circumstances. To kill B’s leaders here and now if they do not
agree to, for example, withdraw their troops or restitute a wrongfully annexed
territory is wrong, for to meet a failure to agree to terms, however justified those
terms are, with lethal force is disproportionate. Therefore, threatening to kill then
unless they comply with one’s demand is morally wrong.
Second, to the extent that the terms of the agreement are declarative, B is under
a duty to accept those terms. Accordingly, it is not bound to comply with the
agreement in virtue of having agreed to terms; but nor is it released from that obli-
gation if its leaders were coerced into signing it. The point applies not just to
issuing threats as a means to obtain that to which one is entitled, but also to mak-
ing an offer to the same ends. That is to say, it applies not just to the claim ‘we will
coerce you in such and such a way if you do not agree to give back the territory you
wrongfully annexed’ but also to the claim ‘we will assist in the reconstruction of
your country in ways which are beyond the call of duty provided you give us access
to your markets under terms which are highly harmful to you but to which we are
independently entitled under the circumstances’.
To put the point differently, with respect to declarative agreements or clauses
thereof, the question of whether coercion or the threat thereof are vitiative of those
clauses is moot. It can only arise with respect to peace agreements whose specifying
or creative clauses structure a state of affairs to which neither party has an inde-
pendently justified claim. Thus, it would have been wrong of either Israel or the
Palestinians, when negotiating the Oslo Accords, to threaten to use force as a
means to get the other party to agree to this, or that, time frame for elections,
among a choice of morally equivalent alternatives; had either done so, the resulting
clause would not have been binding, qua such clause, on the coerced party—
though there might well have been all things considered good reasons for the latter
to give in (essentially, the risk of jeopardizing the peace process). Similar consider-
ations apply to the trade treaties passed by Britain and other Western powers on
the one hand, and China on the other hand, to which I alluded above, in so far as
the latter’s authorities were threatened with military action if they refused to sign
those treaties’ declarative clauses.24
23 Unless the request is a request to, e.g., move away as one wrongfully obstructs an escape route
from an attack—in which case killing the obstructor would succeed in thwarting the attack (since the
victim could now escape).
24 For classic accounts of coercion, see, e.g., A. Wertheimer, Coercion (Princeton, N.J.: Princeton
University Press, 1987); C. C. Ryan, ‘The Normative Concept of Coercion’, Mind 89 (1980): 481–98;
R. Nozick, Anarchy, State, and Utopia (New York: Basic Books, 1974). Some authors (e.g., Wertheimer)
endorse a moralized account of coercion whereby A can be described as coercing B into doing x only
if he wrongfully forces B to do x. As should be clear in the text, I favour a non-moralized account. I
discuss the use of lethal force, at gunpoint, as a means to enforce both just and unjust orders in C.
Fabre, ‘Internecine War Killings’, Utilitas 24 (2012): 214–36. On the issue of surrender under threats,
which is related to the issue of agreeing to peace agreements under threat, see John Deigh, ‘Promises
under Fire’; T. M. Scanlon, ‘Reasons, Responsibility, and Reliance: Replies to Wallace, Dworkin, and
Deigh’; E. Chwang, ‘On Coerced Promises’, in H. Sheinman (ed.), Promises and Agreements (Oxford:
Oxford University Press, 2011).
4 . 4 . C O N C LU S I O N
In this chapter, I have offered an account of the conditions under which a peace
agreement is procedurally valid, and of the grounds upon which its signatories are
obliged to comply with its provisions. I have not offered a full account of what
makes a peace agreement substantively just. Nor have I offered an account of by
whom and how peace agreements can be enforced. In the next five chapters, I do
precisely that. Note however that I only offer a partial defence of substantively just
clauses. Note also that my aim, in what follows, is as much to account for the sub-
stantive justice of peace agreements as to establish what is a justified peace all
things considered.
Let me explain. On the first count, given that a justifiedATC peace requires
that individuals’ basic human rights be respected, I take it for granted that a
peace agreement must stipulate that all parties will respect those rights.
Accordingly, I will not defend the view that a constitutional settlement as
embodied in a peace settlement should yield minimally democratic institutions.
Hard as it may be to achieve in practice, that view flows from the claim, which
I made in s.1.2, that the right to have a formal say over the design of coercive
institutions whose decisions affect our fundamental interest is a basic right.
That claim itself flows straightforwardly from the central tenets of cosmopolitan
morality, and is in tension with Rawls’ view that so-called decent hierarchical
societies which respect human rights except for the political right to a formal
say are equal partners with democratic liberal societies under the Law of Peoples.
The Rawlsian view implies that settling for such a society can promote a justi-
fiedATC peace—though not a just peace simpliciter. On my view, in so far as the
right to a formal say is a basic right and as a justifiedATC peace must respect those
rights, establishing a hierarchical society of that kind does not achieve such a
peace.25
I accept that this is a controversial point. But—to repeat—it is a fairly straight-
forward implication of my account of cosmopolitan justice. Instead of developing
it further, I focus on five philosophically difficult issues, namely the restitution of
property and territory, reparations and reconstruction, the punishment of war
criminals, reconciliatory practices, and supranational enforcement mechanisms
which seemingly impugn the imperative of fostering local democracy.
On the second count, in contemporary practice, those issues tend to be dealt
with through peace agreements. They need not be, however, in order for the result-
ing peace to be all things considered justified substantively. To the extent that a
peace agreement merely states independently justified norms for, e.g., the restitu-
tion of stolen property and the punishment of war criminals, the important philo-
sophical task consists in articulating those norms, and the important political task is
to work out how they can be best implemented, through or outside the framework
25 Rawls, The Law of Peoples. My point here is a reconstruction of Rawls’ view as applied to the
design of post-conflict institutions. I am grateful to an anonymous reader for Oxford University Press
for pressing me on this.
of a peace agreement. That said, those norms need to be specified in ways which
are acceptable to belligerents. Although peace agreements are not a necessary
condition for a justifiedATC peace, they are a particularly useful means towards its
specification and implementation. It is for that reason that, in what follows, we
shall have to keep in mind both their declarative and their specifying functions.
5
Restitution
As belligerents turn their mind to ending the war, attention must be paid to recon-
structing what has been destroyed, restituting what has been wrongfully taken
away, and compensating victims for injuries to life, limbs, and property. The ques-
tions thus raised come under the following headings:
Corrective justice. (1) A has taken from B a good g which was in B’s de facto pos-
session. When, if ever, is A under a duty to restitute g to B? (2) A has destroyed
g. When, if ever, is A under a duty to pay reparations to B for her loss by giving
her the equivalent of g and, if B has incurred further harm as a result of A’s act,
by making good for that harm?
Distributive justice. (1) A has destroyed g, as a result of which B has incurred
further harm. A is unable or unwilling to make good on that loss. When, if
ever, is C under a duty to step into the breach and help B? (2) Individuals
who were neither party to the war nor affected by it do not have prospects for
a flourishing life. When, if ever, is C under a duty to help them, at the expense
of failing to meet B’s reparative claim?
Although there has been an enormous amount of work on corrective and distribu-
tive justice in general, relatively less has been said in the contemporary literature on
post-war corrective and distributive obligations.1 In this chapter, I focus on one
aspect of post-war corrective justice, to wit, the restitution of what belligerents
have taken from one another in the course of the war. I leave the issue of reparative
and distributive obligations until the next chapter.
Belligerents routinely appropriate their enemy’s publicly owned resources, mate-
riel, and territory, as well as their privately owned resources. The question is
whether a belligerent which has taken property in the public and/or private de facto
possession of its enemy is under a duty to restitute it once the war is over. Pace the
maxim ex injuria jus non oritur, in virtue of which injustice cannot confer legal
entitlements on its perpetrators, we shall see that the restitution of what was
wrongfully taken, though often an important component of a justifiedATC peace
settlement between belligerents, not only is not always a necessary condition for it
1 Notable exceptions are E. Barkan, The Guilt of Nations: Restitution and Negotiating Historic
Injustices (New York: Thornton, 2000); Bass, ‘Jus Post Bellum’; May, After War Ends; Orend, The Morality
of War.
118 Restitution
but also is, at least sometimes, unjust.2 Moreover, we shall also see that a justifiedATC
peace may also sometimes require the restitution of what was rightfully taken
during the war. I defend those two conclusions by considering three cases: the
restitution of publicly and privately owned moveable property (s.5.3), the return
of territory (s.5.4), and the return of displaced individuals to their homeland
(s.5.5). Throughout, I highlight feasibility constraints on the implementation
of restitutive principles and suggest all-things-considered justified alternatives.
Beforehand, however, I sketch out a cosmopolitan account of individuals’ property,
sovereignty, and occupancy rights—on which the ensuing account of restitutive
justice after war rests (s.5.2).
In keeping with the book’s overarching aim, I tackle this issue from a cosmopol-
itan perspective. At first sight, one might think there can be no such thing as a
cosmopolitan theory of post-war restitution, for cosmopolitan justice pertains to
individuals’ general rights and duties—that is to say, the rights and duties which
they have against and to one another qua human beings, irrespective of borders.
By contrast, restitutive rights and their correlative obligations are grounded in
the special relationship which unites soon-to-be-erstwhile belligerents. However,
cosmopolitan justice does have a lot to say about restitution, for in so far as correc-
tive justice regulates what should be done when individuals’ rights are violated or
justifiably infringed, it presupposes an account of what those rights are and is thus
parasitic on non-corrective principles of distributive justice.3 If only for that rea-
son, cosmopolitan considerations shape what is owed in restitution.
Three remarks before I begin. First, my aim is not to provide a normative
account of wartime seizures in general,4 but to delineate the grounds upon and
conditions under which that which was taken must be returned once the war is
over. I do so on the assumption (crucially) that belligerents have reached the point
where continuing with the war would be unjust, such that they are under a duty to
negotiate a peace settlement.
Second, I seek to discern what constitutes such a peace as negotiated by erstwhile
belligerents and other parties affected by the latter’s conflict, with respect to resti-
tutions. As we saw in ch. 4, we must distinguish the declarative from the specifying
and creative functions of peace agreements. Here, I offer an account of post-war
restitutive justice whose central tenets are justified independently of the fact that
they may be agreed upon by such an agreement. As we shall see, however, there are
2 The maxim found its best known articulation in US State Secretary Henry Stimson’s warning to
Japan that the US would simply not recognize the latter’s 1931 forcible annexation of Manchuria. For
an exposition of the doctrine, see Neff, War and the Law of Nations, 295–6.
3 For illuminating discussions of the complex relationship between corrective justice and distribu-
tive justice, see D. Butt, Rectifying International Injustice—Principles of Compensation and Restitution
Between Nations (Oxford: Oxford University Press, 2009), ch. 2; C. Kutz, ‘Justice in Reparations: The
Cost of Memory and the Value of Talk’, Philosophy & Public Affairs 32 (2004): 277–312, at s.V. The
claim that corrective justice is parasitic on distributive justice does not entail that it has no independ-
ent moral force. Jules Coleman, for example, agrees with the claim of parasitism yet holds that the
point of corrective justice (when it attends to economic wrongs) is to improve efficiency, which gives
it such force. (See J. L. Coleman, Risks and Wrongs (Oxford: Oxford University Press, 2002), 348–54.)
I am agnostic on this particular point.
4 See, e.g., Grotius, Commentary on the Law of Prize and Booty.
Restitution 119
moral constraints on the ways in which signatories to a peace treaty may specify
restitutive rights and obligations, particularly in so far as their restitutive deci-
sions (notably with respect to the retrocession of territory) adversely affect future
generations. In that sense, the chapter both defends specific post-war restitutive
principles and illustrates my earlier account of peace agreements.
Third, I often use simple one-to-one cases to articulate and buttress restitutive
principles before applying the latter to war. If we cannot provide a justification for
restitutive duties in those simple cases, we have very little hope to do so in the infinitely
more complex case of war. Besides, the harms of war are not merely incurred by
individuals qua members of belligerent communities (or indeed, qua members of
communities which are not implicated in the war but which nevertheless suffer
from its consequences): they are also—in fact mainly—incurred by individuals qua
individuals, whose lives are lost, limbs shattered, and livelihood destroyed.
Although an account of post-war restitutive rights and obligations must attend to
the collective dimension of war, it must also pay attention to its individual losses
and sufferings.
5 . 2 . OW N E R S H I P, O C C U PA N C Y, A N D
S OV E R E I G N T Y R I G H T S
Consider three all too plausible scenarios. A combatant set on looting vacant houses
of enemy non-combatants takes artefacts with him to his homeland. Belligerent
A stations its troops in enemy territory T and, at first in the name of military
necessity but later on as part of an expansionist policy, purely and simply annexes
it to its own. Having forcibly annexed T, it settles its own members there and
prevents enemy non-combatants who fled its troops’ advance from returning. At
issue in those cases are violations of, respectively, ownership rights, sovereignty
rights, and occupancy rights. An account of the grounds upon and conditions
under which wrongdoers are under a duty to restitute to their victims what was
taken from them is thus parasitic on a prior justification of all three kinds of rights.
In this section, I sketch out such a justification and provide a defence of restitutive
justice in general.
5.2.1. Justifying Ownership, Occupancy, and Sovereignty Rights

Let us start with ownership. To say that some agent A is the rightful owner of a
good g presupposes a coherent account of what it means to own something, of who
A is, and of the grounds upon and conditions under which one is the rightful
owner of things. Standardly, to own something—g—means to have the following
set of exclusive rights in respect of g: the rights to use, waste and destroy, and man-
age, as well as the right to derive an income from g (ownership of which income in
turn consists in the same set of incidents); the power to sell, give or bequeath g with
our consent, whose jural opposite is an immunity not to have g taken away from
us without one’s consent; the liability to have g taken away from us without our
120 Restitution
consent—as when the state exercises a power of eminent domain over g; and the
duty to ensure that g does not cause unwarranted harm on third parties.5
Those rights can be held by private and public agents: in private ownership, one
or more agent have control over how g will be used and by whom, with no restric-
tions on the ways in which they do so save for a duty not to harm others; in public
ownership, a political agent (usually a political community of some kind) has such
control and exercises it in its politically and socially defined interest.6
Defending private ownership—be it individual or collective ownership—has
proved the hardest task by far, precisely because to have ownership rights over g is
to have the right to exclude third parties from using and/or controlling it, often to
their considerable detriment. Let me offer a (necessarily) brief defence. As I note in
s.1.2, in stipulating that individuals should have prospects for a flourishing life,
justice requires that they should be able to frame, revise, and pursue a conception
of the good which they regard as theirs over the course of a significant part of their
life. This in turns means not just that they should have the personal capacities to
realize such conception but also the material means to do so. It is hard to see how
individuals could fulfil any conception of the good without having some degree
of exclusive control over those means. Thus, justice requires that, subject to their
having satisfied its distributive and corrective demands, individuals enjoy the rights
to use, destroy, waste, and manage those means, the right to derive income from
them, the power to transfer them, and immunity against forcible transfer.
Importantly in the present context, the claim that ownership rights are constrained
by the demands of distributive and corrective justice implies that rights of bequest
and inheritance are conditional on meeting those demands.
Those points obviously apply to moveable resources, but they also support
the conferral of some ownership rights over immoveable resources such as land.
They also support occupancy rights, which include the right to exercise some con-
trol over one’s place of habitation, as well as the rights to move around, and take
part in the social and economic fabric of, the wider territory on which the house
or land in/on which one lives are situated.7 Those rights protect individuals’ inter-
est in enmeshing themselves in interconnected networks of that kind, without
which they could not frame, revise, and pursue a conception of the good. Although
5 T. Honore, ‘Ownership’, in A. G. Guest (ed.), Oxford Essays in Jurisprudence (Oxford: Oxford
6 See, e.g., J. Christman, The Myth of Property: Towards An Egalitarian Theory of Ownership (Oxford:
Oxford University Press, 1994); S. R. Munzer, A Theory of Property (New York: Cambridge University
Press, 1990); Waldron, The Right to Private Property. Collective ownership is sometimes defined as
ownership by a public body which exercises the relevant control over the resource in the interest of all.
It is analytically clearer to focus either on how many people have such control or on the ways in which
and ends to which it is exercised.
7 On this particular point, I follow A. Stilz, ‘Occupancy Rights and the Wrong of Removal’,
Philosophy & Public Affairs 41 (2013): 324–56; and M. Moore, A Political Theory of Territory (Oxford:
Oxford University Press, 2015), esp. ch. 3. (It should be noted though that Moore offers a much more
sophisticated account of occupancy, which she thinks is a property of groups—unlike residency, which
she thinks is a property of individuals. This is an important distinction for any comprehensive account
of territorial justice. But I do not think it matters for my argument in this chapter—and so I speak of
occupancy to denote both the fact that groups of individuals together occupy a territory qua such
groups, and the fact that individuals reside on that territory.)
Restitution 121
individuals can and indeed often do uproot themselves in search of a better life
elsewhere, to force them to do so is, at least at first sight, to fail to treat them with
the concern and respect which they are owed. Clearly, occupancy rights over land
are not co-extensive with ownership rights over one’s place of habitation: many
people do not and never will own the home in which they live. So long, however,
as they enjoy ownership rights over a reasonable set of material resources as well as
some occupancy rights, they can be said to have prospects for a flourishing life.
With a few modifications, my brief account of ownership rights can be extended
to publicly owned resources. By publicly owned, I mean owned by agents in a polit-
ical capacity—where ‘political’ can denote a village, a county, a sovereign nation
state, or an association thereof. In justifying private ownership rights, I claimed
that individuals cannot enjoy a flourishing life unless they have secure access to, and
some degree of control over, the required resources. The same goes with public
ownership rights. For if it is the case that, as I noted in s.1.2, individuals have the
jointly held right to form political associations, it stands to reason that they should
have the jointly held right to use, control, and manage the resources which they
need in order effectively to shape their common life—consistent of course with
the demands of global distributive justice. This is so whether the resources are
moveable, such as money and matériel, or immoveable such as buildings, land,
and territory. However (and this is crucial), the claim that individuals belonging
to a given political community have the jointly held right to associate with one
another and to determine their own future as a community, while providing sup-
port for the further claim that they have jointly held rights over the required
resources, does not provide support for the further claim that they have jointly held
rights over this or that territory T: after all, it may well be that another group has a
stronger claim to T. On what grounds, then, do those individuals have territorial
rights in respect of T?8
In CW, I defended territorial rights on the grounds that individuals cannot
securely exercise their right to political self-determination if they ‘are constantly
vulnerable to being forcibly expelled from where they live and thus to being torn
asunder from the interconnected networks of relationships—familial, professional,
social—which form the fabric of their lives’ (43). Nor, moreover, can they properly
fulfil their obligations of justice to one another and to distant strangers if they are
vulnerable in those ways. On that view, thus, the fact that individuals are already
there confers on them a prima facie right to staying where they are. But it also
confers on them a right to make decisions about their common life over T—in
other words, territorial rights in respect of T. Occupancy, in other words, is not
enough: it must be protected by exclusive rights of government.
Note that on this account, birth is not what confers the status of citizen, and the
territorial rights which come with it—be it birth from citizens of T, or birth on
T. Jus sanguinis and jus solis are both wholly arbitrary bases, morally speaking, for
8 The literature on territorial rights has become enormous. In addition to works by Stilz and Moore
mentioned in the previous footnote, I have greatly benefited from the following: T. Meisels, Territorial
Rights (Dordrecht: Springer, 2005); C. Gans, The Limits of Nationalism (Oxford: Oxford University
Press, 2003).
122 Restitution
the distribution of goods as essential to human beings as the opportunity to share

in the major decisions affecting the community of one’s residence. On egalitarian
grounds, jus habitationis (understood here not just as residing in one’s house, but
as occupancy of the land in general) should be regarded as a sufficient basis for
citizenship, to the extent that long-term residence signals a willingness and ability
to share in C’s long-term future. Granted, there might well be good practical rea-
sons for regarding both as possible bases for citizenship: checking that every single
person coming of age has fulfilled some residence requirement as implied by jus
habitationis might be practically infeasible, and birth, and birthplace, might thus
serve as acceptable proxies. But birth could only be a proxy (and thus has no moral
significance on its own), and could never supersede residence.
Some might object that my argument is vulnerable to the charge of arbitrariness,
since it assigns special, rights-conferring value, to the fact of being there. However,
and crucially, merely being there is not enough to confer on those individuals
all-things-considered rights over T. Being there matters only in so far as it signals a
willingness and ability on the part of habitual residents of T to share in the life
which they make together. Moreover, account must be taken of the fact that their
presence in T may be the result of wrongful acts, which might then generate resti-
tutive duties. There is a limit to those duties, of course: depending on how far back
in the past those wrongful acts were committed, those individuals may not be
expected to leave T if having to do so would jeopardize their prospects for a flourish-
ing life. Thus, even if we accept the view that a de facto Occupier has a presumptive
prima facie claim over T, this needs to be qualified in the light of the following
countervailing considerations: first, how recently the wrongdoing was committed
and by whom; second, the costs on the resident group of being removed; third, the
costs for members of competing claimants over T not to be given this particular
parcel of T.
Note that in taking those costs into account, I am not suggesting that a simpliciter-
just allocation of territorial rights should give way to an all-things-considered just
allocation—which would cater for the resident group at the expense of the claim-
ant group and vice versa. Rather, those costs are built into territorial justice itself,
for excessive as they are, imposing them on either party would wrongfully under-
mine their prospects for a flourishing life. That said, it may well be that, in fact, it
would not be too costly in that sense for residents to be moved off—so long as
they would have a suitable resettlement package. Even then, we might not be in a
sufficiently robust epistemic position to reach such a judgement. Moreover, imple-
menting such a package might prove so costly to third parties, materially and
humanly, that we ought not all things considered do it. In other words, feasibility
constraints might be such as to dictate against implementing territorial justice
simpliciter and in favour of a justifiedATC territorial compromise.
To recapitulate, rights over resources—whether they are rights over moveable
or immoveable resources—are grounded in the importance for individuals, singly
or jointly, privately or publicly, of having meaningful control over the material cir-
cumstances of their life. So construed, those rights are both justified and constrained
by the demands of cosmopolitan sufficientist justice. On this account, a set of
Restitution 123
individuals who conceive of themselves as belonging to the same sovereign political

association jointly enjoy the same incidents of ownership over that association’s
territory as they enjoy over their privately owned property—such as the right to
use, the right to manage, the power to sell, the right to restrict access, immunity
from outsiders’ interference, and so on. This calls for two further comments. First,
sovereignty—understood as exclusive authority over residents within a given territory
consistent with compliance with cosmopolitan morality—is a form of (political)
ownership in that territory. On corporate conceptions of the sovereign, territorially
bounded state, this is undoubtedly odd. For on those conceptions, territorial rights
are not considered as rights that individual members together have, but as rights
that the state has as an artificial person, over time. Such rights are not passed on
from one generation to the next: they are constantly held, over time, by the state.
At the bar of cosmopolitan justice, however, states are not corporate entities: the
rights that they have are in fact rights held by their individual members qua such
members, protect the interests of those individuals so understood, and are passed
on from one cohort to the next.
Second, while members of political community C cannot be jointly sovereign
over T unless they have jointly held rights of public ownership in T as a whole,
they can be sovereign over T as a whole and at the same time forego opportunities
to own privately all of the resource contained in T, by enabling non-citizens to
become property owners. Should they so wish, they could also forego some occu-
pancy rights over T. One might think that such decisions are first and foremost a
prerogative of sovereignty—just as it is my prerogative as a house owner to rent my
house to some other party and, in so doing, temporarily alienate some of my rights
over it. But this is not so: as we saw above, justice imposes limits on agents’ control
rights over the material resources which are in their de facto possession. This
includes resources such as territory conceived of as whole, and applies to agents
qua citizens. Conversely, individuals who are not members of C but who neverthe-
less reside in T can enjoy ownership and occupancy rights in respect of T’s resources
without being sovereign over T. In short, territorial-sovereignty rights over T in
toto are not strictly coextensive with ownership and occupancy rights over all of T’s
constitutive parts.
5.2.2. Restitution: The Basics

So far, so good. Suppose now that at time t1 A takes a good g which is in B’s de facto
possession. Let us first assume that B was not g’s rightful owner. Even so, it does
not follow that A did have a justification for so acting: the fact that B is not entitled
to g does not establish that A is, and it may well be, therefore, that A is under a
duty to hand over g—if not to B, to some other party.
Let us now assume that B is g’s rightful owner but that she forfeits her rights over
it. For example, she starts using her gun to shoot at some defenceless bystander. In
this case, A is justified in seizing the gun at t1; likewise if B shoots at him. But even
so, A is under a duty to return g at t2 if he does not have a justification for holding
on to it.
124 Restitution
Suppose finally that at t1 B was g’s rightful owner, was not under a duty to give
it away, and (thus) did not forfeit her rights over it. Here we must distinguish
between two scenarios. In the first scenario, there is a lesser-evil justification for
taking g anyway: for example, by taking g away from B, who is not under a duty to
divest herself from it, A would save a number of lives. If that or any other justifica-
tion no longer applies, then A is under a duty to return g to B at t2. In the second
scenario, A was not justified in taking g from B at t1. In so far as A was under a
duty not to take g, it seems that he is under a duty to give it back at t2. For A
wrongfully took from B something that rightfully belonged to her and indeed still
rightfully belongs to her, period. To the extent that B was, and still is, g’s rightful
owner, and that there is no reason why she should not be allowed to use g, g should
be returned to her—not just by A, in fact, but by whomever happens to have g.
In sum, even if dispossessing B of g is morally justified, A does not thereby
acquire ownership of g and is therefore under a prima facie duty to return it, either
to B when B can still be described as g’s rightful owner, or to some other party
when B cannot be so described. More clearly still, if dispossessing B of g is not
morally justified, A is also under a prima facie duty to restitute it to her, for B is still
its rightful owner.
Some might worry that those remarks rely on the viciously circular claim that B
still owns g at t2 following A’s act of seizure at t1. Appealing to the claim that she does
begs the question—or so they might say—for the question is precisely whether that
is the case. I do not think that the argument is circular, but highlighting why sheds
light on an important distinction in the political philosophy of ownership—between
justifying the conferral on agents of (exclusive) rights, powers, liabilities, etc., in
respect of g on the one hand, and ascertaining the conditions under which a given
agent actually does have those rights. Let me explain. Earlier, I justified ownership
rights by appealing to individuals’ interest in durably enjoying and controlling the
resources they need to lead a flourishing life. But from the fact that enjoying and
controlling g would so protect B’s interest, it does not follow that B has ownership
rights over g. To see this, suppose that B has a gun in her possession, with which she
has already shot at A once without justification. Suppose further that A has the
means to defend himself. Clearly, once A fires back, for B to use her gun in her own
defence (at that point) would serve her interest in remaining alive. Yet she ought not
to kill A, even in self-defence: rather, she ought to surrender, which implies that she
does not at that point have the exclusive right to use her gun.
In other words, the flourishing-life justification for ownership rights must be
supplemented with an account of the conditions under which agents acquire (and
forfeit) those rights. I shall assume along Lockean lines that B rightfully owns g if
and only if she has appropriated it ab initio, or produced it out of resources which
she rightfully owned or obtained it from someone who rightfully owned it, and
provided that her having exclusive rights over g does not wrongfully cause some
other party to have a less than flourishing life.9 Now, suppose that B rightfully
9 I lack the space to defend those Lockean conditions here. I hope that they are intuitively plausible
enough to warrant endorsing, even though the theory of justice which underpins my arguments in this
Restitution 125
owns g at t1. By unjustifiably taking it away from her without her consent, A does
not acquire rights over it: he has not produced it, nor was it given to him by its
rightful owner. There is, thus, independent support for the claim that B still right-
fully owned g at t2, and the argument for restitution is not viciously circular.
There is also independent support for the claim that if B’s descendants would
have rightfully owned g at t3, in virtue of having rightfully inherited it, then A
is under a prima facie obligation to give it back to them. However, victims’
descendants, here and now, have an inheritance-right over those objects only if
(a) individuals in general have inheritance rights over their ancestors’ estate and
(b) it is plausible to hold that those descendants would have inherited those
artefacts had it not been for the initial wartime taking. On the first point, subject
to inheritance taxes as required by the imperative to ensure that all individuals
enjoy prospects for a flourishing life, sufficientist justice does make space for both
the right to bequeath and the right to inherit as one important way in which agents
can exercise the personal prerogative to attach greater weight to their own projects
and attachments than those of others. B, that is, might well enjoy the personal
prerogative to bequeath a prized heirloom to a loved one; should she die intestate
with respect to that particular heirloom, her heir might well enjoy the personal
prerogative to inherit it. However, and with regard to the second point, the more
time lapses the less certain one can be that such claims can be made, since the more
time has passed, the more likely it is that the chain of ownership linking current
claimants to their ancestors would have broken anyway through sales and gifts
to outsiders.10
Now is as good a time as any to address an often-rehearsed objection to corrective
claims as made by victims’ descendants. The objection—or non-identity problem—
begins by noting that those descendants would not exist were it not for the initial
rights violation (since their parents were brought together through that initial act,
without which they would therefore not have procreated when they did and would
therefore not have created those particular children). How, then, can they plausibly
claim that they were wronged by that act, and that they are owed reparation?11
Although the objection is raised against reparations in general, it applies equally, if
it does at all, to restitution claims in particular, for example as follows: ‘had A’s
troops not invaded TB and wrongfully appropriated its resources, consequent pop-
ulation movements would not have occurred, and you yourself would not exist,
and so you have not been wronged by the invasion and therefore cannot demand
that A should return what they took’.
book is not itself Lockean. There is no reason why Lockean conditions on the conferral of ownership
rights must, first, imply a commitment to self-ownership and, second, imply a denial of robust (non
rectificatory) rights to resources.
10 For a well-known defence of the familiar claim that time can extinguish rights and obligations,
see J. Waldron, ‘Superseding Historic Injustice’, Ethics 103 (1992): 4–28. Christopher Kutz makes a
similar claim in the specific context of land reparations in central Europe to individuals and groups
who had been despoiled by Communist regimes. See Kutz, ‘Justice in Reparations’.
11 The locus classicus for the non-identity objection is D. Parfit, Reasons and Persons (Oxford:
Clarendon, 1984), chs. 16–17.
126 Restitution
Of all the response which the non-identity objection has generated, one in
articular is worth explicitly rejecting in the context of this book. It goes like this:
p
we cannot say in individuals’ cases ‘had this person’s ancestors not been victims of
discrimination, she would now not be a victim of injustice (since she would not
exist)’; but in group cases, we can say: ‘had members of this group not been victims
of slavery and subsequent discrimination in the past, members of this group, here
and now, would not be victims of injustice qua group members’.12 In so far as war
tends to pit groups of individuals against each other at least as much as it pits mul-
titudes of combatants each individually killing other individual combatants and/or
non-combatants, a group-based account of corrective justice might seem to avoid
the non-identity objection. Unfortunately, however, it grants individuals a claim to
reparations in general, and restitution in particular, not in virtue of their having
been harmed as a result of their ancestors having been wronged, but simply in
virtue of belonging to a given group. In so far as cosmopolitan justice regards
group membership per se as irrelevant to the rights and liabilities of individuals,
one must look elsewhere for a way around the non-identity problem. One must do
so, moreover, in a way which preserves the claim that in failing to satisfy restitutive
demands arising from past wrongful harms, one is derelict in one’s duties to the very
same person who is a victim of injustice here and now.
An attractive response to the objection which I will adopt here and which applies
not just to the restitution of moveable resources but also to territorial restitutions,
the right of return and reparations, has been offered by a number of authors. As
they note, even if an unjust act brings about the existence of a particular individual
and thus does not make that individual worse off overall than she would have
been otherwise (on the assumption that existence is not worse than non-existence),
failure to provide reparations to her parents for that unjust act after she has been
conceived might well have severely harmful effects on her—so harmful in fact as
to constitute an injustice for which she in turn is owed reparation.13 In addition,
to the extent that wrongdoers commit a given injustice not merely at time t1 but
also, and ongoingly, at t2, an individual whose existence is in part caused by the
commission of that injustice at t1 nevertheless has a t2-related grievance against
those wrongdoers. The point, though obvious, bears stressing in the present context,
12 See, e.g., J. S. Fishkin, ‘Justice Between Generations: Compensation, Identity, and Group
Membership’, in J. W. Chapman (ed.), Compensatory Justice (New York: New York University Press,
1991). Fishkin couches the argument in terms of individuals being worse off or better off, and not in
terms of being wronged or not.
13 See G. Sher, ‘Transgenerational Compensation’, Philosophy & Public Affairs 33 (2005): 181–200;
Waldron, ‘Superseding Historic Injustice’; J. Waldron, ‘Redressing Historic Injustice’, The University
of Toronto Law Journal 52 (2002): 135–60; Butt, Rectifying International Injustice; E. F. Paul, ‘Set-asides,
Reparations, and Compensatory Justice’, in J. W. Chapman (ed.), Compensatory Justice (New York:
New York University Press, 1991). Note that this move only blocks the non-identity objection in its
backward-looking invocation (that is to say, in the form: ‘we are not under a duty to compensate you
for past wrongs which led to your existence, since you would not exist but for those wrongs and since
your existence is not worse than your non-existence’). It does not block the objection in its for-
ward-looking invocation (that is to say, in the form: ‘we are not under a duty to you now not to make
such and such decisions some of which consequences will harm you, since not making those decisions
will lead you not to exist, and since your existence is not worse than your non-existence’).
Restitution 127
precisely because as we saw in ch. 2, belligerents do not merely start a war at t1:
they continue to fight it at t2, t3, and so on, and at each stage their decision not
to desist from their initial course of action calls for moral scrutiny. A child born
of citizensB who would not have met but for the fact that they both fled in the
aftermath of A’s wrongful appropriation of resources within TB at t1 might not
have been made wrongfully worse off by A’s regime at t1; but to the extent that A’s
continued appropriation and failure to return what it took impaired her prospects
for a flourishing life (for example, because the loss of those resources undermined
the ability of B’s institutions to deliver justice to citizensB), she does have a legiti-
mate grievance against A. To object to that child’s demand for the return of those
resources on the grounds that she owes her existence to the annexation at t1 will
not do, since at t2 A’s regime could have returned the resources, and yet wrongfully
failed to do so.
Let us take stock. I have argued that ownership, occupancy, and sovereignty-cum
territorial rights protect their holders’ interests in having control over their social,
economic, and political environment, but that sufficientist justice constrains both
the right to bequeath and the right to inherit. Turning to restitutive justice in gen-
eral, I have defended the following theses: wrongful appropriation of what individ-
uals rightfully own generates restitutive duties on the part of the appropriators, so
long as those individuals can still be considered rightful owners of what was taken;
the fact that someone is not the rightful owner of a good does not licence all and
sundry to appropriate that good, and wrongful appropriations in such cases can
generate a duty to return the good; a rightful appropriator is sometimes under a
duty to restitute that which he has taken. In the remainder of this chapter, I bring
those general principles to bear on wartime appropriations.
5 . 3 . R E T U R N I N G M OV E A B L E P RO P E RT Y
Suppose that in the course of their invasion of B’s territory, A’s troops seize move-
able resources owned by citizensB. Suppose further that at the end of the war, A still
has at least some of those resources.14 On what grounds and under what condi-
tions ought a justifiedATC settlement between A and B stipulate the restitution of
those things? As a preliminary step, we ought to distinguish the seizure of resources
by combatantsA for their personal or collective enrichment, from the seizure of
resources needed to carry on with the war. Strictly speaking, combatantsA may have
taken military artefacts to enrich themselves or their community, and non-military
artefacts to carry on with the war effort. For ease of exposition, however, I assume
that non-military artefacts were taken for personal enrichment, while military arte-
facts were taken to assist in the war effort.
Let us begin, then, with enrichment. A well-known example involves combat-
ants seizing paintings and other similarly prized objects from their victims during
14 If it does not, the issue at stake is whether it ought to compensate B for the resulting loss. I postpone
this question until ch. 6.
128 Restitution
the war. In so doing, they act either on their own initiative or at the behest of their
governments, and they do so either to enrich themselves personally, or to fill their
country’s museums, or both. Thousands of such objects crossed international bor-
ders during the Second World War, whether taken by German combatants from
occupied countries or by Soviet combatants on their way to Berlin.15 The case of
German looting in particular is so familiar that I will use it for illustrative pur-
poses, though the conclusion I reach below applies mutatis mutandis to movable
property in general. So, is it a necessary condition for a justifiedATC peace that
those artefacts should be returned to the museums from which they were taken,
or to the individuals—mostly Jewish—from which they were confiscated, or to
those individuals’ heirs?
First, let us assess restitutive claims as made by those individuals from whom the
objects were taken. As we saw in s.5.2.2, only if citizens of occupied countries were
in fact the rightful private or collective owners of those objects can they make a
restitutive claim. But in so far as the regimes of occupied communities themselves
would routinely despoil their own conquered enemies, one may doubt that those
communities’ members were in rightful collective possession of those objects.
Moreover, whether private owners of confiscated artefacts were in morally rightful
possession of the latter at the time of the occupation is doubtful too (at the risk of
being controversial). For under no plausible description of the world as it stood in
1939 can it be held that it was a just world—a world, that is, where all individuals,
wherever they resided, enjoyed prospects for a flourishing life. To bring about a just
world then would have required heavier taxation and considerably more stringent
restrictions on inheritance than were in fact imposed. In so far as there are over-
whelmingly good reasons to believe that those who were in legal rightful possession
of those objects were not in fact their morally rightful owners, it is not clear that
they have a restitutive claim.
Second, let us consider restitutive claims as made by the heirs of those whose
artwork were taken away. Those heirs cannot always claim that they would have
had those objects in their de facto possession had enemy combatants not stolen
them—most obviously in those cases where they would not exist but for the act of
theft. As we saw in s.5.2.2, it is only if the theft’s failure to return the good to their
parents has harmed them to the point of constituting a wrongdoing that they have
a restitutive claim. But in this case it is not clear why the claim should be a claim
to have that object itself returned to them, and not a claim to be compensated for
the wrong which they suffered. In any event, even if heirs can plausibly claim that
they would have the artwork in their possession had it not been stolen away from
their ancestors, it does not follow that their restitutive claims are justified. In par-
ticular, there might be strong reasons at the bar of justice to consider those objects
as a way to pay inheritance tax, particularly when heirs are unable to pay the latter
in hard currency. Thus, in lieu of the $50 million which Picasso’s heirs owed in
estate duties to the French state, they gave the latter first pick of his enormous
collection—now on display at the Picasso Museum in Paris.
15 See, e.g., Barkan, The Guilt of Nations, ch. 4.

Restitution 129
Suppose, then, that those from whom those objects were taken, and/or their
descendants, are not the rightful owners of those artefacts. Even then, it does not
follow that the Germans are entitled to hold on to those objects tout court.
Remember: the claim that someone is not the rightful owner of a good g does not
imply that g may be forcibly taken away from him, or withheld from him, by
whomever so wishes. What, then, are we do to?
Those cases admit of two rather different responses. On the one hand, one could
envisage a global redistribution of stolen artworks away from the art-rich towards
the art-poor. On the other hand, one could also envisage a system whereby the
artefacts which have been wrongfully taken away from a wrongful de facto posses-
sor are sold to the highest bidder, with the proceeds to go either to the relief of
worldwide poverty or to a global reparations and reconstruction fund. Although
access to art is (I believe) a component of a flourishing life, access to the basic
necessities of life is more urgent, which suggests that the latter solution is to be
preferred to the former so long as there still are individuals in the world who are
dying of thirst, starvation, and preventable diseases.
Both proposals, incidentally, also provide a way around feasibility constraints
which stand in the way of implementing restitution principles. Not all victims of
wrongful dispossessions can be identified, and even if they can, the material and
human costs of ensuring that they get back precisely what was theirs are likely
be so high as to detract from the pursuit of other just ends (be they reparative
or distributive.) Inability to garner the requisite knowledge and resource scarcity
operate as, respectively, hard and soft as well as purely practical and moral con-
straints on the implementation of restitutive principles for moveable resources.
Distributing wrongfully taken property in general (and not just works of art) or
selling them on to fill the coffers of global poverty and/or reparations-reconstruction
funds constitutes a compromise on a just peace simpliciter, but an important part
of a justifiedATC peace.
Although I have focused on the restitution of works of art, the points I have
made so far apply to moveable property in general which is taken away for personal
and/or collective enrichment. The case of property taken away with a view to
assisting in the taker’s war effort raises interestingly distinct issues, which I shall
illustrate with the example of the seizure of weapons.16 As a preliminary step, it
would be natural to suppose that if A’s invasion is wrongful (either because it lacks
a just cause or because it violates some other requirement(s) of a just war), its
troops are under an obligation not to seize B’s weapons, and should return them as
part of a just settlement. In some cases, however, this is not so. For a start, and as
we saw in chs. 2 and 3 respectively, A might sometimes have a justification for
continuing at t2 a war which was unjust at t1, and/or for continuing to occupy B
though its initial war against and occupation of B was unjust. Were that the case,
A would have a justification for ensuring that B’s regime and citizens should not be
able to defend themselves, by seizing their military equipment. If, as a result of
having had to fight a now-justified war against B, A’s own military resources are
16 See also May, After War Ends, 191.

130 Restitution
depleted to the point where it does in fact need B’s matériel in order to retain the
ability to protect itself or third parties once the war is over, then it is not under a
duty to return all that it took from B.
Furthermore, suppose that although A’s leaders are in a position to reach a justi-
fiedATC settlement with B’s leaders, there are insurgent groups within B who are still
posing a threat (though not one severe enough to justify continuing with the war).
If returning military equipment to B’s regime would pose too much of a risk (for
example, because insurgents would be likely to seize it themselves), A’s leaders are
not under a duty to return it to B. However, it does not follow that they have the
unconstrained right to hold on to it. Most clearly, they lack that right if they will
go on to use those weapons to unjust ends, or (more controversially) if some third
party C lacks the weapons it needs rightfully to defend itself—in which case A may
well be under a duty to give or at the very least to sell those weapons to C.
So much, then, for restitutive obligations with respect to military equipment as
befall on A, whose war was unjust at t1. B, then, is waging an ex hypothesi just war,
and its troops are morally permitted to seize enemy matériel, on two grounds: as a
means to undermine troopsA’ ability wrongfully to defend themselves, and as a
means to bolster their own ability to do so. But even if B’s war is a just war, it ought
to return the military equipment it has seized once the war is over unless it has
objective reasons for thinking that A will use it wrongfully. To be sure, whether
belligerents are permitted to keep, or are under a duty to destroy or give, the mili-
tary equipment which they have seized, depends not just on what this equipment
might be used for; it also depends on one’s prior stand on the moral permissibility
of arms acquisition in general. Since going to war is sometimes justified, procuring
the weapons required to fight a just war is itself justified. In so far as A might at a
later stage need such weapons to further justified ends, it would be wrong of B to
fail to return equipment to which A had a just title before their war started. There
is more to say of course about arms procurement in general. But the lesson to draw
from this short discussion is that with respect to the restitution of military equip-
ment, the status of belligerents as just or unjust has little bearing on their post-war
obligations to their erstwhile enemy.17 As we shall now see, the same goes with the
restitution of territory.
5 . 4 . R E T U R N I N G T E R R I TO RY
Suppose that belligerent A has invaded territory over which B exercises de facto
jurisdiction and forcibly annexed part of that territory (TB) while the war was
17 I am assuming in this and the previous paragraphs, first, that there can be a duty to provide
weapons to those who need them and, second that there is a such a thing as a right to procure and sell
weapons. I defend the first assumption in C. Fabre, ‘Mandatory Rescue Killings’, Journal of Political
Philosophy 15 (2007): 363–84. For a thoughtful treatment of the ethics of international arms trade,
see J. Christensen, ‘Weapons, Security, and Oppression: A Normative Study of International Arms
Transfers’, Journal of Political Philosophy 23 (2015): 23–39.
Restitution 131
ongoing—as, e.g. the USSR did with the Baltic States in 1940. On what grounds
and when, if at all, ought A return TB to B at the conclusion of the war?
To start with, assume that A’s war against B was just, and that A went to war
precisely because its citizens have legitimate sovereignty rights in respect to TB. In
annexing TB, citizensA, via their military and civilian leaders, recovered what was
theirs to begin with, and they are not under a duty to return it—pending qualifi-
cations to be issued below. The same point holds if A’s war was justified on grounds
other than the recovery of TB and if, in fact, recovering TB would not on its own
have justified that war. The case I have in mind here goes something like this: B’s
regime carries out a genocidal campaign against an ethnic group within its borders;
A mounts a justified war of humanitarian intervention, in the course of which it
seizes TB. The recovery of TB together with the intervention outweighs the harms
caused by the war, but the former alone would not. Yet, the fact that annexation
(through war) would have been unjustified had there not been another just cause
for A’s war does not render citizensA’ claim over TB null and void. In this case too
A has a right to hold on to it. It also has that right in a slightly different case, where
A succeeds in its war of intervention against B but nevertheless continues with the
war as a means to recover and annex TB, even though it would not have been jus-
tified in waging war solely to do so at the outset. Although A’s decision to continue
with the war is unjust (as we saw in s.2.3.1), the fact remains that its citizens are
rightfully sovereign over TB. Contrastingly, if A does wage an overall just war
against B in the course of which it annexes TB though does not, in fact, have sov-
ereign claims over it, then—again, pending qualifications below—it is under a
duty to return it as part of a justifiedATC peace.
Suppose now that A’s war against B, in the course of which it annexed TB, was
unjust—as was the USSR’s invasion and subsequent annexation of the Baltic States
in 1940. It would be tempting to think that A is under a duty to restitute TB in the
sense that it should renounce all claims to it. In general, however, this would be too
quick (though not in the USSR’s case). For the claim that A has wrongfully invaded
B’s territory is ambiguous, denoting as it does either that A lacked a just cause so
to act, or that although A did have a just cause, its invasion breached some other
requirement of the just war. In the first case, A’s leaders and combatants ex hypothesi
have violated citizensB’ jointly held rights to the territorial integrity of their com-
munity. If, in addition, A’s leaders have annexed TB, they have violated citizensB’
jointly held sovereignty rights over it. If citizensB, via their leaders, insist on TB being
returned to them, A must comply, in virtue of the fact that (as we saw in s.5.2.2)
its members do not have rights over TB. By the same token, it would be wrong of
A’s leaders to threaten B’s leaders with not returning TB as a means to get them to
agree to further demands. This is so even if A’s demands are just, for even if they
are just, the fact that A and B are under a duty to reach a settlement implies that
the satisfaction of those (just) demands cannot proceed by way of a continuing
invasion.
In the second case, by contrast, A had a just cause for invading B’s territory,
though ought not all things considered to have done so. Even then, we need to
distinguish between the following two kinds of cases: (1) A had a just cause for
132 Restitution
invading TB because citizensA have jointly held rights over TB; (2) B’s regime had
wrongfully attacked A, and A could successfully repel the attack only if it invaded B’s
territory (over which its citizens do not have sovereignty rights), but A’s defensive war
nevertheless breaches some other requirement(s) of the just war. Type (1) cases are
relevantly similar to cases where A’s war is just in toto but where A continues to fight,
unjustly, in order to recover TB: a justifiedATC peace settlement must declare that A’s
annexation of TB was justified, even though its invasion thereof was wrongful.
In cases of type (2), A may have a justification for occupying TB, under the con-
ditions set out in ch. 3. For as we saw therein, members of a political community
have jointly held rights of sovereignty over that community’s territory, and political
officials acting on their behalf have the fiduciary power to govern, if and only if they
respect one another’s as well as outsiders’ human rights. Should they fail to do so,
they temporarily forfeit their rights of sovereignty over T. Crucially however none of
this is tantamount to justifying A’s annexation of TB—where annexation consists in
a full-blown (unilateral) transfer of all sovereignty rights. In this case, a justifiedATC
settlement must therefore acknowledge that the annexation was wrongful, and, if
not return TB to citizensB, at least remove it from A’s sovereign jurisdiction.
In short, as a first cut, the moral status of a belligerent’s war, and more precisely
the reasons why and ways in which that war is unjust, only have limited bearing
on that belligerents’ obligations with respect to territorial restitution: there are
cases where A’s war might be just but where A nevertheless ought to return TB, and
vice versa.
As a second cut, moreover, we must attend to the composition of TB’s residents
and to their wishes with respect to their territory at the conclusion of the war.
Suppose that a substantial number of those who live on TB at the time of the
annexation do not wish to be incorporated back into B once the war is over: they
would welcome remaining annexed to A, or would wish to become a politically
sovereign, independent community. The most that one can say in those cases is
that A must divest itself of TB if those individuals refuse to be subject to its rule.
Were that the case, the question is whom should A relinquish control to. The answer
to that question depends on whether those individuals within TB have the right to
secede from B. On some views, a group has the right to secede from the commu-
nity which exercises sovereignty over it if and only if its members consent to it. On
other views, it has the right to do so only if its members consent to it and provided
that the resulting newly independent sovereign community would not violate the
rights of its own minorities. On other views still, it has the right to do so only if its
members are subject to persecution from the larger community or are not ade-
quately recognized as a distinct national community by the latter, and subject to
the same proviso.18
18 For a thorough review of various arguments for and against secession, see A. E. Buchanan,
Secession: The Morality of Political Divorce from Fort Sumter to Lithuania and Quebec (Boulder:
Westview Press, 1991). For examples of the arguments mentioned here, see, e.g., M. Moore (ed.),
National Self-Determination and Secession (Oxford: Oxford University Press, 1998); A. Patten,
‘Democratic Secession from a Multinational State’, Ethics 112 (2002): 558–86; C. Wellman,
‘A Defence of Secession and Political Self-Determination’, Philosophy & Public Affairs 24 (1995).
Restitution 133
Providing a full defence of the right to secede is clearly beyond the scope of this
book, let alone this chapter. At the very least, however, it follows from the account
of political legitimacy which I endorse here that persecution from the larger group
provides its victims with a justification for seceding, and that ability and willingness
to respect the human rights of both insiders and outsiders following independence
operates as a constraining condition on the right to do so. This leaves the door
open to more permissive justification for secession—for example, justifications
grounded on consent alone. Without settling that issue, one can at least say that if
citizensB will suffer human rights violations as a result of remaining in B such that
they will lack prospects for a flourishing life, A not only is not under an obligation
to return TB to B’s regime and citizens: it is under an obligation not to do so. Given
that ex hypothesi residents of TB do not wish to be incorporated into A, A is under
a duty to grant them independence. The same would apply, in principle, if the
human rights account of secession is in fact too restrictive and should be replaced
with a more permissive justification.
Interestingly, this human rights–based argument against restitution also applies,
but this time against the wishes of TB’s residents, in the following case. If descend-
antsB would exercise their self-determination rights over T in violation of justice,
for example by using the fact that they are a numerical majority to launch seriously
discriminatory policies against some other group, A’s leaders and those on whose
behalf they act not only would not be under a duty to accept the resulting settle-
ment: they would be under a duty to reject it and to hold on to TB. Under those
conditions, restitution would also be unjust: whether A invaded TB rightfully to
begin with is irrelevant.
In some cases, of course, residents of TB might not want to constitute themselves
into a sovereign, independent political association, or might not yet be able to do
so, and would rather join A. A might be able to obtain from B’s leaders, as a means
to protect those residents, that they not demand the return of TB—a demand
which would be morally unwarranted anyway since their readiness to persecute
their residents implies that they do not have the morally justified power to govern
over it. Assuming that A would be able to incorporate TB into its own territory
without jeopardizing its own members’ prospects for a flourishing life (as stipu-
lated by the no-undue-sacrifice proviso), the requirement to seek and accept a
justifiedATC peace settlement might well imply, not only that it is under a duty not
to cede TB back to B in this case, but also that it is under a prima facie duty to keep
it, in fulfilment of a duty of assistance to citizensB.
So far, I have assumed that TB has remained populated by citizensB. But suppose
now that the latter flee in large numbers in the face of the invasion and are replaced
by members of A who settle there. In the ensuing peace negotiations, is A under a
duty to return TB to B’s jurisdiction? If A lacked a just cause to begin with and if
the peace settlement occurs sufficiently soon after the initial invasion, then yes,
for those settlers are not entitled to exercise sovereignty rights over that territory.
Several difficulties arise however. First, as I noted at the close of s.5.2, public owner-
ship of T as a whole is compatible with only partial private ownership and occupancy
of land and resources in T. One may therefore wonder whether territorial restitution
134 Restitution
must go hand in hand with allowing citizensB to return to TB and/or forcing citizensA
to leave. I shall tackle this issue in s.5.5.
Second, when peace negotiations occur decades after the conflict and following
a long occupation of TB by A, those wrongful first settlers have been replaced by
their descendantsA, and the question, then, is whether those individuals together
with their compatriots who still live on A’s pre-war territory TA should be held
under a collective duty to renounce a share of sovereignty over TB for the benefit
of the descendantsB of their ancestors’ victims. There are reasons to doubt that
descendantsB have exclusive sovereignty rights over TB. As we saw when discussing
the non-identity objection, most descendantsB cannot claim that they would be
sovereign over that territory were it not for the war, for were it not for the war
they would not exist. Those individuals thus cannot claim to have the right to
inherit TB from their ancestors. To be sure, they might be able to claim that their
current predicament (living a less than flourishing life) can somehow be traced to
A’s annexation of TB. But it does not follow that they have the right to exclude
descendantsA from any share of sovereignty over TB. Moreover, and as we also saw
in s.5.2.1, inheritance is only one possible basis for territorial rights (if it is at all):
current and long-term occupancy (jus habitationis) is another, and a much
stronger one at that. Even if descendantsB have an inheritance-based sovereignty
claim over TB (which I doubt), and a fortiori if they do not, their interest in full
sovereignty over TB is overridden by the presumption in favour of allowing pres-
ent and long-term occupiers—descendantsA—to stay where they are and to enjoy
sovereignty rights over their place of residence. To put the point differently, orig-
inal claims in land are overridden, indeed sometimes extinguished, as political,
social, and economic circumstances change in such a way that new generations of
individuals, each with a fundamental interest in a flourishing life, come into
being and that the world in which they live is substantially different from the
world in which the wrongdoing of territorial annexation was committed.19 To
claim otherwise and say, in this particular case, that descendantsA ought to return
TB wholly to B’s exclusive sovereignty is to imply that they would rightly become a
permanent and disenfranchised minority on TB. But that position is incompatible
with justice since permanent disenfranchisement to which one has not consented
(for example, by migrating to another country) vitiates one’s opportunities for a
flourishing life.
At the bar of justice, thus, the permanent disenfranchisement in TB of both
descendantsA and descendantsB is unjust. By implication, in cases such as these, no
peace settlement should confer full, exclusive (of the other) sovereignty over TB to
either party, unless the other party explicitly consents to it via legitimate ratification
procedures. In the language of ch. 4, the declarative clauses of a peace settlement
with respect to territorial restitution must confer at least some sovereignty rights on
all members of all belligerents who claim such rights, on pain of being unjust.
I say ‘at least some’ however, so as to leave the door open for two different kinds of
arrangements: a division of TB into two distinct territories, over which descendantsA
19 See Waldron, ‘Superseding Historic Injustice’.

Restitution 135
and descendantsB respectively enjoy exclusive sovereignty, or a form of power-sharing

between descendantsA and descendantsB over TB in toto. Under the latter proposal,
neither set of agents would enjoy exclusive sovereignty over the whole of TB. Even
if descendantsA rightly refuse to return TB to B’s exclusive jurisdiction, they are
under a duty to accept either one of those two compromises. Which compromise
they must accept depends in part on feasibility considerations in general, and con-
cessions which descendantsB, via their own leaders, are motivated to make in par-
ticular. Stipulating those terms in detail is the work of declarative clauses in the
settlement when this or that set of terms offers better prospects for a justifiedATC
peace; as between morally equivalent terms, it is the work of specifying clauses.
To illustrate, consider the case of Israel and Palestine. I should say at the outset
that I do not aim here to provide a novel normative solution to this particular
dispute (as if!). Infinitely more modestly, my aim is merely to show how an often
mooted and widely accepted normative solution fits in with my overall argument
in this book. So, assume that Jewish settlers and immigrants acted without just
cause after the Second World War by wrestling control of Palestine from the Arabs
and establishing a Jewish state. Assume further that a just peace simpliciter, here and
now, would require that Israeli Jews and Palestinians be willing to live together on the
land of Israel, in a non-Jewish state, under strictly non-discriminatory political
institutions. Evidence strongly suggests that neither community will accede to
this. Some, on the extremist Palestinian side, wish not just for the destruction of
Israel but for the expulsion of all Jews from the land and (for the most extremist
amongst those) their physical death. Others, on the extremist Israeli side, wish
not just to preserve the Jewish character of their state but to make it impossible
for Palestinians not merely to accede to independent statehood and (for the most
extremist amongst those) to enjoy a flourishing life, indeed to survive. Neither
set of demands is an acceptable compromise, for both violate Israeli Jews’ and
Palestinians’ basic human rights. A justifiedATC peace, under those circumstances,
requires a two-state solution, within pre-1967 borders. Palestinians would have
to accept the loss of some of the land which they lost in 1948; Israeli Jews for
their part would have to renounce their sovereignty claims to Gaza, the West
Bank, and East Jerusalem—which of course is the premise of the ongoing peace
process such as it is.
5.5. P O P U L AT I O N D I S P L A C E M E N T S : E X P U L S I O N S
AND RETURNS
So far I have distinguished the return of territory from the return of movable
resources—whether privately or publicly owned. In many cases, however, the two
issues are fused—a complication which the troubling issue of population displace-
ments and their aftermath illustrates. Suppose that A is in the grip of a civil war
between religious-cum-ethnic factions which spills over neighbouring country
B: thousands of citizensA who are at risk of persecution at the hands of one of those
factions flee into B, leaving behind their privately owned land and houses.
136 Restitution
Or suppose that community A, a fledging independent sovereign state, is at war

with community B, whose members live on disputed territory T and either aspire
to statehood or wish to enjoy equal status under A’s jurisdiction. As A’s troops gain
the upper hand they expel some citizensB (under force of arms) and incite others
under duress to leave the area and to find refuge either in neighbouring country C
or in camps dotted around B. Meanwhile, A’s troops, under instruction from or
with the collusion of A’s leaders, appropriate the houses and land left vacant by
citizensB’ departure. Or suppose that as a result of a civil war within A, or of B’s
invasion of A, thousands of civilians leave their homes but remain within A’s bor-
ders. Or suppose, finally, that A has lost its overall grievously unjust war against B,
and that B’s leaders forcibly expel citizensA from land which is outside A’s jurisdic-
tion but on which they have lived for generations, and return them to A.
Population displacements are a scourge of war. According to the UN Refugee
Agency, as of January 2014, there were nearly 12 million refugees under its man-
date, in addition to some 5 million Palestinians and a further 33 million Displaced
Persons within their own countries (IDPs).20 My stylized civil war case is modelled
on, inter alia, the civil war which ravaged Guatemala between 1954 and 1996, and
the Rwandan civil war of 1994. In Guatemala, thousands of people, mostly from
Mayan communities, died at the hands of genocidal squads; many more fled to
Mexico. In Rwanda, close to a million of Tutsis and moderate Hutus died in one
of the worst mass atrocities since the Holocaust; dozens of thousands fled to neigh-
bouring countries, notably Zaire—as in fact did thousands of Hutus, not all of
them extremists, who feared reprisals as the Tutsi Rwandan Patriotic Front ousted
the genocidal regime. The second case is modelled on the seemingly intractable
dispute between Israelis and Palestinians. The third case, involving internally dis-
placed persons, is modelled on more or less any conflict. The fourth case takes
inspiration from the generally overlooked Allied policy, in the immediate after-
math of the Second World War, to expel German minorities from Czechoslovakia,
Hungary, and Poland, and forcibly return them to Germany and Austria. The
policy was carried out at first in a non-systematic fashion, mostly by military and
civilian Soviet leaders in the closing stages of the war; it was then enshrined in the
1945 Potsdam Agreement, in other words after Germany’s defeat, and imple-
mented in an orgy of appalling crimes against its victims—ranging from routine
torture to gang rapes and mass executions.21
20 See http://www.unhcr.org for the relevant data (accessed on 14/01/2016). According to the
1967 Protocol Relating to the Status of Refugees, a refugee is someone who is outside his/her
country of nationality or (if stateless) habitual residence and unable or unwilling to return there
because of well-founded fears of being persecuted for reasons of race, religion, nationality, mem-
bership in a particular social group, or political opinion (Art. I.A.2). Internally displaced persons,
by contrast, have not left their country and are still technically under the jurisdiction of their
government.
21 For the German case, see, e.g., MacDonogh, After the Reich, ch. 4. A classic account of the
Palestinian refugee problem at the time of Israel’s Independence is B. Morris, The Birth of the Palestinian
Refugee Problem Revisited, 2nd edn. (Cambridge: Cambridge University Press, 2004). For a recent
book-length account of the many difficult issues raised by the right of return, see M. Bradley, Refugee
Repatriation—Justice, Responsibility and Redress (Cambridge: Cambridge University Press, 2013). For
a recent compendium on the relevant laws as well as the range of issues raised by forced migration, see
Restitution 137
Was the Potsdam Agreement, by which the United States, the United Kingdom,
and the Soviet Union settled Germany’s fate, unjust in this particular respect?
Should a settlement between Israelis and Palestinians allow Palestinian refugees
who left Israel in 1947–49 to return to the precise parcels of land which they occu-
pied then on pain of being unjust? Were Mayans justified in claiming (more or less
successfully) the right to come back to the houses and land they had left during the
civil war? Does any individual who has been internally displaced as a result of the
war (be it civil or interstate) have a right to go back to his home of origin? Are they
under a duty to do so?
Let us begin with forcible expulsions of those who lived on a territory in de facto
B’s possession (T) at the time the war started (t1), when those expulsions occur in
the aftermath of the war as part of a peace settlement or policy. Suppose, first, that
A did not have a legitimate claim over T and thus lacked a just cause for invading
it at t1. In this case, it seems straightforwardly true that A’s leaders may not demand
the expulsion of citizensB from T as part of the peace settlement. Suppose now,
second, that A did then and still does have a claim over T, but that its war of inva-
sion was nevertheless unjust (for example, because it was not a necessary means to
achieving A’s end of recovering T). Even if a justifiedATC settlement in this case were
to prescribe the return of T to A’s jurisdiction, it would not follow that it must also
include the forcible displacement of citizensB from T for the sake of those of citi-
zensA who might wish to settle there. For even though citizensB ex hypothesi do not
have sovereignty rights over T, they might nevertheless have occupancy rights over
the houses and land in/on which they live. Given that (as we saw in s.5.2.1)
occupancy is crucial to individuals’ prospects for a flourishing life, forcible dis-
placement must meet a particularly high justificatory threshold if it is to count as
a stipulation of a justifiedATC peace settlement: securing occupancy must be practi-
cally unfeasible, or morally undesirable on the grounds that those individuals’ basic
human rights would be violated as a result.
Note that those points also apply in those cases (illustrated by the Allies’ policy
vis-à-vis German residents in central Europe) where A’s war was overall just and
where it forcibly expels citizensB at its conclusion. CitizensB who reside in T at the
conclusion of the war might have had occupancy rights in respect of T at t1. Or
they might have acquired such rights as the war went on, particularly if the war is
long and if, therefore, they have enjoyed occupancy of T long enough to put down
social, emotional, and economic roots there. This latter case is problematic, for
it raises the spectre of moral hazard. To grant citizensB occupancy rights on the
grounds that they have resided there for a long time, when their regime culpably
failed to surrender to A at the earliest opportunity and as a result unjustly pro-
longed the war, is to reward them for that regime’s moral wrongdoing. Accordingly,
one might think that it behoves the latter and B’s war-friendly citizens to cater for
those individuals by resettling them away from T with proper assistance. However,
and to anticipate on my arguments in ch. 6 regarding post-war duties of assistance,
E. Fiddian-Quasmiyeh, G. Loescher, J. Milner, K. Long and N. Sigona (ed.), The Oxford Handbook of
Refugee and Forced Migration Studies (Oxford: Oxford University Press, 2014).
138 Restitution
if B’s regime is unwilling or unable to do so, A is under a duty to step into the
breach by allowing them to remain in T—provided that this would not jeopardize
A’s ability to deliver justice both to them and to its citizens.
One might also think that if citizensB themselves were not rightful occupiers of
T at t1 or if they would pose a threat to the peace process by remaining within T,
the settlement can provide for their expulsion without being unjust, even if A’s war
against B was itself unjust. However, on the first count, even if those individuals
lacked a right to occupy the dwellings and land in their de facto possession at t1, it
does not follow that they are liable to be expelled from the latter for the benefit of
potential settlersA at t2, when peace negotiations are being concluded. For those
settlers might not themselves have had such rights at t1. Furthermore, even if they
did have such rights, circumstances may have changed in such a way that by the
time belligerents come to the negotiating table, citizensB have in fact acquired such
rights—for example in virtue of the fact that, as a result of the ex hypothesi unjust
war which A waged over B their prospects for a flourishing life would be drastically
and durably impaired by expulsion: as we saw in s.5.2, there is a strong presump-
tive claim in favour of current, de facto occupiers.22
On the second count, even if citizensB’ remaining presence in T would jeopard-
ize the peace process, the peace settlement could not justifiably stipulate their
expulsion if the threat originates, not from those citizens themselves, but from
citizensA’ hatred of them. Whilst citizensB might be willing to leave, A’s leadership
ought not to make their leaving a condition of a settlement, for it would be unfair
to ask them to bear the burdens of citizensA’ moral dereliction. Contrastingly, if
citizensB who reside in T themselves pose a threat, for example because they refuse
peace and would rather continue to fight (unjustly) against putative settlersA, A’s
leaders would be entitled to demand their expulsion—provided that expulsion, as
opposed to (e.g.) reconciliation strategies, is the only way to block the threat which
they pose, and that they can be resettled elsewhere with appropriate material
assistance.
So much, then, for forcible displacement as an element of a peace settlement.
Suppose now that citizensB left T as the war was ongoing and are now refugees in
neighbouring countries or (in the case of a civil war) in other parts of B’s territory.
Do they have the right of return to T and to take possession of the land and houses
they vacated? They face three options: returning to their land and houses of origin,
integrating in their host community, or settling elsewhere in T. The view that a
justifiedATC peace settlement must allow those who have already left to return
implies that at the very least they had occupancy rights over the land before they
left, and that they continue to do so. When claims to continued occupancy are
based on private ownership rights, that view implies that individuals have restitu-
tive claims to exactly that house, or that piece of land, which they vacated during
the war. Furthermore, when those ownership and occupancy restitutive claims are
conjoined with demands for sovereignty, that view further implies that individuals
also have sovereignty rights over the territory on which the land and dwellings
22 See also Waldron, ‘Superseding Historic Injustice’.

Restitution 139
protected by ownership and occupancy rights are situated. Thus, Mayans who
left Guatemala during the civil war were allowed to return to their land as an
essential part of the peace agreement, and to take part in the government of their
country. And, needless to say, the right of return to land and houses which they
or their parents and grandparents left in 1947–49 and 1967 is a demand which
Palestinians are unwilling to forego and the Israelis unwilling to accept—though
in this case, unlike in the case of Guatemala, sovereignty over the land has long
been in dispute.
Having considered the return of wrongfully taken territory to the jurisdiction of
victims in the previous section, I am assuming here that those sovereignty claims
are settled. In the case of Israel and Palestine, thus, I am assuming that the
Palestinians have a jointly held right to self-determination in relation to the terri-
tory comprising Gaza, the West Bank, and East Jerusalem. With that assumption
in hand, what can we say about the right of return?
In international law, the right of return instantiates the individual right to free-
dom of movement across the borders of one’s political community, as well as the
right which any sovereign state has to allow its nationals entry into its own territory.23
Consider first the right of return as an instantiation of the right to freedom of
movement. Someone who has left his country of nationality is entitled to come
back; if he is no longer regarded a national of the country to whose territory he
wishes to return, he nevertheless has a right of return just if he used to be a national
(as Palestinians were under the British Mandate), or if that country was his habit-
ual place of residence. The law has got it right morally speaking, for four reasons.
First, the right of return protects individuals’ interests in private ownership.
Ownership rights protect our interests in using and controlling the resources of the
world. Of all those rights, the right to exclusive use is perhaps the most important.
If I have left my house but am prevented from physically returning to it, I cannot
use it. It may be objected at this point that all I need is access to a house anywhere
in the world, not to the house itself which I owned when I left. However, second,
this objection does not account for individuals’ interest in stable occupancy of the
land on which they were living by the time they left. Nor does it account for the
fact that, after all, I still am the owner of that particular house. Third, occupancy
on its own is also protected by the right of return: even if I did not own property
when I left, I may still have a very strong interest in re-enmeshing myself in the
geographical, social, and economic networks which sustained me then. Those
arguments buttress the right of return as held against a foreign regime by refugees;
they also support the right of return as held against their own regime by internally
23 For a fascinating discussion, with focus on the Palestinian refugee problem, of the right to return
and its connection with freedom of movement on the one hand, and rights inherent in nationality
(formal or habitual) on the other hand, see K. Lawand, ‘The Right to Return of Palestinians in
International Law’, International Journal of Refugee Law 8 (1996): 532–68. For an older discussion,
which predates the Oslo Peace Process, see K. R. Radley, ‘The Palestinian Refugees: The Right to
Return in International Law’, American Journal of International Law 72 (1978): 586–614. I am grate-
ful to Guy Goodwin-Gill for drawing my attention to those articles. For a book-length treatment
of repatriation in a historical perspective, see K. Long, The Point of No Return: Refugees, Rights, and
Repatriation (Oxford: Oxford University Press, 2013).
140 Restitution
displaced persons. Fourth, the right of return is also an important element in the
relationship between members of a political community, vis-à-vis outsiders.
Suppose that you and I belong to the same political community, and that we have
determined that a third party C is a fellow member. C has a right to return to the
territory over which we exercise jurisdiction. But we, together with her, have a
right against third parties that she be able to return. The point is crucial in those
cases where a set of individuals have jointly held sovereignty rights in respect of T
and where T is under military occupation, or quasi-occupation, by an actor who
(as Israel does with respect to Gaza) controls not just movements in and out of T
but also and crucially, register of births and, thereby, has oversight of who exactly
is a national in virtue of jus sanguinis.
Note that if those points are correct, displaced individuals have a (prima facie)
right of return even if they left entirely voluntarily, in just the same way as I do not
forfeit my rights over my house just by leaving it and, e.g., moving abroad, what-
ever my reasons for doing so. This is crucial, particularly in the Palestinian context,
for much of the debate on either side centres on whether Arabs who were living in
Palestine as Israeli forces took possession of their land and houses left voluntarily
or were coerced out of the newly independent Israel. Historical evidence tells us,
unambiguously, that most left under the duress of war.24 But even if they left vol-
untarily, this does not in any way establish that they no longer have a (prima facie)
right of return—any more than the claim that, e.g., Jewish American settlers who
left the US establishes that they have forfeited their right to houses and land which
they still own there. (Incidentally, and at the risk of sounding both cynical and
ungenerous, I suspect that we would hear howls of protest from American settlers
in Israel if the US government decided to expropriate them from their US-based
holdings so as to provide housing for the homeless.)
As I have just suggested, however, the right of return is a prima facie right. What
further considerations may we adduce, then, for or against it in such cases? To
begin with, although the fact that refugees left under duress does not confer on
them a right of return which they would not have had anyway, it does make a
difference to the relative costs which they, and those who wronged them, would
have to bear under a no-return or, on the contrary, a return-policy: those who left
under wrongfully imposed duress have a stronger right against wrongdoers than if
they had left voluntarily.
Moreover, the more time has lapsed between wartime dispossessions and peace
negotiations, the more likely it is that circumstances have changed such that the
properties left behind are now occupied by descendants of perpetrators, who are
24 See, e.g., Morris, The Birth of the Palestinian Refugee Problem Revisited; I. Pappé, The Forgotten
Palestinians: A History of the Palestinians in Israel (New Haven, Conn.: Yale University Press, 2011),
ch. 1. Morris and Pappé sharply differ in their interpretation of the events of 1947–49: Morris believes
that atrocities committed by Jewish then Israeli forces against Arab civilians were not part of a delib-
erate campaign of ethnic cleansing; Pappé thinks otherwise. Either way, both claim on the basis of
archival evidence that atrocities were committed and that they contributed to driving out dozens of
thousands of people from areas in which their families had lived for generations and to which most
were prevented from returning, at gunpoint and/or through expropriation. That is all I need for my
purposes in this section.
Restitution 141
themselves causally innocent of the initial wrongdoing; the harder too is it plausi-
ble to hold that, had it not been for the invasion, victims and their descendants
would still now own and occupy the land and dwellings which they left at t1. The
more time has lapsed, in other words, the weaker victims’ claims for returning to
the exact places from which they fled.
In addition, individuals who left voluntarily and who wish to return owe
something to those who stayed behind and took up the fight against unjust
aggressors, and thanks to whose efforts return is now possible. Imagine that peo-
ple who stayed behind to continue the fight against unjust aggressors and occu-
piers are made homeless as a result of the war of resistance. It may well be that
the returnees owe those fighters occupancy of their former and privately owned
homes and land.25
Finally, although the right of return is first and foremost an individual right,
account must be taken of the number of individuals who, in a given case, might
wish to exercise it, and of the likely impact on the social, economic, and political
structures of the host country. In particular, if vast population movements back
to their land of origin were to threaten host citizens’ jointly held sovereignty
rights, those citizens would have a justification for seeking to restrict the right of
return—though they would be under a moral obligation to make some concessions
in return.
Faced with such complexities, the best that one can say in abstraction from the
specific details of cases under consideration is that any decision reached by bellig-
erents must be respectful of the human rights of those subject to the settlement
(including their right to political self-determination if they do indeed have it).
With facts on the table, however, we can make a bit more progress, bearing in
mind the distinction I drew in ch. 4 between the declarative and specifying clauses
of a peace settlement. Let us illustrate by reference to the Palestinian refugee ques-
tion.26 Recall that we are assuming for the sake of argument that Israelis have both
sovereignty rights over at least some of the land historically associated with the
Jewish people, and occupancy rights in a Jewish state. We are also assuming, again
for the sake of argument, that the Palestinians also have sovereignty rights over at
least some of the land which is historically associated with their own culture, and
occupancy rights in a culturally distinct Palestinian state. Under those assump-
tions, Jewish Israelis who currently occupy land vacated in 1947 and thereafter and
who block the emergence of a self-governing Palestinian community lack the right
to be where they are. The same goes for Palestinians who, if they were allowed to
return to the land over which Israelis are sovereign, would work to destroy Israel.
25 I am grateful to an anonymous reader for Oxford University Press for raising this case.
26 To repeat, my aim is not to provide a solution to this problem but, rather, to illustrate some of
my overarching normative commitments. For recent normative discussions of the issue, see, e.g.,
Bradley, Refugee Repatriation, ch. 8; D. Enoch, ‘Whose Right Is It? Reflections on Harel’s Reflections
on Palestinians’ Interest in Return’, Theoretical Inquiries in Law 5 (2004): 367–78; C. Gans, ‘The
Palestinian Right of Return and the Justice of Zionism’, Theoretical Inquiries in Law 5 (2004): 269–304;
A. Harel, ‘Whose Home Is It? Reflections on the Palestinians’ Interest in Return’, Theoretical Inquiries
in Law 5 (2004): 333–66; Y. Peled and N. Rouhana Nadim, ‘Transitional Justice and the Right of
Return of the Palestinian Refugees’, Theoretical Inquiries in Law 5 (2004): 317–32.
142 Restitution
Importantly, however, to deny those individuals the right to return when they
seek to return to property they or their ancestors owned in 1947–49 is not tanta-
mount to denying them all ownership rights over it. All it does is deny them the
right to use it—which is compatible with honouring their rights to, e.g., sell it,
rent it out, leave it vacant, etc. If the property is no longer there, or if it has long
been inhabited by Israelis who are fully settled there and who could not be relocated
without seriously jeopardizing their prospects for a flourishing life, a justifiedATC
settlement might simply require that those Palestinians be appropriately compen-
sated for that loss.
That said, many Palestinians and Israelis are willing to accept that each side have
jointly held rights to self-determination subject to a genuine commitment to
peace. The difficulty, of course, is that contemporary Israelis who now inhabit the
land and dwellings which were vacated in 1947–49 and 1967 are fully settled
there. At the same time, although Israel bears the greatest responsibility for the
refugees’ flight and inability to return, the Palestinian leadership at the time, as
well as Arab states, are also to blame: Israel, for allowing its forces to act in such a
way either to coerce or terrify Arabs into leaving, and for preventing them from
returning in the early 1950s and expropriating them from their unavoidably vacant
properties; Arab leaders within Palestine, for relentlessly relaying accounts of
(at first isolated) atrocities and thereby inducing inhabitants to leave; Arab regimes
outside Palestine for parking them and their descendants in refugee camps without
giving them equal status. What could we say, under those assumptions, about
Palestinians’ right of return?
The following, I think. Clauses of a peace settlement (should there ever be one)
affirming a commitment to balance the interests of both contemporary Palestinians
and their Israeli counterparts whilst acknowledging both parties’ unequal respon-
sibility for the former’s predicament, would be declarative clauses. The devil is in
the details, however. And the devil might well suggest that Palestinians would have
to give up both their private ownership and occupancy rights over the land and
houses which they left in 1947–49 and 1967, and agree to resettle elsewhere, for
example in Gaza and the West Bank rather than formerly Arab villages in Galilee.
Important concessions though those would be, they would nevertheless better ena-
ble Palestinians to exercise their right to political self-determination. By that token,
Israelis would also have to give up on some of their territorial demands. At the very
least, they would have to put a stop to all ongoing and further Jewish settlements
on the West Bank and retreat within their internationally recognized pre-Six Day
War (1967) borders. There are many different concessions those parties might be
willing to offer and accept: which parcel of land to swap, under which timetable,
subject to which resettlement package, etc. To the extent that offering and accept-
ing this or that concession would offer better prospects for a justifiedATC peace, the
relevant clauses would also be declarative. But to the extent that different options
are morally equivalent, in the sense that they all offer such prospects but better
track negotiating parties’ preferences, the modalities of the resulting settlement
would be contained in specifying clauses, whose binding force would partly lie in
parties’ consent.
Restitution 143
So far, I have focused mostly on the right of return. Suppose however that indi-
viduals who have been displaced by the war do not wish to return. They would
rather stay where they have found asylum, or if not there, resettle elsewhere. Do
they have the right to so do, or may they be repatriated forcibly? This is not an
uncommon problem, especially in Western-style liberal democracies which offer
more attractive economic and social opportunities to refugees than a hazardous
return in their community of origin would afford—particularly when the war has
lasted a long time and when its refugees have therefore had time to develop strong
occupancy interests where they are. Dealing with this issue would take us too
deeply (given the scope of this book) into the ethics of immigration.27 Very
broadly, however, three points are worth making. First, international law forbids
the compulsory return of refugees to their country of origins if they would face
persecution therein. Second, however, this principle—known as the non refoule-
ment principle—is too restrictive at the bar of cosmopolitan morality, for the
latter imposes on the well-off duties to provide assistance to those whose pros-
pects for a flourishing life are impaired by dint of their circumstances. There are
many ways of discharging those obligations—resource transfers being one, and
a policy of (relatively) open borders being another. Third, conceding that such a
policy would be burdensome on the host country, the extent to which the latter
ought to open its borders to war refugees partly depends on the degree to which
it shares responsibility for that war, either in the sense that (at least some of ) its
members contributed to it, or in the sense that it failed to stop the war when it
was under a duty to do so, or both. So construed, the duty to welcome war refu-
gees (in other words, the prohibition on forced repatriation) is a reparative duty
as much as a duty of assistance.
5 . 6 . C O N C LU S I O N
In this chapter, I have delineated conditions under which belligerents, ought, or

as the case may be ought not, to return the movable and immovable property
and resources which they seized during the war; I have also shown when they
ought, or as the case may be ought not, to allow people displaced by war to return
to their territory/country of origin. Throughout, we saw that the moral status
of a belligerent as just or unjust has limited bearing on its post-war restitutive
obligations.
That said, although restitution of some kind is likely to be a key component of
most justifiedATC peace settlements, it is unlikely to be enough as far as reparations
go, for two reasons. First, the loss of territory, land, dwellings, and objects leads to
27 I review the main arguments for and against immigration, in contemporary political theory, in
C. Fabre, Justice in a Changing World (Cambridge: Polity Press, 2007), ch. 6. The best defence of the
kind of open borders policy I endorse here, is J. H. Carens, The Ethics of Immigration (Oxford: Oxford
University Press, 2013). For a strong argument against involuntary repatriation, see Long, The Point of
No Return, esp. ch. 7.
144 Restitution
further harms above and beyond the initial act of taking. Second, war destroys
what is in victims’ possession or control—and not just their property (whether
individually or jointly owned) but their lives, health, and limbs, which cannot be
restituted. The question, then, is whether those harms ought to be compensated for,
and if so how, as part of a justifiedATC settlement.
6
Reparations, Distribution, and Reconstruction
Recall the two strands of justice which pertain to what is owed to whom by whom
as a matter of right, and with respect to resource transfers, once the war is over:
Corrective justice. (1) A has taken from B a good g which was in B’s de facto pos-
session. When, if ever is A under a duty to restitute g to B? (2) A has destroyed
g. When, if ever, is A under a duty to pay reparations to B for her loss by giving
her the equivalent of g and, if B has incurred further harm as a result of A’s act,
by making good for that harm?
Distributive justice. (1) A has destroyed g, as a result of which B has incurred
further harm. A is unable or unwilling to make good on that loss. When, if ever,
is C under a duty to step into the breach and help B? (2) Individuals who were
neither party to the war nor affected by it do not have prospects for a flourishing
life. When, if ever, is C under a duty to help them, at the expense of failing to
meet B’s reparative claim?
In ch. 5, I attended to the restitutive prong of corrective justice. In this chapter,
I turn, first, to its second prong—or reparative justice—and, second, to the issue of
the just distribution of resources post bellum. Taken together, reparative and distrib-
utive principles after war provide a normative theory of reconstruction. I defend the
following claims. Strictly speaking, a just peace simpliciter is one in which parties
implement what one may call the Reparative Principle:
Reparative Principle—individuals who have violated or justifiably infringed the
human rights of some other party in or during war are under a duty to compen-
sate their victims for the wrongs which they have done to them, and to assist in
the reconstruction of their community.
The principle is modelled on the simple and intuitively plausible claim that if some
party A wrongfully takes or damages something, g, which is in B’s control, A owes
B reparations both for the loss of g and for further harms accruing to B as a result
of that loss. The Reparative Principle has four important features. First, in keeping
with the individualistic premises which underpin my account of war, the rights,
duties and liabilities which it stipulates are held by and imposed on individuals.
Second, it relies on the possibility of making counterfactual judgements of the
kind ‘had A not invaded B, this particular individual member of B would still have
his farm, here and now’. Third, the principle mandates the disbursement of material
146 Reparations, Distribution, and Reconstruction
resources. Fourth, under conditions of scarcity, it conflicts with non-reparative,

purely distributive demands (as distinct from the distributive demands outlined in
the first part of Distributive Justice).
As we shall see in s.6.2, those four features of the principle—its individualistic
focus, its reliance on counterfactuals, the financial costs it imposes, and its conflict
with distributive justice—together support the charge that implementing it in full
in the aftermath of war is simply not feasible, morally and practically speaking.
Accordingly, we should partly replace it with what one may call a Reconstruction
Principle, which I defend in s.6.3:
Reconstruction Principle—all agents who have been harmed by war are owed
help with the reconstruction of their community, consistent with respecting the
rights of assistance of individuals who were not affected by the war.
The Reconstruction Principle offers a feasible alternative to the Reparative Principle
and delineates non-reparative obligations owed both to war participants and to
outsiders. Furthermore, endorsing the Reconstruction Principle does not entail
jettisoning all talk of reparations—on the contrary. Having fine-tuned both prin-
ciples in s.6.3, I conclude that a justifiedATC peace—as distinct from a just peace
simpliciter– requires implementing the Reconstruction Principle and those ele-
ments of the Reparative Principle which survive the obstacles raised by feasibility
constrains.
Some caveats before I begin. First, by reparations I mean the disbursement of
material resources to make up, as far as possible, for the wrong one has committed.
I leave the issue of punitive sanctions until ch. 7, and the issue of non-material
reparations such as public apologies until ch. 9. Second, in the literature on justice,
what I call reparative justice is often referred to as compensatory justice. But the
word ‘reparations’ is more often used in the context of war or mass-scale injustice:
accordingly, I will follow extant practice with respect to the noun, but use the verb
‘to compensate’ interchangeably with ‘to pay reparations’. Third, classifying the
second question above as pertaining to distributive justice—as opposed to repara-
tive justice under conditions of non-compliance—might seem odd, since distribu-
tive justice (some might hold) only remedies non-anthropogenic needs or
inequalities. Yet, defences of the welfare state, whose remit is both those who are
made wrongfully needy and those who are made needy through natural causes, fall
within the remit of distributive justice. I see no reason therefore not to conceive of
assistance which is owed by bystanders to war victims as falling within the remit of
distributive justice.
6 . 2 . R E PA R AT I O N S
Suppose that some party A takes or damages something, g, which is in B’s control,
and is not able to restitute it—or at least, not in the condition in which it was
when A took it: food he has consumed, property which he has destroyed, B’s body,
her ability to work, her life. Or suppose that A still has g but that the loss of g
Reparations, Distribution, and Reconstruction 147
harms B in ways that merely restituting g will not make good. What, if anything,
does A owe to B or her estate? In restitution cases, as we saw, if A does not have
rights over g at t2 and if g still fully belongs to B, it is incoherent to deny B a right
to have g returned to her. But in reparations cases, g no longer exists, or is no longer
intact, or its return will not suffice, and so we cannot simply advert to B’s contin-
uing ownership of g to justify A’s duty to compensate her. In ss.6.2.1 and 6.2.2, I
take for granted the rough and ready intuition that someone who has been wrong-
fully harmed is owed reparations, and examine, first, when reparations are due and
to what level, and, second, who is owed them. In s.6.2.3, I defend the intuition by
showing why wrongdoers are under reparative duties. In the course of defending
the Reparative Principle as a requirement of a just peace, I show that feasibility
constraints mandate rescinding from implementing such a peace, and I articulate
what a justifiedATC peace both permits and demands.
6.2.1. Reparations for What, and How Much?

Let me start with the question of for what and to what level. Historically, war rep-
arations have been paid for a wide range of wrongful harms suffered by victims,
both in their individual and collective capacity. To give but a few examples:1 offer-
ing grants in cash or kind to individuals who lost their livelihood; compensating
victims of sexual violence for the dignitary harms they suffered; compensating the
wounded and the dependents of combatants who died in war; compensating a bel-
ligerent for having to pay its invalid combatants’ pensions and rehabilitation pro-
grammes; compensating a belligerent for the costs of having had to fight the war;
financing the cleaning up of battlefields; paying for the rebuilding of bridges,
roads, factories, hospitals, and schools; and so on. Peace settlements which set out
the precise modalities of reparations payments specify independently justified
reparative principles. The question, then, is what are general, independently justi-
fied principles for deciding, broadly, what is owed to claimants, bearing in mind
that (as I shall argue below) a peace settlement is substantively justifiedATC only if
it does not prevent its signatories from meeting their basic obligations of justice to
non-parties.
At first sight, one might think that claimants have rights to the exact equivalent of
what they lost, as a means either to restore them to the status quo ante or to bring
them to the level at which they would be had the wrongdoing not occurred.2 But
1 These are drawn from, inter alia, the Treaty of Versailles of 1919; the work of the UN
Compensation Commission which was set up after the First Gulf War to help victims of Iraq’s inva-
sion of Kuwait in 1990; and the 1952 Luxembourg Agreement between Germany and Israel whereby
the former paid extensive reparations to the latter as a way to compensate Jews from the harms suffered
under Nazism. For comprehensive discussions of various reparations programmes, see P. De Greiff
(ed.), The Handbook of Reparations (Oxford: Oxford University Press, 2008); Barkan, The Guilt of
Nations. Disagreements in 1919 between the US, France, and the UK, on which kinds of damage call
for reparations are illuminatingly described in Macmillan, Peacemakers, ch. 15.
2 For the view that reparative justice demands that, if possible, claimants be given the exact equiv-
alent to what they lost, see R. E. Goodin, ‘Theories of Compensation’, Oxford Journal of Legal Studies
9 (1989): 56–75.
even in simple cases, and setting aside non-identity cases for now, it will often not be
possible fully to compensate an agent for the wrongful harm she suffered as a result
of some other party’s act—if only because some wrongful harms, such as the com-
plete loss of one’s sexual integrity, of one’s ability to work, of cherished heirlooms, let
alone of one’s life, can never be fully made good. In post-war context, this difficulty
is compounded by (a) the need to satisfy vast numbers of claimants; (b) the fact that,
if one goes for the status quo ante view, the war in its inherent destructiveness often
makes it difficult for people to prove, e.g., that they did own a business which did
collapse in the aftermath of an invasion; (c) the fact that, if one goes for the counter-
factual view, it will be very difficult to show, in many cases, that claimants would be
better off had it not been for the wrongdoings which they suffered during the war;
(d) the fact that reparative demands are likely to compete with the imperative to help
those whose prospects for a flourishing life have been impaired by non-anthropo-
genic factors, or by non-war related anthropogenic factors.3
In s.1.5.1, I distinguished between different kinds of feasibility constraints: hard
v. soft, directly moral v. morally inflected practical constraints. In this language,
the fact that some wrongful harms cannot be made good is a hard constraint, for
paying reparations simply does not make good for some agent the loss of his life,
or of her sexual integrity. It is also a moral constraint, not in the familiar sense that
we would lose something of moral and greater value by implementing P, but in the
sense that the impossibility of implementing P lies, in this case, in incommensura-
bility of values (the value of money v. the value of life.) By contrast, epistemic
constraints on our ability to determine the extent to which agents were harmed by
the war are morally inflected practical constraints. If we could establish that this or
that claimant would be better off than he is now were it not for the war, we would
be in a position to give him reparations. But given that we cannot establish it, we
have to decide whether to withhold reparations or, on the contrary, whether to
provide him with resources to which he might not, in fact, be eligible. Either way,
there are moral costs to be paid, and those epistemic constraints thus are properly
regarded as morally inflected practical constraints.
Considerations of numbers of war victims on the one hand, and numbers of
needy people on the other hand, are also morally inflected practical constraints.
Given vast numbers of claimants, it often will not be possible to give all of them
what they are owed in integrum, which gives rise to the moral costs of either giving
to some in full to the detriment of others, or to give to all less than what they are
entitled to. Moreover, given vast numbers of individuals who suffer from desperate
poverty and yet are not members of belligerent communities, the latter will not be
able to meet both their reparative obligations to their erstwhile enemies and their
distributive obligations to outsiders.
3 For discussion of some of those difficulties, see, esp., P. De Greiff, ‘Justice and Reparations’, in P.
De Greiff (ed.), The Handbook of Reparations (Oxford: Oxford University Press, 2008); Kutz, ‘Justice
in Reparations’. I do not need to take a stand here on the status quo ante and the counterfactual views;
it is enough for my purposes to show that on either view, the Reparative Principle should be (part)
abandonned. Still, for a good review of discussions of counterfactuals in compensatory justice, see
Butt, Rectifying International Injustice, 102–17.
In the light of those difficulties, a justified peace settlement all things considered
does not insist on full reparations for each war victim. While this might issue in
some of those victims not being given what they are entitled to prior to taking
other people’s claims into account, to meet their claim in full would count as a
reparative injustice to other war victims and a distributive injustice to non-war
needy individuals. We will revisit this issue in s.6.3.
6.2.2. Reparations to Whom?

So who, then, has reparative rights once the war is over? Victorious belligerents do
not have those rights qua victors, since their war might have been unjust. Moreover,
not all individual members of just belligerent communities have such claims either,
since they themselves, in their individual capacity, might have been harmed in the
course of committing wrongful war acts. Post-war reparative rights track wrongful
harms—whether claimants belong to the victorious side, the vanquished side, the
unjust side, or the just side. Consider for example combatants who were fighting
for the Rwandan Patriotic Front in 1994 as extremist Hutus were carrying out a
genocide against Tutsis and moderate Hutus. Many of those combatants were
wounded by the Rwandan Army as they were fighting the latter’s genocidal forces.
Others, however, were wounded as they were committing atrocities against Hutu
civilians. In so far as they were acting wrongfully, they do not have reparative
claims in relation to those wounds—whereas their victims do have such claims in
relation to their own predicament. In this respect, restitutive claims interestingly
differ from reparative claims: for whereas (as we saw in ch. 5) a party may be under
a duty at t1 to restitute a good which they rightfully seized at t2, a legitimate
reparative claim necessarily rests on the commission of a wrongful harm.
That being said, the notion of wrongful harm needs disambiguating. As I noted
in s.1.3, we must distinguish between violating a right and justifiably infringing it.
A violates B’s right not to suffer a loss if B has not done anything to warrant losing
that right and if A lacks a moral justification for imposing that loss on B. Sometimes,
however, A may have such a justification (typically, his life is at stake) even though
B has not done anything to lose his right. Most obviously, individuals whose rights
are being violated have reparative rights. But so do individuals whose rights have
been justifiably infringed. Suppose that some agent A takes from B a good, g,
which is in B’s prima facie rightful possession, but which A needs in order to sur-
vive. In Joel Feinberg’s well-known example, A is a stranded mountain-hiker who
will die unless he finds shelter very soon. He chances upon a cabin, owned by B
and currently vacant, breaks in, eats the food left by B and burns one of the wooden
chairs to warm himself up. Feinberg holds—as does, e.g., Judith Thomson—that
A’s dire predicament furnishes him with a justification for acting as he does, but
that B retains ownership rights over both the cabin and its contents, such that A
owes her reparation for the food he ate and the wood he burned.4
4 See J. Feinberg, ‘Voluntary Euthanasia and the Inalienable Right to Life’, Philosophy & Public
Affairs 7 (1978): 93–123. For discussions of this particular example, see, e.g., P. Montague, ‘Rights
In this particular case, I disagree with Feinberg and Thomson. At the bar of the
ethics of assistance which is part and parcel of my account of cosmopolitan justice,
B is under a duty to A make her cabin available to him and A acts within his rights
by consuming her resources. But consider the following variant of the case, whereby
B is on the scene and is not physically able to open the cabin to A. The only way
for A to get into the cabin is to trample over B, as a result of which she would
become permanently paralyzed from the neck down. In this particular case, in so
far as A ‘merely’ tramples over her and thus is not using her body as a means to get
in, and as he would die if he does not get in, he has a justification for so acting; at
the same time, the no-undue-sacrifice proviso yields the conclusion that B is not
under a duty to let A so harm her and that she thus retains her right that he not
harm her. In so far as she does retain her right against harm, she has a reparative
right against A.
Let us apply this to war cases. Assuming here again that the rough and ready
intuition is correct, victims of rights-violations, who thus have rights to be com-
pensated for the resulting harms, include: combatants and non-combatants who
are deliberately killed or maimed without justification—for example in prosecu-
tion of a war of unjust aggression, or in a war which has a just cause but which
nevertheless breaches some other requirement; individuals (usually non-combat-
ants) who are killed unintentionally, though foreseeably, but without justification;
individuals, whether combatants or not, whose houses, land, livelihood, and col-
lectively owned property have been deliberately and unjustifiably destroyed.
Victims of justified rights infringements, for their part, include: those who suffer
grievous harm, to which they are not liable, as a result of the deliberate killing or
maiming of liable agents (for example, the dependent children of combatants who
fight for an unjust cause); those who suffer grievous harm, to which they are not
liable, as a result of the deliberate destruction of privately and collectively owned
property and infrastructure; those whose life, limbs, and (collective and private)
property are destroyed as justified collateral damage; those whose property, both
privately and collectively owned, is justifiably taken by the enemy in prosecution
of its just war or a justified occupation.
The case of military occupation is particularly interesting. As we saw in ch. 3, a
just belligerent, A, which takes position in B’s territory, may have a powerful justi-
fication for requisitioning and/or exploiting resources which are in citizensB’ pos-
session—food, clothing, dwelling, natural resources, agricultural land, and so on.
Even if some members within B are significantly causally responsible for providing
A with a just cause for war, not all are. Moreover, among those who are not respon-
sible for A’s predicament (for example, because they could in no way influence
their regime’s bellicose policy vis-à-vis A), there are many who are not under a duty
and Duties of Compensation’, Philosophy & Public Affairs 13 (1984): 79–88; J. J. Thomson, ‘Rights
and Compensation’, Noûs 14 (1980): 3–15; M. J. Zimmerman, ‘Rights, Compensation, and
Culpability’, Law and Philosophy 13 (1994): 419–50; N. Davis, ‘Rights, Permission, and
Compensation’, Philosophy & Public Affairs 14 (1985): 374–84. Thomson defends a Feinbergean line,
while Montague and Zimmerman argue, against Feinberg, that the hiker does not owe compensation
to the cabin owner.
to assist A’s troops in their just war against their own regime and unjust compatri-
ots—notably because they would jeopardize their own prospects for a flourishing
life by providing such assistance. By implication, they retain their rights not to be
so harmed, and thus not only have restitutive claims to their land, houses, and so
on (under the conditions set out in ch. 5): they also have claims to share in some
of the benefits which A’s troops derive as a result of extracting wealth from B’s
collectively held resources, as well as reparative claims for the harms they suffered
as a result of A’s requisitions.
Some loose ends need tying. First, I stated above that the dependents of those
who are justifiably and deliberately killed or maimed can have rights to be com-
pensated for the harms which they themselves suffered as a result. The point applies
to descendants of war victims even if they owe their existence to the war itself (as
we saw in s.5.2.2), so long as they can trace the harms from which they suffer to
continuing rights violations or justified rights infringements committed during the
war after they were conceived, and subject to the usual caveat pertaining to the
passing of time, the number of claimants, etc.
Sometimes, however, there are no descendants left, or at least no easily identifi-
able ones—as when whole families are killed in war-related genocides or in system-
atic and destructive bombing. There are two ways to deal with this particular case,
which incidentally was at the heart of reparations negotiations conducted by
(West) Germany, Israel, and representatives of the Jewish diaspora in the early
1950s.5 One can either hold that no one has a reparative right to this particular
package, which can then be used by the duty-holder for non-reparative purposes,
including distributive justice purposes; or one can hold that other victims of this
act of war have a claim to it. On a strict construal of the demands of reparative
justice, only the direct victim of a given wrongdoing and her heirs are owed repa-
ration for this wrongdoing. Suppose, then, that given available resources, a peace
settlement allocates a certain sum of money (say $5000) to the surviving heirs of
those who were wrongfully killed by A in bombing campaigns, and that 5000
people have died heir-less. On that strict construal, those $25 million dollars are
not to be used for reparative purposes, even to help surviving victims of that bomb-
ing campaign (who thus incurred a harm lesser than death and would have been
entitled to less than $5000 when all the dead’s heirs are taken into account).
However, reparative justice really is a genuine moral imperative, and thus dictates
that we confer on those other victims a right to increased payments until the
demands of reparative justice as pertains to this particular act of war have been
satisfied—subject to meeting the competing demands of distributive justice.
Second, victims of a rights-violating regime who benefit overall from third par-
ties’ decision to wage a war of humanitarian intervention do not have reparative
rights against interveners with respect to the justified harms which the war may
nevertheless have caused them (though they do have claims against their erstwhile
regime both for the pre-war harms it caused them and for the in-war harms that its
5 See, e.g., Barkan, The Guilt of Nations, ch. 1; Colonomos and Armstrong, ‘German Reparations
to the Jews after World War II’.
troops also imposed on them by unjustly fighting the interveners.) As Frances

Kamm notes, the claim on the one hand draws strength from the view that bene-
ficiaries of an intervention are sometimes under a duty to share some of the costs
of the latter, and on the other hand is subject to the proviso that interveners did
not violate their rights in the course of fighting that war.6
As should be clear, given the multiplicity of victims and of the harms which they
incur for which reparations are warranted at the bar of the Reparative Principle,
the epistemic hurdles which any institutional scheme for implementing the prin-
ciple face are so high as to be ‘unclearable’. For not only would one need to know
whether the war which claimants’ side fought was just in toto: one would also need
to know whether the specific campaign within the war during which they were
harmed was just, and whether they themselves committed war wrongdoings which
may well cancel out their reparative demands. I mentioned earlier the case of com-
batants who were wounded by the Rwandan Army as they themselves committed
atrocities against Hutus civilians, and averred that they do not have reparative
claims—whereas their victims did. The problem is that the victims of those atroci-
ties might themselves have been génocidaires, and might have violated the rights of
those combatants or their families. Who, then, if anyone, has reparative claims in
such cases?
Those obstacles are not purely practical. To be sure, if we were ever in a position
to know relevant facts about claimants which would support or on the contrary
undermine their claims for reparations, we would be able to implement the
Reparative Principle. But given that we do not know those facts and are unlikely
to be able to acquire such knowledge, we have to decide whether to withhold rep-
arations altogether, at the likely cost of not helping individuals who in fact have a
claim to reparations, or to replace the Reparative Principle with another principle,
at the likely cost of giving reparations to individuals who in fact do not have a
claim to them, and in so doing withdrawing from the reparative pot resources
which could and should go to genuine claimants. The decision we have to make—
of which moral cost we should bear—is of course an eminently moral one. It is also
eminently complex. To answer it, however, we need to have a clearer picture of
who owes reparations, and why.
6.2.3. Who Should Pay and Why?

A First Cut: The Fault Principle
Assume, then, that we can identify who has a legitimate reparative claim. I asked
earlier why, if reparation is owed to B for harm she wrongfully incurred at the
hands of A, it should be paid by A. At the heart of reparative justice is the thought
that, as Shelly Kagan puts it, other things roughly equal there should be an ‘inter-
nal linkage’ between the claim-holder and the duty-holder.7 Accordingly, even if
6 Kamm, The Moral Target, 158–63. I defend that view in CW, s.5.5.
7 See S. Kagan, ‘Causation and Responsibility’, American Philosophical Quarterly 25 (1988): 293–
302. The clause ‘other things roughly equal’ denotes that if someone other than A is better able to
bystander C could and would help B, it is up to A to do so. The question, though,

is why, and the most straightforward answer to that question is that B’s predica-
ment is A’s fault. When violating or justifiably infringing a right of B’s, A upsets the
moral parity that used to obtain between them, whereby they both respected each
other’s rights. Note, furthermore, that the fault principle also covers cases where A
fails to fulfil his duty of assistance to B. If A is under a duty to ensure that B is well
fed or is taken to hospital as a matter of urgency, and if as a result of A’s dereliction
B’s health deteriorates, B’s predicament is in one important sense A’s fault. By
allowing B to be harmed in violation or justified infringement of his right to assist-
ance, A has upset moral parity between them in this case as well. Given that indi-
viduals are under a general prima facie obligation to treat one another as morally
on a par, when they fail so to act, they are under a duty to do as much as is reason-
able to restore moral parity between themselves and their victims. To the extent
that reparation is one way to restore parity, they are under a duty to pay it.
One may wonder what distinguishes rights-violators from justified rights-in-
fringers, since they both owe reparation to their victims. Other things roughly
equal, violating rights is morally worse than justifiably infringing them—a fact for
which a plausible theory of reparative justice must account. One way to do so is to
vary the amount and kind of sacrifices which they may be expected to incur when
fulfilling those duties.
Thus, on Kagan’s account of reparations, which I endorse, reparations are not
justified by appeal to the forward-looking imperative of bringing about reconcilia-
tion between wrongdoers and victims. Even if reparations do as a matter of fact
serve that end, what justifies them is the fact that B has been wronged in some way.
To hold, on the contrary, that A owes B reparations on the grounds that they must
reconcile and that reparations will help achieve that, is to fail to give proper atten-
tion to what was done to B. It is a bit like arguing that A ought not to kill B on the
grounds that killing her will threaten peace overall—whereas the reason he ought
not to kill her is that she has done nothing to warrant it.8
As applied to war reparations, unjust belligerent A, whose troops and leadership
violate the human rights of citizensB by waging war, is under prima facie reparative
duties towards its victims—where the latter include, not just those individuals
whose rights have been directly violated by A’s own troops, but also those individ-
uals whose rights have been justifiably infringed by B’s troops as a result of A’s own
wrongdoings. The latter might include citizensA who are morally and causally
innocent of the war and who are grievously harmed as a collateral damage of B’s
justified war. Similar considerations apply, mutatis mutandis, to a wrongful failure
make good on B’s loss, then (under certain conditions to be set out in s.6.3), that person should step
into the breach. On compensation and the restoration of moral parity, see also G. F. Gaus, ‘Does
Compensation Restore Equality?’, in J. W. Chapman (ed.), Compensatory Justice (New York: New York
8 For the reconciliation-based justification of reparations, see J. Thompson, Taking Responsibility for
the Past: Reparation and Historical Justice (Cambridge: Polity, 2002). More broadly, for a forward-look-
ing justification for reparations, see L. Wenar, ‘Reparations for the Future’, Journal of Social Philosophy
37 (2006): 396–405. For the objection I articulate here, see R. Kumar, ‘Why Reparations?’, in J.
Oberdiek (ed.), Philosophical Foundations of the Law of Torts (Oxford: Oxford University Press, 2014).
to wage a war of humanitarian intervention. Suppose that B’s regime is guilty of

serious rights violations against its own population, so much so that a war of inter-
vention would be mandatory. Suppose further that A, C, and D together are under
a duty to intervene, but that C and D are derelict: A goes in alone, and in the
course of that war is responsible for greater rights infringements vis-à-vis B’s
civilians than would have been the case had C and D done their duty. Whilst it is
true that A’s justification for infringing those rights lies in B’s culpable rights viola-
tions against its own citizens and that (as I argued above), B is the primary repara-
tive duty holder, C and D too are under reparative duties towards the victims of
their own dereliction of duty.9
But now the following two issues arise. First, we need to distinguish between
three cases:
(a) A wrongfully exposes its fellow citizens to the risk of being justifiably killed
by B’s troops as a matter of right. For example, combatantsA mount an unjust
invasion of B’s territory, under serious duress from A’s leaders. Those combat-
ants are liable to be killed by combatantsB.
(b) A wrongfully exposes its fellow citizens to having their rights violated by B’s
troops. For example, A invades B’s territory without just cause, and com-
batantsB defend themselves and their compatriots with pointless or dis-
proportionate force against combatantsA and deliberate target innocent
non-combatantsA.
(c) A wrongfully exposes its fellow citizens to having their rights justifiably
infringed by B’s troops. For example, A unwarrantedly invades B without
just cause, and combatantsB defend themselves and their compatriots with
proportionate force involving both justified deliberate harm on combatantsA
and justified collateral damage on non-combatantsA.
The first issue is that of who owes reparations to whom. In the first case, B clearly
does not owe reparations to combatantsA (or their estates), since it has not acted
wrongly by harming or killing them. In the second case, B no less clearly owes
reparations to civiliansA. Whether it owes reparations to combatantsA depends on
whether the latter have made themselves liable to be harmed merely by unjustifia-
bly attacking B, or whether they are liable only to proportionate and instrumen-
tally efficient harm. I am inclined to accept that they are liable only to proportionate
harm and that B therefore owes them (or their estate) reparations for dispropor-
tionately harming them—thus in violation of their rights. But I am not inclined to
accept that they are liable only to instrumentally efficient force. If so, B does not
owe them reparations—though it does owe reparations to associates of those com-
batants who are disproportionately harmed as a result.10 This is not a needlessly
9 The issue of interveners’ reparative obligations in the face of their counterparts’ culpable fail-
ure to intervene is raised, in somewhat different terms, by M. Walzer, ‘The Aftermath of War’, in E.
Patterson (ed.), Ethics Beyond War’s End (Cambridge: Cambridge University Press, 2012), 39–40.
10 On some views, notably McMahan’s, a lethal attacker is liable to be killed only if killing him is
instrumental to blocking the harm which his attack threatens to inflict and to which killing is a pro-
portionate response. On that view, instrumentality is internal to liability. Suppose that Barbara, whom
pedantic point: the question is whether France owed Germany reparations in 1919
for having pointlessly killed scores of thousands of German soldiers in trench war-
fare. On my account, it did not owe reparations to those soldiers themselves (via
their estate), but it did owe reparations to German civilians who incurred dispro-
portionate harms as a result of those deaths.
In the third case, B have justifiably infringed the rights of innocent civiliansA,
which imposes reparative duties on its members. However, B’s responsibility for
innocent civiliansA’ predicament is shared with A’s leadership. Those members of A
who are themselves responsible for A’s decision unjustly to go to war also owe rep-
arations to those of their compatriots whose rights were (justifiably) infringed by
B—in fact, they ought to shoulder the lion’s share of the reparative bill since their
moral responsibility for those rights infringements is greater than B’s.
The second and related issue is that of conflicting obligations under conditions
of moderate scarcity. Suppose that citizensA are not in a position to meet both their
reparative obligations to those of citizensB whose rights they violated, and their
reparative obligations to those of citizensA whom their decisions unjustly to go to
war exposed to B’s justified right infringement. To whom may or ought they to
give priority, other things roughly equal (such as the number of victims with
reparative claims, the magnitude of the help they require, etc.)? We ought to reject
the view that they may, let alone should, give priority to their compatriots on the
grounds that they have a special duty of care to them which they do not have to
citizensB: at the bar of cosmopolitan morality, they do not, in fact, have such a
duty. Moreover, there is a stronger causal link between A’s acts of war and citizens’B
predicament than between its acts of war and its own citizens’ predicament, since
the latter is due, more proximally, to B’s (ex hypothesi justified) decision to retaliate.
In the ethics of defence which underpins my account of killing and harming in
war, the more significant one’s participation in an unjust war or phase of war, and
thus the stronger the causal link between one’s acts and the predicament of one’s
victims, the greater one’s degree of liability to defensive force at the hands of, or for
the sake of, those victims. By parity of reasoning, the stronger the causal link
Andrew culpably and lethally wounds in the course of raping her, somehow manages to pose a lethal
threat to him though this will not enable her to survive. According to the internalist, she violates his
right not to be killed by so acting. Should she somehow miraculously survive, she would owe repara-
tions to his estate (assuming that, as I argue in this chapter, descendants themselves are owed repara-
tions under some circumstances), even though he is morally culpable for her predicament. Many find
that implausible. I do too. Setting aside this worry about reparations, there are other good reasons
anyway to reject internalism. For recall that, if internalism is correct, an attacker is liable to be killed
defensively only if killing him is instrumental to saving the lives of those he is attacking. By implica-
tion, the more defenceless his victims, the more robust his own defensive rights are, since the more
defenceless they are the less likely they are to succeed in blocking his threat. Surely, though, the more
defenceless one’s victims, the more grievous one’s wrongdoing. The internalist thus holds that (by
implication) the more grievous one’s wrongdoing (on the defencelessness scale), the more robust one’s
defensive rights. I find that deeply puzzling, to the point of incoherence. If I am right that we ought
to reject internalism, combatantsB do not wrong combatantsA for inefficiently killing them and thus
do not owe them reparations. (For McMahan’s view, see McMahan, Killing in War, 9. For other argu-
ments against internalism, see J. Firth and J. Quong, ‘Necessity, Moral Liability, and Defensive Harm’,
Law and Philosophy 31 (2012): 673–701; H. Frowe, Defensive Killing (Oxford University Press:
Oxford, 2014), ch. 4.)
between one’s acts and the predicaments of one’s victims, the greater one’s degree
of liability to reparative burdens for their sake. Accordingly, A should give priority
to citizensB—though, as we saw above, B itself also has some (weaker) reparative
obligations to those of citizensA whose rights it justifiably infringed.
Suppose now that A, which violated or justifiably infringed B’s rights, has to
choose between discharging its reparative obligations to B or its distributive obli-
gations to a non-war needy person C. In so far as A can have distributive obliga-
tions to C only if the latter is below the sufficiency threshold, the dilemma arises
in the following two scenarios: (1) both B and C are below the sufficiency thresh-
old, or (2) a right of B’s has been violated or justifiably infringed by A but B is still
above the threshold while C is below the threshold. In both cases, B and C have
prima facie rights against A. The question then is whose rights A should respect
given that he cannot give satisfaction to both.
Consider the first scenario. If other things are roughly equal—such that C is not
considerably needier than B or that A is not better equipped to help C than to help
B—A ought to give priority to B. For although B and C are on par, the fact that A
is responsible for the former’s predicament while innocent of the latter’s is a mor-
ally relevant basis for differentiating between them: restoring moral parity is not
always decisive, as we saw above, but it sometimes is. If other things are not equal,
considerations such as the degree of need and A’s effectiveness might well tip the
balance in C’s favour. As should also be clear, in adjudicating between the demands
of reparative justice and the demands of distributive justice, A may not appeal to
the fact that B is a compatriot while C is not—since in so doing he would breach
the cosmopolitan requirement that individuals’ basic entitlements should not
depend on their membership in a given political community.
Consider the second scenario. Suppose that A takes wood from B’s cabin while
on a walk not because he wants to avoid freezing to death (he has suitable winter
clothes) but because he fancies a spot of woodcarving; B will not fall below the
sufficiency threshold as a result (she can easily afford to buy more wood). Suppose
further that A chances upon D who is at risk of so freezing. A is under a duty to
give the wood to D instead of returning it to B, on the aforementioned grounds
that in the face of someone’s needs (as opposed to another’s understandable prefer-
ence), B’s interest in recovering what was rightfully hers and restoring moral parity
is not decisive. The same considerations apply if A has for some reason wrongfully
destroyed the wood but can help D by giving him money which would otherwise
have been used to compensate B for the loss of her property. Moreover, B herself is
under a duty to make wood and/or appropriate resources available to D. In provid-
ing those resources to D instead of discharging his prima facie reparative obligation
to B, A is in effect discharging B’s distributive obligation to D on her behalf.
Five Complications: Refining the Fault Principle

Let us take stock. I have argued that if A wrongs B, he has a prima facie obligation
to make reparations to her, unless he can use the required resources to satisfy dis-
tributive obligations to someone in greater need.
However, this simple allocation of reparative burdens needs considerable refine-

ment when we apply it to war, in the light of five complications. First, war is
fought by groups whose individual members have differential responsibility for
rights violations or justified rights infringements committed in resorting to, and
fighting in, that war. The most grievous harms (death, sexual violence, destruction
of property) are directly inflicted by the combatants who fight, not by the generals
and civilian leaders who direct operations from headquarters. At the same time,
many of those war wrongdoings are carried out upon commanding officers’
orders—not least an unjust invasion, an orchestrated campaign of gang rapes as
were rife in, e.g., Bosnia, or a decision deliberately to target civilian centres.
Second, individual wrongdoers might not be identifiable. Thus, a victim of mul-
tiple rapes at the hands of enemy combatants will often not be able to recognize, let
alone name, her attackers—nor will those who witnessed the attack. Moreover, even
if wrongdoers are identifiable (as evidenced by war-crime trials), they might not be
in a position to shoulder reparative burdens, most straightforwardly because they
have already died by the time the peace settlement is negotiated, or because they
simply lack the resources each individually to pay in full.
Third, even if individual wrongdoers are identifiable, their wrongdoing consists
in taking part in a large-scale, collective wrongdoing. Thus, one might be in a posi-
tion to identify, thanks to regimental records, the Serb combatants who wrongfully
took up arms against the newly independent Republic of Bosnia-Herzegovina in
1992; but the contribution of each of those individual combatants to the war in toto
against Bosnia (as distinct from separate war crimes such as the deliberate shooting
of civilians which they may have committed) is so marginal when taken on its own
that one may wonder whether they are liable each to pay reparation to the victims
of the invasion, particularly when doing so would be very burdensome to them.
Fourth, combatants whose war is just in toto often commit serious rights viola-
tions in prosecution of that war, and their enemy, by implication, do not act
wrongly by defending themselves or third parties in those phases of the war.
Relatedly, and as we saw in ch. 2, a belligerent might be under a duty to end at t2
a war which it started justly at t1. Were it to continue with the war, it would act
unjustly towards its previously unjust enemy.
Fifth, not all culpable parties in a war are themselves declared belligerents in that
war, and/or negotiating parties in the peace settlement. To give but one example,
the 1954 coup which removed democratically elected Guatemalan President Jacobo
Arbenz and set up a military dictatorship was backed by the CIA with President
Eisenhower’s permission, with a view to protecting the interest of US company
United Fruit from Arbenz’s policy of land redistribution. The United States con-
tinued to support the dictatorship until the 1980s.11
How, then, can we best deal with those complications? For a start, the fault princi-
ple should be construed so as to apply to all agents who have either violated or justifi-
ably infringed human rights during war, whether or not they are parties to the peace
11 D. Farah, ‘Papers Show U.S. Role in Guatemalan Abuses’, Washington Post. http://www.
washingtonpost.com/wp-srv/inatl/daily/march99/guatemala11.htm (accessed 7/02/2014).
process. Thus, the United States, who were not signatories to the 1996 Peace Agreement
in Guatemala, were under a moral duty to compensate ordinary Guatemalans, most of
whom were Mayans, whose lives had been destroyed by the civil war.12 Likewise, pri-
vate corporations which are complicitous in war crimes are under reparative obligations
to their victims—such as, notoriously, German firms which employed slave labourers
during the Second World War and which entered a reparation agreement with repre-
sentatives (state and non-state) of those victims in 2001.13
In addition, when belligerents on all sides are responsible for grievous rights
violations in the course of the war, it would not make sense for them to allocate to
one another reparative responsibility for victims of their wrongdoings (A pays for
A’s victims, B pays for B’s victims). The best way for each to discharge their repara-
tive responsibility is for each to look after its own, until those violations have
roughly cancelled each other out. If one party has committed a greater share of
such wrongdoings than the other, then it owes further reparation to the latter. This
is the best way to deal with the fact that the French and the Germans owed each
other reparations in 1919, though the latter to a greater degree than the former.
Furthermore, we should conceive of the fault principle in collective-participa-
tory terms, bearing in mind that (as I noted in s.1.3) the right to wage and kill in
war can be held by agents qua members of a collective each of whom participates
in the latter’s wrongful venture. This helps solve the difficulties arising from com-
batants’ inability on their own to compensate their victims for their individual
grievous wrongdoings or for their marginal participation to collective wrongdo-
ings. All perpetrators of a gang rape are liable; all those on whose behalf the unjust
war was fought and who bear some degree of causal responsibility for it can be held
under the relevant compensatory burdens, and so on.
Those qualifications to the fault principle are important, but they will only take
us so far. For a start, if fault remains at the heart of reparative justice, we still need
to know who was responsible for which (part of the) wrongdoing—a task which
remains in most cases insuperable. Moreover, many perpetrators of war crimes die
before the settlement is reached. Finally, the scale of war-caused destruction is
usually so enormous that reparations programmes typically need contributions
from individuals who, though alive and of age when the war occurred, nevertheless
are completely innocent of it; those programmes also need implementing over
several years, indeed a couple of decades, and must be paid for by those who were
not of age during the war. We thus need to establish who, if anyone, should pay in
lieu of the dead wrongdoers, and why, if at all, current generations may be held
liable to the reparative burdens which arise from their predecessors’ wrongdoings.
A Second Cut: The No-Unjust-Benefit Principle

To make sense of the intuition that, up to a point at least, the causally innocent
may be held under reparative burdens, we need to supplement the fault principle
12 On this point, see also May, After War Ends, 192.
13 J. Authers, ‘Making Good Again: German Compensation for Forced and Slave Laborers’, in
P. De Greiff (ed.), The Handbook of Reparations (Oxford: Oxford University Press, 2006).
with a no-unjust-benefit principle, in virtue of which reparative burdens may

befall those who benefit from their contemporaries’ or predecessors’ wrongdoings
even though they are not themselves causally implicated in those wrongdoings.
Consider first the thought that benefitting from a wrongdoing places agents
under reparative obligations towards victims.14 Suppose that A wages an unjust
war or phase of war against B. Many who do not, or did not, participate in that war
in any way will nevertheless benefit from its prosecution—from economic oppor-
tunities incidentally created by the war though not directly related to its conduct,
from infrastructure expenditures which would not have been made but for the war,
from acts of spoliations committed by combatants the proceeds of which are sent
back home, and so on. Those benefits, moreover, may well accrue over the lifespan
of more than one generation. It would be wrong of citizensA, who continue to
enjoy those benefits, not to pay reparations, in virtue of the general principle that
one may not accept the benefits accruing from one’s predecessors’ acts without at
the same time honouring the debts which they incurred in virtue of committing
those acts. Consider a peace settlement which is reached several decades after a
conflict has started (as will be the case with Israel and the Palestinians if they ever
get there, and as was the case in, e.g., Guatemala). The no-unjust-benefit principle
thus permits, indeed mandates, that its negotiators and their contemporaries
should incur reparative burdens for wrongdoings committed at the very beginning
of the war (long before they themselves participated in the latter), so long as they
continued to benefit from the war. Note however that in so far as violating or jus-
tifiably infringing rights is pro tanto worse than benefiting from such acts, the
principle also implies that peace negotiators should strive to ensure that reparative
burdens are born, as much as possible, by the cohorts of wrongdoers before they
have all died. If one includes those who come of age during the war and have sig-
nificantly contributed to its constitutive wrongful acts, a payment schedule might
well cover fifty to sixty years from the end of the war.
At this juncture, three worries arise. First, some might worry that the principle
implies, counterintuitively, that (e.g.), lawyers, doctors, social workers, teachers in
deprived areas, and so on, who benefit from wrongful harm imposition (since their
jobs partly depend on it) are under a duty to make reparations to their clients/
patients. Consider lawyers specializing in prosecuting and defending war criminals,
14 In his influential book on reparations in the international context, Butt worries that the unjust
benefit principle cannot do much work in the context of large-scale reparations such as are mandated
by the resort to and conduct in war. Instead, he defends a version of the fault principle which extends
across age cohorts, and which is similar to George Sher’s arguments (Butt, Rectifying International
Injustice, ch. 4; Sher, ‘Transgenerational Compensation’). For extended discussions of the view that
beneficiaries of injustice are under reparative obligations, see, e.g., R. Fullinwider, ‘Preferential Hiring
and Compensation’, Social Theory and Practice 3 (1975): 307–20 (against); J. J. Thomson, ‘Preferential
Hiring’, Philosophy & Public Affairs 2 (1973): 364–84 (for). See also the more recent special issue of
the Journal of Applied Philosophy, notably the following articles: D. Butt, ‘ “A Doctrine Quite New and
Altogether Untenable”: Defending the Beneficiary Pays Principle’, Journal of Applied Philosophy 31
(2014): 336–48; R. E. Goodin and C. Barry, ‘Benefiting from the Wrongdoing of Others’, Journal of
Applied Philosophy 31 (2014): 363–76; B. Haydar and G. Øverland, ‘The Normative Implications of
Benefiting from Injustice’, Journal of Applied Philosophy 31 (2014): 349–62; A. Pasternak, ‘Voluntary
Benefits from Wrongdoing’, Journal of Applied Philosophy 31 (2014): 377–91.
and who derive considerable professional satisfaction from doing this particular
job. There is a sense in which they derive a benefit from the commission of war
crimes (which is consistent with their wishing it to be the case that no such crimes
were committed). According to the no-unjust-benefit principle, however, it would
seem that qua war crime specialists, they are under reparative duties to the victims
themselves of the suspected war criminals they are prosecuting and defending. This is
counterintuitive—particularly so in the case of prosecution lawyers who, after all,
directly help victims already, but also in the case of defence lawyers who, after all,
indirectly benefit victims by making it possible to hold fair trials. The aforemen-
tioned professionals are already engaged in helping post-war communities to deal
with those particular instances of injustice, and it would indeed be odd to ask them
to do more qua such professionals.
To block this counterintuitive implication, the reparative principle can be for-
mulated as follows: someone B who benefits from the commission of an injustice
done to another person V is under a duty to pay reparations to V unless he is
already engaged in benefiting her. The principle so reconstructed is compatible
with the view that the lawyers may be held under reparative duties either to the
victims of those they are prosecuting or defending, or to other war victims, for
deriving unjust benefit from the war by dint of (e.g.) enjoying greater prosperity as
a result of the war. The principle is also compatible with the view, which I shall
develop in s.6.3, that they are under distributive duties of assistance to war victims.
(By analogy, the fact that someone works in a shelter for homeless people and thus
in some sense benefits from the latter’s plight does not exonerate her from paying
taxes some of which will go towards funding shelters.)
Second, although those benefits have sometimes been received entirely volun-
tarily, more often than not their recipients had no choice but to accept them—for
example when the war has generated non-excludable public goods. Few would
dispute that voluntarily and knowingly receiving a benefit which has come about
wrongfully places one under an obligation to compensate the victim of the wrong-
doer. To many, however, the involuntary receipt of wrongful benefits simply can-
not generate reparative obligations. Nozick, thus, insists that one can only be
obliged to provide a good if one has acted in such a way as to forfeit one’s right to
that good or if one has given prior consent (for example, though a mutual exchange,
or a promise) to being divested of it.15
I lack the space to engage with this long and familiar debate. Let me simply note
that at the bar of sufficientist justice, one can be obliged to divest oneself of
resources for the sake of helping those whose life is less than flourishing, even
though one has done nothing to forfeit rights to those resources nor given one’s
consent to such a policy. Accordingly, the Nozickean claim that the involuntary
receipt of unjust benefits does not confer obligations to compensate remains
unsupported. In any event, there is a positive reason for endorsing the principle—
namely that war victims do have rights to assistance, and we need to identify
duty-bearers. The receipt of unjust benefit from victims’ predicament establishes a
15 See, e.g., Nozick, Anarchy, State, and Utopia, esp. ch. 5.
morally relevant connection between those victims and the recipient, which in
turn warrants the imposition of some duties to pay. In addition, although some
agents have no choice but to receive those benefits, some among those may never-
theless welcome them, which clearly would place them under reparative burdens.
The point applies, note, whether or not A intends to benefit the recipient, which
intention is often formed as a result of a special relationship. Suppose that A inflicts
a wrongful harm on B, as a result of which you end up better off. Whether A does
so for his own sake and without knowing you or caring that you should benefit, or
whether he does so intending that you should benefit from B’s predicament because
you are his friend is irrelevant: either way, if you benefited from B’s being harmed
and know that this benefit proceeds from a wrongful act, you are under a duty to
compensate her. This is highly relevant to wars, for although they are fought by their
actors on behalf not just of their contemporary fellow community members but of
their successors in their transgenerational community, they often benefit outsiders
as well: think of the enormous benefits which are derived over decades by arms
manufacturers and which accrue not just to those companies’ employees and share-
holders, but to surrounding populations (more jobs in the area mean more and
better infrastructure, more opportunities, more tax proceeds for local governments,
etc.) Given that, in some cases, the mere receipt of wrongful benefits, irrespective of
borders, is sufficient to generate reparative obligations, the cosmopolitan claim that
borders are morally irrelevant to individuals’ burdens at the bar of justice dictates
against the view that the relationship of co-citizenship imposes additional burdens
on its parties (as opposed to, for once(!), conferring additional privileges.)
The third worry is this. We might be able to say with some degree of plausibility
that a political community in toto has benefited from an unjust war. We might also
be able to identify specific individuals who have enriched themselves thanks to
such a war—for example, the shareholders and/or board directors of companies
which were awarded lucrative contracts by the leadership. But we clearly cannot
identify most individual beneficiaries. And even if we can, we might not be able to
design a reparation scheme which is sensitive enough to degrees of fault and bene-
fiting. Under those circumstances, it is hard to see how we can implement the
Reparative Principle as it stands. The moral costs I highlighted earlier in relation to
the fault principle also apply here: implementing reparative justice necessarily
amounts to extracting reparations payments from individuals who, because they
neither committed wartime wrongdoings nor benefited from them, are not under
reparative duties. Imposing such payments on them wrongs them since they still
have ownership rights over the extracted resources.
In reply, it would be tempting to hold that the rights of the innocent, though
infringed, are infringed justifiably. But this will only take us so far. For not only is
it exceedingly hard to discern amongst dozens of thousands of individuals who
actually benefited from the war or were at fault for some of its injustices: in many
cases, it is even harder to discern among those who did benefit or were at fault
those who were also victims; to make it worse, even if we could reach the judge-
ment that this, or that, agent was (e.g.) in some respect a beneficiary of the war and
in others a victim thereof, it is still harder to work out whether the fact that he is a
beneficiary outweighs the fact that he is a victim, and if not, how much exactly he
ought to pay. This seems clearly true of individuals who, because the war takes
place where they live, experience it at close hand and often have no choice but to
be caught in it, as perpetrator, victim, or both. But it is also true, to some extent,
of those who live away from the theatre of war yet feel its effect. Consider a janitor
working at the headquarters of a US contractor whose main business is based in
Iraq in 2003. There is a sense in which that janitor benefits from the unjust war
which the US waged in that country. At the same time, his company health insur-
ance is patchy and does not cover the costs of the life-saving treatment he needs to
be cured of cancer. The costs of the war on US taxpayers are such that, even if the
will were there to provide universal healthcare, it would be financially impossible
to do so. There is a sense, thus, in which that janitor, who will die in six months, is
a victim of the war. Why should he pay up?16 Generally put, it may well be that in
extracting reparation payments from agents who we guess are innocent of the war,
we are in fact extracting payments from victims. It is much less plausible to claim
that victims’ rights are justifiably infringed, than to claim that the rights of the
innocent are justifiably infringed.
Put in general terms, the worry is this. At the bar of cosmopolitan morality,
principles for the imposition of harm are necessarily individualistic: whether we
are justified in harming another person depends on what that individual specifi-
cally did, on whether he benefited from another person’s predicament, and on
whether he in fact had a claim not to be harmed. These are very fine-grained judge-
ments which simply cannot be reached in the case of large-scale wrongdoings such
as wars, involving as they do multiplicities of actors whose status—as wrongdoers
or victims—is more often than not unclear.17 What, then, are we to do?
More Comments on Fault and No-Unjust-Benefit

Let me close this section with three final remarks on the fault- and no-un-
just-benefit principles for the allocation of reparative duties. First, the principles
differ in their specific implications. In particular, merely receiving unjust bene-
fits is not as bad as being causally responsible for the rights violations or rights
infringements from which those benefits accrue. Accordingly, the reparative
duties which the no-unjust-benefit principle imposes should be less onerous
than the reparative duties imposed by the fault principle—at least other things
16 According to a study by Joseph Stiglitz and Laura Blimes, the Iraq War and subsequent occupa-
tion of Iraq will have cost the US three trillion dollars. (J. Stiglitz and L. Bilmes, The Three Trillion
Dollar War: The True Cost of the Iraq Conflict (London: Penguin, 2008).) Even accounting for the fact
that some of those expenses would have been incurred anyway (such as replacing obsolete military
equipment), think of what could have been done with those resources instead. I hasten to add that the
costs—human, economic, social, etc.—on Iraqi civilians have been far worse. My point is merely that
someone can be a victim of a war fought in a distant land.
17 For a clear articulation of this worry, see S. Lazar, ‘Scepticism about Jus Post Bellum’, in L. May
and A. T. Forcehimes (ed.), Morality, Jus Post Bellum, and International Law (Cambridge: Cambridge
University Press, 2012). Unlike Lazar, however, I think (and argue in s.6.3 below) that there is still
space for reparative considerations in a theory of justice after war.
roughly equal, as someone who is only minimally causally responsible for a rights
violation is under lesser reparative duties than someone who derives sizeable
benefit from it.18
Second, both principles apply to wrongdoers and victims, irrespective of nation-
ality or, more widely, membership in state or non-state organizations.
Third, both principles justify the imposition of reparative duties on individuals
whose wartime state was dissolved as a result of the war. To give a well-known
example, the 1919 Treaties of St. Germain and Trianon imposed reparative obliga-
tions on the republics of Austria and Hungary, even though those two republics
were created in 1919 following the fall and dissolution of the Austro-Hungarian
Empire. Changes in statehood pose significant problems for public international
lawyers, precisely because the law confers personhood on states. At the bar of cos-
mopolitan justice, however, states’ rights and duties are the rights and duties of
their individual members. They endure even when the state does not. Whatever
one thinks of the substance of the reparative obligations imposed on citizens of
Austrian and Hungary by those treaties, one cannot reject them on the grounds
that the belligerent imperial state was no longer in existence.
6.3. T H E R E C O N S T RU C T I O N P R I N C I P L E : F I L L I N G
T H E G A P S O F R E PA R AT I O N S , B R I N G I N G A B O U T
DISTRIBUTIVE JUSTICE
6.3.1. Setting out the Reconstruction Principle

To recapitulate, I have argued that if A is at fault for or benefits from B’s predica-
ment, he owes her reparations. As we have seen throughout, however, epistemic
and resource constraints bedevil the implementation of the Reparative Principle to
post-conflict cases. In this section, I defend a so-called Reconstruction Principle,
to the effect that members of communities destroyed by war in general are owed
help with their reconstruction. Some of this help will be owed as reparation, some
of it as simply assistance.
Let me restate what those constraints are. As we have just seen, epistemically, we
are not in a position to know exactly whether and how this or that person was
wronged by the war and thus has a reparative claim, and/or whether they them-
selves committed or benefited from war wrongdoings and thus are under repara-
tive obligations. What we do know, however, is that the overwhelming majority of
individuals caught in war experience enormous suffering and deprivation.
Resource constraints arise as a result of the two-pronged fact that there are likely
to be too many war victims for full compensation to be possible and that not all
desperately needy individuals are war victims. Accordingly, we have no choice but
to forsake the Reparative Principle. For at the bar of fundamental equality, and
other things equal, there is no reason to hold that any given war victim has a
18 See Butt, Rectifying International Injustice, 117–30. See also Pasternak, ‘Voluntary Benefits from
Wrongdoing’.
stronger claim than other war victims. Moreover, there are good reasons to think
that in some cases, a war victim lacks a claim to be given priority over the non-war
needy. Let us assume for the sake of argument that belligerentA owes reparation to
war victimsB, and let us suppose that A cannot compensate those victims fully for
the wrongs they suffered and at the same time contribute to secure prospects for a
flourishing life to individuals who were not part of this or indeed any other con-
flict—call them C. Assume that A is under the latter, distributive obligation. B is
not always justified in insisting that A give citizensB precedence over citizensC when
providing material assistance. Let us suppose that fully compensating citizensB for
the harms they wrongfully suffered at the hands of combatantsA would enable
them to enjoy a maximally flourishing life. The fact that there are members in C
who are in such desperate need of help as to warrant holding third parties under an
obligation of assistance suggests that citizensB, at the time at which the treaty is
concluded, are under a prima facie obligation to help C. If the only way for them
to help C is by giving them part of that which A took from them, or the equivalent
thereof, then they are not in fact entitled to get back all that they lost during the
war, even if they were entitled to those resources when the war started.
Thus, inability to garner the requisite knowledge and resource scarcity operate
as, respectively, hard and soft as well as practical and moral constraints on the
implementation of the reparative principle. In the light of those difficulties, a just
peace cannot be achieved and a compromise in favour of a justifiedATC peace is
morally required. Remember: this might issue in some of those victims not being
given that to which they are prima facie entitled by way of reparations; but to meet
their claim in full would count as a reparative injustice to other war victims and a
distributive injustice to non-war needy persons. A justifiedATC peace, I suggest, is
one where we partly replace the Reparative Principle with a Reconstruction
Principle, to the effect that members of communities destroyed by war in general
are owed help with their reconstruction. I say ‘partly replace’, rather than ‘replace’
tout court, for as we shall see throughout this section, there still is room for repara-
tive considerations in setting out reconstruction duties.
The claim that those who have suffered rights violations or justified rights
infringements during war are owed post-war help is uncontroversial. More contro-
versial is the claim that agents who are sufficiently responsible for those wrongdo-
ings as to be held under some reparative duties are also owed some help. Yet they
too retain claims to have their basic needs met: analogously, even criminals right-
fully convicted for the most heinous crimes ought to be treated with minimal
concern and respect, and thus have rights to the basic necessities of life. Non-
combatants who share responsibility for heinous war crimes (such as the directors
of the Nazi-era industrial group IG Farben) by implication have such rights, as a
fortiori do non-combatants who are only marginally responsible for such crimes or
who are significantly responsible for lesser crimes. This is not to say that they have
rights to as much, by way of post-war assistance, as war victims; nor is it to say that,
if resources are simply not available to meet the basic needs of wrongdoers and
comply with the demands of cosmopolitan sufficientist justice vis-à-vis non-war
victims, the latter and the former ought to be regarded as exactly on a par. Rather,
my point is that those in a position to help rebuild ought to ensure as far as possi-
ble that resources are available to both.
6.3.2. Reconstruction and Conditional Aid

Assuming that I am right, may we subject the provision of reconstruction help to
conditions? One might think that in so far as victims have independently justified
rights to restitution, reparation, and/or reconstruction assistance, actors who are
under the correlative duties may not make their fulfilment of those duties condi-
tional upon recipients’ willingness to further their interests. This seems particularly
true of wrongdoers, since they are at fault for victims’ predicament and thus are not
in a position (one might think) to impose conditions on their victims. However,
although victims have those rights, they may not deploy the resources they are
thereby given to unjust ends. If a peace negotiator A has objectively good reasons
to believe that the other party B will use the restituted military equipment and/or
the repaired infrastructure towards a programme of unwarrantedly aggressive mil-
itary expansion or towards the repression of its own population, it is justified in
refusing to discharge its reparative and/or distributive obligations—though not to
the point where individual members of B would not have their basic needs met—
unless the relevant assurances are both given and enforced.
Conditionality, here, is understood narrowly, to mean that a beneficiary of
reconstruction help ought not to use the help so given as a means to promote unjust
ends. Suppose, however, that B would not directly use those resources themselves
to unjust ends (schools would be built, hospitals would be refitted, roads would be
repaired, individual victims of war would receive reparation for the loss of their
livelihood, etc.) but would thereby be in a position concurrently to carry out
unjust policies: thanks to the resources which B would receive as reparations or
non-reparative assistance, its regime could devote other resources to, e.g., rebuild
an aggressive military force, fund and direct human rights–violating police and
judicial institutions, etc. May A, in this case, impose justice-promoting conditions
on the fulfilment of its duties? Yes, to the extent that receiving reconstruction help
would facilitate B’s pursuit of injustice.
The foregoing remarks call for two qualifications. First, it seems (tentatively)
that parties to a peace agreement which are themselves unwarrantedly bellicose
and/or violate their members’ or outsiders’ human rights while negotiations are
ongoing do not have the standing to withhold the assistance which they are under
a prima facie duty to provide to war victims on the grounds that the latter’s regime
is likely to be derelict at the bar of justice. Second, when conditionality is legiti-
mate as suggested in those cases, its stipulations are nothing more, and nothing
less, than what justice requires. In setting out those conditions, peace negotiators
merely declare that they will do what they ought to do in any case. But consider
now cases where, thanks to its having won the war, peace negotiator A imposes on
its counterpart B conditions which B is not under an independently justified duty
to meet. Suppose, for example, that A, whose members are under reparative duties
towards war-victimsB, agree to set up a reparation package for the latter provided
that B’s regime sets up a bilateral trade agreement between the two countries such
that B’s firms would buy material and/or natural resources from A’s firms. In effect,
A says to B: ‘We will give you the money we owe you to rebuild the houses we
unjustifiably destroyed, but on condition that you buy the concrete from us’.
Consider the reconstruction programme undertaken by the United States in Iraq
following its invasion in 2003. Admittedly, it was not the object of a peace settle-
ment between the two regimes for the simple reason that there was no Iraqi regime
left then. Yet it is not implausible to suppose in the light of the ways in which the
programme was set up that, had Saddam Hussein’s regime survived the war and
embarked on peace negotiations with the US, such conditions would have been
imposed on them. Or suppose that donors make it a condition of their providing
assistance that aid recipients should engage in privatization reforms which will
have a devastating impact on the poorest members of those communities.19 Is a
settlement of that kind unjust? Yes, on the grounds that it is both wrongfully
exploitative and wrongfully coercive. Standardly, a transaction between A and B is
wrongfully exploitative, in A’s favour and to B’s detriment, if (a) A benefits from
the transaction; (b) the outcome of the transaction is harmful or (in the case of a
mutually advantageous transaction) unfair to B; and (c) A takes advantage of a
feature of B or of his situation to get B to agree to the transaction. A transaction is
not merely wrongfully exploitative, but in addition wrongfully coercive, if A has
wrongfully brought about B’s predicament. In some post-war settlements, of which
the putative Iraq–US settlement I am imagining here would have been a good
example, conditional reconstruction is wrongfully exploitative, as follows: belliger-
ent A takes advantage of B’s situation (military defeat), which it has wrongfully
brought about, to impose on it treaty clauses which are beneficial to A and harmful
or unfair to B. In cases such as these, conditionality vitiates the peace settlement,
in the sense that B is not under a duty to A to comply with the settlement (though
it may of course be under an all-things-considered obligation to abide by its terms,
for example if noncompliance would trigger a new war).
6.3.3. Who Owes Reconstruction Duties?

So much, then, for reconstruction rights. Who, then, owes the relevant duties? To
the extent that they are reparative duties, they are owed by rights violators, justified
rights-infringers, and unjust beneficiaries. To the extent that they are distributive
duties, they are owed by parties who do not fall in either categories, whether they
were bystanders to the conflict or participants therein. But perhaps this is too
quick. Perhaps—some might object—we need to distinguish between the case
where C, though innocent of B’s predicament, himself has violated or justifiably
infringed some other person’s right or has derived unjust benefit from some other
19 For examples of conditional reconstruction aid as provided by international organizations—

some of which proved controversial for reasons indicated here—see R. Caplan, International
Governance of War-Torn Territories: Rule and Reconstruction (Oxford: Oxford University Press,
2004), 139.
person’s wrongful predicament, and the case where C is neither a rights violator or
justified infringer nor has derived unjust benefits from anyone else’s wrongdoing.
In the former case, it is plausible to hold that C, together with other similar act-
ors, is under a duty to contribute to victims’ reparation funds in general—as in the
US, where so-called ‘offenders fees’ fill the coffers of such funds.20 In the latter
case, some think that C, whom we can describe as a bystander, is not under a duty
to step into the breach, on the grounds that it would be unfair to ask him to bear
the costs of another party’s dereliction of duty.21 But this proves too much. For by
that token, no one is under an obligation to help a child whose parents have failed
in their duty of care towards her, or an employee whose employer wrongfully fired
him and cannot for whatever reason be made to pay him reparation, or someone
who suffers from a debilitating injury as a result of someone else’s careless conduct.
Exempting C would undermine the legitimacy of much of welfare provision, and
is not compatible with justice.22 In fact, there are many examples of outsiders’
involvements in reconstruction programmes. Transitional administrations, as we
shall see in ch. 8, are a primary contemporary example: the funds transferred to
East Timor, Kosovo, Bosnia-Herzegovina and others amount to billions of dollars
and have been provided by the United Nations, many of whose members were
neither parties in nor benefited from the conflicts which ravaged those communi-
ties though were signatories to the peace settlements which set up those pro-
grammes. I see no reason to regard such disbursement as unjust—on the contrary,
they are justified by straightforward appeal to more general duties of assistance.23
To be sure, there is a limit to those distributive duties, which is set by the extent
to which B has been harmed. If B’s prospects for a flourishing life have been under-
mined by A’s act such that B now stands below the sufficiency threshold, C is under
a duty to her to secure those prospects when A defaults. But if A’s act has not
undermined those prospects and yet harmed B relative to the state of affairs she
would have been in otherwise (assuming, as we sometimes can, that we are able to
discern the relevant counterfactual state of affairs), reparative payments are owed
to her either by wrongdoers other than A or via other parties’ ex ante contributions.
If those contributions are not sufficient, the taxes which B pays can be regarded as
insurance against third parties’ dereliction of reparative duties.
Importantly, the same party might be under reconstruction duties at the bar of
both reparative justice and the ethics of assistance. Consider the 1947–52 Marshall
Plan, by which the American administration embarked on one of the largest
20 See http://www.nacvcb.org/ (accessed on 28/10/2015).

21 See, e.g., D. W. Brock, ‘Defending Moral Options’, Philosophy and Phenomenological Research 51
(1991): 909–13.
22 For extended discussions of agents’ duties under situations of partial and full compliance, see,
e.g., T. Mulgan, ‘Two Conceptions of Benevolence’, Philosophy & Public Affairs 26 (1997): 62–79;
L. B. Murphy, ‘The Demands of Beneficence’, Philosophy & Public Affairs 22 (1993): 267–92.
23 This is not to deny that, as a matter of fact, such reconstruction efforts were hampered by cor-
ruption, greater attention paid to the wants of international staff than to the needs of bereft popula-
tions, and general incompetence. For a particularly withering critique, see, e.g., S. Chesterman, You,
The People—The United Nations, Transitional Administration, and State-Building (Oxford: Oxford
University Press, 2004), ch. 6.
post-war reconstruction programme ever seen. Under its terms, the United States
were to provide free goods to European countries annually, which would then be
sold to the latter’s populations, with the proceeds being used as each national gov-
ernment thought fit in agreement with the US government. The aim of the Plan
was to incite European governments to embark on long-term structural invest-
ments with emphasis on growth and efficiency. According to some estimates, were
a similar Plan carried out nowadays, it would cost, roughly, 200 billion dollars.24
Most of the Plan’s beneficiaries had been whole or part victims of Nazi Germany’s
policies: at the bar of reparative justice, Germany, not the United States, should
have paid for the reconstruction of (inter alia) Britain, Belgium, and the
Netherlands. In fact, Germany should also have paid reparations to the United
States for the wrongful harms inflicted on the latter’s combatants. To be sure, the
United States themselves (with their Allies) had committed war crimes against
German civilians (through carpet bombings), for which they did owe reparations.
But the Germans’ reparative burdens were considerably more onerous. Yet
Germany was utterly unable to pay reparations to those countries, destroyed as she
herself was by the war. Moreover, insisting on extracting reparations from her, as
the French initially sought to do, would have jeopardized Europe’s revival and
prospects for a durable peace. Under those circumstances, insisting on holding
wrongdoers to account would in fact have been morally wrong; but for the United
States to refuse help would have been morally wrong as well (as well as prudentially
misconceived).
The foregoing remarks constitute an alternative justification for extracting
reconstruction payments from those who are not liable to the imposition of
reparative burdens. But there is an additional justification for holding bystanders
under reconstruction duties, which is modelled on a powerful argument in favour
of the welfare state. In paying taxes, C protects himself against the risk that he
might one day have a reparative claim against someone who will not be able or
willing to pay, and that he might also, one day, be under reparative duties to
someone whom he has wrongfully harmed yet will not be able to discharge. So
construed, welfare provision is not mandated by either reparative or distributive
justice but, rather, serves as a prudentially justified insurance scheme. Analogously,
however removed a given community is from military conflicts here and now, it
would be condemnably foolish on the part of its members not to anticipate either
that war wrongdoings might one day be carried out with the consent of substan-
tial enough a part of the population, or that they will one day benefit from such
crimes. As Larry May suggests in After War Ends, given that a great many individ-
uals at some point have been or will be connected to war wrongdoings through
fault or unjust benefiting, it makes sense to impose on all of them, via whichever
coercive institutions under whose jurisdiction they live, a duty to contribute to an
international reparation fund out of which the reconstruction of war-torn com-
24 T. Judt, Postwar—A History of Europe since 1945 (London: Pimlico, 2007), 91. Chs. III and IV
of that book provide a clear account of the Plan and, more generally, of the issues of reparations and
reconstructions in Europe following the Second World War.
munities could be rebuilt.25 Note that the duty to pay reparations and assist in
reconstruction programmes is an institutionalist duty—that is to say, a duty
which individuals discharge through the relevant institutions, rather than in their
daily lives. The rationale I adduced in s.1.2 in favour of the institutionalist
approach to duties of distributive justice in peacetime thus applies to duties of
distributive and reparative justice after war.
That said, May believes that this insurance scheme should be construed as a
no-fault scheme and used to pay all reparations and reconstruction costs. I worry
that his proposal rides roughshod over the fact that in some cases at least, an iden-
tifiable actor is more significantly responsible for a particularly destructive unjust
war or unjust phase of war, or has benefited significantly more than other mem-
bers of the international community from it. For example, while it is incontro-
vertible that France supported the Rwandan Hutu regime while its genocidal
intentions were already well known, and that France therefore is morally liable for
some of the costs of rebuilding Rwanda, the fact remains that the genocide itself
was committed by Hutu extremists at the behest of their regime. Likewise, while
some members of the international community backed Syrian President Bashar
al-Assad for a number of years before the uprising of 2011, the best available
evidence suggests that his forces are responsible for the lion’s share of atrocities in
a conflict which at the time of writing this shows no sign of ending.26 A no-fault
insurance scheme would deem France and Assad’s backers to be as liable for those
depredations and killings as the Hutu regime and Assad’s regime respectively
were. This strikes me as a deeply troublesome and in fact unnecessary retreat from
a more fine-grained, and therefore less unjust, way of handling multiple responsi-
bilities for war crimes. On that more fine-grained model, if, in a given case, some
actors are significantly more responsible for, or significantly stronger beneficiaries
of, such crimes, they could be held liable to extra reparative burdens, via (e.g.)
systematic confiscation of their privately owned assets, or via supplementary rep-
aration payments, or both—in addition to payments drawn from the interna-
tional reparations fund.
There is one final worry to consider. Admittedly, the Reconstruction Principle is
less vulnerable than its reparative counterpart to the charge that epistemic con-
straints render its implementation impossible, since it does not to the same degree
rely on the possibility of identifying who is at fault for or benefits from which war
wrongdoings. However, it too is faced with epistemic difficulties which, some
might think, are insuperable for the kind of moral theory which underpins my
account of post-war justice: after all, providers of reconstruction help will usually
not be able to differentiate between those who have a claim to more than having
their needs met (a claim, that is, to compensation above and beyond needs, for the
25 May, After War Ends, 194–5. Gains realized from the wrongful seizure of territory should be
reversed into the fund (see s.5.2.2).
26 Various UN reports in support of this claim can be found at http://www.ohchr.org/EN/
HRBodies/HRC/IICISyria/Pages/AboutCoI.aspx On France’s support for the Rwandan genocidal
regime, see, e.g., L. Melvern, Conspiracy to Murder: the Rwandan Genocide (London: Verso, 2006),
38–9.
wrongdoing they suffered) and those who do not have such claim because they
were wrongdoers. This is because, to repeat, establishing the relevant degree of
responsibility of this or that non-combatant for an unjust war, phase of war, or
isolated war wrongdoings, is more or less impossible except for (most) young
children, the very elderly, the severely incapacitated, etc. Moreover, some non-
combatants who contribute to war wrongdoings often also contribute to just acts
within the war—as, for example, do workers who build weapons used by the same
army for a just cause as well as for grievously unjust ends, or those who work in
dual-use (military and civilians) industries. If their livelihood is destroyed in the
latter stages of the war, we might hold that they are owed nothing more than what
is required to meet their basic needs, on the grounds of their previous unjust con-
tributions; but we might also hold that their contribution to just parts of the war
confers on them rights to what they need in order to lead a flourishing life. It is
hard to see how reconstruction programmes can arrive at the right judgements in
such cases. Finally, reconstruction often consists in delivering and providing
non-excludable public goods such as roads, bridges, hospitals, schools, factories,
and so on, which will benefit both wrongdoers and the innocent.
The worry, then, is this—that in just the same way as mass reparations pro-
grammes simply cannot operate on the basis of such fine-grained judgements, nor
can mass reconstruction programmes. But if, as I and others maintain, individuals’
rights, duties, and liabilities do not depend on their membership in this or that
group per se but are grounded in relevant factors such as what they have done, or
the extent of their need or whether they have knowingly benefited from a wrong-
doing, it would seem that those programmes will necessarily be unjust, and that
not much has been gained by the move from the Reparative Principle to the
Reconstruction Principle.
By way of reply, I do not deny that implementing principles for a fully just
reconstruction settlement is unfeasible, for the reasons just mentioned. It will una-
voidably help those who do not in fact have a claim to be helped to that extent—
such as wrongdoers who benefit from the public goods of a rejuvenated industry,
of an educated youth, etc. Whatever we do, we will incur moral costs. Either we
will err on the side of unwarrantedly helping wrongdoers as an unavoidable side-
effect of helping the innocent; or we will err on the side of withholding or restrict-
ing reconstruction help as a means to ensure that wrongdoers do not get what they
do not deserve but with the consequence that the innocent will not get that to
which they have a right. In the light of the grievously wrongful harms incurred by
war victims, of the fact that a war is just only if it helps bring about a durable
peace, and of the fact that allowing dozens of thousands of people to live in desper-
ate destitution surely is not conducive to such a peace, we should err on the side of
helping those who do not in fact have a right to assistance, and not on the side of
not helping those who do have such a right. The fact that wrongdoers will get help
to which they are not entitled undoubtedly renders the peace unjust stricto censu;
it is however justified all things considered—which under the circumstances is the
best that we can hope for.
6 . 4 . C O N C LU S I O N
To conclude, I have defended the following claims. First, a peace settlement, if it is

to be justifiedATC, requires that, under some circumstances, agents who caused
wrongful harms during the war, and/or who did or still do derive unjust benefit
from the commission of such harms, compensate victims and their descendants. In
so far as the imposition of reparative duties proceeds from what agents have done,
or of what they benefited from, irrespective of their membership in this or that
political community, it is of a piece with cosmopolitan morality. Second, a peace
settlement whereby war victims do not have their prospects for a flourishing life
fully restored thanks to reparative measures is justifiedATC provided that the reason
why it does not give full satisfaction to some of the claimants is that other agents
are below the sufficiency threshold—whether those agents’ predicament can be
traced to wars or to non-anthropogenic factors. Third, when wrongdoers are una-
ble or unwilling to meet their reparative obligations, outsiders to the conflict are
under distributive duties to step into the breach. Fourth, post-war duties of assist-
ance are owed to all war victims, whether they happen to belong to a just or unjust
belligerent community. Fifth, war victims do not have priority over victims of
non-anthropogenic harms.
This concludes our inquiry into the norms which ought to govern the disburse-
ment and (re)allocation of material resources after war. In the next chapter, I con-
sider another set of issues arising from the commission of wrongdoings in war, to
wit, the grounds upon and conditions under which their perpetrators may justifi-
ably be punished.
7
Punishment
The view that punishment for some crimes of war is warranted is beyond dispute,
as witnessed by, e.g., the Nuremberg and Tokyo trials, and the establishment of
the International Criminal Tribunal for the former Yugoslavia (ICTY), of the
International Criminal Tribunal for Rwanda (ICTR), and of the International
Criminal Court (ICC). Although the Nuremberg Trials are often regarded as the
founding act of international criminal law, they were not the first attempts at
bringing to trial political actors deemed responsible for waging a wrongfully
aggressive war or for committing crimes against humanity. Famous, and ill-fated
precedents, include the Allies’ attempts to try the Kaiser and other German war
criminals after the First World War (as stipulated in articles 227 and 228 of the
Treaty of Versailles), and attempts by the British to bring Turkish leaders to account
for the 1915 Armenian Genocide at the so-called Constantinople Trial. Both
ended in failure but paved the way for Nuremberg.1 Addressing the issue of pun-
ishment in the context of a cosmopolitan ethics of war and peace requires deline-
ating the cosmopolitan underpinnings of punishment for war-related crimes. This,
in effect, is the task which I set myself in this chapter.
I proceed as follows. In s.7.2, I sketch out what I take to be the most plausible
defence of punishment. In s.7.3, I offer an account of which war-related wrongdo-
ings call for punishment. In s.7.4, I argue that agents who participate in such
wrongdoings are liable to punishment simpliciter, and that the precise punishment
which they are liable to incur must fit both their individual contribution and the
fact that what they are participating in is a collective enterprise. In s.7.5, I discuss
two kinds of punishment, in addition to imprisonment, which might be thought
to fit war-related crimes—to wit, lustration and financial penalties. In s.7.6,
I defend a revised version of the principle of universal jurisdiction, whereby (roughly
put for now) all human beings (for short, humankind) have primary jurisdiction
over grievous human rights violations in general and therefore have jurisdiction
over war-related crimes. Punishment for war crimes, I conclude, is a requirement
of a just peace. As I also suggest at various junctures, it may not be achievable and
we might therefore have to compromise in favour of a justifiedATC peace. One such
1 See G. J. Bass, Stay the Hand of Vengeance: The Politics of War Crimes Tribunals (Princeton, N.J.:
Princeton University Press, 2002), esp. chs. 3–4.
Punishment 173
compromise consists in granting an amnesty to wrongdoers. I offer a qualified

defence of this practice in s.7.7.
Before I begin, I need to clarify the terminology I employ here, and to say some-
thing about the relationship between the laws of war and the morality of war. On
the first count, throughout this chapter and unless otherwise stated, I use the
phrases ‘war-related crimes’ and ‘war-related wrongs’ interchangeably to refer to
human rights violations committed in the course of a war—of which war crimes
tout court (a specific category in international criminal law) are a subset. By ‘crimes
committed in the course of a war’, I mean those crimes which are committed by
combatants and civilians against one another qua actors in the war. A combatant
who murders his own relatives because, e.g., he wants their house for his own use
is not a war criminal—unlike a combatant who murders his own relatives by mar-
riage because they belong to a different ethnic group which is being persecuted by
that combatant’s regime in the context of a civil war. As I suggest, however, the
norms which govern punishment for war-crimes are not substantially different
from the norms which govern punishment for domestic offences. The point is
worth stressing, for in many cases, notably when combatants and their leaders are
enmeshed in criminal activities such as money laundering, drug trade, and human
trafficking, the line between war-related crimes and ‘common’ crimes is often
blurred. Accordingly, the distinction I have just drawn between war-related crimes
and other crimes is a matter of expository convenience and has no deep normative
significance.
On the second count, many proponents of the view (of whom I am one) that
unjust combatants who kill just enemy combatants are committing an act of mur-
der usually deny that they should be punished for so acting—and thus affirm that
the laws of war ought not to track (at least, on this point) the morality of war. It is
not always clear, however, whether they mean exactly that, or whether they mean,
more precisely, that unjust combatants who have murdered just combatants ought
not to be punished in the same way as domestic murderers ought to be punished.
The latter claim is compatible with the view that they ought to be punished in
some way—in other words, that the laws of war ought to regard such killings as a
crime. One of my main aims, in this chapter, is to defend that view and to show
that there nevertheless are good reasons not to treat most of those combatants as
‘ordinary’, domestic murderers. (Though I shall also suggest that those reasons also
provide grounds for reforming many of our domestic punitive practices.)2
2 McMahan is one of the main, neoclassical proponents of the view that the laws of war ought not
to track the so-called deep morality of war as far as the unjust killing of just combatants is concerned.
See his J. McMahan, ‘The Morality of War and the Law of War’, in D. Rodin and H. Shue (ed.), Just
and Unjust Warriors (Oxford: Oxford University Press, 2008). I used to accept the distinction
between deep morality and laws of war in CW, though I now think that it is not well drawn. For
criticisms of McMahan’s view, see H. Shue, ‘Do We Need a “Morality of War”?’, in D. Rodin and
H. Shue (ed.), Just and Unjust Warriors (Oxford: Oxford University Press, 2008). This issue has mor-
phed into a lively debate. For a particularly thoughtful, and recent, take on the relationship between
morality and law in this area, see A. A. Haque, ‘Law and Morality at War’, Criminal Law and
Philosophy 8 (2014): 79–97.
174 Punishment
7 . 2 . J U S T I F Y I N G P U N I S H M E N T
Punishment involves the imposition by some party (P) on some other party (W) of
a burden, harm, or cost which W has a prima facie right not to incur, in response
to a wrongdoing which W committed against some agent (V). Justified punish-
ment is necessarily a response to some wrongdoing, and a comprehensive justifica-
tion for it—which is beyond my remit here—must thus delineate a plausible set of
wrongdoings the commission of which warrants it. For my purposes in this book,
it suffices to say that, if some wrongdoings do warrant it, then culpable attacks on
other parties’ human rights clearly do. Those rights, you recall, protect individuals’
interests in the freedoms and resources they need to lead a flourishing life. Secure
enjoyment of those goods requires that individuals have reasons to expect that
others not only will desist from harming them wrongfully, but will also be willing
to attach sanctions to such acts.
That last clause points to yet another task which a comprehensive justification
for punishment sets itself—namely to explain why it is important that those acts
should be sanctioned. Such an explanation will show, in particular, why punishing
W does not wrong him (or, put differently, why W is liable to punishment). But it
will also show that it is all things considered a justified thing to do. Those are dif-
ferent claims: it may well be that punishment does not wrong W, in the sense that
W is liable to be punished and lacks a grievance if he is, but that there are counter-
vailing (moral) considerations for not doing it (for example, punishing him would
impose disproportionate harm on third parties, such as his dependent children).
Finally, a comprehensive theory of punishment must provide an account of who
may punish (P).
In the remainder of this section, I shall defend (briefly) an expressivist account
of punishment whereby the latter is justified as a means to express one’s commit-
ment to victims’ equal moral standing. Notwithstanding important criticisms of
expressivism in general, and the strengths of rival accounts, this strikes me as the
most promising view. It is worth noting, however, that much (though not all) of
what I then go on to say about punishment for war-related crimes in ss.7.3 to 7.6
is compatible with those rival accounts. Accordingly, my treatment of this particu-
lar feature of peace after war is more ecumenical than might appear at first sight.
7.2.1. Expressivist Punishment: A First Cut

There are many candidate-bases for liability to punishment. To name but a few,
according to so-called desert retributivism, W is liable to punishment because he
deserves to be punished and people should get what they deserve. According to
fair-play retributivism, by violating V’s rights, W breaches rules of social coopera-
tion from which he himself benefits (in the sense that the fact that other people do
not violate his rights makes it easier for him to go about his life, including his
wrongful endeavours); punishing him is a way to cancel out the unfair advantage
which W has thereby gained. There are many other kinds of retributivism—all of
which have it in common that they regard punishment as backward-looking:
Punishment 175
whatever good it brings about, punishment is justified solely by reference to W’s

wrongdoing. By contrast, deterrence or protection theories justify punishment by
appealing to forward-looking considerations. W is liable to punishment, on those
views, only if punishing him serves to protect other people from rights-violations
at his hands, or only if it serves to deter putative wrongdoers from committing
similar crimes. Pace an often-rehearsed objection to deterrence-based justifications,
the latter need not deny that W is liable to punishment only if he has committed
a serious wrongdoing. Pace retributivists, however, they insist that committing a
serious wrongdoing is not, on its own, a basis for liability to punishment, and that
it must be supplemented by forward-looking considerations. Note also that hybrid
justifications for punishment can be constructed out of both retributivist and pro-
tection-based (or protectivist) arguments. One can conceivably hold, for example,
that W is liable to punishment only in so far as he deserves it or has gained an
unfair advantage by violating V’s rights, and that punishing him is all things con-
sidered morally justified only if it also serves protective ends.3
With retributivists and some protectivists, I believe that W’s wrongdoing is the
basis for his liability to punishment (modulo considerations to do with his frame of
mind when acting wrongly, as we shall see in s.7.2.2)—though I do not think that
desert or fair play are particularly promising justificatory routes. With protective
theories, however, I also believe that all-things-considered justifications for punish-
ment must attend to forward-looking considerations. Let me elaborate. By violat-
ing V’s rights, W has forfeited his right not to incur the very specific kind of
burden which punishment, as distinct from compensatory or indeed defensive
harms, constitutes. Granted, all three kinds of harm imposition are responses to
W’s wrongdoings. Unlike the imposition of compensatory and defensive harm,
however, punishment has an inherently expressive dimension: that is, it expresses
the view that by violating V’s rights, W has treated her as less than his moral equal
whereas in fact she is and always was entitled to be treated by him as such; it also
expresses the view that W’s conduct was not of the kind one can expect of a rational
and moral agent. Imposing such a harm is justified, in turn, by the importance
of explicitly communicating that there always was moral parity between W and
V. Failure to sanction W’s wrongdoing is on the one hand a failure to take V’s rights
seriously and on the other hand a failure to regard W as the moral and rational
agent he is. Merely disapproving of W openly will not do, for it will not adequately
express to V, or indeed to W and all and sundry, that W acted grievously wrongly
3 For a classic typology of retributivist theories, see J. Cottingham, ‘Varieties of Retribution’, The
Philosophical Quarterly 29 (1979): 238–46. For a classic retributivist account of punishment, see
M. S. Moore, A Theory of the Criminal Law (Oxford: Oxford University Press, 2010), ch. 3. For stand-
ard criticisms of retributivism, see, e.g., D. Dolinko, ‘Some Thoughts About Retributivism’, Ethics
101 (1991): 537–59; R. Shafer-Landau, ‘The Failure of Retributivism’, Philosophical Studies: An
International Journal for Philosophy in the Analytic Tradition 82 (1996): 289–316.
The locus classicus for protective theories in contemporary discussions is W. Quinn, ‘The Right to
Threaten and the Right to Punish’, Philosophy & Public Affairs 14 (1985): 327–73. To my mind, the
best articulation and defence of protectivism is V. Tadros, The Ends of Harm: The Moral Foundations of
Criminal Law (Oxford: Oxford University Press, 2011).
176 Punishment
by violating V’s rights. Harming W, by contrast, does express that point, in that it
reaffirms V’s moral status as well as his.4
Crucially, the expressive value of punishment does not lie in the fact that pun-
ishment often expresses anger, repugnance, and disgust at what offenders did.
However understandable or even justified those emotions might be, expressing
them through our punitive practices is likely to increase the risks that offenders,
alienated as they will be made to feel, will reoffend and harm their prospects of
rehabilitation. It is also likely to detract attention from victims for the sake of
debasing perpetrators. This is particularly problematic, because expressing com-
mitment to victims’ moral standing is as important as communicating the extent
to which perpetrators did do wrong. In addition, offenders, though unquestiona-
bly responsible for their acts, nevertheless often act under less than ideal agential
and epistemic circumstances, suffering as many do of various psychological disor-
ders. Our reactive attitude to their wrongdoings ought to reflect both the fact that
they did commit grievous wrongdoings and the fact that there were non-trivial
obstacles to their desisting from so acting. As Nicola Lacey and Hanna Pickard
persuasively argue, we may, indeed should, deem them blameworthy, but in a
detached way that recognizes that they are responsible for their actions (failing
which we culpably show lack of respect for their agency), and not in an affective
way which overlooks the conditions under which they acted (as a result of which
we culpably show lack of compassion for their humanity, and risk imposing on
them punishment which would neither fit nor be proportionate to the crime).5
To recapitulate briefly, W is liable to be punished to the extent that and only if
he has committed a grievous wrongdoing, as a way to communicate to him, as well
as to all and sundry, the magnitude of his wrongdoing. As we saw above, however,
it does not follow from the claim that W is liable to punishment that he may
4 The justification for punishment which I develop in this paragraph is best described as an expres-
sive interpretation of the forfeiture view. For defences of the forfeiture view to which much of what
follows owes, see A. H. Goldman, ‘The Paradox of Punishment’, Philosophy & Public Affairs 9 (1979):
42–58; C. H. Wellman, ‘The Rights Forfeiture Theory of Punishment’, Ethics 122 (2012): 371–93.
Prominent communicative, or expressive, theories of punishment are: see R. A. Duff, Punishment,
Communication, and Community (Oxford: Oxford University Press, 2001); J. Feinberg, ‘The Expressive
Function of Punishment’, The Monist 49 (1965): 397–423; A. von Hirsch, Censure and Sanctions
(Oxford: Oxford University Press, 1993), esp. ch. 2; J. Hampton, ‘The Moral Education Theory of
Punishment’, Philosophy & Public Affairs 13 (1984): 208–38. Hampton herself articulates, better than
anyone I can think of, the view that punishment expresses a commitment to V’s and O’s moral parity
before the wrondgoing, and is constrained by respect for the enduring dignity of the offender. See her
contribution to J. Hampton and J. G. Murphy, Forgiveness and Mercy (Cambridge: Cambridge
University Press, 1988), esp. ch. 5. For a contemporary defence of the view that a failure to punish W
is a failure to treat him with respect, see C. Bennett, The Apology Ritual (Cambridge: Cambridge
University Press, 2008). Finally, for an interesting critique of expressivist (or communicative) theories,
see Tadros, The Ends of Harm, ch. 5.
5 Broadly, Lacey and Pickard’s point is that clinical attributions of responsibility (which requires of
the clinician that she treat patients with respect and compassion at the same time as she holds them
accountable for their wrongdoings) can usefully be imported into the penal world. See N. Lacey and
H. Pickard, ‘From the Consulting Room to the Court Room? Taking the Clinical Model of
Responsibility Without Blame into the Legal Realm’, Oxford Journal of Legal Studies 33 (2013): 1–29.
The view that punishing is a way to mark respect for offenders’ agency is developed by H. Morris,
‘Persons and Punishment’, The Monist 52 (1968): 475–501.
Punishment 177
justifiably be punished all things considered. Advocates of forward-looking theo-

ries of punishment hold that punishing W is all-things-considered justified only if
it serves some future good—such as protecting third parties either from W’s reof-
fending or from other putative wrongdoers, or such as rehabilitating W into the
community. I agree, particularly in so far as punishment involves the deployment
of considerable resources which could be directed to other ends. Thus, we may not
punish W if, as a result, he would commit further and equally serious or worse
wrongdoings, or if others would so act. Nor may we punish him if the harms
incurred by third parties as a result either of the punishment itself or of the opera-
tion of the judicial system are disproportionate relative to the value of expressing
our condemnation of W’s wrongdoing and affirming V’s dignity.
7.2.2. Qualifying the Theory: A Second Cut

Let me make four points, by way of elaboration. First, on my account, there are
inherent moral limits on the infliction of punishment: punishment manifests to
the wrongdoer that his victim is not inferior to him; but given that (it is assumed)
wrongdoers never lose their standing as moral agents whom it is warranted to treat
with dignity, punishment so conceived must not degrade the wrongdoer to such an
extent as to impair his dignity.
Second, the educational and remembrance value of trials in general, and par-
ticularly war crime trials, forms no part of my justification for punishment. They
certainly have such value, offering as they do opportunities to record and archive
witness statements, defendants’ justifications for their deeds, and victims’ suffer-
ing. However, as Gerry Simpson and William Schabas note in separate works, the
demands of punitive justice are in tension with the demands of remembrance.
Punitive justice should focus on what the accused actually did, which rules out the
introduction of evidence that is not directly related to it such as background his-
torical context—precisely the kind of evidence which remembrance requires. For
that reason, both the Nuremberg Trial of 1945–46 and Eichmann’s Trial in
Jerusalem singularly failed as a purely punitive exercise. Moreover, punitive justice
requires that the court should treat both prosecution and defendant impartially, so
that, for example, expert witnesses introduced by the latter should be treated on a
par procedurally speaking with expert witnesses introduced by the former. In some
cases, as when Holocaust deniers are brought in as expert witnesses in defamation
trials, this flies in the face of the truth-tracking requirement of war remembrance
(a point to which I shall return in ch. 10). The worry with justifying punishment
by appealing to the trial’s remembrance and educational value is that neither the
defendant nor historical truth will be done justice.6
6 G. J. Simpson, Law, War and Crime: War Crimes Trials and the Reinvention of International Law
(Cambridge: Polity, 2007), ch. 4; W. Schabas, Unimaginable Atrocities: Justice, Politics, and Rights at
the War Crimes Tribunals (Oxford: Oxford University Press, 2012), ch. 6. On Nuremberg and the
Eichmann trial, see, e.g., A. Tusa and J. Tusa, The Nuremberg Trial (New York: Skyhorse Publishing,
2010); and H. Arendt, Eichmann in Jerusalem: A Report on the Banality of Evil, rev. and enl. edn.
(London: Penguin, 2006).
178 Punishment
Third, my stipulation that attacks on V’s human rights should be the product of
culpable agency is worth developing in the context of my account of defensive
killing. In CW (s.2.2) I argue that morally innocent attackers such as Thomson’s
psychotic gunman who unjustifiably pose a wrongful threat of lethal harm to some
other party lose their right not to be killed—in other words, are liable to defensive
harm. However, even though what they do—their conduct—is the kind of act
which warrants punishment (what the law calls actus reus) they are not ipso facto
liable to it. Justified punishment requires that the wrongdoer should have acted
culpably (what the law calls mens rea)—in other words, that he should have been a
morally and legally responsible agent when committing the crime and thus in full
possession of his physical and mental faculties. This speaks to the expressive dimen-
sion of punishment: it is partly because it is a means to convey the grievousness of
the wrongdoing and the blameworthiness of the agent that it cannot be meted out
to those who were morally innocent when committing the wrongdoing or, for that
matter, to those who, whilst morally responsible when offending, are mentally
unfit to stand trial.
Finally, to claim that one may justifiably punish W is one thing. To hold that
one ought to do so—that there is a duty to punish—is another. But if one is com-
mitted to the view that all individuals are under strong duties to respect one anoth-
er’s human rights, then one is also committed to the view that all individuals are
under a prima facie duty to establish and support punitive institutions. Failing
that, one’s commitment to human rights is vacuous.
7.2.3. Who May Punish?

To claim that W may or must be punished does not settle the question of who
may punish him.7 Before answering that question, we must take a conceptual-
terminological detour. In the ethics of defence, the claim that W is liable to be
killed by V standardly implies that V has the right to kill him—which means that
W, and the world at large, are under a duty not to interfere with her defensive steps
by trying to kill her. Likewise, mutatis mutandis, with the imposition of reparative
harms. Punishment is different, however, for the right to punish is not just a claim
that third parties not interfere with the imposition of the relevant harm: it is also
and primarily a Hohfeldian power to change W’s bundles of rights, duties, permis-
sions, and liabilities. In that respect, it is similar to the power to govern. Moreover,
in just the same way as the power to govern is protected by prima facie claims
against interference with its exercise (as I argued in CW, s.1.4), so is the power to
punish. When I speak, for short, of the right to punish, I shall mean the protected
power to do so.
7 For illuminating discussions of this issue, see A. J. Simmons, ‘Locke and the Right to Punish’,
Philosophy & Public Affairs 20 (1991): 311–49; C. Wellman, ‘Rights and State Punishment’, The
Journal of Philosophy CVI (2009): 419–39. For a concise Hohfeldian analysis of the right to punish,
which I endorse here, see A. Chehtman, The Philosophical Foundations of Extraterritorial Punishment
(Oxford: Oxford University Press, 2010), esp. ch. 2.
Punishment 179
Now, ex hypothesi W has culpably violated human rights of V’s and thus has
forfeited his right not to be punished. At the very least, then, V herself has the right
to punish him—as a way to openly condemn his grievous wrongdoings towards
her and restore herself as his moral equal. Moreover, her interest in his being pun-
ished is protected not just by a right to punish but also by a power to transfer her
right to third parties.8 Finally, other parties also have the right to punish W, as a
means to express their commitment to V’s, and indeed anyone else’s, human rights.
Some might dispute that individuals have the right to punish wrongdoers, and
hold on the contrary that only political institutions can have that right. On that
view, those institutions, or rather their officials, can have rights which do not super-
vene on individuals’ similar, pre-institutional rights. However, as I noted in s.1.2,
the right to govern is a protected fiduciary power (for short, right) which public
officials exercise on behalf of the individuals who are subject to their directives, and
which supervenes on and flows from the relevantly similar rights held by those
individuals. To the extent that public officials have the right to punish offenders,
then, they have it in virtue of the fact that individuals themselves have it.
That said, there are good reasons to hold that individuals ought not to exercise
their right to punish but instead ought to entrust it to public officials—in other
words, that the moral right to punish should be turned into a legal right and
devolved to relevant institutions. It is precisely because officials’ directives provide
agents with better conditions for the protection of their moral rights and the ful-
filment of their moral duties than they would have in the absence of any political
institutions or than if they were included in some or other feasible political insti-
tution that agents have a reason for regarding those directives as binding. Victims
(indeed, anyone who has a stake in ensuring that human rights do not go unpun-
ished) have the right to mete out punishment and are under a duty to mete it out
justly. There are overwhelmingly good reasons to think that suitably designed
political institutions are likely to do a better job than private individuals at ena-
bling the latter to exercise that right and fulfil that duty. For political institutions,
unlike private individuals, have the material and human resources to set up judicial
structures which can determine in a fair, transparent, impartial way whether W
culpably violated V’s rights. Unlike individuals, moreover, political institutions are
more likely to have the authority publicly to denounce W’s grievous wrongdoing
on behalf of all. Finally, recall that punishment ought not to express anger, repug-
nance, and more generally negative emotions at what war criminals did, partly out
of respect for the latter, partly to ensure that the punishment is both fitting and
proportionate. Yet it is sometimes entirely appropriate for victims to experience
those emotions, in which case they would not be able to mete out justified punish-
ment. Public officials by contrast are in a better position to convey detached blame.
With the bare elements of a justification for punishment in place, we are now in
a position to conclude, at a fairly high level of abstraction, that violations of human
8 Likewise, one’s interest in not being killed unwarrantedly is protected by a right to kill one’s
attackers in self-defence as well as a power to transfer that right to putative rescuers. See Cosmopolitan
War, s.5.2.
180 Punishment
rights warrant punishment, that those who commit those crimes with the requisite
mental state are liable to be punished, but that there might be countervailing con-
siderations which dictate against punishing them. Furthermore, the right to pun-
ish is vested first and foremost in victims and then in parties who have a stake both
in preventing the occurrence of such wrongdoings and in communicating their
seriousness. Finally, when punishment is morally justified all things considered, it
is also morally mandatory all things considered. In the remainder of this chapter,
I develop those points into a defence of punishment for war-related crimes.
7 . 3 . WA R - R E L AT E D C R I M E S
In s.7.2, I averred that if one kind of wrongdoing warrants punishment, human

rights violations certainly do. In this section, I offer an account of which war-
related wrongdoings ought to be regarded as criminal offences in the limited sense
(at this juncture) that their perpetrators are liable to some form of punishment.
According to international criminal law, whose foundational texts are the judge-
ments reached by the International Military Tribunal at Nuremberg and the 1998
Rome Statute establishing the International Criminal Court, the following four
categories of wrongdoings fall within the remit of war crime tribunals: ad bellum
crimes against peace, in particular the crime of unwarranted aggression against a
political community’s territorial integrity and sovereignty; violations of the laws of
war in bello, notably the deliberate killing, torture, and rape of innocent civilians
and prisoners of war, and the use of weapons of mass destruction; crimes against
humanity committed in the course of the war; genocide.9
Crimes against peace and crimes against the laws of war have proved relatively
easy to define, but crimes against humanity have resisted consensual treatment.
According to the Rome Statute, a crime against humanity is an act, such as murder,
torture, rape, or enslavement (including sexual slavery) which is ‘committed as part
of a widespread or systematic attack directed against any civilian population, with
knowledge of the attack’ (art. 7.1). The widespread or systematic attack, in turn,
must be ‘pursuant to or in furtherance of a state or organised policy to commit [it]’
(art. 7.2(a)). Some crimes against humanity, such as religious persecution or the
crime of apartheid, are attacks on individuals in virtue of their group membership;
others, such as forced population displacements, are not. Some are committed by
the state or a political organization, or with the acquiescence thereof, whilst others,
9 Larry May’s three volumes provide a useful map of what has become an enormous body of liter-
ature. See L. May, War Crimes and Just War (Cambridge: Cambridge University Press, 2007); L. May,
Crimes against Humanity: A Normative Account (Cambridge: Cambridge University Press, 2005);
L. May, Aggression and Crimes Against Peace (Cambridge: Cambridge University Press, 2008). The
term ‘crime against humanity’ appears as early as the eighteenth century, notably in Voltaire’s
Dictionaire Philosophique (in the entry ‘Crimes and Offenses’), and was used in the context of war
during the Second World War (in condemnation of Germany’s war of aggression, and of the Armenia
genocide). See Bass, Stay the Hand of Vengeance, chs. 2–3, and Schabas, Unimaginable Atrocities, 6–7,
51–2; Voltaire, Philosophical Dictionary. ed. J. Morley and T. Smollett (New York: E.R. DuMont,
1901 [1764]).
Punishment 181
such as enslavement in general and sexual slavery in particular, need not be. Finally,
whereas violations of the laws of war are war-related crimes by definition, crimes
against humanity can be carried out within or outside a war.
In the face of such eclectic categorization, it is hard to get a sense of what it
means to commit a crime against humanity. As a first cut, these are crimes which
ought to shock the conscience of all human beings (for short, humankind) because
they destroy, or at any rate seek to destroy, that which is specifically and nobly
human in their victims (notably, their self-respect and dignity) and are committed
in defiance of that which is or at any rate ought to be specifically human in their
perpetrators. As such, they embody a failure to respect and protect that in us which
separates us from non-human animals and which we should cherish—namely, the
capacity of rational and moral agency, the ability reflectively to engage in social
relationships of various kinds, the ability reflectively to experience both positive
and negative emotions such as anger, jealousy, hatred, empathy, and kindness, and
to decide whether and when it is appropriate to act on those emotions. Murder,
torture, persecution, rape, enslavement, indeed severe and persistent abuse, as well
as deliberate failure to save lives out of callous indifference, at best seriously impair
and at worst wholly destroy their victims’ ability reflectively to direct their life in
morally and emotionally appropriate ways; in their sheer horror, moreover, they
display a failure in their perpetrators to do what humans are uniquely (as far as we
know) capable of doing, namely choose to treat one another with reflective com-
passion and kindness.
So construed, crimes against humanity are not crimes which perpetrators com-
mit, not just against their victims directly, but against the whole of humankind.
Rather, they are crimes against humanity understood as that which is specifically
human.10 If this is correct, we should endorse a more expansive account of crimes
against humanity than are found in recent literature. In particular, the dehuman-
izing treatment of combatants, be it by their own leaders or the enemy, also counts
as a crime against humanity. In fact, some categories of combatants are already
re-classified as ‘civilians’, such as unarmed prisoners of war and peacekeeping forces
(see, esp. art. 7.2 of the Rome Statute) and it makes sense, therefore, to regard
them as victims of crimes against humanity if that which is done to them is similar
to treatment meted out to civilians and condemned as such a crime. My larger
point, however, is that if dehumanization is at the core of a crime against human-
ity, whether the crime targets an agent in uniform who is still fighting or someone
who is not is irrelevant: what matters is what the crime actually does to him.11
10 Or, as Massimo Renzo puts it, against humanness. I use the word ‘humanity’ to denote ‘human-
ness’ and the word ‘humankind’ to refer to all human beings. He too develops an expansive account
of the notion of crimes against humanity, with which I am very much in agreement. See M. Renzo,
‘Crimes Against Humanity and the Limits of International Criminal Law’, Law and Philosophy 31
(2012): 443–76.
11 See, e.g., the case of Argentinian conscripts who were tortured by their own officers during the
Falklands War, and demands that this should be regarded as a crime against humanity. See http://
news.bbc.co.uk/1/hi/8373942.stm (accessed on 16/12/14). I am grateful to A. Chehtman for drawing
this case to my attention.
182 Punishment
If this is correct, we should not stipulate by definitional fiat (as extant works
are sometimes wont to do) that such crimes can only be committed by or with the
acquiescence of political organizations; nor should we stipulate that they constitute
rights-violations on groups of individuals; nor should we define them as attacks
against individuals on the basis of some group feature or other. On my account,
thus, Soviet officials, to the highest level, repeatedly committed crimes against
humanity when they seized crops from Ukrainian farms in the late 1920s–early
30s, as a result of which millions died. On one interpretation of the famine, the
Soviet authorities intended to exterminate the kulaks as a class. On another inter-
pretation (which many would regard as excessively generous), the Soviet authori-
ties regarded those deaths as a side effect of a policy meant to speed up the
industrialization of the USSR thanks to grain exports. Either way, they showed at
best extraordinarily callous disregard for individual suffering as such and it is that
which makes their wrongdoing a crime against humanity. It is in virtue of that very
same feature that those officials can also be charged with crimes against humanity
for sending people to the gulag merely for having the audacity to think for
themselves.12
In sum, and to put the point in the language I used in s.1.2, crimes against
humanity are violations of basic human rights, period. This does not occlude the
fact that there is a morally relevant difference between, e.g., a belligerent using rape
as a systematic policy of terror, as in Bosnia, and a platoon of combatants gang-
raping civilians of their own initiative—any more than regarding a one-off killer
and a serial killer as both guilty of murder occludes the fact that there is a morally
important difference between them. But it does have the merit of not drawing
arbitrary distinctions between civilians and combatants, and between victims of
state policy and victims of combatants acting randomly.
Importantly, not all war-related crimes are crimes against humanity: some
war-related crimes consist in violating non-basic human rights (for example, as
when a bombing campaign causes economic hardship and brings civilians below
the sufficiency threshold, but does not result in starvation). One may wonder what
hinges, normatively speaking, on the classification of a crime as a crime against
humanity. In international criminal law, crimes against humanity by definition
may be prosecuted by international institutions. However, as we shall see in s.7.6,
12 On the Ukrainian famine, see, e.g., R. Conquest, The Harvest of Sorrow (London: Hutchinson,
1986). On the Gulag, see A. Applebaum, Gulag—A History (London: Penguin Books, 2004). For a
classic account of crimes against humanity, see Arendt, Eichmann in Jerusalem. For more recent
accounts, see, e.g., D. Luban, ‘A Theory of Crimes against Humanity’, Yale Journal of International
Law 85 (2004): 85–168; R. Vernon, ‘What is a Crime Against Humanity?’, Journal of Political
Philosophy 10 (2002): 231–49; M. A. Drumbl, Atrocity, Punishment, and International Law
(Cambridge: Cambridge University Press, 2007). The main difference between Luban’s and Vernon’s
accounts on the one hand, and mine on the other hand, is that they believe that a crime against
humanity must by definition be committed by a state or a political institution, or agents acting on
behalf thereof and pursuant to their plans. On their view, what separates out this particular kind of
crime from others, such as war-related crimes, is that it represents a perversion of what states, or polit-
ical institutions, are meant to do—to wit, exercising their power to protect people, not to dehumanize
them. Dumbl, for his part, argues that crimes against humanity are sui generis and require a different
penology.
Punishment 183
all war-related crimes fall within the jurisdiction of humankind as a whole, via the
requisite institutions. Accordingly, the notion of crime against humanity—whether
construed widely as I do here or narrowly as does orthodoxy—does not have dis-
tinctive jurisdictional mileage. All that it does is identify the most grievous of pos-
sible crimes, which, other things equal and taking into account considerations
of (e.g.) feasibility, would warrant the harshest punishment. That really is all there
is to it.
Whether a war-related crime ought to be classified as a crime against humanity
or not, a comprehensive account of such crimes needs expanding in the light of the
just war requirements. A war is just, you recall, if and only if (a) it has a just cause,
where a just cause consists in the violation, backed by the threat of lethal force, of
some party’s human rights; (b) it is a proportionate response to the injustice which
the belligerent has suffered; (c) it is not fought and won through the deliberate and
indiscriminate targeting of innocent non-combatants; (d) it stands a reasonable
chance of succeeding by military means which do not breach the requirements of
proportionality and discrimination; and (e) it is the only way to pursue the just
cause. Those requirements yield the following account of war-related crimes.
First, consider breaches of the just cause requirement. A party has a just cause
for war in so far as its members are subject to violations of their human rights
backed by lethal force or the threat thereof. The crime of aggression is the paradig-
matic case of such a breach. However, those breaches also include acts committed
by a regime against its own people—which provide third parties with a just cause
for a war of humanitarian intervention—and failures to respect subsistence
rights—which provide rights-holders with a just cause for waging a subsistence
war. Turning to intrastate violations, there is every reason to regard them as a crime
of aggression when they threaten a people, or subset thereof, in their communal
integrity. To claim, on the contrary, that this regime would be guilty of the crime
of aggression if it attacked a foreign political community but is not guilty of that
crime if it attacks a subset of its population is to confer on political borders a nor-
mative weight which they do not have at the bar of cosmopolitan morality.13
Moreover, even if it is inappropriate to regard intrastate violations of human rights
as constituting a crime of aggression, it is certainly appropriate to regard them as a
war-related crime, which at first sight warrants punishment. As for violations of
subsistence rights, some might think that they are not to be classified as war-related
crimes—unlike acts of unwarranted military aggression—since they do not take a
military form. But this would be too quick: for to the extent that such rights vio-
lations consist in or are backed by lethal force, and to the extent that they can be
described as tantamount to an aggression (as I argue in CW s.3.2.2), there is no
reason to exclude them from the category of war-related crimes in general.
Second, consider the requirements of proportionality (d) and necessity (e).
Suppose that although B’s leadership and military threatened, with lethal force, the
13 The point applies not simply to rights violations which give rise to just causes for war, but to
war-related crimes in general. It is in tension with article 8 of the Rome Statute of the International
Criminal Court, which differentiates crimes committed in an international conflict and crimes com-
mitted in a civil war.
184 Punishment
human rights of citizensA, diplomatic routes were available to A; or suppose that

the war inflicted disproportionate harm both to combatantsB and citizensB. Ex
hypothesi, A had a just cause but acted wrongly towards those individuals. At the
very least, its combatants and leaders violated the rights of those of B’s members
who were innocent of their own regime’s wrongdoing, and are thus liable to pun-
ishment—even though members of B who did commit the wrongdoings which
provided A with a just cause for war are also liable.
Third, we saw in ch. 2 that continuing at t2 with a war which one was justified
in starting at t1 is sometimes unjust. If so, it constitutes a wrongdoing whose per-
petrators are prima facie liable to punishment. For if unwarrantedly starting a war
of aggression at t1 is a crime, so is continuing with a war which is now unjust at t2,
since at t2, that war has now become a war of unwarranted aggression. To claim
otherwise is tantamount to allowing the timing of an unjust aggression (as being
unjust ab initio, or as becoming unjust) to determine aggressors’ liability to pun-
ishment, which is morally arbitrary.
Fourth, in ch. 3, we examined peacekeeping operations and military occupation
in the transition from war to peace. Human rights violations which are regarded as
war-related crimes if committed before or during hostilities should be similarly
regarded if they are committed by or against peacekeeping and occupying forces
qua such forces, or by belligerent parties notwithstanding the presence of peace-
keepers, or by local agents who collaborate with unjust occupiers. The point is
worth stressing in the light of the exactions committed by occupying armies dur-
ing the Second World War, the US-led coalition in Iraq, UN forces in some of
their missions, or the Israelis in Gaza and the West Bank.14 Moreover, continuing
with an unjust occupation should also be regarded as a war crime, in just the same
way as an unwarranted aggression.
Fifth, suppose that (as I argued in CW, ch. 5), some party is under a duty to
wage a war of humanitarian intervention yet fails to do so. One may wonder
whether that failure should be regarded as a crime of war. If (as I also argue else-
where15) morally wrongful failures to provide assistance ought, under some cir-
cumstances, to be turned into criminal offences, then so should a morally wrongful
failure to go to war.
Sixth, according to what one may call the orthodox account of punishment for
war-related crimes, killing combatants who fight for an unjust cause is legally and
morally permitted, and thus not a punishable offence. But the neoclassical ethics
14 For exactions committed by the German forces in Europe during the Second World War, pick
any account of the war. One of the worst examples of exactions committed by American combatants
in Iraq is the Abu Ghraib scandal of 2004, where combatants on duty at the prison tortured, abused,
and humiliated Iraqi detainees (see, e.g., http://edition.cnn.com/2013/10/30/world/meast/iraq-
prison-abuse-scandal-fast-facts/ accessed on 15/01/2016). For examples drawn from Israel’s occupation
of the West Bank and Gaza, see, e.g., https://www.amnesty.org/en/countries/middle-east-and-north-
africa/israel-and-occupied-palestinian-territories/ (accessed on 15/01/2016). For UN cases, notably
involving sexual exploitation and abuse of civilians in the Central African Republic, see, e.g., http://
www.un.org/News/dh/infocus/centafricrepub/Independent-Review-Report.pdf (accessed on
15/01/2016).
15 C. Fabre, Whose Body is it Anyway? Justice and the Integrity of the Person (Oxford: Oxford
University Press, 2006), ch. 2.
Punishment 185
of war killing which inform this book prescribes a radical departure from ortho-
doxy. For on the neoclassical account, combatants who participate and fight in an
unjust war or phase of war are liable to be killed, whereas combatants who partic-
ipate and fight in a just war or phase of war are not liable. In keeping with the
orthodox account, killing or maiming unjust combatants should not be deemed a
criminal offence; at variance with it, however, killing just combatants should—in
just the same way as in domestic contexts killing someone without justification
should be a criminal offence.
7 . 4 . WA R C R I M I N A L S
The claim that war-related crimes warrant punishment does not tell us who,
exactly, of all the participants in such crimes, is liable to punishment. In this sec-
tion, I argue that rank and file combatants, and not just leaders and generals, are
liable to punishment for participating not just in what the laws of war labels war
crimes and crimes against humanity, but also and more controversially, in the
crime of aggression.
7.4.1. Officers, Combatants, Leaders, Citizens: The Problem of

Participation in War-Related Crimes
As we saw in s.7.2, an agent W is liable to punishment for a wrongdoing W only
if he has committed W (the conduct—or actus reus requirement) and if he was
morally culpable for so doing (the fault—or mens rea—requirement).
War-Related Crimes and the Actus Reus Requirement

Let us apply that point to war-related crimes, starting with the actus reus. Some
war-related crimes are committed by combatants each acting on their own, as
when a combatant deliberately kills or tortures some innocent civilians or prison-
ers of war in his care. In those cases, it is easy to hold that the combatant meets the
conduct requirement. However, far many more crimes are committed by combat-
ants acting jointly, as when an entire platoon takes part in the deliberate killing of
civilians, or when a regiment mounts an attack on an illegitimate target, or when
a whole army unwarrantedly invades another country. More widely still, unjust
wars are waged by armies with the assistance, forced or voluntary, of large sections
of civilian populations. Finally, crucial decisions in war are taken by agents—the
belligerents’ leaders—who do not themselves directly harm the enemy. The defence
of punitive liability which I sketched earlier must thus account for two features of
war-related crimes. First, they are committed by a multitude of agents whose
respective individual contributions differ widely. Second, leaders seemingly com-
mit a lesser wrongdoing and thus fare comparatively well—relative to ordinary
combatants—at the bar of the conduct requirement (they are not the ones who kill
and rape). Yet they are the most likely to intend the commissions of the harms
186 Punishment
which they are instigating or, at least, the most likely to be reasonably expected to
know that those harms will ensue as a result of their decisions; they are also the
most likely to be reasonably expected to know that those harms constitute war-
related crimes. Accordingly, they fare comparatively badly at the bar of the fault
requirement. By contrast, ordinary combatants commit the more grievous wrong-
doings (they kill and rape) and fare comparatively badly—relative to leaders—at
the bar of conduct, and yet they often do not intend to commit the larger harm of
which many of their individual acts are a part; or they are the least likely to be
reasonably expected to know both that they are committing the larger harm and
that in so doing they commit war-related crimes.
Taken together, those two features of war-related crimes generate the following
dilemma, in which one cannot honour both actus reus and mens rea.16 If one con-
fers greater weight on conduct when determining liability to punishment, leaders
will be punished rather lightly compared to ordinary combatants and civilians,
which, given that they stand in a relationship of authority to them, seems unjust.
But if one confers greater weight on fault, combatants will be punished rather
lightly compared to their leaders, which, given that their acts (killing, raping, etc.)
are more directly harmful to their victims, also seems unjust.
How then can we best justify the intuition that leaders, some combatants, and
indeed civilians who assist them are liable to some form of punishment for war-
related crimes? The task is relatively simple for crimes such as the deliberate targeting
of civilians or the deliberate torture of prisoners of war. When those crimes are
committed by several combatants jointly, they can each easily be described as tak-
ing part in a wrongful collective enterprise. Thus, combatants who killed civilians
during the My Lai massacre are each properly described not as ‘each killing a few
civilians’ but as ‘taking part in a massacre’, which might carry out a stiffer sentence
than ‘ordinary murder’. Moreover, if the massacre is ordered by a commanding
officer who did not himself take part in it, or if it happens without his orders but
thanks to his negligent failure to intervene, it would be appropriate to hold him as
accountable for it as his men: for to the extent that they acted upon his orders or
that he could have stopped them but did not, his contribution to the death of
those civilians is highly significant. Likewise, mutatis mutandis, with torture.
Those are micro-threats—which together form the macro-threat of the war
itself. In view of the vast numbers of agents involved and how varied their individ-
ual contributions, establishing individual liability to punishment tout court, and to
which punishment in particular, for that macro-threat is harder. Suppose that A’s
army wrongfully invades B’s territory. Combatants and civilians on A’s side who
contribute to that invasion are properly described as participants in that war crime.
However, combatants actually carry out the invasion, whereas civilians’ participa-
tion typically takes the form either of ordering and planning the invasion
(as high-ranking officials), or of providing the material means for its execution
16 See Larry May’s three-volume treaty on punishment for war wrongdoings (see fn 9), which in
effect seeks to provide a solution to that dilemma. His solution consists in privileging the mens rea to
the detriment of the actus reus. I set out my reasons for disagreeing with him below.
Punishment 187
(as taxpayers, weapons factory workers, etc.) Allowing a small part of one’s income
to be directed, through taxation, towards the war effort is not as grievous a wrong
as planning, let alone ordering, an invasion. At the same time, the former is still a
wrong. Moreover, among combatantsA themselves, some actually kill a number of
combatantsB who were not liable to be killed, while others, e.g., carry out recon-
naissance and intelligence-gathering missions as preparation for the invasion. In
this particular case, killing non-liable agents is more grievous a human rights vio-
lation than spying. Carrying out reconnaissance missions in preparation for an
invasion, however, is wrong in a way that spying to satisfy one’s curiosity is not.
And so on.17
Aggressive war is not the only macro-threat individual liability for which is hard
to establish. So is genocide. As the ICTY recognized in its judgement against
Goran Jelisic, an individual acting on his own who kills (e.g.) members of a par-
ticular ethnic group can be charged with genocide.18 However, such a crime usu-
ally cannot occur without the participation of huge numbers of agents—from
those who instigate it at the highest level, to those who carry out its constituent
acts of killing, to the camp guards who oversee the prisoners before they are exe-
cuted, to the clerks who maintain records, the train drivers who ferry victims to
their final destination, the manufacturers who supply the gas, bullets, or machetes
with which they are killed, the citizens whose taxes fund all of the above, and so
on. In order to apportion liability to punishment among all those agents, we need
to refine our understanding of a collective enterprise and participation therein.
In CW, drawing on Christopher Kutz’s account of complicity, I defended the
following view: to the extent that a combatant or a civilian intentionally contrib-
utes in various ways to an unjust war and that he can reasonably be expected to
know that other agents, combatants, and civilians are also contributing and share
an understanding of what they are doing, that combatant or civilian can be said to
share individual responsibility for the occurrence of the war—as can any other
combatant or civilian in a similar epistemic and agential situation.19 Irrespective of
their specific individual contributions, those agents can under those conditions be
described as taking part in the war or phase of war. If the war or phase of war is
unjust—in the sense that they consist in the violation of some other party’s human
rights—all participants are liable to some form of punishment. The phrase ‘some
form of punishment’ is crucially important. Participation in an unjust war, where
17 For discussions of the liability of high-ranking civilian officials for the crime of aggression, with
reference to the Nuremberg trials, see S. Levinson, ‘“Responsibility for Crimes of War”’, Philosophy &
Public Affairs 2 (1973): 244–73. See also Walzer, Just and Unjust Wars, ch. 18; May, Aggression and
Crimes Against Peace.
18 See Jelisic (IT-95-10-T), Judgement, 14/12/1999, art. 100. At http://www.icty.org/sid/7712
19 Cosmopolitan War, 29–31. See C. Kutz, Complicity: Ethics and Law for a Collective Age
(Cambridge: Cambridge University Press, 2000); as well as C. List and P. Pettit, Group Agency—The
Possibility, Design, and Status of Corporate Agents (Oxford: Oxford University Press, 2011). For a rich
and nuanced account of different kinds of complicity, see Goodin and Lepora, On Complicity and
Compromise. The account of complicity provided by Kutz strikes me as more helpful than criminal law
doctrines of complicity and, for that matter, of conspiracy (not least because those doctrines get mired
in the difficulties of distinguishing co-principals from accomplices).
188 Punishment
participation involves some contribution, constitutes the conduct in virtue of

which they are liable to punishment simpliciter. Which kind of punishment
depends on what they actually did in pursuit of that collective venture.
War-Related Crimes and the Mens Rea Requirement

I shall return to this issue in s.7.5. Beforehand, we need to attend more closely to
the frame of mind in which combatants must be in order to be liable to punish-
ment. As I noted in s.7.2.2, liability is subject to two conditions: the wrongdoer
was in full possession of his intellectual and physical capacities when committing
the wrongdoing, and he must be in full possession of his intellectual capacities
when standing trial. The second condition precludes punishing senile or otherwise
mentally impaired wrongdoers even if they were not so impaired during the war.
For example, it precludes punishing senile old Nazis here and now.20
The first condition stands in greater need of discussion. Some war-related crimes
are isolated individual incidents, as when a combatant rapes a civilian. Most such
crimes however are collective wrongdoings which are committed pursuant to a war
by participants therein. The question, then, is whether wrongdoers are liable to
punishment only if they intend to contribute to the war or to a specific plan within
the war by so acting, or whether it is enough that they actually know that they are
contributing to the war or to those plans, or whether it is enough (as well as nec-
essary) that they can reasonably be expected to know on the basis of the available
evidence that they are contributing. To illustrate, consider those crimes against
humanity which differ from the ‘mere’ deliberate killing of innocent civilians in
that they are part of a systematic and widespread plan to persecute a particular
group on the basis of disability, ethnicity, gender, religious affiliation (or lack
thereof ), or sexual orientation. Those crimes are deemed to be morally worse than
non-discriminatory crimes against humanity precisely because they are discrimina-
tory. And yet, those who commit such crimes do the same things to their victims
as those who, e.g., ‘merely’ deliberately rape and kill innocent civilians. The ques-
tion, thus, is whether they must act with the intent to persecute a group in order
to be liable to punishment for a crime of this severity, or whether it is enough that
they actually know that their acts are part of a larger discriminatory and wrongful
plan, or indeed whether it is enough that they should reasonably be expected to
have such knowledge. Likewise, mutatis mutandis, with crimes against peace.
Generally, it strikes me as plausible to hold that W meets the mens rea require-
ment (and is liable to punishment) only if he meets either one of the following two
conditions. Either he intends to do something which he could reasonably be
expected to know constitutes a crime; or he could reasonably be expected to know
that he was doing something which he could also reasonably be expected to know
constitutes a crime. On this disjunctive view, neither intentions (to violate rights)
20 On which point I agree with Hampton’s discussion of such cases. See Hampton and Murphy,
Forgiveness and Mercy, 135.
Punishment 189
nor actual knowledge (of the fact that one acts in a way that, one also knows,
constitutes a rights violation) are necessary for mens rea.
Indeed there are good reasons to reject the views that intentions or actual knowl-
edge are necessary requirements for fault and therefore for liability to punishment.
The view that the intention to persecute or invade is necessary implies that those
who are only grievously negligent are not liable. This is hard to accept. In fact,
domestically, individuals who take part in what they know is a criminal conspiracy
though do not intend to do so are liable to punishment—and rightly so. Pending
arguments to the effect that there is something radically different about the cir-
cumstances of war, there is no reason to hold war participants to a more lenient
fault requirement than ordinary wrongdoers.21
We should also reject the view that actual knowledge, of what one is doing and
of the fact that one is thereby committing a crime, is necessary for fault. For on that
view, actors who deliberately shield themselves from the knowledge of what they are
actually doing are not liable. Again, in domestic settings, the fact that I do not know
that I am taking part in, e.g., a conspiracy to murder (‘But he did not tell me why
he wanted me to give him a gun!’) will not and should not shield me from prosecu-
tion, unless I can show that I could not reasonably be expected to have known: what
does the work is reasonable expectation of knowledge. Agents who generally are of
sound mind can be expected to take steps so as to inform themselves of what it is
that they are doing, to the best of their abilities. To reiterate, pending arguments to
the contrary, the fact that they wear the uniform, or more widely take part in a war,
should make no difference to the standards to which we hold them.22
7.4.2. Three Objections

The Nuremberg Defence
And yet, combatants are generally not treated as garden-variety murderers, even
when we think (rightly) that they kill wrongfully. Nor are civilians who participate
in unjust wars and unjust killings therein treated as garden-variety accomplices in
murders. I can think of three arguments to the effect that those agents should not
be treated on a par with ordinary criminals. The first holds that combatants act
under orders and that only those who gave those orders (whether combatants or
civilians) are liable, even if the order is immoral and even if the agent who receives
the order satisfies the requirement of mens rea. In brief, however, the so-called
21 I thus strongly disagree with Larry May’s account of punishment for crimes against humanity
(indeed, for punishment for war wrongdoings in general), which requires intent to persecute. I shall
have more to say on this in s.7.4.2 below. (See in particular May, Crimes against Humanity, 109.) For
an effective rebuttal of his argument, see D. Luban, ‘Fairness to Rightness: Jurisdiction, Legality, and
the Legitimacy of International Criminal Law’, in J. Tasioulas and S. Besson (ed.), The Philosophy of
International Law (Oxford: Oxford University Press, 2010), 582–3. For an illuminating discussion of
this issue as it has been dealt with by various international criminal tribunals, see Schabas, Unimaginable
Atrocities, ch. 5.
22 For a thorough discussion and rejection of the view that ignorance itself should not be regarded
as a defence against criminal charges, see A. Ashworth, ‘Ignorance of the Criminal Law, and Duties to
Avoid it’, The Modern Law Review 74 (2011): 1–26.
190 Punishment
‘superior orders’ defence was rejected at Nuremberg and is not admitted in inter-
national criminal law for reasons which are well-known and which I will not
rehearse here—other than to stress, once more, that combatants are not automa-
tons who are exonerated from the burdens of autonomous moral agency once they
put on their uniform.23
The Duress Objection

The second objection holds that, to the extent that combatants commit war-related
crimes whilst not in full possession of their mental capacities, they are not liable to
punishment. The objection has a point. As we saw, it is not a sufficient condition
for a wrongdoer to be liable to punishment that he should commit a particular
crime: it is necessary that he should meet the fault requirement. By implication,
combatants who commit war-related crimes while unable to exercise autonomous
agency (an extreme example of which are child combatants who are subject to
severe physical, psychological, and sexual abuse at the hands of their commanders)
are not liable to punishment. A harder case is that of combatants who are ordered
so to act under extreme duress, for example at gun point, or under threat that their
relatives will be killed otherwise. In domestic contexts, however, being told at gun
point to kill or severely maim an innocent person does not render the agent wholly
immune from criminal charges—and rightly so, for the agent nevertheless does
commit a grievous wrongdoing. A fortiori, duress provides no justification for
committing even minor wrongdoings in domestic contexts either. Likewise, then,
in war—not just for acts of killing, but for other kinds of and even minor war
crimes such as (e.g.) unjustly keeping as prisoners of wars combatants who ought
to remain free, crossing over the border as part of an unjust aggression, or tighten-
ing screws on weapons.
However, the claim that the agent is liable to some punishment—in the sense
that punishing him would not be wrongful—is compatible with the view that, in
the light of the duress under which he was placed, he had an excuse for so acting,
which should be taken into account when deciding which punishment he should
suffer. The often-discussed case of Dragan Erdemovic, a Croat fighting with the
Yugoslav National Army in the 1990s and indicted by the International Criminal
Tribunal for the former Yugoslavia (ICTY) for his participation in the massacre at
Srebrenica, illustrates the point well: Erdemovic’s claim that he would have been
shot by his officer alongside Bosnian civilians had he refused to obey the order to
kill the latter was regarded by the court as factually likely but not a justification for
killing. Erdemovic was first sentenced to imprisonment for ten years—reduced to
five years on appeal precisely on account of duress.24
23 For a rejection of the ‘superior orders’ defence, see Walzer, Just and Unjust Wars, ch. 19. For a
sympathetic take on the defence, see, e.g., May, Crimes against Humanity, 181–4. For a clear account
of the defence as it is treated in international criminal law, see, e.g., A. Cassese, International Criminal
Law, 2nd edn. (Oxford: Oxford University Press, 2008), 268–80.
24 See, e.g., May, War Crimes and Just War, 285–8; R. E. Brooks, ‘The New Imperialism: Violence,
Norms, and the “Rule of Law”’, Michigan Law Review 101 (2003): 2275–340, at 2318–21. Note that
Punishment 191
My response to the duress objection yields one further point. In s.7.4, I stressed
that in holding wrongdoers accountable for what they did, and in expressing pub-
licly the view that what they did was grievously wrong, one must also attend to the
conditions under which they acted. War is one of the most brutalizing experiences
there is, and it is not surprising therefore that combatants should report high levels
of stress, various kinds of affective disorders, and heightened aggressivity—all of
which diminish, without fully impairing, their ability properly to assess the evi-
dence available to them.25 In punishing them, one ought to take those factors into
account—not to exculpate them fully from wrongdoing (for they are morally
responsible agents and failing to treat them as such would be a mark of disrespect),
but in mitigating recognition of their agential circumstances. In some cases, this
should lead us to endorse the imposition of a reduced sentence if imprisonment
is called for. In others, and as we shall see in ch. 9, this should lead us to adopt
reconciliatory practices.
The Epistemic Objection

According to the third objection, although ordinary, rank-and-file combatants can
reasonably be expected to know that they are taking part in the wrongful killing of
innocent civilians, they cannot reasonably be expected to know that they are taking
part in a wrongful crime against peace when they are told, for example, to invade
another country.26 Nor, for that matter (proponents of that objection may be
tempted to press) can ordinary citizens be expected to know that they are support-
ing an unjust war. However, this objection can only go so far. Whilst it may hold
of conditions under oppressive regimes where neither ordinary combatants nor
ordinary citizens have access to open media, it does not hold for broadly demo-
cratic regimes. At any rate, individuals under these regimes can reasonably be
expected to do as much as they can to inform themselves of the justness of the war
(or phase of war) which they are asked to fight and support. If it turns out that they
could not possibly be expected to have known that the war or phase of war was
indeed unjust, then they should be exempt from punishment. Otherwise, they are
liable—with the proviso that the less cognizant they can reasonably be expected to
on my account of crimes against humanity, whereby such crimes ‘merely’ are violations of basic human
rights, Erdemovic was himself the victim of such a crime. For (I take it) all human beings have a basic
human right not to be forced on penalty of having their other human rights violated (here, the right
not to be killed) to murder another person. This of course does not justify what he did. (By the same
token, moreover, forcing people to fight unjust wars on penalty of similar wrongs is also a crime
against humanity. I am grateful to an anonymous reader for Oxford University Press for raising this
interesting point.)
25 A Google search conducted on 15 January 2016 with the keywords ‘ptsd in soldiers’ generated
over 1.4 million results. I won’t even attempt to list a selection of articles—both scholarly and from
the media—on the issue.
26 See, e.g., Walzer, Just and Unjust Wars, 312. There is a serious inconsistency between Walzer’s
claim that combatants are not automatons and thus cannot invoke the superior order defence in jus-
tification or excuse for committing in bello crimes and his claim that they are not to be punished for
ad bellum transgressions.
192 Punishment
be of the moral status of their acts, the more excusable their conduct, which in turn
will determine the severity of the punishment to which they are liable.
One final point. To the extent that those three objections are not deployed
against the view that common murderers are liable to punishment, they are parts
of a broader view which holds that the norms which regulate agents’ violent con-
duct in war are sui generis—different from the norms which regulate agents’ violent
conduct outside war. The neoclassical account of the ethics of war which underpins
my theory of the just peace argues, on the contrary, that agents are subject to the
same norms irrespective of the context (war, or peace) in which they kill. It does,
however, need to account for the view that, in general, combatants who return
from a war widely held to be unjust ought not to be treated as if they were murder-
ers: they ought not to be subject to prosecution, let alone sent to jail, for the
wrongful killing of enemy combatants; and they ought not to be affectively blamed
(at least by non-victims) for what they did.
In the next section, I shall explore reasons for not sending those combatants to
jail and suggest appropriate punitive sanctions for their wrongdoings. Regarding
the attribution of blame, the account of punishment I articulated earlier is of a piece
with the neoclassical view: if it is appropriate (at least on the part of non-victims)
not to blame ordinary combatants affectively but, on the contrary, to blame them
in a way that simply acknowledges their responsibility for what they did, then it is
appropriate (at least on the part of non-victims) not to blame common murderers
affectively. In just the same way as the norms for war killings are not sui generis in
relation to the norms for common murder, norms for blame attribution in the
realm of domestic criminal law are not sui generis in relation to the norms for
blame attribution in the realm of war.
7.5 . W H I C H P U N I S H M E N T ? I M P R I S O N M E N T,
LU S T R AT I O N , A N D P U N I T I V E S A N C T I O N S
7.5.1. Imprisonment
Assume, then, that some individual W is liable to punishment for war-related
crimes. But what kind of punishment, though? It is a common trope that the pun-
ishment should both fit, and be proportionate to, the crime. Fittingness and pro-
portionality are not the same. Thus, one might think that imprisonment is a fitting
punishment for assault, but that condemning someone to life imprisonment is a
disproportionate punishment for a single (non-lethal) assault. Conversely, one
might think (justifiably) that asking someone to pay a fine for killing another per-
son is not on its own a fitting punishment (though punitive damages might be
warranted, alongside a jail sentence), but that, if it were a fitting punishment,
asking someone to give their victims n per cent of their financial resources might
be proportionate.27
27 On the connections between compensation and punishment, see A. Ashworth, ‘Punishment
and Compensation: Victims, Offenders and the State’, Oxford Journal of Legal Studies 6 (1986): 86–122.
Punishment 193
What counts as fitting and proportionate punishment is one of the thorniest

difficulties which any full-blown account of punishment must tackle. I cannot do
so here. Instead, I shall take for granted that the more grievous the rights violation,
the greater the punishment—subject to upper limits set by W’s right (which he
may never be deemed to have lost) to humane treatment. Recall, however, that
even if W is liable to be punished in particular ways, it does not follow that we are
justified in punishing him all things considered. In particular, the requirement not
to undermine goods such as third parties’ interests in being protected from W
himself or from putative wrongdoers delineates conditions for justifiably punish-
ing him (see s.7.2.1). Suppose that W belongs to a powerful gang of criminals at
war against the state, and that if we punish him, his associates will launch a cam-
paign of murderous violence against business owners who refuse to pay protection
money. Punishing W might not be justified in this case, on the grounds that it
would lead to the deaths of many innocent people.
Not only, then, must punishment be proportionate to the crime itself; it must
also be proportionate in the sense that the goods it brings about must not be out-
weighed by the harm it will cause. To put the point differently, punishment, like
defensive force, is subject to both narrow and wide proportionality. Narrow pro-
portionality refers to the fit between his wrongdoing and the amount and kind of
punishment to which he is subjected, whereas wide proportionality refers to the
impact on third parties of punishing him. Thus, W is liable to punishment p only
if p is narrowly proportionate to his wrongdoing; but inflicting P on W is justified
all things considered only if p is widely proportionate.
The distinction between narrow and wide proportionality is notably and regret-
tably absent from contemporary debates about appropriate sentencing.28 In the
present context, it is absolutely crucial. To see this, consider punishment for signif-
icant contributions to violations of rights to life, bodily integrity, and property. On
the plausible assumption that they generally warrant imprisonment in domestic
contexts, one might think that they also warrant imprisonment when classified as
war-related crimes. However, there are at least three very good reasons not to
impose it other than in the clearest, most egregious cases such as massacres of civil-
ians, or the instigating and planning of unjust aggression and concomitant unjust
killings—for two reasons. First, punishing combatants merely for fighting might
provide them with incentives to continue to fight so as to avoid falling into the
hands of the enemy—at the cost of securing prospects for peace. The point is
familiar, though it is not made in domestic settings, at least for violent crimes. For
example, we do not hear it said that human trafficking should not be turned into
a criminal offence lest it should provide traffickers with incentives to commit fur-
ther rights violations in order to avoid punishment (e.g., kill victims who show
signs that they might go to the police, corrupt police officers, which in turn weakens
On proportionality, see A. Ashworth, Principles of Criminal Law, 5th edn. (Oxford: Oxford University
Press, 2006), 19–20.
28 In domestic contexts, wide proportionality refers, for example, to harms accruing to a criminal’s
dependent relatives as a result of his imprisonment. For a brief discussion of wide proportionality and
(domestic) state punishment, see Tadros, The Ends of Harm, 358–9.
194 Punishment
the legal enforcement of rights, etc.) The only reason why we might want so to
distinguish between war and non-war criminals is the sheer numbers of the former,
relative to the latter: we might be able to bear the costs of further rights violations
for the sake of punishing non-war criminals to the degree to which they are liable;
but bearing the costs of the vastly greater rights violations committed by combat-
ants desperate to avoid punishment seems morally undesirable.29
Second, as we saw in s.7.4.1, many participants contribute to unjust wars or
unjust phases therein in ways which do not warrant as severe a punishment as
imprisonment—such as the citizen who knows that parts of her taxes fund an
unjust war of aggression, the journalist who writes in favour of the war, the engi-
neer who works in dual-use facilities, and so on. In so far as imprisonment would
be disproportionate in the narrow sense, those wrongdoers are not liable to it.
Third, even if imprisonment is narrowly proportionate in some cases, it is likely
to be widely disproportionate anyway. Remember that it is only if one can be
under no reasonable doubt that the defendant committed the crime and either
intended so to act or could reasonably be expected to know what he was doing
(and the moral status of his act) that one may harm him to that extent. Given the
sheer numbers of agents involved, establishing judicial institutions which would be
able to reach a verdict on both actus reus and mens rea would be extraordinarily
costly. Assuming that financial resources could justifiably be made available to that
end, imprisoning all those found guilty would also be extraordinarily costly. And
assuming that this could be done, the human costs of so doing would themselves
be enormous, since thousands of individuals would be sent to jail who could take
part in the rebuilding of their community. To put the point in the language of
feasibility, even if it were practically feasible to put on trial, judge fairly, and
imprison every single agent who is strongly suspected of having participated in
grievous war-related crimes, the financial and human costs attendant on doing so
would detract from the pursuit of other just ends, whether post bellum ends such as
the reconstruction of war-torn communities or extra bellum ends such as the pro-
vision of non–war related essential goods and services. Partly for those reasons, this
would also jeopardize prospects for a justifiedATC peace, and it is in that sense that
it would be widely disproportionate. The wholesale imprisonment of suspected
war criminals, in other words, is morally unfeasible given scarcity of material
resources. The case of Rwanda illustrates the point well: seven years after the gen-
ocide, over 100,000 people, each suspected of having taking part in genocidal
killings, were still in jail awaiting trial.
29 Note that even if this point is correct, it does not undermine my earlier claim that war punish-
ment is not sui generis in relation to domestic punishment. In principle, we must also take wide pro-
portionality into account in both contexts. The point thus merely holds that the proportionality
calculation is likely to differ across those contexts, such that war punishment is likely to be widely
disproportionate while domestic punishment for roughly similar wrongdoings is not. By implication,
then, if, in a given pairwise comparison, the harms attendant on punishing combatants for serious
violations of human rights would be widely proportionate while the harms attendant on punishing
domestic wrongdoers such as, e.g., gang members would be widely disproportionate, punishment
would be warranted in the former case while not in the latter.
Punishment 195
The most that we can hope for, thus, is the imprisonment of high-ranking offi-
cials and civilian participants such as directors of implicated companies (e.g.,
Krupp, Flick, and IG Farben at Nuremberg) for large-scale war-related crimes, and
of those who directly killed or raped, or directly gave orders for micro-threats such
as massacres or torture. In fact, sometimes we ought not to punish those agents at
all, particularly in the context of a civil conflict. Even if we may punish them, this
leaves us with the question of what to do with other agents—many of whom (let
us not forget) did commit wrongdoings which are justifiably punished with impris-
onment within domestic contexts. In that vein, the Rwandese authorities released
most of those suspects and had them face non-adversarial community courts mod-
elled on traditional modes of conflict resolution. I shall discuss this case in some
detail in s.9.4. In the remainder of this section, I examine two punitive options—
lustration on the one hand, and financial penalties on the other hand. Both have
been available to international criminal courts, as stipulated both in the Rome
Statute for the International Criminal Court, and in the Statutes establishing the
ICTY and the ITR. Both may provide a solution to the difficulty raised by the case
of those who are liable to be imprisoned but whom we should not put in jail all
things considered, and those who are not liable to be imprisoned but who are nev-
ertheless liable to some form of punishment. In all those cases, the claim that
wrongdoers are so liable, conjoined with the claim that punishment of some kind
is in order (whether that to which they are liable or a lesser form thereof ), yields
the conclusion that their war-related wrongdoings ought to be regarded as criminal
offences—that there is a sense in which, thus, that the laws of war should track the
morality of war.
7.5.2. Lustration
In some cases it is appropriate to deprive those individuals of some of their civil
and political rights—what is sometimes called lustration.30 But we need to distin-
guish between different kinds of lustration, and justifications for it. Political lustra-
tion consists in depriving some agent of his right to exercise office, whereas civil
lustration consists in, e.g., barring them from exercising their former profession or
working in the public sector, or denying them the pension and other economic
privileges they had been granted under the previous regime. Furthermore, there is
a difference between lustrating someone on grounds of incompetence, and lustrat-
ing them punitively. The former is uncontroversial—not least in domestic, peace-
time contexts, let alone in the transition from war to peace. In fact, as I argued in
s.3.3.1 and will reiterate in s.8.3, officials and citizens of an unjust belligerent can
sometimes be deemed to have forfeited their rights to exercise powers of govern-
ment by acting in such a way during the war that they cannot be trusted post
30 C. Offe, ‘Disqualification, Retribution, Restitution: Dilemmas of Justice in Post-Communist

Transitions’, Journal of Political Philosophy 1 (1993): 17–44. See also Y. Chiu, ‘Liberal Lustration’,
Journal of Political Philosophy 19 (2011): 440–64. For lustration in Europe following the Second
World War, see A. Rigby, Justice and Reconciliation After Violence (Boulder, Co.: Lynne Riener: 2001),
ch. 1.
196 Punishment
bellum to govern justly. A similar claim can be made about combatants who have
killed in pursuit of an unjust war, or about police officers and agents of security
services who have taken part in atrocities against fellow members or occupied pop-
ulations: to the extent that their deeds during the war gives their opponents good
reason to believe that they cannot be trusted with exercising their profession justly,
they may justifiably be expelled from the army or the police, without being
wronged, once the war is over—either by a transitional administration or by a new
regime constituted by opponents to the old regime.
My focus here, however, is on punitive lustration. To see the difference between
those two kinds of lustration, consider the well-known cases of German and
Austrian musicians who performed in front of Nazi officials between 1933 and
1945, were hauled in front of de-Nazification commissions after the war and (in
some cases) were banned from performing for a while. Bad musicianship was not
and could not have been the basis for preventing them from working as profes-
sional musicians. In such cases, one must look elsewhere for a reason permissibly
to disqualify. One such reason, and a powerful one at that, is precisely the expres-
sive dimension punishment. Temporarily and publicly preventing wrongdoers
from exercising the profession thanks to which they were able to ingratiate them-
selves to the unjust belligerent and through which they helped serve its unjust
ends, in however minor a way, is a way to express one’s commitment to justice and
peace, and one’s condemnation of what they did. Moreover, in cases such as these,
involving a public profession for which one is normally lauded, lustration serves
to communicate that there is in fact rather little to be applauded in those
individuals.31
Note that this is not a proposal in favour of lustrating all agents who have com-
mitted any kind of war-related crimes. Dismissing someone from their occupation
does not require evidence beyond reasonable doubt of their guilt. But it does
require some evidence thereof, which in turn requires proper procedures. Wholesale
dismissal, if done fairly with proper consideration of each individual’s guilt, is
likely to be as costly and as detrimental to prospects of a justifiedATC peace as
wholesale criminal punishment, and therefore as widely disproportionate not least
because dozens of thousands of unemployed young men are a recipe for the
resumption of violence. The comprehensive purges which the US-led Coalition
Provisional Authority conducted in Iraq within the Baathist Party, the armed
forces, and the police, following the overthrow of Saddam Hussein, are a case in
point.32 Rather, the proposal rests on the claim that those individuals would not
be wronged if dismissed. Moreover, it should be restricted to cases where there is
not enough evidence to indict the wrongdoers in a criminal court, but enough
evidence to rescind their employment contracts. Finally, with respect to those who
are dismissed, it is an important part of post-war reconstruction efforts that residents
31 I owe this latter point to V. Tadros. The best-known cases are Furtwängler’s and Karajan’s. See
R. Osborne, Herbert von Karajan—A Life in Music (London: Random House, 1998), chs. 20–1. Both
Furtwängler and Karajan were cleared of all criminal charges.
32 See Allawi, The Occupation of Iraq, esp. chs. 8 and 9; J. Meierhenrich, ‘The Ethics of Lustration’,
Ethics & International Affairs 20 (2006): 99–120.
Punishment 197
of war-torn communities should ultimately enjoy prospects for a flourishing life.

This applies not just to victims, but also to wrongdoers whose crime was not such
as to warrant a long jail sentence: the claim that a combatant may justifiably be
dismissed from his army as punishment is compatible with the view that he should
be able to make a living by taking on an occupation which, whatever it is, will not
afford him opportunities to resume fighting, and in such a way as to lack incentives
for embarking on other criminal activities.
Thus, subject to those caveats, lustration is one way in which wrongdoers may
justifiably be punished when it would be impossible, or disproportionate in the
narrow and/or wide sense, to bring them to a criminal trial.
7.5.3. Financial Penalties: Punitive Sanctions and Confiscation

Another way to do so, in some contexts, is via financial penalties, of which there
are broadly two kinds: the confiscation of property on the one hand, and punitive
sanctions on the other hand. I reject the latter, but cautiously endorse the former.
Punitive sanctions extracted via taxes might seem a better way to distribute the
punitive costs of participating in an unjust war through a democratic citizenry.33
They differ from reparations in that they go beyond what is owed to victims on
ground of fault or unjust benefit. In principle, adding them to the reparation bill
might seem a morally justified way to reflect the judgement that many more agents
than can feasibly be punished through war-crime trials are responsible for war-re-
lated crimes.
I am sceptical, for reasons similar to those adduced against implementing the
Reparative Principle (s.6.3.2). Taxpayers who are wholly innocent of participating
in war-related crimes will inevitably be harmed by punitive sanctions, at least to
the extent that the taxation system will not be able properly to discriminate
between those who are liable to punishment and those who are not. This will be so
whether the tax base is contributors’ ability to pay or (regressively) a flat levy on
purchases. Either they will have to pay more taxes for the same level of state provi-
sion than they would have had to pay absent punitive sanctions, or they will pay
the same amount of taxes but will receive less by way of state provision than they
would have done otherwise. Being innocent of war-related crimes, those taxpayers
have not forfeited their rights to those resources and thus are not liable to be so
harmed.
To be sure, as a matter of fact, it may be possible to identify taxpayers who are
liable. I shall argue presently that their estate ought to be confiscated. When iden-
tification is not possible, the question is whether it is permissible all things consid-
ered to catch the innocent in the net of punitive sanctions. Here, we are presented
with two options. Either we respect the rights of the non-liable and do not impose
punitive sanctions at all, at the cost of letting wrongdoers go unpunished; or we
confer priority to the punishment of wrongdoers at the cost of infringing the rights
of the non-liable. We encountered a similar problem in s.6.3, when discussing
33 See, e.g., Walzer, Just and Unjust Wars, 296–7.

198 Punishment
non-punitive sanctions. There, I argued that some taxpayers are likely to have been
victims of the war—and not just innocent of its wrongdoings—and that it is not
appropriate to ask them to pay up. The point applies here too.
Contrastingly, punitive confiscation is not vulnerable to this charge. It can range
from complete confiscation such that the property owner permanently loses all her
rights over the property, to partial confiscation where she temporarily loses some
of her rights (notably rights to use and to derive income from the property) but
retains other rights, or temporarily loses all and regains them a few years later
either unconditionally or against payment. Partial confiscation of the latter kind
was carried out by the English Parliament against Royalists during the Civil War,
partly as a way to finance the war, partly as a punitive measure against those it
called ‘delinquent’ but who were allowed to recover their estates subject to paying
for it at (a rather generous) 10 per cent of their market value. Wrongdoers who are
liable to imprisonment but whom there are good (aforementioned) reasons not to
send to jail are not wronged by the confiscation of their property (so long as they
are not left wholly destitute). Wrongdoers who are liable to punishment lesser than
imprisonment may well be liable to the loss of some or all of their property rights.
In principle, there seems to be nothing particularly problematic about this particu-
lar punitive measure.
It is however a targeted measure, whose implementation depends on being able
to identify who committed which wrongdoings, how many assets they have, and
so on. The example of the English Civil War is particularly illuminating on this
point: the two committees set up for, respectively, organizing confiscation and
processing back-purchases morphed into a rather cumbersome bureaucracy funded
by resources which might and should have been better deployed elsewhere, for
example towards the relief of poverty.34 Hence a final, cautionary note: punitive
confiscation is a morally justified alternative to imprisonment so long as its imple-
mentation does not unduly divert resources away from the reconstruction effort.
7 . 6 . W H O S H O U L D P U N I S H ? D E F E N D I N G
UNIVERSAL JURISDICTION
7.6.1. Universal Jurisdiction: A Cosmopolitan Archipelago

So far, so good: we now have accounts of which wrongdoings committed in war
warrant punishment, of which agents are liable to be punished, and of what kind
of punishment they may justifiably incur all things considered. It remains to be
seen by whom, if any one, they may or should be punished. In s.7.2.3, I argued
that victims have the right to punish those who have wronged them, but that there
are good reasons for entrusting the exercise of that right to political institutions—
put differently, that the moral right to punish should be turned into a legal and
34 For a discussion of this example, see D. L. Smith, Constitutional Royalism and the Search for
Settlement (Cambridge: Cambridge University Press, 1994), 202–8. I am grateful to Noel Malcolm for
drawing my attention to this case.
Punishment 199
institutionalized right. The question then is which political institution(s) have the
right to punish war criminals. International criminal law gives a contrasting pic-
ture. Whereas the International Criminal Tribunal for the former Yugoslavia
(ICTY) and the International Criminal Tribunal for Rwanda (ICTR)—both ad
hoc tribunals—had primacy over national courts, the 1998 Rome Statute for the
establishment of the International Criminal Court (ICC) gives primacy to sover-
eign states. Sovereign states have the presumptive right to punish wrongdoers in
general, and those who commit so-called international offences as well as war crim-
inals in particular, just if they are connected to the crimes in either one of the fol-
lowing ways: the defendant is a national of that state (nationality principle); the
victim is a defendant of that state (passive personality principle); the crime threat-
ens that state’s vital interests (protection principle); the crime was committed on
the territory of that state (territoriality principle). If a state which is connected to
a given war crime in those ways is either unable or unwilling to prosecute and
punish its authors, an international institution such as the ICC acquires jurisdic-
tion over it (though the case has to be referred to it by the UN Security Council).
In addition, the principle of universal jurisdiction stipulates that any state has the
right to punish war-related crimes, irrespective of perpetrators’ and victims’ nation-
ality, the territory where they were committed, and the extent to which they threat-
ened that state’s interests. To the extent that the Rome Statute is now the bedrock
of international criminal law, the latter endorses what one may call a double-com-
plementarity principle, whereby the ICC on the one hand, and third parties nation
states on the other hand, complement nation states which are connected to a given
war crime in either one of the aforementioned ways and which thus have primary
jurisdiction over those crimes.35
Before I proceed, three points are worth making. First, I speak of a presumptive
as opposed to a monopolistic right. Some deny that anyone other than a sovereign
state has the right to punish crimes committed by one of its nationals or on its
territory. In so far as this view relies on the Westphalian understanding of state
35 For classic discussions of the principles listed in the text, see J. Crawford and I. Brownlie,
Brownlie’s Principles of Public International Law, 8th edn. (Oxford: Oxford University Press, 2012),
chs. 21 and 30; Cassese, International Criminal Law, ch. 16; B. Broomhall, International Justice and
the International Criminal Court: Between Sovereignty and the Rule of Law (Oxford: Oxford University
Press, 2004); L. Reydams, Universal Jurisdiction: International and Municipal Legal Perspectives
(Oxford: Oxford University Press, 2003). For an excellent discussion of states’ right to punish war-
related crimes, with numerous references to relevant legal sources, see Chehtman, The Philosophical
Foundations of Extraterritorial Punishment, ch. 3. On complementarity more specifically, see W. W.
Burke-White, ‘Proactive Complementarity: The International Criminal Court and National Courts in
the Rome System of International Justice’, Harvard International Law Journal 49 (2008): 53–108;
Wellman, ‘Rights and State Punishment’, esp. 436–7; A. Altman and C. Wellman, A Liberal Theory of
International Justice (Oxford: Oxford University Press, 2009), ch. 4; Drumbl, Atrocity, Punishment,
and International Law, esp. ch. 7. My position is in some ways similar to Altman’s and Wellman’s: I
shall argue below that international institutions have jurisdiction over human rights violations.
However, unlike them (and, for that matter, unlike Dumbl) I reverse the principle of complementa-
rity. In this particular respect, my defence of universal jurisdiction closely resembles Luban’s. But
unlike him, I embrace a non-political view of war-related crimes, with the effect that universal juris-
diction has a much broader reach on my account than his. (See Luban, ‘A Theory of Crimes against
Humanity’, and s.7.2 above.)
200 Punishment
sovereignty as immune from any kind of interference in its domestic affairs, and as
I explicitly reject that view throughout this book, I will not discuss it here in rela-
tion to punishment.
Second, one might think that punitive sanctions are best imposed through peace
settlements and enforced by a designated institution, rather than meted out by
criminal courts. For courts do not, or at any rate ought not, prosecute and punish
a collective of citizens, but only individual agents. In so far as this particular kind
of punishment is collective, it is not part of the arsenal of those judicial institutions—or
so one might think. The point admittedly has some force; at the same time,
although a criminal court would perhaps be a blunt instrument for the imposition
of punitive sanctions, it might be less blunt or at least more impartial than the
settlement’s negotiating parties. At any rate I will not take a stand on this issue. My
argument in this section applies to those kinds of punishment, whichever they are,
which may justifiably be imposed by a criminal court.
Third, as set out in articles 100–10 of the 1982 UN Convention on the Law of
the Sea, the crime of piracy is a well-established exception to the Westphalian view
that the punishing sovereign state must be connected in at least one of the afore-
mentioned ways to the crime. Pirates may be apprehended and punished by any
state in the world, anywhere on the high seas, irrespective of their nationality, that
of their victims, and the extent to which if at all they threaten some interest of the
punishing state. In effect, my claim is that war-related crimes are relevantly similar
to the crime of piracy, pace the jurisdictional structure set out by the Rome Statute.
Now, the Rome Statute issues from a multi-state treaty and (unlike the
International Criminal Tribunal for Rwanda and the International Criminal
Tribunal for the former Yugoslavia) not from a Security Council decision. Given
that the consent of a number of states was required as a matter of political expedi-
ency to establish a permanent international criminal court, and that we live in a
world of sovereign states, it is not surprising that the Statute should confer primary
jurisdictional competence over war-related crimes on states. But there is a norma-
tive basis for that view, namely that, subject to their satisfying basic conditions
for legitimacy, states, or rather their citizenries, enjoy presumptive rights of self-
government, of which the right to punish is a part. Accordingly, even though war
crimes are typically regarded as international crimes, there should be a presump-
tion in favour of conferring on citizenries, via their state, the right to punish war
criminals if they are somehow connected (in the aforementioned ways) to the com-
mission of those crimes. So construed, the right to punish comprises the power to
change wrongdoers’ bundles of entitlements, a claim against being interfered with
by third parties when meting out punishment, and an immunity from having third
parties impose their domestic law in the realm of punitive justice.
At first sight, this view is of a piece with the conception of sovereignty which
I endorse in this book. For on that conception, so long as they abide by their uni-
versal duties to all other human beings, agents have jointly held rights to form
political associations over territorially bounded territories and to enjoy the benefits
of their mutual cooperation. Upon closer inspection however, and as I shall now
argue, a genuinely cosmopolitan account of punishment for war-related crimes is
Punishment 201
committed to a revised interpretation of the principle of universal jurisdiction

coupled with a reversal of the complementarity principle.
As we saw in s.7.2.3, war victims are not the only agents to have the pre-institutional
right to punish perpetrators of war crimes: third parties do as well. This conclusion
is supported by both victim-centred and victim-neutral considerations. Take, for
example, an atrocity committed against some civilians. Those victims have an
interest in protecting themselves from the commission of similar crimes, and an
interest in openly communicating their sense of the grievousness of the wrong
which they suffered. Taken together, those interests are important enough to con-
fer on them a power to punish those perpetrators as well as (inter alia) a claim not
to be interfered with unwarrantedly when doing so. Suppose however that they are
not in a position to exercise their right in compliance with the constraints of puni-
tive justice, either because (most obviously) they are dead, or because they simply
lack the wherewithal to mete out punishment. Under those circumstances, their
multifaceted interest in perpetrators being punished will be adequately respected
only if their right to punish vests in some third party.
The question, then, is which third parties. At this juncture, some might be
tempted to draw on the various principles which, in international criminal law,
ground a state’s right to punish, and hold that victims’ right to punish must neces-
sarily vest in individuals who are connected to the crime in virtue of the nationality
of the perpetrators, the nationality of the victims, or the location of the crime.
Those individuals would then exercise that right via their own, domestic state insti-
tutions. However, at the bar of cosmopolitan morality, membership in a given
political community cannot be an appropriate basis for restricting the range of
third parties in whom victims’ right to punish can vest, from the point of view of
both victims and defendants. For states are uneven in their willingness and ability
to punish suspected war criminals: in particular, they are less willing perhaps than
they should be when the latter are their own nationals, and more willing perhaps
than they should be when the defendants are foreigners and the victims their own.
To embrace the ‘statist restrictive view’ is to render victims and suspects vulnerable
to the vagaries of punishing states’ effective exercise of that right. The problem is
particularly acute when victims and defendants are stateless persons and thus at the
mercy of, respectively, defendants’ state or victims’ state. But it also arises when
both victims and defendants enjoy the benefits of state membership. To be sure, it
is part and parcel of membership in a given sovereign community that one is at risk
of being less well treated than those who are members of a different community.
However, the rights of defendants to be judged fairly, and of victims that they
should receive justice, are fundamental rights which should not be left to this par-
ticular kind of chance. Accordingly, any actor who is able to mete out punitive
justice fairly and irrespective of political membership can, in principle, exercise the
right to punish on behalf of war-related crime victims. It is in that sense that juris-
diction over war-related crimes is universal.
Those are victim-centred arguments in that they appeal to war victims’ interest
in the punishment of their murderers, rapists, and torturers. But there also are
victim-neutral reasons for endorsing universal jurisdiction. War-related crimes
202 Punishment
have a universal dimension, not least to the extent that they have profoundly
adverse effects on outsiders. The point is not just the obvious one that (e.g.) the
commission of a genocide might lead to further genocides, or that the wrongful
annexation of a country or area by another is likely to destabilize a whole region.
Irrespective of those contingent connections between past and future wrong
doings, the point is also that in so far as we ought to conceive of one another,
irrespective of borders, as one another’s moral equals, any such crime, committed
anywhere in the world, is of concern to us all. This really, is what cosmopolitanism
commits us to—to the view that, as John Donne puts it better than anyone else,
anyone’s suffering diminishes, or should diminish, every single one of us because
we are all involved in humankind.36 To hold, thus, that only states which have a
special connection between either perpetrator, victim, or location have the right to
punish those crimes is not to do justice to what is universally wrong about them.
Those victim-centred and victim-neutral considerations support the view that
all human beings, irrespective of borders, have jurisdiction over war-related crimes
on behalf of victims as well as on their own behalf—whether those crimes are vio-
lations of basic and non-basic rights. Not only do they have the right to punish:
they are also under a (prima facie) duty to do so. More precisely, recall that the right
to punish consists, first, in a power to change defendants’ bundles of rights, duties,
etc., and, second, in a claim against third parties that they not interfere with the
exercise of that power. In the context at hand, this claim correlates with duties not
to (e.g.) kidnap and kill officials as they exercise their judicial functions, not to
shield individuals accused of war-related crimes from prosecution, and so on. In
addition, the power to punish is protected by a claim against third parties that they
enable the provision of punitive justice, by establishing and funding properly
designed criminal courts. Those various duties clearly are universal in scope. For to
claim, contrariwise, that (e.g.) Serbs are justified in protecting fellow Serbs from
the long arm of the ICTY but are under a duty to hand over non-fellow Serbs to
that tribunal, or that American citizens are not under a duty to provide material
support for non-American courts should the latter prosecute American combat-
ants for war-related crimes, is to allow some crimes to go unpunished purely on the
basis of nationality. This is clearly not tenable from a cosmopolitan point of view.
That said, although all human beings have the pre-institutional right and duty
to punish war-related crimes, only properly designed political institutions can
properly deliver punitive justice. If so, it remains to assess which kind of institu-
tions best does that. At first sight, it might seem that only supranational institu-
tions should be entrusted with that task. But this would be a mistake. Suppose
that by supranational institutions, we mean several states acting together for the
purpose of prosecuting and punishing war-related crimes committed in a specific
war to which they were parties, as the Allies did when they constituted the
36 ‘No man is an island, entire of itself; every man is a piece of the continent, a part of the main; if
a clod be washed away by the sea, Europe is the less, as well as if a promontory were, as well as if a
manor of thy friend’s or of thine own were; any man’s death diminishes me, because I am involved in
mankind, and therefore never send to know for whom the bell tolls; it tolls for thee.’ John Donne,
Devotions Upon Emergent Occasions, Meditation XVII.
Punishment 203
International Military Tribunal at Nuremberg. This variant of the supranational

view is vulnerable to the same objections as the statist view. To block those objec-
tions, then, we might mean the world community of sovereign states establishing
a permanent, International Criminal Court—as was done through the Rome
Statute. This would solve the problems inherent in the statist view—in particular,
problems arising from belligerent states’ difficulties (in the aftermath of a war) to
mete out fair and impartial punishment. It would also have the advantage of
entrusting the task of punishing war-related crimes to institutions which can gen-
uinely claim to represent humankind as whole. However, this purely internation-
alist approach has one serious weakness, namely that the sheer volume of cases
and the enormous attendant difficulties in processing witnesses, gathering evi-
dence, and so on, makes for such a slow judicial process as to fall foul of the
requirement that punitive justice should be delivered expeditiously. It would be
similarly burdensome, I suspect, to set up an ad hoc international tribunal when-
ever some individual somewhere commits a war-related crime. Pure internation-
alism, in other words, is morally unfeasible. At the same time, it might be more
feasible for domestic courts in some cases to access the necessary evidence, organ-
ize judicial proceedings, and so on. Save but for the most egregious cases, a divi-
sion of labour between international and domestic institutions is more likely to
bring about punitive justice in a greater range of cases.
That said, the division of labour argument yields different options. First, one
could set so-called hybrid tribunals, whose personnel (defence lawyers, judges,
prosecutors, administrators) are drawn from both international organizations and
a state connected in relevant ways to the crime, its authors and/or victims, and
which apply both international and national criminal laws.37
Second, we might think that, where international institutions cannot or will not
go, only the state on whose territory the crimes were committed, or whose national
committed the crimes, or whose nationals were victims thereof, may go. This is
most obviously so for practical reasons: in fact, as we shall see in s.9.4, building or
rebuilding domestic judicial institutions in the wake of a conflict is part and parcel
of a just reconstruction programme. But practical feasibility is not the only reason
for entrusting that state with that particular punitive task. Consider the case of
Sergeant Alexander Blackman, a British Royal Marine who was given a lifelong jail
sentence by British courts in November 2013 for having deliberately shot to death
an unarmed and wounded Taliban fighter. So long as Britain deploys its forces
37 On hybrid tribunals, see Schabas, Unimaginable Atrocities, 18–21; A. Chehtman, ‘Developing
Local Capacity for War Crimes Trials: Insights from BiH, Sierra Leone, and Colombia’, Stanford
Journal of International Law 49 (2013): 297–329. On the interplay between international, hybrid, and
national courts, see W. W. Burke-White, ‘A Community of Courts—Towards a System of International
Criminal Law Enforcement’, Michigan Journal of International Law 24 (2002–2003): 1–101; Drumbl,
Atrocity, Punishment, and International Law. Incidentally, one should not overstate the point that pure
internationalism is morally unfeasible. David Scheffer, who was President Clinton’s ambassador-at-large
for war crimes, notes that the total costs for the five international criminal tribunals which operated
between 1993 and 2008 totalled $3.43bn—which was what the American military spent on its mili-
tary operations in Iraq in two weeks. (D. Scheffer, All the Missing Souls—A Personal History of the War
Crimes Tribunals (Princeton, N.J.: Princeton University Press, 2012), 28.) Still, it would be prohibi-
tively expensive for an international criminal tribunal to prosecute all war-related crimes.
204 Punishment
abroad, particularly in support of transitional administration or occupying regimes,

it behoves on its leadership to show to local civilian populations and to its own
combatants that it will not leave those crimes unpunished, both as an expression
of a commitment to human rights and as a means to deter combatants who fight
on its behalf and at its behest to deter them from committing similar crimes. To
have that combatant prosecuted and punished by the International Criminal
Court at the Hague would not have achieved that. Likewise, it mattered that
Germany herself should bring to trial, in 1963–65, individuals for their genocidal
work at the complex of Auschwitz-Birkenau. To be sure, one may justifiably worry
that war crime trials are sometimes (often?) a means for the punishing state to
exonerate itself from all wrongdoings by singling out the defendant as an isolated,
rogue, deviant agent. Still, it might be important for the sake of peace that the very
same political community whose policies lead to or at any rate facilitated the com-
mission of war-related crimes should demonstrate willingness to punish offend-
ers.38 Clearly, if it cannot do so in the immediate aftermath of the war (because its
institutions are too weak, or because the wounds opened by the war are too raw to
allow for just punishment), then other states, or an international criminal court,
should do it (unless, as we shall see in ch. 9, reconciliatory processes are called for).
But if wrongdoers are apprehended long after the alleged facts, there might well be
good reasons for entrusting a political community whose past regime shared in
responsibility for war crimes with the task of punishing them.
The foregoing remarks do not reintroduce through the back door the principle
that states have the monopolistic right to punish in war-related matters which are
of particular concern to them (however one defines that concern). Nor do they
imply that states have the right to punish war criminals only if properly authorized
to do so by international institutions. Punishment for war-related crimes is
devolved to states which are connected to the war-related crimes in some way (via
the nature of the crimes, the nationality of defendants and/or victims, or the loca-
tion of the crimes) only if doing so can be justified by appeal to expressivism and
so long as it does not undermine other goods such as prevention and rehabilita-
tion.39 This is entirely compatible with the view that a state may justifiably punish
war criminals, when that condition obtains, without being authorized to do so.
38 At a Court Martial hearing held in October–December 2013, Blackman was given a life sen-
tence with a mandatory minimum term of ten years minus time already served. In April 2014, the
Court of Appeal maintained the life sentence but reduced the mandatory term to eight years. See
https://www.judiciary.gov.uk/wp-content/uploads/2014/05/r-v-sergeant-alexander-wayne-blackman.
pdf (accessed on 15/01/2016). On the Frankfurt-Auschwitz Trial, see, e.g, D. O. Pendas, The Frankfurt
Auschwitz Trial 1963–65: Genocide, History, and the Limits of Law (Cambridge: Cambridge University
Press, 2006); R. Wittmann, Beyond Justice: The Auschwitz Trial (Cambridge, Mass.: Harvard University
Press, 2005). High-ranking officials such as camp commandants were tried by Poland in the immedi-
ate aftermath of the war. On war crime trials as an instrument for distortingly validating the punishing
state, see Simpson, Law, War and Crime, 88–91.
39 Larry May, you recall, argues that perpetrators of war-related crimes are liable to international
prosecution only if they can be shown to have intended to commit a wrongdoing. If they only had
knowledge that their acts furthered their regime’s policy of crimes against peace or of crimes against
humanity, then they must be tried by domestic courts. May’s rationale for this position is that there is
a difference between intentionally participating in a discriminatory policy, and merely knowingly
doing so—a difference which would not be captured by having both crimes fall under the remit of
Punishment 205
Save for those cases where it matters morally speaking that a specific political
community should punish specific perpetrators (as illustrated a couple of para-
graphs ago), neither of those first two institutional solutions cohere sufficiently
tightly with cosmopolitan morality. Instead, I suggest that any state can exercise
the right to punish when international institutions wrongfully will not, or cannot,
or ought not to do so, so long as it does not fall foul of the aforementioned condi-
tions, meets basic procedural standards, and thus can be trusted to act fairly and
impartially. From a cosmopolitan point of view, this is the view one should hold.
For not only does it permit punishment by actors who are more likely to be impar-
tial than parties to the conflict in which those crimes were allegedly committed; it
also instantiates the view that membership in this or that state does not have deep
normative significance, that victims whoever they are have a claim that wrongdoers
whoever they are be punished, that alleged wrongdoers whoever they are also have
claims to be treated fairly—irrespective of where the crimes were committed; and
that those crimes, precisely because of what makes them so deeply wrongful, are of
concern to humanity as a whole understood as all of us, here and now.
However, in so far as expressing condemnation is the function of punishment,
those who punish must have the standing to do precisely that. If they shield their
own agents from prosecution and punishment, their standing is correspondingly
diminished. Accordingly, the view ought to be qualified so as to rule out or at the
very least severely constrain the practice by states of punishing foreign war crimi-
nals, in the following three cases: when they themselves were parties in that war;
when their forces are plausibly suspected of having committed war-related crimes
in that conflict; and when they are unwilling to prosecute and punish the latter (or
to facilitate their prosecution and punishment by credible other parties).40
Note that the point is not that an unjust belligerent, whether ad bellum or in
bello, thereby lacks standing to punish war criminals; rather, it is that an unjust
belligerent’s attitude vis-à-vis its own forces has a bearing on its standing as a pun-
ishing party. It is partly for this reason, then, that one may question the legitimacy
of the Nuremberg International Military Tribunal. Soviet troops committed exac-
tions not just in Germany but in countries or provinces which they ‘liberated’ from
German occupiers (not least the Ukraine). Britain and the United States flattened
German cities in relentless bombings. French Gaullist troops also committed exac-
tions against German civilians in 1945 (notably gang rapes). All those crimes went
unpunished, and one should at the very least have considerable moral qualms at
the thought that those four powers should have sat in judgement of their enemies.
In that vein, Judge Radhabinod Pal, sitting at the International Military Tribunal
for the Far East (or so-called Tokyo Trials), argued in his dissenting o pinion that
the Tribunal, failing as it had done to indict the United States for the nuclear
destruction of Hiroshima and Nagasaki (a war crime if there is one) was little more
international institutions. (See May, Crimes against Humanity, 125ff; May, War Crimes and Just War,
ch. 11.) While I agree that those two crimes so differ, I simply don’t see why they should be subject to
different jurisdictions.
40 For the view that a party which is itself guilty of violations of the natural law may not punish
those who act against it, see Grotius, The rights of war and peace, Book II, ch. XX.
206 Punishment
than victor’s justice.41 More recently, in the 2000s, Belgium punished some
Rwandese nationals for their participation in the genocide. In so doing, it invoked
its own legislation, which enables its authorities to punish any one of its residents,
irrespective of nationality, for crimes against humanity and war crimes committed
anywhere in the world against any victim. Yet, Belgium’s own policy in Rwanda
while the latter was part of its colonial empire is widely held to have been a con-
tributing factor to the genocide. Had Belgium authorities acknowledged that fact
when punishing those individuals, as well as the failures of its peacekeeping forces
during the genocide to do anything, those trials might have been seen as an appro-
priate exercise in punitive justice. As it is, it is hard not to think that they were a
convenient currency with which to buy a collective good conscience.
To be sure, had those trials not taken place, even more war-related crimes com-
mitted by the Nazis, Japanese armed forces, and Rwandese nationals would have
gone unpunished than was in fact the case. It is true too that those trials, for all the
difficulties they raised, nevertheless were characterized by a commitment to due
process. Under the highly charged circumstances of the war’s immediate aftermath,
the Nuremberg trials in particular, though falling short of full procedural fairness,
were in that respect rather remarkable. Given the options on the table, they were
the lesser of two wrongs. Nevertheless, they were a wrong of some kind, and it is
precisely in the recognition of that fact that one finds the strongest case for grant-
ing to international tribunals or, should this not be possible, to impartial states, the
right to punish.
7.6.2. An Objection to Universal Jurisdiction: The Problem

of Domestic Criminal Codes
Let us take stock. I have argued that humankind at large, via international institu-
tions such as international criminal tribunals, has primary jurisdiction over war-
related crimes; when international institutions cannot, will not, or ought not to
punish, third-party states have the right to punish war criminals even though they
are not connected by nationality or territory to those crimes. It is precisely because
the right to punish is divorced from territoriality that the principle is aptly labelled
‘universal jurisdiction’. In effect, it consists in turning on its head the complemen-
41 On Belgium’s Rwanda trials, see Drumbl, Atrocity, Punishment, and International Law, 83–4.
On the Tokyo Trial, see, e.g., R. H. Minear, Victor’s Justice—The Tokyo War Crimes Trial (Princeton,
N.J.: Princeton University Press, 1971). For discussions of victors’ justice in the context of Nuremberg
in particular, and in general, see, e.g., Tusa and Tusa, The Nuremberg Trial; Bass, Stay the Hand of
Vengeance, ch. 5; Simpson, Law, War and Crime, ch. 5; Schabas, Unimaginable Atrocities, ch. 3. In
international law, Judge Pal’s claim is known as the tu quoque defence. The defence has three strands.
As used at Nuremberg (relatively effectively) by Admiral Dönitz, it holds that the prosecuting party in
fact interpreted the law as did the defendant and that the crime for which the latter is being tried is
not, in fact, a legal crime; it is also sometimes used to justify the commission of the crime, as in ‘you
did it, therefore I could do it too’; finally, it is also used to deny the prosecuting party standing so to
act, as in, ‘you did it too, therefore you may not punish me for it’. I am agnostic on the first reading,
wholly reject the second, and am sympathetic to the third. For a thorough discussion of the defence,
see S. Yee, ‘The Tu Quoque Argument as a Defence to International Crimes, Prosecution, or
Punishment’, Chinese Journal of International Law 3 (2004): 87–134.
Punishment 207
tarity principle as set out by the Rome Statute. According to the principle, you
recall, sovereign states have primary jurisdiction over the punishment of wrong
doers, and the international criminal court has jurisdiction when states cannot or
will not punish. On my account, by contrast, states have jurisdiction over those
crimes only when international judicial institutions will not go or when it is better
from the point of view of justice after war (for reasons mentioned a couple of par-
agraphs ago) that they should not go. In doing so, however, states would simply act
on behalf of humankind—and not exercise a prerogative of sovereignty.
The principle of universal jurisdiction as I have expounded and defended it
might seem vulnerable to a number of objections, only one of which I will address
here.42 According to that objection, given that the rationale for the principle is that
violations of human rights are of concern to humankind at large, there is no reason
to exclude from its remit such violations as are committed in peacetime and in
purely domestic contexts—for example Austrian citizen Josef Fritzl’s repeated
crimes of rape and imprisonment against his daughter Elizabeth and the seven
children she bore him. And yet to hold that, in principle, an institution like the
international criminal court, as opposed to the Austrian state, has primary jurisdic-
tion over him, is to undermine individuals’ jointly held right to punish wrongdoers
for what are essentially domestic crimes albeit violations of human rights. Given
that the Austrian state was deemed able to prosecute and punish him in compli-
ance with basic standards of due process and fairness (which by all accounts it did),
to regard international institutions as having ultimate jurisdiction in this case is
tantamount to holding that the Austrians are not (jointly) immune from interfer-
ence with the operation of their criminal law system. Put bluntly, it is a violation
of their sovereignty over their own territory.
From a cosmopolitan point of view, the claim that the conferral of a given right
stands in tension with sovereignty rights is not in and of itself worrisome. After all,
sovereignty rights are largely instrumentally valuable as a way to bring about jus-
tice; and to the extent that they have value above and beyond that, they are severely
constrained by the moral imperative to bring about justice across borders. That
said, the objection sheds light on how best to understand crimes which fall within
the primary jurisdiction of international institutions or the secondary jurisdiction
of states. Members of a territorially bounded and sovereign community have the
right to punish human rights violations, which they entrust to state officials, in so
far as doing so is instrumental to the realization of punitive justice across borders.
This allows for some leeway in the design of domestic criminal law. The distinction
I drew in ch. 4 between the declarative and the specifying clauses of a peace agree-
ment helps explain why. Declarative clauses—for example, a clause stipulating that
belligerents will respect each other’s rights of self-government—merely state bellig-
erents’ independently justified rights and duties vis-à-vis one another, and are
binding on belligerents in virtue of a general duty to do what justice requires. By
contrast, specifying clauses—for example, a clause specifying that elections will be
42 For other objections and rebuttal thereof, see Chehtman, The Philosophical Foundations of
Extraterritorial Punishment, 134–9.
208 Punishment
held in belligerent community A within three years of the conclusion of the peace
process—render more specific such rights and duties. By parity of reasoning, a
clause in a criminal code which deems it an offence to (e.g.) have sex with someone
who is not in a position validly to consent is declarative in so far as it merely states
the independently justified obligation not to rape another person. But what counts
as valid and invalid consent remains to be established. Within a reasonable range,
there are broadly equivalent (morally speaking) conceptions of what counts as
invalid consent which it is up to the relevant political community (be it a sovereign
territorially bounded state, or a state within a federal union) to formulate. Not pro-
tecting children as young as nine from having sexual intercourse forced upon them
by adults is a violation of those children’s basic human rights to sexual integrity and
autonomy—such that anyone who commits such act ought to be deemed guilty of
rape. Contrastingly, setting the age of sexual consent at seventeen rather than six-
teen is not (I think) a moral imperative given that as far as we know, whether they
are better able to consent at seventeen than at sixteen is a tough call. Put differently,
having (consensual) sex with a sixteen-year-old is not a human rights violation; but
nor is making it a criminal offence to prevent someone from having sex with a six-
teen-year-old a human rights violation. It is therefore a prerogative of a political
community whose members have the right and power, via their officials, to settle on
the age of consent at sixteen or seventeen via their criminal code.
To see why we can thus block the objection to universal jurisdiction, consider
someone—call him Andrew—who knowingly and willingly has sex with a sixteen-
year-old. In Turkey, he would be guilty of rape; in England, he would not. The
worry elicited by the objection is that Andrew would be liable to punishment
under the Turkish penal code for having sex with a sixteen-year-old English girl in
Britain and that Turkey would have a prima facie right to punish Andrew, even
though neither he nor his partner is Turkish. But the principle of universal juris-
diction does not have this implication. For so long as England’s criminal code
issues in a reasonable interpretation of the age at which teenagers can be presumed
validly to consent to sex, its citizens have the protected power via their officials to
enshrine such an interpretation in its criminal code and to punish Andrew accord-
ingly; they also enjoy immunity from being interfered with, in this endeavour, by
Turkey. In that sense, England’s power to punish Andrew can go hand in hand
with an immunity from interference by other punishing parties.
Those considerations do not impugn my defence of the reversal of the comple-
mentarity principle. For it still remains the case that humankind has primary juris-
diction over all human rights violations. Thus, the principle of universal jurisdiction
as I construe it does hold adults who knowingly have sex with a nine-year-old
guilty of a gross human rights violation and liable to punishment at the hands of
humankind at large.43 From a cosmopolitan point of view, to repeat, this is not to
43 Another way to put my point is this. (See, e.g., Duff, Punishment, Communication, and
Community, 64ff.) Some offences are mala in se—offences irrespective of time and space. Others are
mala prohibita—offences in virtue of being made so by a given legal system. Offences which are mala
prohibita ought to be divided into two categories: specifications of mala in se (along the lines suggested
in the main text) on the one hand, and wholly new offences on the other hand. My claim here, in
Punishment 209
be regretted. For given the nature of their crime, those individuals are not wronged
by the imposition of punishment; nor are their compatriots wronged by other
states’ power to refuse to regard their refusal to punish those wrongdoers as dispos-
itive. Moreover, in practice, the principle of universal jurisdiction so conceived
does not entail that UN-mandated police squads should whisk off any suspected
paedophile to The Hague or to any country whose criminal code explicitly prohib-
its paedophiliac acts and is properly enforced. There are in fact very good, morally
directed pragmatic reasons to resist such a move. What the principle does entail,
though, is that should such an individual be liable to prosecution and should it be
possible to apprehend him without incurring such widely disproportionate costs
(be they financial or reputational), punishment would not wrong him.
7.7. F RO M A M N E S T I E S TO A M N E S I A : S A C R I F I C I N G
P U N I T I V E J U S T I C E AT T H E A LTA R O F P E A C E ?
7.7.1. Defending Amnesties: A First Cut

Let us take stock. At various junctures throughout this chapter, I have argued that
victims of war-related crimes are morally justified in seeking the punishment of
perpetrators, and that third parties are justified, indeed sometimes under a duty, to
mete out punishment on victims’ behalf. As I have also argued, however, punishing
all perpetrators of such crimes is not feasible: all things considered, a justifiedATC
peace may demand that we forego punitive justice in at least some cases.The feasi-
bility constraints I have discussed so far have been epistemic, financial, and social.
But there is a rather different kind of constraint on just punishment which I have
not yet discussed —namely the fact that perpetrators of war crimes will at best fail
to support, and at worst actively and unjustly derail, the peace process if they are
brought to trial. Faced with such prospects, what may proponents of do?
Cases abound of communities whose members chose to let bygones be bygones
for the sake of peace. In European countries which had been invaded and occupied
by the Nazis, and many of whose citizens had collaborated with the latter, initial
purges, kangaroo courts, summary executions, and (to be fair) some properly con-
ducted trials gave way to amnesty laws and the cloak of silence. In Spain, the newly
elected democratic regime which succeeded General Franco’s forty-year-long dic-
tatorship did not prosecute a single one of the officials who, during that dictator-
ship, had participated in the systematic oppression, torture, and assassination of
opponents to the regime. Likewise, in Brazil, Uruguay, and El Salvador, newly
installed democratic regimes passed amnesty laws to appease the junta. Similarly,
the 1999 Lomé Peace Accords which brought an end to the civil war in Sierra
Leone stipulate that no action (judicial or otherwise) will be taken against any
military or political officials for acts which he or she committed in pursuit of their
effect, is that offences which are purely mala in se (no sex with a nine-year-old), as distinct from spec-
ifications thereof (no sex with someone aged seventeen or less) are subject to universal jurisdiction.
210 Punishment
war, up to the signing of the Agreement. In Spain, Brazil, and Uruguay, the transi-
tion occurred through a negotiated handover of power from a military dictatorship
to an elected civilian government, with the military agreeing not to take up arms
in exchange for exemption from punishment. Non-state agents have also been
granted amnesties under the terms of a peace agreement, such as rebels in Sierra
Leone under the terms of the Lomé Peace Accords and rebels in Uganda under the
latter’s 2000 Amnesty Law.
Amnesties are offered partly and sometimes because some wounds are thought
to be too deep to withstand scrutiny: so construed, they are an alternative to the
reconciliatory processes I shall describe in ch. 9. Most often, however, they are
granted because there are considerable obstacles to bringing suspected war crimi-
nals to trials. Capturing them while they are on the run exposes one’s combatants
to risks to life and limbs. Punishing parties’ combatants and civilians might also be
subject to reprisals while the trials are ongoing. Those concerns were at the heart of
Western powers’ initial reluctance to commit troops to capturing and bringing to
trials Slobodan Milosevic and Radovan Karadzic. In all those cases, wrongdoers’
unwillingness to commit to a just peace made the latter morally unfeasible.44
Under those circumstances, a justifiedATC peace is one in which those wrongdoers
are granted an amnesty for the sake of ensuring that individuals’ basic rights are
secure and that non-basic rights are not under imminent threat of war.
How, then, are we to evaluate those practices? For a start, there is an important
difference between not punishing and granting an amnesty. In granting a perpetra-
tor an amnesty for having committed grievous wrongdoings which would nor-
mally warrant prosecution, the competent authority explicitly foregoes its right to
punish that perpetrator: to grant an amnesty, thus, is not merely to desist from
punishing. If not punishing tout court is morally problematic, explicitly declaring
that one will not do so seems in some respects even worse. In addition, there are
different forms of amnesties. Some are selective (only for some crimes), condi-
tional (upon meeting certain conditions and applying for it), and partial (from
criminal, but not civil, liability.) A blanket unconditional full amnesty protects all
alleged perpetrators without conditions, from both civil and criminal liability.45
44 For discussion of the cases of post-Second World War Europe, Spain, Brazil, and Uruguay, see
Rigby, Justice and Reconciliation After Violence, chs. 1–3. For the case of Europe, see also I. Deak, J. T.
Gross and T. Judt (ed.), The Politics of Retribution in Europe—World War II and its Aftermath (Princeton,
N.J.: Princeton University Press, 2000). For the case of El Salvador, see, inter alia, P. B. Hayner,
Unspeakable Truths: Transitional Justice and the Challenge of Truth Commissions, 2nd edn. (London:
Routledge, 2011), 102–4. For the case of Sierra Leone, see, e.g., Schabas, Unimaginable Atrocities,
ch. 7. On the difficulties inherent in bringing war criminals to trial, see Bass, Stay the Hand of
Vengeance, ch. 6. For an interesting discussion of the practice of granting amnesties to rebels in civil
wars, see F. Megret, ‘Should Rebels be Amnestied?’, in J. S. Easterday, J. Iverson and C. Stahn (ed.),
Jus Post Bellum (Oxford: Oxford University Press, 2014). See also A. G. Reiter, ‘Examining the Use of
Amnesties and Pardons as a Response to Internal Armed Conflict’, Israel Law Review 47 (2014):
133–47. My factual points in this section are all drawn from those sources.
45 For a clear typology, see K. Greenawalt, ‘Amnesty’s Justice’, in R. I. Rotberg and D. F. Thompson
(ed.), Truth v. Justice—The Morality of Truth Commissions (Princeton, N.J.: Princeton University Press,
2000), 195. See also M. Freeman, Necessary Evils—Amnesties and the Search for Justice (Cambridge:
Cambridge University Press, 2009), 13–15, for an overview of different kinds of amnesty and a dis-
cussion of the differences between amnesties and, inter alia, statutes of limitation, pardons, and pros-
Punishment 211
Now, I have argued that victims and non-victims (humankind at large) have the
pre-institutional right to punish perpetrators, but ought to entrust the exercise of
that right to properly designed punitive institutions. I have also argued that
humankind at large is under a duty to punish. Furthermore, I have claimed that
the right and the duty are grounded in the expressivist imperative to condemn
wrongdoings. But if all those points are correct, how could amnesties be justified,
given that (let us not forget) their beneficiaries are, in some cases, mass torturers,
mass rapists, mass murderers?
Easily so, one might retort—by appealing to the necessity of forestalling war,
which in fact tells not just in favour of conditional and targeted amnesties of the
kind offered by the South African administration to Apartheid officials in the mid-
1990s, but also in favour of blanket, unconditional, full, amnesties—so long as
they are a measure of last resort: the examples described at the outset of this section
speak for themselves. And if those wide-ranging amnesties are justified on those
grounds and to those ends, then a fortiori so are more limited amnesties.
I am sympathetic to that argument, as I am sympathetic to the view that the
United Nations, in explicitly condemning the granting of amnesties for the sake of
punishing agents guilty of crimes against humanity and serious war crimes, are
insufficiently attentive to the serious risks posed to victims’ life and limbs by the
drive to punish. From the comfort of one’s study, it is easy to underestimate how
much people are justifiably willing to sacrifice for the sake of putting an end to
violence.
In fact, moreover, victims are sometimes under a duty to accede to demands for
amnesty—and are not simply justified in so doing. Consider the case of South
Africa. The Post-amble to the 1993 Constitution provided for the establishment of
a Truth and Reconciliation Commission (TRC), and, controversially, granted
amnesties to individuals willing to testify, subject to the condition that no amnesty
could be granted for non-political crimes. The amnesty provision was framed as a
compromise between the ANC’s demand for full criminal accountability, and the
National Party’s threat to derail the transitional process by force of arms if wide-
spread prosecutions were undertaken. Given that the National Party still controlled
the coercive apparatus of the state, the threat was not to be taken lightly. To be
sure, had the ANC held on to its commitment to full and widespread prosecution,
the NP would have acted grievously wrongly had they taken up arms. Whether or
not the ANC would have had a just cause for retaliating in kind, it is overwhelm-
ingly likely that it would have had no chance of succeeding, for precisely the rea-
sons which rendered the NP’s threat credible. Apartheid would have continued,
which means that a majority of the population would have continued to be subject
to grievous rights violations. National Party officials and their allies in the police,
military, and intelligence services would have been primarily responsible for those
wrongs. But one cannot wholly escape responsibility for the wrongs committed by
ecutorial immunity. As Amia Srinivasan pointed out to me, an explicit decision not to punish has the
merit (over a failure to do so) to be transparent and open. In that respect, it is better. I wholly agree.
But in another respect (that of communicating to victims that punishment is not warranted), it seems
worse.
212 Punishment
an intervening agency. Under those circumstances, in other words, it would have

been wrong for the ANC to refuse the compromise. The point, note, is not that
they prudentially had to do so as a matter of realpolitik: the point is that under
the circumstances, they were under a moral obligation to do so as a matter of
justice.46
Here is an objection to the point—which I believe is what Frances Kamm has in
mind, when arguing against Barbara Herman’s interesting discussion of the South
African Truth and Reconciliation Commission—though she does not put it in
quite those terms.47 At the heart of my argument is the thought that victims do not
have the right to cause (albeit through intervening agency) the death of innocent
people. But if that is correct, then by implication they no longer have the right to
punish wrongdoers in the first instance, since in so doing they would cause inno-
cent people to die and thus not act within their rights. And if that is correct, then
it does not make sense to grant wrongdoers an amnesty, since one can do so only
to those who are liable to being punished. Or so I take Kamm to argue.
Has something gone wrong there? I believe so. The claim that V has the right to
punish W is ambiguous. It can be taken to mean that V has a right vis-à-vis W to
punish him, such that W is liable to (proportionate) punitive burdens and is under
a duty not to resist punishment. But it can also mean that V has a right vis-à-vis
third parties that they not interfere with her punishing W (for example by helping
him flee the country). The first right is a right ad personam, in that it is held against
W only; the second right is in rem, in that it is held against the world at large as
well as against W. Usually, the claim that V has that right is taken to mean both.
However, even if V does not have the in rem right to punish W (on the grounds
that punishing him would jeopardize prospects for peace), she may still have the ad
personam right to do so, in the sense that she would not wrong W himself for doing
so and that W is under the relevant duties. One can therefore hold both that V
lacks the in rem right to punish W and is thus under a duty to W’s putative future
victims to desist from punishing him, and that W nevertheless is liable to be pun-
ished and is under a duty not to resist even if V, in fact, punishes him in violation
of her duties with respect to the peace process. Given that all that is required to
escape the charge of incoherence is for W to be liable to punishment at the point
46 For a clear and concise account of the establishment of the TRC and of the (conditional and
limited) amnesty granted to Apartheid officials, see A. Boraine, ‘Truth and Reconciliation in South
Africa’, in R. I. Rotberg and D. F. Thompson (ed.), Truth v. Justice—The Morality of Truth Commissions
(Princeton, N.J.: Princeton University Press, 2000). Strictly speaking, South Africa under Apartheid
does not count as a war since it does not meet the COW requirement of 1000 battle-related death per
year of its duration. I use this particular example purely for illustrative purposes. On a point of sub-
stance, and controversially perhaps, I do not think that the ANC would have had a just cause, for the
NP’s threat was not to uphold Apartheid no matter what: it was a threat to uphold apartheid if the
ANC went for full and widespread prosecution. Given that there were other options on the table—
ranging from no Truth Commission to Truth Commission + amnesty, neither of which required
maintaining the Apartheid—to say that the ANC would have had a just cause is to say that the value
of punitive justice warrants killing those who oppose it as well as inflicting lethal collateral damage on
a bystander. It is not clear to me that it does.
47 See B. Herman, ‘Contigency in Obligation’, in B. Herman, Moral Literacy (Cambridge, Mass.:
Harvard University Press, 2007); Kamm, The Moral Target, 140–1.
Punishment 213
at which he is given an amnesty, the argument in favour of the duty to grant an

amnesty is not incoherent in a way Kamm seems to suggest.
7.7.2. Expanding the Defence: A Second Cut

To recapitulate, amnesties are sometimes morally justified as a necessary means to
bring about a justifiedATC peace, albeit at the cost of punitive justice. I would like
to end this section with four questions/points.
First, are there nonetheless crimes such that their perpetrators ought not to be
granted amnesty? In some cases, notably in South Africa, crimes committed pursu-
ant to a conflict and/or upon order such as interrogational torture were protected,
whereas similar wrongdoings committed for personal gain or out of sadistic grati-
fication under the cloak of war were not. In other cases, as articulated by the
United Nations, crimes against humanity, war crimes, and (notably in domestic
politics) serious human rights violations such as enforced disappearances are not
protected. Finally, again as per the United Nations, there is a difference between
international and civil conflicts. Whereas amnesties ought not to cover crimes
against humanity in the former cases, they may do so in the latter case. According
to the 1977 Protocol Additional II to the Geneva Conventions (art. 6.5), in the
aftermath of civil conflicts, ‘the authorities in power shall grant the broadest possi-
ble amnesty to persons who have participated in the armed conflict, or those
deprived of their liberty for reasons related to the armed conflict, whether they are
interned or detained’.48
Given how difficult it is to discern a torturer’s motives, and more generally to
distinguish criminal violence from war violence, I am not convinced that the dis-
tinction drawn by the South African’s TRC works. The second distinction seems
more promising, accounting as it does for the intuition that certain crimes are so
serious that their perpetrators ought to be punished no matter what. However, and
to adapt a point made above, if granting amnesty for genocide is necessary for and
effective at preventing another genocide, there is a strong prima facie reason for
endorsing such a move. Finally, reconciliation in the aftermath of a civil conflict
may well require amnesties to a greater degree than reconciliation in the aftermath
of an interstate conflict, since victims and perpetrators have to live in closer prox-
imity to one another in the former case than in the latter case. Note, however, that
this claim is entirely contingent on the fact of that proximity: there is nothing
inherent in a civil conflict qua civil conflict which dictates in favour of amnesties,
and there is nothing inherent in interstate conflicts qua interstate conflict which
dictates against them. Thus, if the prospect of being prosecuted for (e.g.) the crime
of aggression or incitement to genocide leads a negotiating party credibly to threaten
to resume full-scale interstate war against its neighbours, or credibly to threaten
to undermine reconstruction efforts (for example by withholding reparations),
48 For discussions of the United Nations’ position on the one hand and more generally of eligible
and ineligible crimes on the other hand, see, e.g., Freeman, Necessary Evils, 88–108 and 145–51,
respectively; see also Hayner, Unspeakable Truths, ch. 8, on the UN and amnesties.
214 Punishment
amnesty for those crimes might be justified too. In both cases, what justifies amnes-
ties is the importance of bringing about a justifiedATC peace within and beyond
borders.
Second, when conditional amnesties are sometimes justified, on what factors
may they be conditioned? Standardly, on telling the truth about one’s deeds, since
amnesties conditional on truth-telling can deliver a form of accountability which
need not go hand in hand with punishment.49 Other possibilities include the will-
ingness to go into exile, issuing apologies, and making material reparations. Note
that in all those cases except exile, to impose the condition is to impose a burden
which the agent is already under a duty to incur irrespective of what victims may
demand: wrongdoers are in any case under duties to apologize (as we shall see in
s.9.6), to pay reparations (as we saw in ch. 6), and to testify (or so I shall argue in
s.9.5).
Exile is somewhat different. To claim that wrongdoers are under a duty to leave,
irrespective of whether they are asked to do so, is to claim that they have forfeited
their occupancy rights in the community of which they are members or residents
just in virtue of having committed war crimes (whether against other members or
strangers). But on the assumption that wrongdoers have rights to the necessities of
life, they have a right to live somewhere. Unless their fellow community members
in general and/or their victims in particular ask them to leave, it is hard to see why
they are under a duty to do so of their own volition. It is in that sense that impos-
ing exile as a condition for granting amnesty is not the same as making the latter
conditional on apologies, confessions, and reparations. The question then is
whether one may justifiably grant an amnesty conditional on exile. It seems that
one may. In fact, it seems that exile tout court, unhinged from amnesty, is some-
times permitted as an extreme form of lustration, either on the grounds that
wrongdoers must somehow be punished, or on the grounds that their past deeds
have rendered them untrustworthy—so long as wrongdoers would not, in leaving
and settling elsewhere, be subject to harsher burdens than the punishment which
they are liable to incur anyway. If that is correct, exile in exchange for the granting
of an amnesty is a fortiori justified.
Third, the rationale I have offered for sacrificing punishment to peace via amnes-
ties has interesting implications for two related issues: the issue of who, of foreign-
ers and compatriots implicated in the same war crime, may be granted an amnesty;
and the question of which, of a domestic or foreign jurisdiction, may grant it.
Consider the first issue. Amnesties are justified on the grounds that their benefi-
ciaries would otherwise derail the peace process. Suppose that in the course of a
civil war in community C, C’s leadership authorize its own military forces to com-
mit grievous human rights violations against civiliansC, with the assistance of for-
eign actors. Suppose now that, post-conflict, those leaders, and not those foreign
actors, pose a threat to the peace process. C’s new regime would be justified in
49 L. Mallinder, ‘Amnesties in the Pursuit of Reconciliation, Peacebuilding and Restorative Justice’,
in J. J. Llewellyn and D. Philpott (ed.), Restorative Justice, Reconciliation, and Peacebuilding (Oxford:
Oxford University Press, 2014); see also Philpott, Just and Unjust Peace, 246ff.
Punishment 215
granting an amnesty to the former, and not to the latter. In fact, it would be justi-
fied in prosecuting and punishing the latter (subject to other feasibility constraints
such as financial costs). This of course is not a case of permissible partiality to one’s
compatriots qua compatriots: what permits C’s regime so to act is not the fact that
those former leaders are C’s members; rather, it is the fact that, in virtue of their
membership in C, they pose a threat to peace. By that token, granting amnesty to
some perpetrators but not others simply on the basis of shared nationality, for a
crime which they committed jointly, is prima facie wrongful. In a particularly egre-
gious example of such partiality, the French Parliament voted in 1953 to exempt
from punishment French, but not German, members of the SS platoon who
together had killed almost all inhabitants of Oradour-sur-Glanes in June 1944.
Given that neither the French (former) soldiers nor those who opposed their being
brought to trial were posing a threat to peace in 1953, this is an utterly condemn-
able decision from a cosmopolitan point of view.50
Consider now the issue of which, of a domestic or foreign jurisdiction, may
grant an amnesty. The claim that amnesties are sometimes morally justified, indeed
mandatory, for the sake of peace has two variants. In its strong variant, it might
mean that their beneficiaries, for all that they have perpetrated serious human
rights violations, nevertheless ought not to be punished by anyone, including exter-
nal actors acting under the principle of universal jurisdiction. In its weak variant,
it might mean that reasons not to punish only apply to domestic actors, thus leav-
ing open the possibility of foreign prosecution. The case of Chile’s General
Pinochet, who toppled democratically elected President Allende in 1973 and lost
power in 1989 is interesting in that regard. His regime was marked by terror and
repression in the course of which thousands of people were tortured to death, sent
to jail without trial, and/or executed. Pinochet and the military and judicial estab-
lishment on which he relied agreed to hand over power to a democratically elected
government (no doubt helped along the way by growing international pressure
and the withdrawal of American support), but on one condition, couched as a
threat which Pinochet himself issued: ‘No one touches anyone. The day they touch
one of my men, the rule of law ends. This I say once and will not say again’.51
Under those circumstances, Chile’s civilian government and its citizenry had very
good reasons to grant amnesty to Pinochet and his henchmen.
But now suppose that there had been an international criminal court at the
time, with the power to punish defendants in absentia. Indeed, suppose that it
would have been possible to kidnap Pinochet and the key members of his regime
for trial abroad, either by that court, or in the countries of some of his victims—
Spain and the United States, for example. Pinochet’s crimes were such as to fall
under the remit of the principle of universal jurisdiction as set out in s.7.6. In
principle, then, the fact that amnesties are granted by domestic actors does not
50 See S. Farmer, ‘Postwar Justice in France: Bordeaux 1953’, in I. Deak, J. T. Gross and T. Judt
(ed.), The Politics of Retribution in Europe: World War II and its Aftermath (Princeton, N.J.: Princeton
51 In T. Rosenberg, ‘Overcoming the Legacies of Dictatorship’, Foreign Affairs 74 (1995): 134–53,
at 134.
216 Punishment
preclude prosecution by other parties.52 If Pinochet’s threat would have been

empty in the face of those prosecutions, conducted as they would have been by
outsiders, the claim that amnesty in that case would have been justified holds only
on its weak variant—that is to say, as far as domestic prosecution is concerned. In
s.7.6, I argued that it often matters that a state whose members committed a war
crime should act as the punishing party. By reverse reasoning, it often matters that
the perpetrators’ state should not be the one to do so.
Finally, once amnesties are granted, they should be honoured, for if perpetrators
know or even only have reasons to believe that, a few years down the line when
their position is much weakened, they will be brought to trial, they are far less
likely to accept the amnesty here and now when they are in a position to derail the
peace process. However, this creates a real risk that amnesties will lead to collective
amnesia, which in turn might be detrimental for peace: in fact, the word itself,
‘amnesty’, has its etymological origin in the Greek ἀμνηστία, shares the same root
as amnesia, or ἀμνησία, and in its original sense denotes oblivion. Now, it is one
thing for victims not to see their murderers, torturers, rapists in jail; it is another
for those victims to be altogether forgotten. To be sure, many might deeply wish
not to relive their own tortured past. But many others might insist on a public
acknowledgment of their suffering, and/or private access to some of its details—for
example, where their relatives were buried, who tortured them, who assassinated
their son, that their father is dead. It is no coincidence, then, that the granting of
amnesties has often been accompanied by requests for truth commissions (notably
in Chile), or by initiatives on the part of civil society actors such as human rights
organizations, local Churches, and victim movements (notably in Brazil and
Argentina) to uncover and publicize information about past abuses.
Admittedly, there might be cases where perpetrators will resume war unless the
facts of their grievous wrongdoings are wrapped in a veil of silence or publicly
replaced by a distorting imaginary past in which they are as much sinned against
as sinning. In such extreme cases, collective amnesia within war-torn communities
might be the only way to secure peace. But there is no reason why outsiders should
remain silent—on the contrary. Thus, the fact that Turkish leaders and citizens are
for the most unwilling publicly to admit that their forebears committed a genocide
against the Armenians in 1915 is no reason for others—other communities, histo-
rians, the world at large—not to say so. In fact, it furnishes them with stronger
reasons still to make it possible for the Turks, in time, to come to terms with their
past, by preserving knowledge thereof. In s.9.5 when tackling truth commission,
and in ch. 10 when offering a cosmopolitan argument for war remembrance,
I shall return to the importance of uncovering and recording the past in the after-
math of conflict. Suffice it so say for now that failing to honour those requests, or
quelling those initiatives, when doing so would not jeopardize prospects for peace,
would add to the injury of being made to come to terms with unendurable loss
52 That point was in fact asserted by the Special Court for Sierra Leone (a hybrid of UN and
domestic actors) in connection with the amnesties granted to perpetrators of atrocities, on both sides
of the civil war, in the Lome Agreement of 1999. I am grateful to Joanna Rozpedowski for bringing
this case to my attention.
Punishment 217
whilst perpetrators go free, the insult of being made to do so purely in private, as

if it were no one’s concern but one’s own.
7 . 8 . C O N C LU S I O N
The view that agents who significantly contribute to the commission of war-related
crimes may, indeed ought to, be punished is consensually held. However, justifying
it is harder than appears at first, in the light of the sheer complexity—at the bar of
both conduct and fault—of those crimes. In this chapter, I have sought to show
which war-related crimes warrant punishment and under what conditions. In so
doing, I have attempted to address two of the main challenges encountered by the
neoclassical account of killing in war. The first challenge is that war wrongdoers’
individual contributions to the macro-threats of war are so marginal when taken
on their own that they seemingly do not warrant punishment. The second chal-
lenge is that, on the face of it, the neoclassical account treats combatants who kill
enemy combatants in prosecution of an unjust war as murderers and thus as
deserving punishment, even though most of us in fact do not claim that those
combatants all belong in jail and do not behave around them as if they were gang
members; nor do we so describe, and relate to, civilians who actively support an
unjust war. To block those challenges, I have defended forms of punishment which
fall short of imprisonment, such as punitive lustration and confiscation—though
I have rejected punitive reparations. I have also conceded that amnesties are some-
times morally justified, indeed mandatory, for the sake of peace. Finally, in keeping
with the cosmopolitan underpinnings of my account of war, I have defended a
version of the principle of universal jurisdiction.
Throughout, I have claimed that punishing war wrongdoers is not merely justi-
fied as a matter of right: it is also a moral imperative, failure to respect which is,
more deeply and at least prima facie, a failure properly to respect individuals’
human rights. It is in that sense that punishment is a key feature of a just peace
simpliciter. At various junctures in this chapter, however, I have suggested that
amnesties, and some processes of reconciliation, rather than or at least alongside
punishment, might sometimes offer better prospects for a justifiedATC peace. I shall
address reconciliation in ch. 9. Beforehand, however, we must consider the institu-
tional conditions under which punitive justice, alongside its restitutive, reparative,
and distributive counterparts, may justifiably be implemented and enforced.
8
Transitional Foreign Administrations
Restitution, reparations, and punishment are all constituents of a justifiedATC

peace. Throughout those chapters, I have touched on some of the institutional
questions raised by (inter alia) the moral imperative to rebuild war-torn commu-
nities, conflicts between the demands of punitive justice and the requirement not
to jeopardize peace, and so on. In this chapter, I address a different institutional
question (though again from a normative point of view), namely that of which
actors are prima facie morally justified in enforcing peace in general and peace
agreements in particular.
That question is both related to, and significantly different from, the issue of
peacekeeping and military occupation, which we tackled in ch. 3. There, I argued
that a political actor—be it a state, a coalition thereof, or an international organi-
zation such as the UN—has the right to conduct a military occupation or deploy
peacekeeping forces on the territory of a community if a section of that communi-
ty’s population presents a serious threat to the human rights of some other party.
My focus was on military and peacekeeping operations which occur while a war is
ongoing or in its immediate aftermath—in other words, prior to the conclusion
of a peace settlement. My concern here is somewhat different. It is with what the
empirical literature on peace has called peace-building operations, which occur
following a peace settlement and which aim to turn war-torn communities into
stable, liberal democratic societies.
The most comprehensive of those operations issue in so-called transitional
foreign administrations (TFAs). In law, TFAs differ from both occupying admin-
istrations and peacekeeping operations of the kind studied in ch. 3, precisely
because they occur following a peace settlement. Furthermore, and as we saw in
ch. 3, Occupiers may replace extant local laws only if the latter violate the laws of
war, if the preservation of law and order in the occupied territory requires that they
should pass new laws, or if their war effort outside the occupied territory (hence-
forth, OT) requires new laws. However, TFAs assume the legal power wholly to
overhaul the legal system in place at the start of their mission if necessary—though
they do not become sovereign, since territorial sovereignty vests in individuals
themselves. Rather, they take on the full range of executive, legislative, and judicial
competences traditionally exercised by sovereign states, until such time as the
community over which it so governs is able to do so itself. This often includes, not
just writing up constitutions and reforming and training the police, but also (inter
Transitional Foreign Administrations 219
alia) running hospitals, and rebuilding and managing basic transport, water, and
electricity infrastructure.1
Notwithstanding those differences between occupation and transitional foreign
administrations, I aim to develop and extend my earlier argument for the former
into a defence of the latter. I thus set aside less ‘invasive’ missions undertaken
elsewhere by the UN: if transitional foreign administrations are justifiable, then a
fortiori and pending arguments to the contrary, so are lighter touch approaches
to peace-building; and so is a decision not to remove the defeated regime while at
the same time containing it with measures such as trade restrictions, disarmament,
and so on.2
At first sight, this task, which has been almost entirely neglected by just war
theorists, should be relatively easy.3 For if, as cosmopolitans aver, national and
political borders are largely irrelevant from a moral point of view, and if military
occupation is justified as a means to thwart threats to human rights, then it does
seem that national governments and international organizations have a prima facie
justification for turning war-torn communities into self-governing and legitimate
entities. The case seems all the stronger in the light of the fact, stressed by advocates
of the so-called liberal peace thesis, that liberal societies on the whole do not go to
war against each other, and (notwithstanding the invasions of Afghanistan in 2001
and Iraq in 2003 by the United States and their allies) are less likely than illiberal
societies to go to war tout court. In fact, if the liberal peace thesis is correct, it might
seem, more strongly still, that populations of illiberal societies are under a duty to
subject themselves to transitional administrations so long as the latter can help
turn them into peace-promoting societies.4
1 See Bellamy, Griffin and Williams, Understanding Peacekeeping, 255. For an interesting account
of the concept of ‘suspended sovereignty’, or ‘sovereignty in abeyance’, as deployed in public interna-
tional law, see Yannis, ‘The Concept of Suspended Sovereignty in International Law and Its
Implications in International Politics’. On the legal status of territories administered by transitional
administrations, with specific reference to the concept of sovereignty, see R. Wilde, International
Territorial Administration: How Trusteeship and the Civilizing Mission Never Went Away (Oxford:
Oxford University Press, 2008), esp. ch. 3.
2 See, for example, the cases of Afghanistan in the early 2000s, in Namibia in 1989–90, or in
Cambodia in 1991–93, discussed in (e.g.) Bellamy, Griffin and Williams, Understanding Peacekeeping.
One of the difficulties with this less invasive approach—sometimes dubbed the ‘light footprint’
approach—is that it favours the entrenchment of local hierarchies. It can also serve to justify con-
demnable underinvestment by international actors.
3 I know of only one book-length substantially philosophical treatment of the issue, in the form of
C. Hermanson’s doctoral thesis. See C. Hermanson, ‘Duties in the Wake of Atrocity: A Normative
Analysis of Post-Atrocity Peacebuilding’ (D.Phil thesis, University of Oxford, 2012). For a much briefer
defence, with specific reference to Iraq post-2003, see N. Feldman, What We Owe Iraq—War and the
Ethics of Nation Building (Princeton, N.J.: Princeton University Press, 2004).
4 For a classic late twentieth-century defence of the liberal peace thesis (which has its theoretical
origins in Kant’s moral and political philosophy, and its political roots in Wilsonian internationalism),
see M. W. Doyle, Ways of War and Peace: Realism, Liberalism, and Socialism (New York; London:
Norton, 1997); M. W. Doyle, ‘Kant, Liberal Legacies, and Foreign Affairs’, Philosophy & Public Affairs
12 (1983): 205–35; M. W. Doyle, ‘Kant, Liberal Legacies, and Foreign Affairs, Part 2’, Philosophy &
Public Affairs 12 (1983): 323–53; M. W. Doyle, ‘Liberalism and World Politics’, The American Political
Science Review 80 (1986): 1151–69. For a highly critical stand on the thesis, see, e.g., S. Rosato, ‘The
Flawed Logic of Democratic Peace Theory’, American Political Science Review 97 (2003): 585–602. As
220 Transitional Foreign Administrations
However, the view that TFAs are a morally acceptable means towards a justifiedATC
peace is vulnerable to the charge of neocolonialism. That cosmopolitanism should
elicit this particular criticism is not surprising—though the criticism is often
deployed under the label ‘liberal imperialism’. For cosmopolitanism, it is often
averred, is insensitive to the distinct identity and salience of national cultures, rides
roughshod over individuals’ culturally situated understandings of the demands of
morality, and, when enforced by political and military might, offers no promise
but that of a contemporary form of colonialism. At the very least, sceptics will
press, even if the charge is overstated, TFAs nevertheless rest on a paradox, that of
imposing democracy and liberalism by force. Moreover, although the enormous
political, economic, and social difficulties which those administrations aim to
solve are exacerbated by war, they do not always originate in it. To advocate the
direct governance of a given community by largely foreign institutions is thus not
to advocate a return to the status quo ante bellum: more often than not, it is to
advocate a profound reshaping of that community to something which it was not
before the war—in the same way as colonial powers would seek to transform
their colonies.5
This raises a difficulty for cosmopolitans, for however willing they might be to
advocate outside interference in the affairs of political communities whose regimes
do not abide by the fundamental tenets of cosmopolitan justice, qua cosmopoli-
tans they ought to reject any principle which ultimately rests on, or leads to, the
denial of equal concern and respect. To the extent that colonialism exemplifies
such denial, they must either rebut the charge or renounce TFAs as a morally jus-
tified means to bring about a justifiedATC peace.
Faced with this particular challenge, I proceed as follows. I begin with a histor-
ical overview of the practice (s.8.2), and move on to offering a preliminary justifi-
cation for it (s.8.3), which I then qualify in the light of an extended discussion of
the colonialism objection (s.8.4).
A few caveats before I start. First, the label I use, ‘transitional foreign adminis-
tration’, is unusual. The empirical literature on peace-building tends to use either
the words ‘transitional administration’ or the words ‘international administration’.
For my purposes, however, neither will do. Strictly speaking, ‘transitional adminis-
tration’ can denote a domestic institution tasked with carrying out post-conflict
constitutional reforms. (A good example is the provisional government of the
French Republic, which governed over France between 1944 and 1946 and over-
saw that country’s transition from the Vichy collaborationist dictatorship to the
Fourth Republic.) The label ‘international administration’, for its part, implies that
a transitional administration counts as such only if it is carried out by an interna-
tional institution such as the UN. Whilst in practice transitional administrations
Rosato notes, the fact that the US supported military coups in Brazil (1964) and Chile (1973), which
were then fledgling though left-leaning democracies, casts some doubt on the coherence of the thesis.
5 For expositions and discussions of the colonialism objection in the context of transitional admin-
istrations, see, e.g., Chesterman, You, The People, 1; Wilde, International Territorial Administration,
ch. 8; D. Zaum, The Sovereignty Paradox: The Norms and Politics of International Statebuilding (Oxford:
Oxford University Press, 2007), ch. 6; Caplan, International Governance of War-Torn Territories.
since 1945 have fit that model, and whilst (as we shall see) there are reasons to
think that international organizations are better suited to the task, my concern is
with the view that a given community may be restructured, shaped, and gov-
erned by foreign actors, whether foreign national governments or international
organizations.
Second, although I scrutinize post-war transitional foreign administrations, this
particular mode of governance is sometimes used in order to forestall conflict.
Thus, in 1962, the UN set out to administer the province of then West Irian (now
West New Guinea) as a means to settle an escalating dispute between the Dutch
and the Indonesian governments over its status—a status which had been left
undecided pending further negotiations when the Netherlands had recognized
Indonesia’s independence in 1949.6 A full study of those administrations, which is
outside my remit, would need to tackle preventive cases.
Third, my aim is not merely to show that TFAs are a morally justified means to
enforce the substantive requirements of a justifiedATC peace, which I defended in
chs. 5–7. As I noted at the close of ch. 4, given that a justifiedATC peace requires
that individuals’ human rights be respected, a peace agreement must stipulate that
all parties will respect those rights, including liberal and democratic rights, on pain
of being unjustified all things considered. Accordingly, I defend TFAs not just on
the grounds and to the extent that they can under some circumstances enforce a
justifiedATC peace, but also on the grounds and to the extent that they can under
some circumstances bring about a liberal democratic regime—in other words,
regime change.
Finally, as the empirical literature shows, the success of a given mission is largely
contingent on specific local circumstances such as the size of the territory under
administration, the ethnic composition of its population, the nature, length and
severity of the conflict, and the features of its pre-war institutions. One may won-
der, therefore, what a normative inquiry could possibly do, other than articulate a
one-size fits-all, and therefore unfit-for-implementation, blueprint for a just tran-
sitional foreign administration. However, local factors are in part shaped by outsid-
ers, not least foreign interveners or peacekeepers whose conduct and decisions may
well have a lingering impact on local populations and institutions.7 It thus pays to
articulate the moral norms which ought to regulate transitional administrators’
acts and decisions.
8.2. M A N D AT E S , T RU S T E E S H I P, A N D T R A N S I T I O N A L
F O R E I G N A D M I N I S T R AT I O N S
Virtually all book-length empirical studies of TFAs offer an overview of the prac-
tice and note that its contemporary instantiations, such as the UN Transitional
Administrations in Bosnia, Kosovo, and East Timor have their roots in colonial,
6 See Wilde, International Territorial Administration, 13.

7 See Caplan, International Governance of War-Torn Territories, 27.
or quasi-colonial, administrations going as far back as the Roman Empire (to
name but one long-past example) and, more recently, the Mandates system estab-
lished under the auspices of the League of Nations following the First World War
and the Trusteeship system set up by the United Nations following the Second
World War.
I shall review the case of colonial administrations when scrutinizing the coloni-
alism objection below. To avoid pre-empting that discussion, I begin my historical
overview with the Mandates and Trusteeship systems and compare them with the
aforementioned three UN missions, whose central features are a recurrent illustra-
tive theme of the chapter.8
One of the most important and difficult tasks which the US, Britain, France,
and Italy took on at the 1919 Paris Peace Conference was to decide what to do
with the colonial empires of their defeated enemies. While President Wilson held
that German colonies should be administered by the to-be-formed League of
Nations, his three counterparts took the rather different view that those colonies,
as well as the Ottoman Empire, should be carved up between the winners (not
forgetting Australia, New Zealand, and South Africa, all of whom had designs on
this or that territory) on the basis of the good old-fashioned ‘spoils of war’ princi-
ple. The outcome of protracted, often bitterly divisive negotiations between the
allied powers was a two-pronged system: the government and administration of
some of those colonies were entrusted under the authority of the League to a given
state (Britain, thus, ‘got’ Palestine, while France got the best part of Cameroon,
and Belgium—at the last minute—was given Rwanda), whilst some territories
(notably the Saar Basin and the city of Danzig) were governed directly by the
League via an international commission or administrator.
The Mandates system served different purposes. On the one hand, it provided a
solution to intractable territorial disputes. Thus, removing the city of Danzig from
the jurisdictions of both Poland and Germany conferred protection on the city’s
largely German population while safeguarding Poland’s access to the sea. Likewise,
the Saar Basin, which was still legally part of Germany, was administered by the
League as a means to enable the French securely to exploit its mines without
threatening German jurisdiction over it. On the other hand, the system was meant
to provide secure governance to communities (in Africa and the Middle East)
deemed unable to govern themselves, and until such time as they could properly
exercise their right to self-determination. Of course, the system was but a fig leaf
for the old practice of outright annexation after war and soon made itself vulnera-
ble to the charge of Western imperialism. Legitimate concerns about the efficiency
of international administrations established by the League may well have driven
Western states to disfavour the system. That said, there is little doubt that those
states did regard the newly acquired territories as just that: acquisitions. In fact,
only some of the territories held as Mandates were earmarked for independence
8 What follows in this section heavily draws on the works mentioned in fn. 4. For a clear and con-
cise account of the establishment of the Mandates system by the Paris Pace Conference, see Macmillan,
Peacemakers, ch. 8.
(so called Class A mandates, principally located in the former Ottoman Empire

and deemed very close to self-sufficiency), while the others came to be seen as parts
of the British and French colonial empires.
Although the League of Nations did not survive the outbreak of the Second
World War, the aspirations which it embodied led to the creation of the United
Nations. The Mandates system itself partly survived through the Trusteeship sys-
tem enshrined in Ch. XII of the UN Charter. As Simon Chesterman notes, in
1945 roughly a third of the world’s population lived on territories which were
regarded, legally speaking, as non-self-governing—some of which were still parts
of a colonial empire (notably Britain’s and France’s) and some of which were placed
under the Trusteeship system.9 Under ch. XII, all non-self-governing territories are
placed under the oversight of the United Nations, provided that they meet one of
the following conditions: they were held in mandate following the First World
War, they were separated from their original state as a result of the Second World
War, or (relevantly to contemporary world politics) they have been placed under
the system by states currently responsible for their administration (art. 77). Under
Trusteeship, and in contrast with the Mandates system, supervising authorities
must aim to enable all non-self-governing territories to achieve independence
(which in fact all of the eleven territories held in trust in 1945 eventually did).
Moreover, the legal status of trusteeship cannot be conferred on ‘territories which
have become Members of the United Nations, relationship among which shall be
based on respect for the principle of sovereign equality’ (art. 78).
Territories which were held in trust in 1945 were not governed directly by the
UN but, rather, by the Mandatory power in place at the outbreak of the Second
World War. By contrast, the UN Trusteeship Council was tasked with supervising
the governing powers, receiving complaints from local populations, and sending
inspectors to ensure that the former were fulfilling their legal obligations to the
latter. Although the governing power exercised the right to govern over the terri-
tory, it did so in trust and on behalf of its holder, to wit individual members of the
community in question, and having been authorized to do so by the international
community. A similar principle informed the UN’s view of yet another category of
non-self-governing powers, namely colonies held by France, Britain, Spain, and
Portugal. At the time of writing, there remain seventeen such territories—totalling
a population of c.1.6 million.10
All the territories regarded as non-self-governing by the UN were, or are, admin-
istered by a sovereign state. In only two cases, that of Trieste and Jerusalem, was the
possibility of direct UN administration considered—and abandoned. Nevertheless,
the Mandates legacy of direct administration by international institutions in the
1920s and 1930s, creative interpretations of ch. XII of the UN Charter, and
the extensive enforcement powers granted to the Security Council by Ch. VII of
9 Chesterman, You, The People, 37.

10 So-called non-self-governing territories fall within the remit of Ch XI of the UN Charter. For
data on those territories, see http://www.un.org/en/decolonization/nonselfgovterritories.shtml (accessed
on 15/01/2016).
the Charter, paved the way for the conduct of complex, state-building missions in
the late 1990s/early 2000s—most notably in East Timor, Bosnia, and Kosovo.
East Timor—now Timor Leste—which lies to the south-east of the Indonesian
archipelago—gained its fragile independence in 1974 from Portugal when the
latter’s regime was overthrown by the so-called Carnation Revolution. In 1975
however, Indonesia invaded East Timor, over which it ruled by terror until the
newly elected President of Indonesia renounced his country’s claim to the territory
in 1999. The status of East Timor had already been at the heart of difficult, halting,
and unsuccessful negotiations between Indonesia, Portugal, and the UN—from
which the East Timorese resistance parties had been excluded. Following President
Habibie’s announcement, the three negotiating parties agreed on a date for a ref-
erendum on independence to be monitored by the UN, and on the deployment of
a peacekeeping mission to oversee the transition towards self-government should
this prove necessary. The East Timorese voted overwhelmingly in favour of inde-
pendence, following which the Indonesian Army withdrew en masse while paramil-
itary groups opposed to the referendum’s outcome launched a devastating scorched
earth policy. By September 1999, two thirds of all buildings had been destroyed,
no administration remained, and over two thirds of the population was on the
move. At first, the UN authorized the deployment of an international force led by
Australia and made up of troops from South East Asia, invoking as the basis for its
decision Ch. VII of its Charter, in virtue of which the Security Council may
authorize the use of military force as a means to maintain peace and security. Once
stability of some kind was restored, however, it became clear that the Timorese
lacked the infrastructure, funds, and, crucially, human resources to govern them-
selves, and that outside help and intervention were necessary. This took the form
of the United Nations Transitional Administration for East Timor (UNTAET),
which, with the support and growing involvement of the local population, gov-
erned the country until 2002 under the terms of Security Council Resolution
1272. Backed by a UN peacekeeping force of several thousand combatants,
UNTAET oversaw elections to a constitutional assembly, the reform of the legal
system and the judiciary, the restructuring of the civil service, and the rebuilding
of the infrastructure.11
The task of international administrators was rendered less complex for the fact
that the East Timorese themselves were not locked in a civil war and had united
before UNTAET’s arrival in favour of independence. Moreover, there had already
been an unambiguously clear referendum on that very issue, which provided tran-
sitional administrators with a clear mandate for their mission at the outset—to
wit, enabling East Timor to become a self-governing, human rights–respecting
independent state.
11 See http://www.un.org/en/peacekeeping/missions/past/etimor/etimor.htm (accessed on 15/01/2016),

and Howard, UN Peacekeeping in Civil Wars, ch. 8; O. Tansey, Regime-Building: Democratization and
International Administration (Oxford: Oxford University Press, 2009), ch. 3; Zaum, The Sovereignty
Paradox, ch. 5. Another case of explicitly and intended direct governance by the UN is Eastern
Slavonia. See Howard, UN Peacekeeping in Civil Wars.
Consider now the case of Bosnia and Herzegovina, whose population in 1995,
at the conclusion of the civil war which followed the break-up of the former
Republic of Yugoslavia, comprised Serbs, Croats, and Bosnian Muslims (Bosniaks).
Having voted in favour of independence from Yugoslavia in 1992 against the
wishes of the majority of Bosnian Serbs, the Republic of Bosnia and Herzegovina
was attacked by Serbia in the spring of that year. It was under threat of breaking up
when, in 1993, the secessionist Croatian province of Herzeg-Bosnia launched a
war of independence with the backing of Croatia. Following intense diplomatic
and military pressure from key members of the UN and NATO, the parties agreed
to reach a peace settlement. The General Framework Agreement for Peace in
Bosnia and Herzegovina, or Dayton Agreement, was signed by President Slobodan
Milosevic of Yugoslavia on behalf of Bosnian Serbs (in the absence of their President
Radovan Karadzic), President Franco Tudjman of Croatia on behalf of Bosnian
Croats, and President Alija Iztebegovic for the Republic of Bosnia-Herzegovina.
The French and American presidents, the British and Russian prime ministers and
the German Chancellor signed the Agreement as witnesses.
The clauses of the Agreement are far-ranging: they provide, inter alia, for the
establishment of an international peacekeeping force, for the conduct of demo-
cratic elections under supervision by the Organisation for Security and Cooperation
in Europe (OSCE), and for the conferral of extensive powers to NATO regarding
the demilitarization of the country. In addition, Annex 4 in effect serves as the
country’s new constitution, along liberal-democratic lines; other Annexes cater for
the preservation of human rights, the establishment of an international police task
force, a commission to preserve national monuments, and various commissions for
transports and utilities. Crucially for our purposes here, Annex 10 stipulates that,
‘In view of the complexities facing them, the Parties request the designation of a
High Representative, to be appointed consistent with relevant United Nations
Security Council resolutions, to facilitate the Parties’ own efforts and to mobilize
and, as appropriate, coordinate the activities of the organizations and agencies
involved in the civilian aspects of the peace settlement’.12 Initially, then, the coun-
try was to be governed by and through its local institutions and officials, in consul-
tation and with the help of an array of international organizations, and under the
supervision of the High Representative and the so-called Peace Implementation
Council, comprising representatives of the states and international organizations
involved in the peace process. However, when distrust amongst the three ethnic
groups and tensions between local and international officials threatened to derail
the process, the Council met in summit in Bonn and endorsed the High
12 See http://avalon.law.yale.edu/subject_menus/daymenu.asp (accessed on 15/01/2016) and

http://www.un.org/en/peacekeeping/missions/past/unmibh/ (accessed on 15/01/2016).
For clear accounts of international actors’ involvement in Bosnia, see, e.g., Chesterman, You, The
People, esp. 128–31, for the High Representative’s powers; Tansey, Regime-Building: Democratization
and International Administration, ch. 5; Zaum, The Sovereignty Paradox, ch. 5; Bellamy, Griffin and
Williams, Understanding Peacekeeping. 266–73; Fox, Humanitarian Occupation, esp. 74–84; E. M.
Cousens, ‘From Missed Opportunities to Overcompensastion: Implementing the Dayton Agreement
on Bosnia’, in S. J. Stedman, D. Rotchild and E. M. Cousens (ed.), Ending Civil Wars—The
Implementation of Peace Agreements (Boulder and London: Lynne Rienner Publishers, 2002).
Representative’s decision to use his existing powers of interpretation of the Dayton

Agreement more broadly—so broadly in fact as to make binding decisions (includ-
ing removal from office) against local officials deemed to act in violation of the
Agreement. Unlike Resolution 1272, the Dayton Agreement did not explicitly pro-
vide for a transitional foreign administration. But institutional paralysis on the
grounds, subsequent mission creep, and the conferral on the High Representative of
so-called Bonn Powers in effect turned Bosnia over to the UN Mission (UNMIBH),
notwithstanding the continuing and growing participation of local elites.
In those two cases, East Timor and Bosnia, the transitional administration’s
objective was clear. The case of Kosovo is markedly different. There the UN admin-
istration embarked on its institution-building enterprise without having settled the
question of the legal status of the province—a question which remains open at the
time of writing. Kosovo, whose population is Albanian in its majority, had enjoyed
a certain degree of autonomy within the former Republic of Yugoslavia since 1969,
but when its self-declared parliament announced the province’s independence in
1990, Yugoslav and Serb President Milosevic revoked its privileges. Following a
few years of growing violence opposing the Serbian Army and Albanian resistance
movements, the international community sought to end the conflict through the
so-called Rambouillet Accords, which demanded that Serbia restore Kosovo’s
autonomy and allow peacekeeping forces to enforce the ceasefire. Faced with
Serbia’s refusal to sign the Accords and its ongoing policy of ethnic cleansing in
Kosovo, NATO carried out a bombing campaign in the spring of 1999 (without
authorization from the Security Council), which lead to the withdrawal of Serb
forces from Kosovo. Under the terms of Security Council Resolution 1244, Kosovo
was placed under a UN-led transitional administration (UNMIK) backed by a
NATO-led military force. But whereas a majority of Kosovo’s population did wish
for independence and full sovereignty, resolution 1244 reaffirms that Kosovo is to
remain part of Yugoslavia. In defiance of resolution 1244, Kosovo proclaimed its
independence in 2008 and has been recognized as an independent state by over
one hundred UN members (in particular the permanent members of the Security
Council with the [unsurprising] exception of Russia). It is not a member of the
UN. Its legal status is still under dispute (despite the International Court of Justice’s
advisory opinion to the effect that its declaration of independence does not violate
international law), not least because Serbia, with Russia’s backing, still considers it
to be part of its territory.13
UNMIK’s mandate was to steer the three wartime regimes to build self-governing,
human rights–promoting institutions which would enable Kosovo to enjoy auton-
omy within the (then) Republic of Yugoslavia. Although the overall stated aim of
the mission was to hand over the administration of the territory to Kosovars, the
latter were given only a consultative (and, according to critics of the process, inef-
fectual) role at first. UNMIK took responsibility for drafting a constitutional
13 See http://www.un.org/en/peacekeeping/missions/unmik/ See also, for detailed studies of
UNMIK’s transitional work between 1999 and 2008, Tansey, Regime-Building: Democratization and
International Administration, ch. 4; Zaum, The Sovereignty Paradox, ch. 4; Murphy, UN Peacekeeping
in Lebanon, Somalia and Kosovo.
framework, preparing for elections, establishing the rule of law, rebuilding infra-
structures destroyed by years of conflict, overseeing economic reconstruction, and
providing conditions for political reconciliation.
In the case of East Timor, the transitional foreign administration assumed the
powers and privileges of government as part of a process of decolonization. In
Bosnia-and-Herzegovina and Kosovo, it did so as part of a peace process to which
international institutions and other states had been parties. In other cases, such as
Congo (1960–64), Somalia (1993–95), and Sierra Leone (1999), the UN stepped
into the breach left by the near collapse of domestic state institutions.14 Across
those differences, it is an explicit aim of transitional foreign administrations as they
have operated in practice that they should enable the target community not only
to become self-governing (albeit not necessarily as an independent, sovereign state,
as the example of Kosovo attests), but also to be a liberal society. In the remainder
of this chapter, I argue that, at the bar of cosmopolitan justice, and notwithstand-
ing understandable worries raised by the colonialism objection, transitional foreign
administrations of that kind are (sometimes) morally justified.
8.3. J U S T I F Y I N G T R A N S I T I O N A L F O R E I G N
A D M I N I S T R AT I O N
8.3.1. TFAs as a Form of Occupation

I noted above that transitional foreign administrations are deployed following a
peace settlement to which belligerents typically (viz. Bosnia and Herzegovina and
Kosovo) though not always (viz. East Timor) consent. To the extent that the estab-
lishment of the administration is a component of the settlement, and to the extent
that belligerents take part in the peace negotiations and sign on the resulting set-
tlement, as was the case with the Dayton Agreement, parties can be deemed to
consent to the administration. To be sure, more often than not, they do so under
the threat of coercion. However, as we saw in s.4.3, the fact that parties to a peace
settlement have been coerced by military action to come to the negotiating table
does not in itself invalidate their consent to the settlement if its terms are inde-
pendently justified. The crucial question, then, is whether transitional foreign
administrations admit of an independent justification.
In s.1.2, I argued that states are legitimate if and only if, through the laws which
they vote and enforce and the executive decisions which they make on the basis of
those laws, they respect the fundamental rights of both the states’ members and
outsiders. Moreover, it is precisely because their directives provide agents with bet-
ter conditions for the protection of their moral rights and the fulfilment of their
moral duties than they would have in the absence of any state or than if they were
included in some or other feasible state that agents have a reason to regard those
14 See Chesterman, You, The People, 83–7, for a brief discussion of these three cases. I set them aside
here as my concern is with transitional administrations following a peace settlement.
directives as binding. The point applies whether the right to govern is exercised by
domestic or foreign political actors. To the extent that a transitional foreign admin-
istration provides the services which governments normally ought to provide, and
that in so doing it enables those who are subject to its directives better to exercise
their rights and fulfil their obligations, and thereby to enjoy conditions for a flour-
ishing life, its officials have the morally justified power to govern. One might think
that, in so far at the administration is foreign, and that members of the target
community thus do not enjoy full political rights, which are a subset of human
rights, its officials do not have that power. However, it is precisely because those
individuals are not as yet in a position to take on the reins of government that the
foreign administration is called in. If it does provide conditions for effective
self-government by local actors, then it fulfils the legitimacy condition for govern-
ment in general.
Let me put the point somewhat differently. Sometimes, the regime of a war-torn
community is not yet able to secure for its members the conditions under which
they can have a flourishing life. Granted, there are cases where the evidence in
support of that claim is not clear-cut. Were that so, and even if as a matter of fact
that regime is indeed unable to do what it must do in order to have the power
to govern, outsiders ought not to intervene unless specifically asked to do so.
Sometimes, however, the evidence is there to see: inability of erstwhile enemies to
muster enough trust in each other to govern effectively, rampant corruption at all
levels of government, skyrocketing rates of unemployment, particularly among
young men who thus become prime recruits for criminal gangs, endemic violence,
rising mortality rates as a result of a complete breakdown of the healthcare system,
and so on.
The human rights violations which are part and parcel of this kind of political
and social breakdown provide outsiders with a just cause for assuming authority
over it. The claim that there can be such a just cause draws strength from justifica-
tions for so-called transformative occupations, of which Germany between 1945
and 1949 and Iraq in 2003–04 are paradigmatic examples. In surrendering,
German leaders agreed on behalf of their fellow citizens, though without the lat-
ter’s formal authorization, not just to a military occupation but to the Allies’
‘supreme authority’ over their country. This included any measures which the
occupying powers deemed necessary for the preservation of peace and security.
More recently, and admittedly far more controversially, the US-led coalition’s
occupation of Iraq following the defeat and fall of Saddam Hussein’s regime in
2003 saw the so-called Coalition Provisional Authority embark on a wide range
of reforms.15
It is precisely because it consists in reshaping the occupied community politi-
cally, economically, and socially—above and beyond the constraints set out by the
15 The terms of the surrender were set out in their final form by F. D. Roosevelt, W. Churchill, and
J. Stalin at the Yalta Conference in February 1945. See http://avalon.law.yale.edu/wwii/yalta.asp
(accessed on 15/01/2016). For a concise account of the measures taken by occupying forces, see
Chesterman, You, The People, 25–37. For a detailed account of that occupation, see MacDonogh, After
the Reich. For the case of Iraq, see (e.g.), Allawi, The Occupation of Iraq; Fox, ‘The Occupation of Iraq’.
laws of military occupation which we examined in s.3.3—that this kind of

occupation is called transformative. Measures taken by transformative occupiers in
Germany and Iraq have included the provision of a constitutional framework, the
establishment or rebuilding of a financial sector, reforms of the police and judicial
system, and the abolition of oppressive institutions such as child labour.16
That there sometimes is a just cause for transformative occupation flows from
arguments in favour of military occupation traditionally understood. As we saw
in s.3.2, a military occupation is justified in the first instance as a self-defensive
measure against the occupied community whose regime conducted an unjust war
against A, or as a measure to protect oppressed minorities from the resurgence of
unjust violence at the hands of their compatriots. On both counts, it is justified as
a means to respect and promote human rights. If transformative measures of the
kind just described are a necessary means to respect and promote human rights,
then they are justified.
Furthermore, transformative occupation can also be justified by appeal to cosmo-
politan justice. At the bar of cosmopolitan justice, you recall, all human beings wher-
ever they reside, have rights to the freedoms and resources which they need in order
to lead a flourishing life; moreover, territorially bounded state institutions are an
important mechanism through which to discharge universal obligations of justice;
finally, once those obligations are discharged, individuals have the right to shape
their collective future through membership in political associations. Accordingly,
they enjoy civil, welfare, and political rights, not merely against their own govern-
ment, but also against foreign actors, including foreign states. Accordingly, under
post-conflict circumstances A, as Occupier, must to the best of its ability act in the
same way towards non-liable occupied civiliansB as it does towards its own. Thus, in
so far as justice requires of it that it should secure the material needs and fundamental
freedoms of its citizens, it also requires of it that it secure the material needs and
fundamental freedoms of those over whom it exercises temporary jurisdiction.
Cosmopolitan justice thus has no difficulty with a transformative occupation aiming
to restore the state in war-torn communities, so long as the institutions which
Occupier establishes are an effective conduit for the realization of justice across
borders and thereby effectively promote a justifiedATC peace.
At this juncture, some loose ends need tying. First, it might seem that if an
occupier is justified in issuing transformative directives to occupied populations,
transformation itself can be a just cause for war. To many, this would be a worrying
16 For scepticism on transformative occupation, see, e.g., N. Bhuta, ‘The Antinomies of
Transformative Occupation’, European Journal of International Law 16 (2005): 721–40; S. Chesterman,
‘Occupation as Liberation: International Humanitarian Law and Regime Change’, Ethics &
International Affairs 18 (2004): 51–64; K. Z. Marten, Enforcing the Peace: Learning from the Imperial
Past (New York: Columbia University Press, 2004). For a more sympathetic take, see M. Sassòli,
‘Legislation and Maintenance of Public Order and Civil Life by Occupying Powers’, European Journal
of International Law 16 (2005): 661–94; A. Roberts, ‘Transformative Military Occupation: Applying
the Laws of War and Human Rights’, The American Journal of International Law 100 (2006): 580–622.
There is an interesting debate in the legal literature on the degree to which the law of occupation can
accommodate transformative occupation. See the sources already mentioned here, as well as Scheffer,
‘Beyond Occupation Law’; for an extensive review of those cases, see Fox, Humanitarian Occupation,
ch. 3 and ch. 7/s. III.
implication of the view defended in this chapter, since it would pave the normative
way for more wars than would seem justified. However, the claim that an occupa-
tion can be justifiably transformative neither implies nor entails that waging war to
transformative ends is also justified, for war, unlike occupation, almost necessarily
involves killing, and the justificatory bar is thus far higher for war than it is for
occupation. In fact, as I argued in CW (ch. 5), a regime change of that kind is not
itself a just cause for a war of intervention.
Second, consider the case of a transformative occupation which proceeds from
an unjust war. Suppose that although Occupier lacked a justification for occupying
B in the first instance, it nevertheless brings about not just an end to endemic violence
but in addition confers considerable benefits to the occupied population—such as
liberal democratic institutions, a tolerant culture, welfare provision, etc.17 Is it
justified in so doing? I suspect so. For if the occupation would bring those benefits,
and if the (temporary) harms of the occupation were proportionate to those bene-
fits, Occupier may well acquire such justification. In fact, this is how some
who opposed the invasion of Iraq in 2003 subsequently sought to justify the
occupation—on the (overly optimistic) grounds that it would turn Iraq into a
liberal democratic state at the acceptable cost for the Iraqi of their temporary ina-
bility to exercise their sovereignty rights. That the coalition conducted itself very
badly as a transformative occupier is beyond dispute. However, the thought that an
initially unjust Occupier might nevertheless enable those which are de facto subject
to its directives to do better at the bar of their fundamental rights and duties than
they would under any other available alternatives is still worth considering.
8.3.2. The Difference a Treaty Makes

If those points in favour of transformative occupation are correct, they suggest that
there sometimes is a prima facie justification for transitional foreign administrations,
precisely because the latter carry out the same governmental tasks as transformative
occupiers. However—and this is crucial—to the extent that, unlike transformative
occupiers, those administrations are called upon to discharge powers of government
following a peace settlement, the foregoing all-things-considered justification for
transformative occupation prior to a peace settlement applies without further ado
only if the existence of the settlement has no normative bearing on the moral status
of those administrations. To see whether this is so, we must distinguish cases where
the transitional foreign administration is established by the settlement itself with a
mandate to enforce the latter, and cases where it is not established by the settlement
but is called upon ex post to enforce it.
Let me begin with the first case. In s.4.2, I distinguished between declarative,
specifying, and creative treaty clauses. Declarative clauses set out independently
justified rights and duties. Specifying clauses render more precise and allocate
those rights and duties, in ways which are not themselves generated by independ-
ent principles (such as the ‘the best qualified person must do it’ principle). Creative
17 I am grateful to David Rodin for pressing me on this.

clauses establish rights and obligations entirely de novo. As we saw therein, declar-
ative clauses are binding on the parties irrespective of the latter’s consent, whereas
specifying and de novo clauses are binding only if the parties consent. The question,
thus, is whether an agreement by belligerents to establish a transitional foreign
administration and endow it with powers of government is declarative, specifying,
or creative. In the latter two cases, the parties’ consent to the settlement is a necessary
condition for the settlement to be all-things-considered justified; in the former
case, it is not.
So stated, the question is ambiguous as between the following three variants: (a) is
an agreement to establish a transitional foreign administration declarative, specifying,
or creative? (b) Is an agreement to establish this or that transitional foreign administra-
tion declarative, specifying, or creative? (c) Is an agreement to give this or that mandate
to a transitional foreign administration declarative, specifying, or creative?
Institutions entrusted with coercive, regulatory, and coordinating powers and
competences are necessary to ensure that individuals are able to exercise their
human rights and, equally importantly, to fulfil their duties to one another. To
the extent that, faced with the inability or unwillingness of domestic actors to
govern their community, entrusting a foreign actor with the power to do so is
the only way to bring about those ends, a clause to that effect is declarative.
Accordingly, local populations’ consent is not necessary to render that clause
morally justified. To be sure, ensuring that local populations consent to and
actively support the foreign administration might well be necessary for the lat-
ter’s success. In this case foreign administrators would be under a moral obliga-
tion to seek such consent on pain of failing to meet the requirement (to which
I shall return presently) that their mission should stand a reasonable chance of
success.18 My point, however, is that those individuals would not be wronged if
a transitional foreign administration were imposed on them without their con-
sent via a peace settlement, so long as that administration fulfils the legitimacy
conditions outlined above.
That said, the declarative clauses relating to the establishment of a transitional
administration need to be rendered more precise, via further clauses. In particular,
a peace agreement needs to confer on a specific actor the power to govern the target
community. In s.3.3.2, I argued in favour of what I called a ‘heavily qualified mul-
tilateral’ approach to military occupation: multilateral institutions, I claimed, are
more likely than national governments acting alone to be effective occupiers and
to protect civilians. A fortiori, the point applies to the infinitely more complex cases
at hand, involving as they do the task of rebuilding a community. Governmental
effectiveness and the ability to protect civilian populations thus constrain the speci-
fication of the latter’s rights and duties. It does not follow, however, that an agreement
to establish this or that transitional administration is necessarily declarative too.
For the phrase ‘foreign actor’ encompasses a wide range of possible institutional
arrangements—from complex, multi-international-agencies missions coordinated
18 This is amply shown by L. Howard’s empirical study of local participation and consultation in
transitional administrations (Howard, Un Peacekeeping in Civil Wars).
and overseen by the UN via the latter’s so-called High Representative, to missions
carried out by a coalition of foreign states or, indeed, by one state alone should
international institutions be unwilling or unable to step into the breach. If several
of those institutional arrangements fulfil the aforementioned legitimacy condi-
tions equally well, then there is scope for local actors to negotiate with other parties
over the designation of the administrator(s). In this kind of case, the relevant
clauses are specifying clauses, and thus are binding only if local populations or
those negotiating on their behalf consent to them. By implication, of course, if
only one outside actor fulfils those conditions, the relevant parts of the settlement
are declarative and do not depend for their justifiability and bindingness on the
consent of the parties.
So much, then, for a decision to set up this or that administration. What about
the decision to give it a particular mandate? As we already saw in s.4.2, many of the
clauses relating to mandates are specifying rather than declarative clauses. Consider,
for example, the Rambouillet Accords over Kosovo. Article I.1 mandates that
Kosovo will govern itself democratically, whereas Article II mandates that it will
have an Assembly of 120 members, to be elected every three years.19 In so far as
there is no general, independently justified principle which dictates the number of
representatives and the frequency of elections, this article is a specifying clause, and
consent to it is a necessary condition for it to be morally justified and binding.
Finally, should a peace settlement contain creative clauses, the consent of those
subject to it is also required.
That very last point needs qualifying. For the claim that consent is a necessary
condition for justifiability and bindingness in those cases does not itself elucidate
whose consent must be sought. The case of the Rambouillet Accords is particularly
interesting in that context. Milosevic refused to sign it on behalf of Serbia. Assume,
for the sake of argument, that the Serb authorities had not carried on with their
campaign of ethnic cleansing following their departure from the negotiations
(which provided NATO with a justification for the bombing), and that they had
submitted to the Accords without resistance. Would the Accords’ specifying clauses
have been morally justified, absent the Serbs’ consent? As we saw in s.4.3, some
leaders/belligerents are not competent to negotiate and enforce peace settlements
on account of their wartime wrongdoings. Milosevic’s regime was responsible for
some of the worst atrocities committed in Europe since the Second World War. It
did not during the war fulfil any of the minimum requirements for state legiti-
macy, and could not be trusted to do so after the war. Consequently, the consent
of its officials and of those of their compatriots whom they were representing was
not a necessary condition for the moral validity of the Accords’ specifying clauses,
whilst Kosovars’ consent was.
So far, I have assumed that the transitional foreign administration is established
by the peace settlement itself. Consider now cases where the settlement does not
provide for it, where some of the signatories refuse to abide by the settlement’s
19 See http://peacemaker.un.org/sites/peacemaker.un.org/files/990123_RambouilletAccord.pdf
clauses and where the question thus arises as to how peace will be enforced. Imagine
for example that the Dayton Agreement had not contained Annexe 10 and that
Serbs, Croats, and Bosniaks, riven as they were by intercommunal conflicts, had
not been able to implement it. In this counterfactual case, in so far as the parties
would not have explicitly consented to the establishment of a transitional foreign
administration via the peace settlement, they could not have been presumed to
consent to its peace-enforcing intervention ex post. However, outsiders would have
had as strong a prima facie justification in this counterfactual case for enforcing the
agreement via a transitional foreign administration as they did in actuality, on the
grounds that individuals who are subject to a peace settlement both have a right
that the latter be enforced and a duty to ensure that it be enforced. If they fail to
enforce it themselves and if such failure results in widespread human rights viola-
tions, they lose their claim not to have the settlement enforced by third parties who
are in a position to do so. The point is particularly relevant to the enforcement of
peace settlements in the aftermath of civil wars, in so far as insurgents have good
reason to trust that, absent enforcement by outsiders, the government will renege
on the settlement as soon as they have laid down their arms.20
8.3.3. TFAs as Morally Mandatory

To recapitulate, I have argued that a transitional foreign administration is justified
as a means to provide war-torn communities whose members are unable or unwill-
ing to govern themselves with the conditions under which they can come to enjoy
their human rights and discharge their duties to another and to outsiders. To claim
that a transitional foreign administration may so act is one thing. Whether outsid-
ers are under a moral obligation to set up a transitional foreign administration in a
war-torn community is another. At the bar of cosmopolitan justice, they are.21 As
I argued in CW (s.5.3), potential interveners sometimes are under a duty to carry
out a war of humanitarian intervention against regimes who commit grievous
human rights violations against their own members—a duty which finds its justi-
fication in a more general duty to provide assistance to those in need. Whether
there is a duty to help another thwart a lethal threat, in other words, does not
depend on the source of the threat as natural or anthropogenic: there is no reason,
in fact one ought not, to restrict the scope of the duty to help to non-anthropogenic
devastation, for to do so would render the conferral of the relevant right dependent
on factors which are wholly independent of victims’ control. Likewise, mutatis
mutandis, with the duty to rebuild by foreign intervention. For absent a properly
functioning police and judiciary, properly run schools, hospitals, sanitation sys-
tems, transport systems and energy-generating infrastructures, individuals simply
do not lead a flourishing life. Were a country entirely and similarly devastated
by a natural catastrophe, arguments in favour of a duty to assist would clearly
20 For a discussion of this problem, see, e.g., Doyle, Making War and Building Peace, 48.
21 For powerful defences of that view, see Hermanson, Duties in the Wake of Atrocity, esp. chs. 2 and
4; J. Pattison, ‘Jus Post Bellum and the Responsibility to Rebuild’, British Journal of Political Science
45 (2013): 635–61.
be deemed to apply. If, then, outsiders are under a duty to help rebuild a commu-
nity whose economic and political infrastructures have been destroyed by natural
causes (as cosmopolitan justice dictates that they are), by parity of reasoning they
are under a duty to do so once the conflict is over.
8.3.4. Constraining TFAs

So far, we have seen that enabling a war-torn society to become a self-governing
human rights–respecting community provides outsiders with a just cause for estab-
lishing a transitional foreign administration on that society’s territory. However,
other conditions must be met, in addition to consent to the specifying or creative
clauses of a transitional foreign administration’s mandate, in order for the admin-
istration to be justifiable all things considered—in just the same way as war and
peacekeeping operations are not just simply in virtue of their having a just cause.22
To reiterate, a transitional foreign administration is legitimate only if those subject
to their directives do better by complying at the bar of their rights and duties than
they would do in the absence of that administration or under any other feasible
alternative. By implication, a transitional foreign administration loses its legiti-
macy if its officials systematically abuse their position by (for example) committing
acts of sexual abuse or taking part in organized criminal organizations, and if the
institutions to which those individuals are accountable fail to prosecute them.
Similar constraints as apply to the resort to war apply post-conflict.23
Moreover, transitional foreign administrations are enormously costly endeav-
ours. To give but one example, the UN Transitional Administration in East Timor
comprised over 1500 police officers, and close to 10,000 troops and 4000 civilian
staff. Add administrative, military equipment, and reconstruction costs—and the
total bill amounts to billions of dollars.24 But at the bar of cosmopolitan justice, all
human beings wherever they reside have prima facie rights to the material resources
they need in order to lead a flourishing life, not just against their compatriots but
also against distant strangers. The vast sums allocated to transitional foreign admin-
istrations could also be allocated elsewhere, for example in fulfilment of the duty
to help victims of natural disasters, or in fulfilment of a duty to help war-torn
communities which do not require such extensive help. It may be that priority
ought to be given in a given case to victims of war. Even so, in the light of the
harms which would accrue to other victims as a result, directing resources to the
former at the expense of the latter could be justified only if the mission stands a
reasonable chance of succeeding—failing which those resources would have been
diverted in vain from third parties in need.
22 The consent requirement applies even though outsiders are under a duty to establish a transi-
tional foreign administration. Similarly, I may well be under a duty to provide you with food if you
are dying of starvation, but I am not thereby justified in force-feeding you against your will (assuming
that you are morally competent to decide whether to live or die).
23 There is a long history of such abuse amongst UN peacekeeping forces. See, e.g., Bellamy,
Griffin and Williams, Understanding Peacekeeping, ch. 16.
24 See Howard, Un Peacekeeping in Civil Wars, 260.
This, of course, raises the question of what counts as success. Since providing ab
initio or restoring the conditions under which members of war-torn communities
can enjoy their human rights is what provides the relevant actors with a just cause
for establishing a transitional foreign administration, one might think that success
means precisely that. However, in the light of the enormous difficulties faced by
those administrations, so to define success seems overly demanding. Contrastingly,
to define success as merely putting an end to violence (or ‘negative peace’) is unduly
lenient. As we saw in s.1.3 when discussing the requirement that the resort to war,
and the way in which the war is fought, should not jeopardize chances for a justi-
fiedATC peace, we must distinguish between a just peace simpliciter, a justified peace
all things considered, and peace as the absence of war. There I noted that pursuing
a just peace simpliciter might not be possible, and that the best that we can both
hope for and expect to obtain is a justifiedATC peace. Likewise, a transitional foreign
administration might not be able to achieve peace simpliciter; at the same time, it
ought not to be content with peace as the absence of war. It is successful, in other
words, even if (through no fault of its officials and backers) it secures less than
optimal conditions for the enjoyment of all human rights but nevertheless brings
about a post-conflict state of affairs where basic rights are secure and when non-
basic rights are not under imminent threat of war.
To summarize the foregoing remarks, then, just as war and peacekeeping are
subject to five requirements, fulfilment of which are singly necessary and jointly
sufficient to render it just, so are transitional foreign administrations, modulo rele-
vant revisions along the following lines: (a) it has a just cause, where a just cause
consists in the absence of a justifiedATC peace; (b) the deployment of financial,
human, and military resources towards establishing it constitutes a proportionate
response to this state of affairs; (c) transitional administrators do not deliberately
cause wrongful harms to innocent parties as a means to secure the success of their
mission; (d) the administration stands a reasonable chance of success (as defined
above) by means the deployment of which violates neither (b) nor (c), where suc-
cess consists in establishing a justifiedATC peace; and (e) it is the only way to get
erstwhile belligerents to move to a justifiedATC peace. As in previous iterations of
those conditions, the justifiedATC peace requirement is part of both the just cause
and the proportionality requirements.
8.3.5. Some Loose Ends

Before addressing the colonialism objection to transitional foreign administrations,
some loose ends need tying. First, as we saw in s.9.2, there is a crucial difference in
law between communities which are not recognized by the international commu-
nity as state members of the UN when the conflict starts, and communities which
are so recognized. As per art. 78 of the UN Charter, the former may be adminis-
tered under the Trusteeship system, while the latter may not: one does not treat a
party which is already sovereign and thus one’s legal equal in the society of nations
as if it were not so. But the way the law treats those different political communities
will not do, for two related reasons. For a start, I argued in s.1.2 that individuals
each have a jointly held right to form political associations, whose standing as a
source of special obligations depends on the degree to which they (and officials
acting on their behalf ) respect the human rights of their members as well as of
outsiders. Such rights as groups or institutional actors have ultimately flow from
and supervene on the rights of individual members of those groups. On this view,
sovereignty ultimately lies in individual members of a political community. If a set
of individuals have a moral right to self-government, whether or not they are
already constituted into an independent state is irrelevant to the determination of
their community as warranting foreign administration. Conversely, in so far as
sovereignty ultimately lies in individuals qua members of political associations,
whether or not an actor has forfeited (albeit temporarily) her sovereignty right does
not depend on the characteristics of her political community as stateless or, on the
contrary, as a legally sovereign state. Moreover, the rationale for establishing such
administration, lying as it does in the imperative to provide all individuals, wher-
ever they reside, with the conditions for a flourishing life, applies whether or
not the beneficiaries of the administration are already constituted into a legally
recognized sovereign state (albeit one which ex hypothesi has failed). To grant or
withhold the benefits of this kind of administration by outsiders on the basis of
membership in this, or that, political community, would be in flagrant violation
of cosmopolitan justice.
Second, and relatedly, the foregoing considerations also support the view that
the nature of the conflict, as civil or interstate, has no bearing on whether the
establishment of a transitional foreign administration is justified. In practice,
transitional foreign administrations have tended to rebuild communities torn by
civil war. My point here is that the norms which regulate the resort to, and con-
duct of, those administrations apply to the aftermath of interstate as well as of
intrastate wars.25
Third, just as the literature on peacekeeping regards the issue of the use of force
by peacekeepers as both a major concern and a moral imperative if defenceless
civilians are under threat, so does the literature on peace-building in general and
transitional foreign administrations in particular. The three UN-authorized transi-
tional foreign administrations which I described in s.8.2 all comprised contingents
of military personnel entrusted with the task of protecting international adminis-
trators and local populations from lethal harm at the hands of those unwilling to
accept the settlement. Moreover, in the early stages of their missions, they took on
policing burdens which local police forces were unable or unwilling to shoulder.
Scepticism as to whether the military is best placed to act as a police force in unstable
25 See also CW, ch. 4. The claim that there is far less to distinguish morally speaking between both
kinds of conflicts draws support from the observation that a war which starts off as a civil conflict can
soon spill over the borders of that community and, in so far as outside states become involved, morph
into an interstate war—and vice versa. Angola’s and Namibian’s wars of independence against their
colonial masters (Portugal and South Africa respectively) were thus part of a larger regional conflict in
Southern Africa. Likewise with the Congo Wars of the 1990s. See, e.g., Doyle, Making War and
Building Peace, 40–2.
post-conflict societies is a recurrent theme in academic and policy circles.26 My

concern here is with the grounds upon which transitional administrators widely
understood so as to include the military may use force against local populations,
and vice-versa. It is a central claim of this book, as well as the cosmopolitan ethics
of war which I defend in CW, that the use of lethal defensive force in war is con-
strained by the same norms as regulate the use of lethal defensive force in domestic
cases—including cases traditionally regarded as a matter of police enforcement. In
post-conflict situations, transitional administrators exercise powers of government,
which include the power to coerce individuals to obey independently justified
laws, in just the same way as indigenous state officials do in a fully politically sov-
ereign community. Across all those politically varied contexts, individuals are liable
to lethal force if, and only if, they unjustifiably pose or contribute to posing a
wrongful lethal threat to some other party as a means to, or in the course of, vio-
lating that party’s rights. Accordingly, transitional administrators tasked with
enforcing the administration’s directives have the right to use lethal force against
agents who unjustly and violently resist the administration and attack its officials
as well as other civilians. Conversely, individuals who are subject to the adminis-
tration’s authority have the right to use lethal force against its agents if the latter
themselves use lethal force as a means to or in the course of violating their or some
other party’s rights. To return to an earlier example, if members of a transitional
military force sexually abuse the vulnerable civilians whom they are meant to pro-
tect and resist the latter’s attempt to defend themselves by using lethal force, they
are liable to be killed: that they are transitional officials is neither here nor there.
To summarize briefly, under conditions such as described here, foreign actors are
justified, indeed are under a duty, to govern war-torn communities with a view to
providing their members with the conditions under which they have prospects for
a flourishing life—in other words, with a view to turning or restoring those com-
munities into, broadly, liberal democracies.
8 . 4 . T H E C O L O N I A L I S M O B J E C T I O N
As I intimated at the outset of this chapter, transitional foreign administrations

often elicit the charge that there is an inherent tension between their ends (estab-
lishing secure democratic self-government) and the means which they employ (in
effect, those of a benevolent dictatorship).27 According to those critics, transitional
foreign administrations are a thinly veiled form of Western imperialism, or, as one
commentator puts it, ‘colonialism redux’.28
26 For a good overview of that debate, see Caplan, International Governance of War-Torn Territories,
ch. 2.
27 Chesterman, You, The People, ch. 8; Zaum, The Sovereignty Paradox.
28 See Chesterman, You, The People. For discussions of the objection, see Marten, Enforcing the
Peace; Wilde, International Territorial Administration, ch. 8; Brooks, ‘The New Imperialism: Violence,
Norms, and the “Rule of Law”’. Brooks, it should be noted, does not query that forcible intervention
to promote human rights in communities which do not enjoy them is sometimes justifiable: her con-
cern is with the way in which Western powers have done so, and in particular with the view, at the
There are three ways of handling the objection. One can accept it and conclude
that transitional foreign administrations are unjustifiable. Alternatively, one can
bite its painful bullet and go as far as to claim, following some commentators, that
transitional foreign administrations in practice have not been colonial enough
and ought to have gone further than they did in imposing their rule on war-torn
communities. Or one can attempt to show that transitional foreign administra-
tions need not be ruled as colonies, and that while the objection perhaps hits one
target—transitional foreign administrations as they have sometimes operated—it
nonetheless misses the main one—transitional foreign administrations as they
ought to operate.
Unsuprisingly perhaps I shall go down the third route. Beforehand, however, let
me provide a working definition of colonialism. This is not an easy task, for the
practices and institutions which are labelled ‘colonial’ are extraordinarily diverse:
they include largely mercantile enterprises as carried out by the Portuguese and the
Dutch in Asia in the fifteenth and sixteenth centuries; European white settlers
emigrating, acquiring land in, and governing the Americas, Africa, and Oceania
from the sixteenth century onwards; British expansion, without en masse settle-
ments, in India; French expansion, with such settlements, in North Africa; Russia’s
conquest of Central and North East Asia from the fifteenth century onwards; the
Han Chinese’ expansion into territories (such as Manchuria) which are now part
of contemporary China. Still, those practices and institutions are usually character-
ized by proponents and foes as the political, cultural, and economic subjection of
a people by another: political subjection, in the sense that the majority of colo-
nized populations not only have no control over who governs them and how they
are governed but in addition are governed by actors who are alien to them and
perceived as such;29 cultural subjection, in that they are coerced into endorsing the
norms and values of the colonizers; economic subjection, in the manifold sense
that they are either enslaved or have little choice but to work for very low salaries
in colonizer-owned businesses, that they are expropriated from their land, and that
the natural resources which colonizers extract from the land are then sold at a low
price to colonizers’ markets. Whether or not colonialism differs from imperialism
need not detain us here. Whereas the latter, unlike the former, normally denotes
that those who govern over the subject population also take possession of the lat-
ter’s territory, there are good reasons to think that territorial acquisition is not what
heart of UN missions, that establishing the rule of law in war-torn communities will necessarily lead
to a decrease in criminal violence. Similarly, see M. Ignatieff, Empire Lite—Nation-Building in Bosnia,
Kosovo and Afghanistan (London: Vintage, 2003).
29 I say ‘the majority’ for local indigenous political and economic elites actively collaborated with
colonizers: otherwise Britain would not have been able to rule India, whose population totalled 250
million at the apex of the British Empire, with a corps of only 1000 administrators and an army of
100,000 soldiers. But the fact that the colonizers are alien to the colonized, are so perceived by the
latter, and themselves conceive of themselves as alien, thus distinguishes colonial power from (e.g.) the
power which a dictator exercises over his own people. The exceedingly short description I give here of
colonial practices and institutions is drawn in part from W. Reinhard, A Short History of Colonialism
(Manchester: Manchester University Press, 2011). For an illuminating account of Britain’s colonial
expansion, see J. Darwin, Unfinished Empire: The Global Expansion of Britain (London: Penguin
Books, 2012).
is central, and according to its critics, centrally wrong, about colonialism. Rather,
what is wrong, if anything at all, is that one people should subject another people
to political, cultural, and exploitative economic control without their consent,
whether or not such subjection is accompanied by or occurs through taking over
the latter’s territory.30
My concern here is not to settle whether the objection’s description of coloni-
alism as a threefold subjection of one people by another is historically accurate.
I suspect that it is not, not least because it presupposes that a colonial power is a
unified entity with a unified and systematic will to oppress, and that a colonized
population is a unified entity wholly unable to resist oppression. There is good
historical evidence to reject this kind of oversimplifying narrative.31 However,
I will accept it for the sake of argument. If that is really the best way to understand
colonialism, then colonialism is indeed grievously wrong for the aforementioned
reasons.
Now, to assess the objection in the present context, we must first consider its
premise, to the effect that transitional foreign administrations are really just like
colonial powers. In order to do so, we must determine whether, functionally, tran-
sitional foreign administrations carry out the same tasks and policies, and in the
same ways, as did colonial powers. We must also determine whether, from a justifi-
catory point of view, the best case for this particular mode of governance is in effect
the same as justifications standardly advanced in support of colonial enterprises.
Given that the colonialism objection draws whatever strength it has from Western
(and notably French and British) colonial practices in Asia and Africa from the
eighteenth century onwards, I shall use the latter as my basis for comparison.
At first sight, there are important similarities between the French and British
administration of their Asian and African colonies on the one hand, and transi-
tional foreign administrations on the other hand. In just the same way as former
colonial powers enacted laws and administered justice over colonized populations,
oversaw the building and provision of basic infrastructures such as roads and
sanitation, and took a hand in the delivery of (some form) of education and
healthcare—so have transitional foreign administrations as a matter of fact,
and so should they do as a matter of principle (or so I have argued). More impor-
tantly still, in just the same way as colonial powers conceived of their colonies
and their populations as essentially foreign both territorially and culturally, tran-
sitional foreign administrators are outsiders in those same respects in relation to
the target post-conflict communities.
In functional terms, thus, there is something to the premise of the colonialism
objection. Upon closer inspection, however, there are profound differences between
30 In this respect, I accept Lea Ypi’s argument about the wrongness of colonialism, though shall
dissent from it in other respects below. See L. Ypi, ‘What is Wrong with Colonialism’, Philosophy &
Public Affairs 41 (2013): 158–91. For good definitions of colonialism, see, e.g., D. Butt, ‘Colonialism
and Postcolonialism’, in International Encyclopedia of Ethics (Oxford: Blackwell Publishing Ltd, 2013);
M. Kohn, ‘Colonialism’, Stanford Encyclopedia of Philosophy (2012) (available at http://plato.stanford.
edu/archives/sum2012/entries/colonialism).
31 With respect to the British colonial empire, Darwin makes that case in Unfinished Empire, esp.
ch. 7 (on ruling methods) and ch. 9 (on cultural subjection).
those two modes of government. First, without wanting in any way to occlude the
fact that foreign officials have been guilty of grievous wrongdoings in Bosnia,
Kosovo, and East Timor, transitional foreign administrations have not been accom-
panied by the large-scale killings, starvation, exploitation and all-round misery
that are characteristic of colonial rule as imposed by the French and the British.32
The question, then, is whether transitional foreign administrations as they ought
to operate—fairly, transparently, in a human rights–respecting way—are simply an
idealized form of fair, transparent, human rights–respecting colonialism. Perhaps
they are, but then, it is hard to see what is morally objectionable about the ways
either would, so idealized, operate. (I shall turn to the justificatory dimension of
the colonial objection presently.)
Second, as a matter of fact, transitional foreign administrations have overseen
extensive programmes of aid and reconstruction in target communities, have
helped build local penal institutions, and provided assistance with reconciliatory
measures such as I will describe in ch. 9. In a similar vein, they have sought to
reform taxation and fiscal policies in such a way as to enable emerging regimes to
become financially self-sufficient. This is not to deny the corruption (on the part
of both local and international actors) which often accompanies such reforms. Nor
is it to deny that multinational corporations, with the support of their own govern-
ment and international institutions such as the World Bank or the International
Monetary Fund, have often succeeded in pushing for the kind economic reforms
in post-conflict communities which would best serve their own, and not local
populations’, interests.33 But even if economic measures to the benefit of partici-
pating actors are necessary to garner the requisite support, it is neither a defining
feature of those administrations, nor part of the best normative justifications for
their establishment, that they should allow for self-serving economic reforms. To
the extent that the French and British governments of the day embarked on colonial
expansion with a view to seek and control outside markets for the benefit of their
own people and assisted in their nationals’ exploitative practices, in that respect at
least, transitional foreign administrations are not a colonial enterprise.
32 See Darwin, Unfinished Empire, ch. 5.

33 On economic reforms and transitional foreign administrations, see Caplan, International
Governance of War-Torn Territories, ch. 6. The best-known example of private sector involvement in
post-conflict reconstruction is that of Iraq, where US corporations were awarded contracts worth bil-
lions of dollars. There are other cases though, some of which are described in, e.g., M. Bhatia,
‘Postconflict Profit: The Political Economy of Intervention’, Global Governance 11 (2005): 205–24;
A. Gerson, ‘Peace Building: The Private Sector’s Role’, The American Journal of International Law 95
(2001): 102–19. There is an interesting debate in the empirical literature on the order in which, and
speed at which, transitional administrators should proceed: should they first set up properly function-
ing political and judicial institutions? Should they go for rapid or slow democratization? Should they
engage in a shock therapy of economic and social liberalization? For an influential defence of the
‘institutions first, go-slow’ view, see R. Paris, At War’s End—Building Peace After Civil Conflict
(Cambridge: Cambridge University Press. For criticisms of that view, see D. Chandler, Empire in
Denial—The Politics of State Building (London: Pluto, 2006). Chandler’s point is that, in practice, the
approach has led international actors such as the US and the EU to pretend that they were building
up independent states in, e.g., Bosnia, whereas in fact, they are still in effect governing from far above
without proper accountability. He may be right, but this does not impugn my normative, in-principle
case for transitional foreign administrations.
Third, it is inherent in a transitional foreign administration that it should not

consist in territorial and jurisdictional annexation. To be sure, its officials exercise
powers of government over individuals who are not their compatriots or, if they
are, who have not authorized them as their compatriots to do so in the same way as
citizens of a democratic community elect their compatriots into political office.
Moreover, to the extent that local officials assist transitional administrators in the
task of political and economic reconstruction, in the initial stages, at least, their
role is, precisely, to assist and not to lead—in the same way as local elites assisted
French and British colonizers in their expansionist enterprise. However, a morally
justified transitional foreign administration is not sovereign over the community
which it governs in the way Britain was sovereign over India and France over
Algeria. Sovereignty still resides in individual members of that community who
have not forfeited their claim collectively to govern themselves. It is precisely for
that reason that transitional foreign administrations are under an obligation to
provide the conditions under which those individuals will be able to exercise the
sovereignty rights which they already have. This means, inter alia, establishing or
consolidating the rule of law by attending not just to its formal aspects (constitu-
tional mechanisms, judicial procedures, etc.) but also and equally importantly by
attending to existing social-cum-legal norms which might undermine or on the
contrary buttress it; empowering those individuals to support and serve in demo-
cratic institutions at the regional and local level; conducting impartial and fair
elections; holding itself accountable to those which it governs; and being open to
the possibility of reconciling the imperative of democratization with seemingly
conflicting local agendas.34
Furthermore, and crucially, it means working from the very beginning to a
clearly defined exit timetable. When elections should be held and when the transi-
tional administration’s exit should take place cannot be settled in the abstract,
though there are good reasons to think, on the basis of available evidence, that
holding elections too early and exiting as soon as elections are conducted might
lead to renewed instability.35 The key point here is that a transitional foreign
administration by definition is established and structured with a view to relin-
quishing control over the target community. Neither Britain nor France began and
continued to govern their respective colonies with that particular end in sight.
34 On the rule of law under TFAs, see in particular Brooks, ‘The New Imperialism: Violence, Norms,
and the “Rule of Law”’. For an overview of tensions between global peace-builders and local agents, see, e.g.,
G. Millar, J. Ø. van der Lijn and W. Verkoren, ‘Peacebuilding Plans and Local Reconfigurations: Frictions
between Imported Processes and Indigenous Practices’, International Peacekeeping 20 (2013): 137–43;
M. R. Freire and P. D. Lopes, ‘Peacebuilding in Timor-Leste: Finding a Way between External Intervention
and Local Dynamics’, International Peacekeeping 20 (2013): 204–18; S. Hellmueller, ‘The Power of
Perceptions: Localizing International Peacebuilding Approaches’, International Peacekeeping 20 (2013):
219–32. Hellmueller’s case study—the UN Mission established in Congo in the late 1990s (MONUC, and
now MONUSCO) —is particularly interesting: on her account, UN peacekeepers concentrated on
state-building and eschewed the task of helping the Congolese to resolve the land conflicts which had, in
part, triggered the war in 1996 and which were still regarded as the main obstacle to peace by the latter.
35 See, e.g., Chesterman, You, The People, ch. 7; Caplan, International Governance of War-Torn
Territories, ch. 10; D. Zaum, ‘The Norms and Politics of Exit: Ending Postconflict Transitional
Administrations’, Ethics & International Affairs 23 (2009): 189–208.
Finally, although it is true that the language of human rights informs and justi-
fies transitional foreign rule, and sometimes is inconsistent with local norms and
practices, to claim that target populations are victims of a form of ideological and
cultural subjection in the way (e.g.) non-Christians in India were expected to con-
vert to Christianity and renounce their own cultural roots, is to do poor justice to
the fact that the moral requirements encapsulated by human rights are some of the
most basic requirements under which human beings are held—not to kill, not to
torture, not to rape, to provide for basic sustenance, healthcare, and education,
and to provide for a formal say in the political organization of one’s society. This is
not to deny, of course, that the language of human rights has often been used by
Western powers as a cloak for their own exploitative practices. In this respect, the
colonialism objection in its functional variant does hit this particular target; but
unless it is wedded to a form of moral relativism which I reject at the outset, it
misses the target which really matters, to wit, transitional foreign administrations
as they ought to operate.
Proponents of the colonialism objection might perhaps resist my rebuttal on
the following grounds: even if, functionally, transitional foreign administrations
are not guilty of the crimes of colonialism, the ways in which they are justified are
sufficiently similar to justifications for colonialism for the objection to stand. As
I shall now argue, however, of the arguments which have been offered in support
of colonial expansion, two stand out, only the second one of which provides sup-
port for the objection—and exceedingly limited support at that.
The first argument—to be found inter alia in some of Alexis de Tocqueville’s
writings on Algeria—holds that a nation may appropriate land and subjugate a
people if it serves its interests. Anything that the mother country does to or for
colonized populations must promote those interests.36 Now, as a matter of fact,
transitional foreign administrations have been justified by appealing to the neces-
sity of preserving peace and promoting human rights in the target territories—as
well as by the imperative of preserving, if not international, at least regional peace.
This at any rate is one way to interpret the UN’s use of Ch. VII as a basis for its
governance intervention in East Timor. There is a sense, thus, in which justifica-
tions for transitional foreign administrations have appealed to the interests of out-
siders. However, this is not enough to buttress the colonialism objection, for two
reasons. First, European settlers’ interests in subjecting African and South East
Asian communities to their political rule and in exploiting their resources were not
important enough to be protected by rights: by contrast, regional instability of the
kind witnessed in, e.g., Bosnia issues in severe rights violations outside the target
community, which in turn may well justify outside governance. Second, even
36 See, especially, Tocqueville’s 1841 ‘Essay on Algeria’, in P. T. Coleman, M. Deutsch and
E. C. Marcus (ed.), The Handbook of Conflict Resolution—Theory and Practice, 3rd edn. (San Francisco,
Ca.: Jossey Bass, 2006). The 1847 reports which he wrote on Algeria are more nuanced: while consid-
erations of France’s national interest still frame his recommendations for good colonial governance of
the province, he insists that colonial administrators and their Paris’ masters not lose sight of the
well-being of colonized populations. For a fascinating insight into British and French political thought
on colonialism from 1770 to 1850, see J. Pitts, A Turn to Empire: the Rise of Imperial Liberalism in
Britain and France (Princeton, N.J.: Princeton University Press, 2005).
when appeals to outsiders’ interests partly shape justifications for transitional for-
eign administrations, the case for the latter also and centrally rests on the interests
of administered populations—unlike the Tocquevillian argument under study.
It is at this point that the justificatory strand of the colonialism objection
really comes into its own. Colonialism, it holds, has mostly been justified on
paternalistic lines, as a ‘civilizing mission’ undertaken by enlightened liberal, capi-
talist, developed, and Christian nations for the sake of backwards, authoritarian,
agrarian, underdeveloped and heathen communities. This is a familiar argument,
of which J. S. Mill, who worked for the East India Company until it lost its
Charter in 1858, is one of the best-known proponents in British political
thought. It has enough similarities, at least superficially, with justifications for
transitional foreign administrations to deserve serious consideration.37 For
although endorsing such administrations need not imply a commitment to the
view that the individuals who are so administered are backward and culturally
(let alone biologically) inferior, it assumes that they are not yet able to govern
themselves and need help to do so. Replace ‘we must help Indians and Africans
to become civilized peoples able to govern themselves—and against their will if
necessary’ with ‘we must enable war-torn populations to become human rights–
respecting communities—against their will if necessary’: thus said, there seems
to be little to distinguish colonial administrations from the administrations
under study here. Or so the objection goes.
However, the ‘civilizing mission’ rationale for colonialism is not the rationale
for transitional foreign administrations. It conceives of the target people as a
corporate entity which is unable to govern itself and which it is appropriate to
force to be free for its own good, in just the same way as a child is unable to lead
his life rationally and autonomously and must be forced by his parents down the
path of an ultimately autonomous life. However, the justification for transitional
foreign administrations which I offered in s.8.3 takes a rather different view.
To begin with, as I argued there, many institutional arrangements which taken
together comprise a transitional foreign administration must elicit the consent of
those subject to their directives if those directives are to be binding—in so far as
those arrangements specify declarative rights and obligations. The colonialism
objection does not work against those facets of the case for transitional foreign
administrations.
More importantly still, that case flows from the two-pronged recognition that a
community so administered comprises agents who, if left to their own devices,
would commit, or continue to commit, grievous wrongdoings towards their fellow
community members and/or other parties, and that institutions are needed to enable
individuals to fulfil their duties towards one another—which institutions transitional
37 Mill’s defences of colonialism appear in a plethora of letters and newspapers articles, notably his
‘A Few Words on Non-Intervention’ as well as in his Considerations of Representative Government
(esp. ch. 18). For thoughtful analysis, see, e.g., Pitts, A Turn to Empire, ch. 5; D. Bell, ‘John Stuart Mill
on Colonies’, Political Theory 38 (2010): 34–64. For a superb critical discussion of the notion of
‘forcing a people to be free’, see A. I. Applbaum, ‘Forcing a People to Be Free’, Philosophy and Public
Affairs 35 (2007): 359–400.
foreign administrations help set up. Setting up such administrations without the
consent of those subject to their directives does not always wrong them. Remember:
when the agreement which establishes a transitional foreign administration is
declarative, consent is not a necessary condition for its validity (though it might
have to be sought as a matter of expediency). Moreover, to the extent that directives
issued by a transitional foreign administration aim to thwart or punish grievous
human rights violations as carried out by some members of the target populations
against others, the former’s consent is not required either to render those directives
binding. To put the point in more policy-oriented terms, involving local officials
in the tasks of governance and reconstruction might in fact worsen the plight of
those officials’ compatriots if those officials—for example, police officers—become
involved in corruption, discrimination on ethnic grounds, human trafficking. This
would provide a justification for not involving them—in other words, for lustrating
them not punitively, but defensively.38
To put my reply differently, the rationale for outside non-consensual interfer-
ence is this: some members of the target community have during the war denied
equal status to fellow members and (in some cases) committed grievous wrongdo-
ings not just against them but also against outsiders, and would continue to do so
if unimpeded. As a result, those agents have (at least temporarily) forfeited their
claim to take part on an equal footing in the governance of their community.39
This is a far cry from holding that beneficiaries of transitional foreign administra-
tions must be protected from themselves thanks to institutional arrangements
which deny them the status of equal partners. On the contrary, it is precisely
because they are deemed as worthy of concern and respect as everyone else in the
world and are endowed with the same human rights and duties that transitional
foreign administrations are (at least sometimes and subject to aforementioned
conditions) morally justified.
8 . 5 . C O N C LU S I O N
In this chapter, I have defended the view that entrusting a transitional foreign
administration with the task of governing a war-torn community is sometimes
justified, on the grounds that if properly conducted, it can better enable commu-
nity members to enjoy their human rights and fulfil their fundamental obligations
to one another and to third parties than available and feasible alternatives. Pace the
colonialism objection, endorsing this view need not commit one to denying the
principle of fundamental equality—on the contrary, in fact.
38 For illustration of this point, see Caplan, International Governance of War-Torn Territories, 51–4.
On non-punitive lustration in transformative occupations (and, by implication, in transitional foreign
administrations), see Meierhenrich, ‘The Ethics of Lustration’.
39 Thus, Lea Ypi’s powerful critique of colonialism as involving a denial of equality and reciprocity
presupposes that all members of all subject societies have a claim to be treated on a footing of equality.
It thus has little to say about outside interference against wrongdoers. See Ypi, ‘What is Wrong with
Colonialism’.
Finally, if transitional foreign administrations are morally justified (subject to

the conditions delineated below), then a fortiori so are less invasive modes of
foreign involvement in post-conflict institutional rebuilding whereby foreign

actors assist, rather than lead, local agents’ reconstruction efforts—provided of
course that the latter do tend towards a justifiedATC peace.
9
Reconciliation
As we have seen throughout chs. 5 to 7, epistemic uncertainty about who commit-

ted which wrong against whom, scarcity of resources, and the sheer number of
perpetrators and victims makes it very hard, and indeed undesirable, to try and
implement a just peace simpliciter. More precisely, wholesale punishment is not
feasible, which makes it impossible for war-torn communities to rebuild them-
selves and weakens belligerents’ ability and willingness to live together peacefully
after the war. For obvious reasons, those worries are particularly acute following
civil wars, but they also arise in the aftermath of interstate conflicts. Scarcity of
resources compounds the issue, for the reparative and reconstruction demands of
war victims compete with the claims of those whose desperate needs do not origi-
nate in war, or at any rate not in this war, and are not always overriding.
Hence the need to compromise in favour of a justifiedATC peace, which one may
legitimately seek to implement and enforce through a transitional foreign admin-
istration if need be, and a fortiori through less invasive governance structures.
Those structures are unlikely to suffice, however. For the fact that so many wrongs
unavoidably go unaddressed and so many needs unnecessarily go unmet is likely to
deepen the feelings of anger and hatred which the war itself aroused on all sides;
this in turn is likely to undermine (imperfect) restitutive, reparative, and punitive
processes, and to lead to war again.
The answer to this volatile cocktail, I argue here, partly lies in reconciliation, to
wit, bringing together enemies in a joint endeavour to work towards peace on a
footing of moral and political equality, and through properly designed processes
and practices. In s.9.2, I locate reconciliation in the overall project of cosmopolitan
justice after war and argue that reconciliation, for all that it is a political process,
often requires that its participants undergo attitudinal changes vis-à-vis their ene-
mies. Those changes are both cognitive and affective and need accounting by any
plausible normative theory of reconciliation. In s.9.3, I scrutinize two cogni-
tive-cum-affective attitudes which are often thought to be key to, indeed necessary
for, reconciliation—forgiveness on the one hand, and trust on the other hand.
I argue that forgiveness is not key to political reconciliation, while trust is. In the
aftermath of bitter conflicts, however, trust emerges and develops only if appropri-
ate institutional settings are in place. In ss.9.4 and 9.5, I turn to two institutional
mechanisms which invite perpetrators and victims to testify to what they did and
suffered during the war, namely nonjudicial fora such as the gacaca hearings in
Reconciliation 247
Rwanda, and truth and reconciliation commissions along the South African
model. In s.9.6, I examine the practice of apologies for past wrongdoings. In those
three sections, I show that those various mechanisms and practices can facilitate
the emergence of political trust after war and thereby buttress reconciliation; they
also help remedy some of the moral gaps left by the impossibility of meeting all
restitutive, reparative, and punitive demands.
Some caveats before I begin. First, the chapter is not meant to be exhaustive: it
may well be that practices which are left out of my inquiry are also conducive to
reconciliation. I hope, however, that the broad justification given here for the prac-
tices I explicitly defend also apply mutatis mutandis to those I set aside. Second,
I use the word ‘reconciliation’ more narrowly than the literature does. There it refers
to all the processes, practices, and institutions which are required for peace after
war—including, for example, restitution, reparations, reconstruction, and punish-
ment. However, my concern here is with what may or ought to be done in the
aftermath of a conflict given that so many wrongdoings will remain unaccounted
for, that victims are likely to be deeply and understandably aggrieved as a result,
and that feelings of hatred, anger, and resentment are likely to undermine pros-
pects for peace. Reconciliation, in this chapter, thus refers to the attitudes which
are supportive of, or indeed necessary for, those other tasks, and to institutional
processes which elicit and instantiate those attitudes.1
9.2. R E C O N C I L I AT I O N A N D C O S M O P O L I TA N
J U S P O S T B E L LU M
Reconciliation after war is to post-conflict studies what motherhood and apple pie
are to American home life. And, after all, why not? Loosely adapted from the
Oxford English Dictionary, to reconcile means to restore friendly relations after a
period of estrangement or enmity; to make peace. To reconcile is not simply to be
willing not to fight and to accept that one’s enemy might have a claim to be left
alone—what I described in s.1.3 as negative peace. Rather, it is characterized by
and requires the willingness to work together towards common political and/or
economic goals on a footing of equality. In the present context, the goal is to estab-
lish a just peace or, if it cannot be achieved, a justifiedATC peace. This is not to
underestimate how much of an achievement negative peace often is, particularly
following mass atrocities. But reconciliation is the aim in sight. And what can
possibly be bad about that?
There are lesser and greater modes of political reconciliation, of course—just as
there are lesser and greater modes of reconciliation in interpersonal relationships.
1 For recent book-length normative treatments of post-conflict reconciliation, see, e.g., Philpott,
Just and Unjust Peace; C. Murphy, A Moral Theory of Political Reconciliation (Cambridge: Cambridge
University Press, 2010). For an excellent analysis of political reconciliation as both institutional and
attitudinal, see D. Moellendorf, ‘Reconciliation as a Political Value’, Journal of Social Philosophy 38
(2007): 205–21. For a historical account of transitional justice, see J. Elster, Closing the Books:
Transitional Justice in Historical Perspective (Cambridge: Cambridge University Press, 2004).
248 Reconciliation
Getting back together after a relationship break-up merely for the sake of the chil-
dren and/or because divorce is economically impossible for either partner is one
thing: doing so for those reasons and because both partners still love, desire, and
respect each other notwithstanding the wrongdoings that led to the estrangement
is quite another. By analogy, for belligerent communities to renew or build coop-
erative relationships through (e.g.) trade and/or political agreements merely
because this is the only way to ensure that they do not go to war again within five
years is one thing; for them to do so at least partly out of a genuine and deep com-
mitment to build a peaceful future together is quite another.
Note that I wrote ‘to renew or build’. In intimate relationships, reconciliation
always means to go back to a pre-existing loving or amicable relationship. In con-
flict cases, however, there often was no friendly relationship to begin with, or at
least not within living memory. The plight of the Mayans in Guatemala, who died
in their dozens of thousands at the hands of successive governments, is a case in
point, as is the decades-long parlous state of enmity between Palestinians and Jews
in Israel. Reconciliation in those cases means to undo the bad relationships which
existed and/or were created by the war, and to build non-existent amicable rela-
tionships more or less from scratch. Equally often, however, reconciliation
post-conflict can mean restoring enemies to amicable relations. The examples of
communities which together made up the former Yugoslavia, and of the Hutus
and Tutsis in Rwanda, come to mind.
As some of the examples just given suggest, and indeed in line with the individ-
ualistic morality which underpins this book, reconciliation seemingly must oper-
ate not just between groups but also between individuals. For enemies conceived
of as groups—the French, the German, the Hutus, the Tutsis—to reconcile with
each other, it often must be the case that a significantly high number of their indi-
vidual members should be willing to reconcile with each other as individuals. This
might seem fallacious: it does not follow from the claim that a relationship between
two groups has a property P that relationships between individual members of that
group also have that property. Many historical cases bear that point out. Assume
(plausibly, I think) that the steps which France, Italy, the Benelux, and Germany
took towards economic and political cooperation in the late 1940s/early 1950s
were instrumental to and an instantiation of European reconciliation in general
and Franco-German reconciliation in particular in the aftermath of the Second
World War.2 Assume also (again, plausibly) that one important reason as to why
yet another war between France and Germany is utterly unthinkable is precisely
that those two communities are long-term partners in the EU. I would wager that
the overwhelming majority of French people have never had a conversation with a
German. Whilst it is quite possible that the EU would be in better shape now if
French and Germans in their majorities were united by personal bonds of friend-
ship, it is by no means a foregone conclusion. Political and social reconciliation
thus does not always require interpersonal reconciliation.
2 See, e.g., Judt, Postwar, 153–9.

Reconciliation 249
Often, however, the former cannot occur without some degree of the latter—
and it is on those cases that I will focus in this chapter. Following mass atrocities,
it is hard to see how political reconciliation understood as a joint endeavour to
promote peace could obtain without some degree of interpersonal reconciliation
between perpetrators and victims, at least when they have no choice but to live side
by side. By lack of choice, I mean any number of the following: victims and perpe-
trators simply lack the wherewithal to remove themselves from the wider commu-
nity; or they are not morally permitted to leave since leaving would mean, for
example, leaving their children even more vulnerable than they are; or sovereign
communities have no choice, by dint of their geographical location, but to share
borders, lakes, oil fields, gold mines; or they are not morally entitled to sever all
relationships with each other because (e.g.) they are under obligations of repara-
tions and reconstruction towards each other. Moreover, combatants often have no
choice once they have lain down their weapons but to return to their villages,
where they often have committed atrocities; their former neighbours, turned ene-
mies, often have no choice but to face them daily. Most are both perpetrators and
victims. All have to cope with untold physical and psychological wounds as they
painfully go about rebuilding their lives—a struggle made all the harder by daily
encounters with those whom, rightly or wrongly, they hold responsible for their
predicament. Under those circumstances, political reconciliation is not conceiva-
ble unless most are willing to turn their personal relationship from one which is
suffused with hatred and anger to one which is grounded in mutual acknowledg-
ment of their equal status.
One might think, accordingly, that interstate reconciliation and intrastate rec-
onciliation should be characterized by markedly different practices and governed
by markedly different norms. In particular, one might think that interstate recon-
ciliation merely calls for appropriate institution-building, the punishment of the
worst of the war criminals, the payment of reparations, and the forging of multi-
lateral or even supranational institutions, whereas reconciliation after civil conflicts
necessarily calls for most of all of those plus a transformation of individuals’ per-
sonal relationships with, and therefore inner attitudes towards, their former ene-
mies. There is some truth to that: those reconciliatory processes are bound to differ
for the simple reason that enemy populations in former cases do not, but in latter
cases do, have to live side by side. Yet, drawing such a sharp distinction between
civil and interstate conflicts with respect to post-war reconciliation is misleading.
For it matters too whether the conflict (civil or not) was marked by mass atrocities,
how long it lasted, and how destructive it was of social, political, and economic
life. It also matters whether one can relatively clearly describe some of its actors as
only victims, others as only perpetrators, others still as both. Moreover, interstate
conflicts sometimes trigger civil conflicts (as was the case in some mainland
European countries during the Second World War, where partisans of Nazism and
Resistants bitterly fought each other). Conversely, civil conflicts sometimes morph
into interstate conflicts (as happened following the Rwandan genocide). In such
cases, reconciliation might best operate along those two dimensions. To give a
predictive example: it is not implausible to surmise that a durable peace between
250 Reconciliation
Israelis and the Palestinians, when and if it finally happens, will require internal
reconciliatory processes within each community, as well as reconciliation between
them. Within Palestine itself, Palestinians who have opposed the occupation and
those who have collaborated with Israeli forces (many of whom have been tortured
and murdered by their compatriots) will need to come to terms with this highly
divisive feature of their collective past. Within Israel itself, tensions will need to be
eased between supporters and opponents of the peace agreement. By dint of their
long-intertwined and bloody history, as well as geographical contiguity, those two
peoples will have no choice but to confront the legacies of external and internal
violence.3 Finally, even when erstwhile enemies do not have to live side by side
after the war, the bloodier, longer, and more encompassing the conflict was, the
more do reconciliatory processes require attitudinal changes—even in the absence
of pre-existing personal relationships between individuals and thus even in the
absence of interpersonal reconciliation.4 This is why reconciliation, particularly
following civil conflicts marked by mass atrocities but also following similarly atro-
cious interstate conflicts, is more often than not rooted in individual reactive atti-
tudes and dispositions.
Reconciliation, thus, is ‘a good thing’. But is it a requirement of cosmopolitan
post-war justice? It might seem not, on two distinct grounds: first, on the grounds
that cosmopolitanism really has little to do with it; second, on the grounds that
there often is no duty to reconcile. I reject the first point, and qualifiedly accept the
second.
Why might one think that cosmopolitanism has little to do with post-war rec-
onciliation? Only parties in the conflict can reconcile with each other—by defini-
tion.5 To the extent that the primary rationale for reconciliatory processes lies in
their ability to improve chances that war-torn communities will be able to live side
by side and work together towards the realization of a justifiedATC peace, the rele-
vant rights and obligations vest only in those individuals: unlike the rights and
3 For discussion of Palestinians’ collaboration with Israel between 1967 and 2000, see Rigby, Justice
and Reconciliation After Violence, ch. 7.
4 A case in point is reconciliation between Germany on the one hand and, respectively, France,
Israel, Poland, Hungary, and the Czech and Slovak Republics on the other hand. Those might be
regarded as successes. A not so successful one is the case of Japan and China after the Second World
War. For discussions, see, e.g., L. G. Feldman, ‘The Principle and Practice of “Reconciliation” in
German Foreign Policy: Relations with France, Israel, Poland and the Czech Republic’, International
Affairs (Royal Institute of International Affairs 1944–) 75 (1999): 333–56; Y. He, The Search for
Reconciliation—Sino-Japanese and German-Polish Relations since WWII (Cambridge: Cambridge
University Press, 2009); A. L. Phillips, ‘The Politics of Reconciliation: Germany in Central-east
Europe’, German Politics 7 (1998): 64–85.
5 For a particularly pithy point along those lines, see S. Levinson, ‘Trials, Commissions, and
Investigating Committees—The Elusive Search for Norms of Due Process’, in R. I. Rotberg and D. F.
Thompson (ed.), Truth v. Justice—The Morality of Truth Commissions (Princeton, N.J.: Princeton
University Press, 2000), 219. A similar view underpins Daniel Philpott’s and Jennifer Llewellyn’s
appeal to a relational theory of justice as the best normative framework for understanding reconcilia-
tion after war. See J. J. Llewellyn and D. Philpott, ‘Restorative Justice and Reconciliation: Twin
Frameworks for Peacebuilding’, in J. J. Llewellyn and D. Philpott (ed.), Restorative Justice,
Reconciliation, and Peacebuilding (Oxford: Oxford University Press, 2014); Philpott, Just and Unjust
Peace.
Reconciliation 251
obligations of cosmopolitan justice, which are general rights of all against all, the
rights and obligations of reconciliation are special rights.
In reply: cosmopolitanism still has something to say about reconciliation, and
humankind at large still has a stake in its constitutive processes. On the first count,
although reconciliatory obligations inhere in special relationships, intra- and inter-
state reconciliatory practices are constrained by the same norms. On the second
count, humankind, via the requisite institutions, is under a duty at the bar of
cosmopolitan justice to assist them with those processes. Most obviously, it can do
so by providing the human and material resources which many reconciliatory
institutions require. In addition, international involvement with those institutions
might help foster a sense amongst parties that they are appropriately impartial. To
anticipate somewhat, consider the cases of the truth commissions established in
Guatemala on the one hand, and East Timor/Timor Leste on the other hand, fol-
lowing the former’s civil war and the latter’s war with its colonial master Indonesia.
As per the agreement signed by the Guatemalan government and the opposition in
1994, the Guatemalan Commission for Historical Clarification (1997–99) was to
be chaired by a non-Guatemalan, to be appointed by the then-UN Secretary
General. In East Timor, the UN Transitional Administration (UNTAET) oversaw
the establishment of the Commission for Reception, Truth and Reconciliation
(2001–05). Reconciliation, thus, is an entirely appropriate object of normative
inquiry for a cosmopolitan theory of post-war justice.6
It remains to be seen how strong the duty to reconcile is. There are cases where
reconciliation is not morally required—both as a political process and as a personal
process upon which the former might depend. If, by reconciliation, one means the
restoration or creation of something like the political relationship that existed or
should have existed before the war, then a community which, post-war, has the
right to secede is not under a duty to restore or build a relationship of co-sover-
eignty with its former enemy. Similarly, members of a persecuted minority might
not be willing to stay in their country of origin after the war; if they all leave,
political reconciliation of that kind between that group and the majority will not
obtain either. This might conceivably have happened had all German Jewish survi-
vors of the Holocaust left Germany after the war. On the plausible assumption that
individuals generally have the right to emigrate and a fortiori so when they were
victims of mass atrocities at the hands of their compatriots, those individuals sim-
ply are not under a moral duty to reconcile, politically speaking, with the latter.
Similar considerations apply, mutatis mutandis, to reconciliation following inter-
state conflicts. Consider two belligerents A and B who were partners in an overall
justified alliance before the war started—for example, a justified trade agreement.
Suppose that A invaded B and that its combatants in the course of the war com-
mitted mass atrocities against civiliansB. Although both parties are under a duty to
work towards the establishment of a justifiedATC peace, B in this case is not under
a duty either to revive the old trade agreement or to accede to a new agreement
6 For discussions of those two cases (and, in fact, of forty truth commissions), see Hayner,
Unspeakable Truths; Rigby, Justice and Reconciliation After Violence, 166–72.
252 Reconciliation
with better guarantees against its being breached. The point is not just that no one
is under a duty to enter in creative relationships (as per the taxonomy of ch. 4) with
anyone else, whether as individuals or qua group members. It is also that one has
even less of a reason to do so following a conflict in which trust has been so severely
undermined.
As we saw above, however, former enemies often have no choice but to live
together, or alongside each other, as partners in peace. If reconciliation is necessary
for a justifiedATC peace, and given that individuals are under a duty to work towards
such a peace, they are under a duty to reconcile—politically and, if necessary,
personally too. But if there is a duty to reconcile (under some circumstances) and
if reconciliation both consists in and requires affective attitudinal changes, it would
seem that affective states—for short, emotions—are the kind of things one ought
(or not) to have. In other words, there comes a point where anger, hatred, and
resentment undermine prospects for peace and therefore are no longer morally
appropriate.
To some, the latter point will seem vulnerable to the objection that our emo-
tions, unlike our behaviour (at least in most cases) are beyond our control, and
thus are neither blameworthy nor praiseworthy, let alone objects of judgements of
rightness or wrongness, since moral judgements must pertain to what is within our
control. Entering this particular fray would require engaging with 2000 years of
ethical debate, beginning with Aristotle’s claim that we can in fact be judged by our
emotions. Let me buttress the Aristotelian thought with admittedly sketchy claims.
First, parents teach their children to be kind, forgiving, and generous; many people
undergo psychotherapy in the hope that they will not simply learn to live with
whatever emotions they are experiencing, but also in some cases change those emo-
tions; most people expect criminals to feel remorse for what they did. If the claim
that we have absolutely no control over our emotions is true, then they all are
deluding themselves. Yet it does not seem that they are: children do acquire those
emotional dispositions from their parents, some patients do change emotionally as
a result of psychotherapeutic work, and some convicts do learn to feel remorse.
Second, even though compulsive behaviour is sometimes exempt from moral
judgement, it is not always so: the claim ‘but I could not help it’ will not always
exonerate the agent from moral evaluation even if it happens to be true (think of
the persistent sexual harasser). Third, whilst we may not be able to transform parts
of our emotional landscape, we are (most of us, at least) able to at least try to do
so. Perhaps we are not under a duty not to feel angry, jealous, resentful: but we
certainly are under a duty to try to school ourselves out of those emotions, when
they are unwarranted. Fourth, and relatedly, even if we cannot help feeling angry,
jealous, and resentful however hard we try not to, it still is possible to say that our
anger, jealousy, and resentment are morally unjustified in the sense that the target
of our negative feelings has not done anything to elicit them. If only in that weak
sense, emotions are amenable to moral appraisal.
To hold that view is one thing. Discerning which attitudinal changes and thus
affective redirection are required for political reconciliation is another. To this
issue, I now turn.
Reconciliation 253
9.3. E M OT I O N S , P O L I T I C A L M O R A L I T Y, A N D
R E C O N C I L I AT I O N
I spoke earlier of suffering, of unendurable loss. Without falling prey to cheap

sentimentalism, reconciliation is painful. Attending to its ethics thus requires
attending to its moral-psychological features—more precisely, to the affective redi-
rection which (it is claimed) erstwhile enemies must sometimes undergo if they are
to work together towards a justifiedATC peace. In this section, I first reject the view
that forgiveness is central to reconciliation and argue that trust, rather, is its key
(though not sufficient) condition.
9.3.1. Forgiveness after War

Having chaired the South African Truth and Reconciliation Commission (TRC)
after decades of indefatigable fight against the Apartheid, Archbishop Desmond
Tutu famously wrote that there is ‘no future without forgiveness’. One should
always be wary of slogans. Still, on Tutu’s account, forgiveness is central to recon-
ciliation. It is an attractive, indeed powerful thesis. But I think that there are good
reasons to resist it. First, though, we need an account of forgiveness. Suppose that
John, whilst on deployment with his army in a foreign country, rapes and tortures
Fatima as part of a campaign of intimidation against defenceless civilians. For her
to forgive him is to overcome (most of ) the negative feelings which his wrongdo-
ing elicits in her towards him, to accept that he is not reducible to his wrongdoing,
and to allow this acceptance to shape her attitude to him—without foregoing the
judgement that he wronged her. Importantly, this is compatible with her still feel-
ing sorrow when remembering what he did, and perhaps even anger at having been
raped: forgiving does not imply overcoming all relevant negative feelings. The
point though is that in forgiving John (assuming that she does) Fatima no longer
feels angry with John himself; she is no longer filled with rage when thinking about
him, and no longer sees him as just her rapist; should she encounter him, she
would not behave in an angry, resentful, hateful, way; nor would she use her
expression of forgiveness as an instrument for humiliating him, for to forgive
another for his wrongdoing is not to diminish him for it.7
7 In thinking about forgiveness, I have hugely benefited from discussions with Annalise Acorn,
Hanna Pickard, Cheyney Ryan, and Amia Srinivasan. The phrase ‘no future without forgiveness’ is,
of course, the title of Tutu’s best-known book on the topic (D. Tutu, No Future without Forgiveness
(London: Rider Books, 1999). For the view that political forgiveness can promote reconciliation
after war, see, e.g., Philpott, Just and Unjust Peace. Out of a vast philosophical literature, my remarks
on forgiveness owe much to the following works: L. Allais, ‘Wiping the Slate Clean: The Heart of
Forgiveness’, Philosohpy and Public Affairs 36 (2008): 33–68; J. Butler, ‘Sermon VIII. Upon
Resentment and Forgiveness of Injuries’, in J. Butler, Fifteen Sermons Preached at the Rolls (London:
J. and J. Knapton, 1726); Hampton and Murphy, Forgiveness and Mercy; P. Hieronymi, ‘Articulating
an Uncompromising Forgiveness’, Philosophy and Phenomenological Research 62 (2001): 529–55;
J. G. Murphy, Getting Even—Forgiveness and its Limits (Oxford: Oxford University Press, 2003);
J. G. Murphy, Punishment and the Moral Emotions—Essays in Law, Morality, and Religion (Oxford:
Oxford University Press, 2012); C. L. Griswold, Forgiveness—A Philosophical Exploration (Cambridge:
Cambridge University Press, 2007).
254 Reconciliation
Forgiveness so construed implies that Fatima has harboured negative feelings

towards John, and has had to work towards no longer letting those negative feel-
ings determine her attitude towards him. Instant forgiveness is not really forgive-
ness; nor is taking a pill which eliminates those feelings of anger; nor is indifference
as wrought by the passage of time; nor, in fact, is the overcoming of negative feel-
ings for prudential reasons, or because (one thinks) God commands us to forgive.
Forgiveness consists in doing all of the above out of recognition that the wrongdoer
is not reducible to his acts.
Although forgiveness is typically granted and withheld in interpersonal rela-
tions, it is easy to discern why many see it as central to political reconciliation after
war, particularly in the aftermath of mass atrocities in civil conflicts and when
perpetrators have no choice but to return to live side by side, literally, with their
former victims. Indeed it is hard to imagine that those victims and perpetrators can
actually work together towards rebuilding their community—both local and less
local—unless the former somehow undergo the change of heart towards the latter
which is at the heart of forgiveness. Examples such as Nelson Mandela, who
famously forgave his jailers, further buttress the thesis.
However, there is a normative risk that the unquestionable moral authority of
Tutu and Mandela should lead us to over-emphasize the role of forgiveness in rec-
onciliation. To begin with, even if interpersonal forgiveness is necessary to political
forgiveness in those cases, it is far less likely to be so of other post-conflict situa-
tions where victims and perpetrators are not intermingled to the same extent.
Moreover, even in those cases, there are difficulties with the view that forgiveness is
key to reconciliation. In the remainder of this section, I reject one but endorse
three arguments against it.
The argument I reject goes like this. Our concern is with reconciliation in the
wake of war in general, and serious war crimes in particular. Those atrocities—
genocide, rape campaigns, widespread torture and abduction, widespread extor-
tions and pillages causing population displacements with their attendant
traumas—are, one might think, so appalling, so evil, as to be unforgivable. And
yet, those communities do need rebuilding, with most perpetrators in situ. If
reconciliation is possible—and in some cases, such as Rwanda post-genocide, there
are reasons to believe that it has happened, if to a limited extent—it is not thanks
to the granting of forgiveness, since there cannot be forgiveness.
I am sceptical. For what it is worth, the claim that some deeds are unforgiveable
is badly put, for a deed itself, divorced from the (morally responsible) agent who
committed it, is not an appropriate object of forgiveness: only the agent is. Rather,
to say that a crime is unforgivable is to say that it would be morally inappropriate
for anyone, even victims, to forgive perpetrators for committing it. A victim who
forgives, on that view, fails properly to understand and account for what was done
to her, as well as properly to conceive of her relationship with her tormentor. It
seems to me that we ought to reject that view. For a start, to claim that there are such
deeds is to condemn victims who do not wish to forget to a lifetime of negative
feelings (should those feelings not go on their own with the help of time passing).
Should they decide to forgive, they would be aptly described as mistaken about the
Reconciliation 255
nature of their predicament and their conception of themselves in relation to their

wrongdoer. This seems to heap further indignity on them.8 Moreover, I shall argue
below that only victims have the standing to forgive perpetrators for what the latter
did to them. If I am right, only victims have the standing to withhold forgiveness.
Given that, as we have just seen, victims may grant it, even for such crimes, there is
no wrong for which perpetrators may not be forgiven. By implication, the unfor-
giveability argument is misguided and thus does not undermine Tutu’s claim.
Still, there are other reasons to doubt that the claim is true. First, it does not
accommodate other forms of affective redirection when forgiveness is not the
appropriate reactive attitude. As we saw above, forgiveness can only be directed at
wrongdoers. However, the commission of a wrongdoing alone is not enough to
warrant the thought that forgiveness is appropriately granted (or withheld): the
negative feelings which the wrongdoing elicited must themselves be justified.
Rights violations appropriately elicit those feelings, and thus are forgiveness cases.
This is so even if the wrongdoer suffered from diminished agency yet remained
responsible for his actions. However, some wrongdoings, such as non-culpable
rights violations or justified rights infringements, do not warrant those negative
feelings (though having those feelings might be excusable, of course). Consequently,
these are not forgiveness cases.9 Yet, both non-culpable rights violations and justi-
fied rights infringements cause an enormous amount of suffering in war. If there is
no future without forgiveness, then there is no future, no prospect of reconcilia-
tion, for those agents, which seems deeply implausible. In fact, reconciliatory pro-
cesses are both important and clearly possible in those cases, but not in a way that
gives forgiveness a central place.
Second, when one says that forgiveness is key to post-conflict reconciliation, one
might mean not personal forgiveness, whereby victims personally forgive perpetra-
tors, but rather and more plausibly, political forgiveness, whereby the act of forgiv-
ing is undertaken by political actors on behalf of victims. But political forgiveness
is problematic: it raises one of the most difficult issues in the ethics of forgiveness,
to wit, that of vicarious forgiveness. Simon Wiesenthal puts the point better than
anyone I know in his often-mentioned book The Sunflower. One day, while work-
ing in a labour camp for mostly Jewish detainees on the Eastern Railway, he was
8 See T. Govier, ‘Forgiveness and the Unforgivable’, American Philosophical Quarterly 36 (1999):
59–75. Note that my objection to the unforgiveability claim is compatible with two views: the view
that there are deeds such that their perpetrators ought never to forgive themselves for having commit-
ted them; and the view that there are some deeds—such as rape, torture, murder—for which perpe-
trators have no claim to be forgiven. For the latter position, see, for example, the position taken by
Resistant Jean Améry, who was tortured by the Gestapo and sent to concentration camps during the
Second World War. J. Amery, At the Mind’s Limits—Contemplations by a Survivor on Auschwitz and its
Realities (London: Granta Books, 1999). For other sceptics of the Forgiveness Thesis, along the lines I
articulate here, see N. Eisikovits, ‘Forget Forgiveness: On the Benefits of Sympathy for Political
Reconciliation’, Theoria 52 (2004): 31–63; Murphy, Getting Even; M. Minow, Between Vengeance and
Forgiveness—Facing History after Genocide and Mass Violence (Boston: Beacon Press, 1998), ch. 2.
9 On non-culpable wrongdoings, I follow Jean Hampton in Hampton and Murphy, Forgiveness
and Mercy, esp. 52 and 55. For the view that excusable wrongdoings are not forgiveness cases, see
Murphy, Getting Even, 13; and J. North, ‘Wrongdoing and Forgiveness’, Philosophy 62 (1987):
499–508.
256 Reconciliation
taken to the deathbed of a SS combatant who wished to confess to a Jew—any

Jew—his participation in an earlier horrific crime against Jewish civilians. The
dying SS asked Wiesenthal for his forgiveness on behalf of his victims, without
which, he claimed, he could not die in peace. Wiesenthal walked away, not having
said a word throughout the encounter, and ends the story, in the book, with the
question: ‘What would you have done?’10
It is important to be clear as to what vicarious forgiveness is. By it, I do not
mean forgiveness bestowed on behalf of a victim by someone whom she, the vic-
tim, has mandated so to act. In such cases, the mandatee would not appropriately
say, ‘I forgive you for what you did to her’. He would have to say, ‘she asked me to
say that she forgives you’. Rather, I mean that the forgiver forgives a wrongdoer for
acts which the latter committed against others, with no stipulation that those vic-
tims have, either explicitly or tacitly, authorized the forgiver so to act.
Did Wiesenthal, then, have the standing vicariously to forgive the SS? In a lan-
guage made popular by Trudy Govier’s important work on reconciliation and for-
giveness, he was not a primary victim of this particular SS: he himself was not
targeted in that episode. Nor were any of his friends or relatives, so he was not a
secondary victim either. As a Jew, however, he was one of the SS’s tertiary victims,
in that the SS would have regarded his Jewishness as a reason for killing him had
he been there with the others.11 This is precisely why the SS asked him, and not a
Gentile, for forgiveness. Wiesenthal’s question, thus, is whether someone who is
not a victim of a grievous wrongdoing has the standing to forgive perpetrators
vicariously, on the victim’s behalf, for what the perpetrator did to the latter.
Wiesenthal answers in the negative, and I agree with him—though this is com-
patible with having the standing to forgive (or not) perpetrators for what they did
to us, secondary or tertiary victims, by wronging their primary victims. Whilst
Wiesenthal had the standing to forgive the SS for the fact that, had he (Wiesenthal)
been present on that day and in that place, he would have been killed alongside the
others, he lacked the standing to forgive the SS on behalf of the latter’s primary and
secondary victims. To claim otherwise is to deny respectful treatment to those who
were themselves wronged: for who after all are we to forgive (e.g.) génocidaires, serial
rapists, willful and lethal aggressors on behalf of their victims, and particularly
when those victims themselves appropriately refuse to forgive them? By implication,
there can be no official granting of forgiveness by political actors on behalf of the
victims of war: the then-President of the World Jewish Congress had no greater
standing than Wiesenthal to forgive the SS. And if that is correct, then I here and
10 S. Wiesenthal, The Sunflower (London: W. H. Allen, 1970).

11 See T. Govier, Forgiveness and Revenge (London: Routledge, 2002), esp. ch. 5. (Tertiary victims
include putative victims, as I describe Wiesenthal here, but also those who, because their leader is
killed, are harmed in their struggle for justice—for example, as Govier claims, Black Americans were
tertiary victims of Martin Luther King’s assassination.) She too argues against vicarious forgiveness;
likewise Murphy, Getting Even, at 14; Philpott, Just and Unjust Peace, ch. 12. The view that only vic-
tims can forgive perpetrators for the wrong done to them implies that murderers cannot be forgiven
for the wrong they did to those whom they killed. Some find that implausible, but I am willing to bite
that particular bullet. For a nuanced defence of vicarious forgiveness, see Griswold, Forgiveness,
117–22.
Reconciliation 257
now have no standing to forgive any war criminal on behalf of his or her victims—
even if the suffering of war victims—of any victim—is something which we can-
not and ought not to wash our hands of, indeed are appropriately angry at.
In summary, to the extent that political forgiveness inevitably involves some
degree of vicarious forgiveness, and in so far as vicarious forgiveness is impossible,
political forgiveness cannot be instrumental to reconciliation. Yet, reconciliation
there must be—and so we must look elsewhere for the kind of affective redirection
which might strengthen it.
It might be objected that third parties can in fact forgive wrongdoers, and thus
that political forgiveness is possible, without doing so on behalf of victims. Indeed
it might seem that cosmopolitans ought not to be troubled by third-party forgive-
ness—in fact, ought to embrace it. For after all, did I not claim in s.7.6, drawing
inspiration from John Donne’s Sermon, that anyone’s suffering diminishes every
single one of us—involved as we all are in humankind? And is it not central to
cosmopolitan morality that one should have negative feelings—anger, resentment,
hatred—at the commission of those wrongdoings, wherever they occur? And, if so,
are we not in a position to overcome those feelings? Do we not have the standing
to forgive? This is a powerful challenge.12 Yet, though I am appropriately angry
with the nameless, faceless murderers of Pol Pot’s killing fields, at the architects of
the Holocaust, at ISIS fighters who think little of massacring anyone who stands
in their way, I find it hard to see how I, whom they did not and do not harm, could
plausibly say, ‘I forgive you’—indeed, for that matter, ‘I refuse to forgive you’.
Perhaps I have a blind spot here. Perhaps third-party forgiveness is in fact coher-
ent. But even so, there remains a worry with the view that forgiveness in general is
central to political reconciliation. Forgiving someone who has wronged us is com-
patible with refusing to live with them—as when a German Jew forgives the Nazis
but would live anywhere in the world except in Germany. In many such cases,
crucially, the refusal to reconcile or to see oneself as partners in relationship stems
from the fact that wrongdoings of that magnitude destroy trust between victims and
perpetrators when there was trust to begin with (which there might not have been),
or renders the building of trust de novo extremely unlikely. Forgiveness, thus, even
if it is crucial to reconciliation, is not enough: trust matters too. I wager that Tutu
would agree. He would also agree, I think, that victims of wrongdoings can develop
trust in the social and political institutions of their community, and be thus able
and willing to live with members of groups to which wrongdoers belong, without
forgiving those wrongdoers themselves. But then it seems that the key to reconcili-
ation is trust, not forgiveness. In the next subsection, I explore and defend that view.
9.3.2. Trust and Reconciliation

To claim that trust is crucial to reconciliation is not to eschew the broader claim
that political reconciliation is attitudinal and requires affective redirection, as
12 See, e.g., N. Lacey and H. Pickard, ‘To Blame or to Forgive? Reconciling Punishment and
Forgiveness in Criminal Justice’, Oxford Journal of Legal Studies 35 (2015): 665–96.
258 Reconciliation
much as it is institutional: if you betray me, I will not be able to trust you again so
long as I still harbour strong feelings of anger, bitterness, and resentment.13 To
trust someone to do x is to have the expectation that when doing x, she will be
motivated to do so out of respect and concern for us. So conceived, trust can only
be granted and withheld by moral agents, to and from moral agents who are
responsive to the right kind of reasons. Likewise, it can be nurtured or destroyed
only by such agents. Moreover, trust differs from reliance, for in reliance, it matters
not what my reasons are for thinking that the person will do x: for example, it
might be that doing x is in her interest, and I believe that she will act on the basis
of her interest, so I rely on the fact that she will do x. For trust to exist, however, it
is crucial that the trustor should believe that the trustee acts at least in part out of
respect and concern for her. Relevantly to the present context, there is evidence to
suggest that, for peace negotiators to trust each other, they must believe that each
is committed to the peace process because the latter is conducive to their constitu-
ents’ interests. Equal concern and respect matter, but are not enough.14
Reconciliation as a process involves building trust between opponents. As an
outcome—a state of affairs where opponents are reconciled—it also requires trust:
reliance is not enough, for to conceive of my enemy as a partner in peace, I need
to know that he genuinely sees me as such a partner as well. Of course, when trust
has been shattered, the most that can be expected at the outset of reconciliatory
processes is that individuals should be able to rely on others not killing, maiming,
or robbing them.15 But building a justifiedATC peace on reliance alone does not
seem enough over time. Remember, more than 40 per cent of conflicts resume
within five years of their ending. It is not implausible that were trust to be properly
rebuilt between erstwhile belligerents, that figure would be lower. Note too that
trust is important not just instrumentally—for the benefits it brings about—but
intrinsically so as well: it is a sign that parties do think of one another as worthy of
equal concern and respect.16
In political cases of the kind which concern us here, trust is instantiated in the
following kinds of cases: victims trust that state officials will treat them with con-
cern and respect out of recognition of and commitment to the fact both that they
are worthy of such treatment and that they rely on such treatment; victims trust
13 This point does not reintroduce through the back door the claim that forgiveness is central to
reconciliation: for although forgiveness consists in overcoming those feelings, not all cases where those
feelings disappear are forgiveness cases: to repeat, the passage of time might do the trick.
Important accounts of trust in general, on which I draw here, are K. Jones, ‘Trust as an Affective
Attitude’, Ethics 107 (1996): 4–25; A. Baier, ‘Trust and Antitrust’, Ethics 96 (1986): 231–60;
M. Walker, Moral Repair—Reconstructing Moral Relations after Wrongdoing (Cambridge: Cambridge
University Press, 2006), ch. 3; P. Pettit, ‘The Cunning of Trust’, Philosophy & Public Affairs 24 (1995):
202–25; R. Hardin, Trust and Trustworthiness (New York: Russell Sage Foundation, 2002). For an
illuminating account of different kinds of trust in divided societies (though he focuses on relatively
peaceful societies), see D. Weinstock, ‘Building Trust in Divided Societies’, Journal of Political
Philosophy 7 (1999): 287–307. See also J. Dunn, ‘Trust and Political Agency’, in D. Gambetta (ed.),
Trust: Making and Breaking Cooperative Relations (Oxford: Blackwell, 1988).
14 See H. C. Kelman, ‘Building Trust Among Enemies: The Central Challenge for International
Conflict Resolution’, International Journal of Intercultural Relations 29 (2005): 629–50.
15 Kelman, ‘Building Trust Among Enemies’.
16 See also Murphy, A Moral Theory of Political Reconciliation, ch. 2.
Reconciliation 259
their neighbours, who were complicitous in mass atrocities, not to attack them at
the first sign of discord; erstwhile enemy leaders trust each other not to violate the
terms of the peace agreement at the first opportunity; and so on.
Trust, thus, is both horizontal and vertical (it is horizontal between citizens
themselves, or between community leaders/peace negotiators; it is vertical between
citizens and public officials). Further, trust is societal as well as being political, in
so far as it operates between individuals in their social, cultural, economic relation-
ships. At the same time, it can be both personal and political/societal: my reasons
for not trusting you to behave morally in our personal relationships might generate
reasons for not trusting you to support, as a citizen, institutions which would pro-
mote the kind of conduct which you eschew in your private life: if I do not trust
you not to attack me on the grounds that I am gay and that you think gays are
scum, I have little reason to trust you, on the face of it, not to vote for a political
party whose manifesto includes harshly repressive measures against homosexuals. I
might be mistaken in the end: perhaps you have reasons not to vote for that party
which trump your enthusiasm for its homophobic manifesto. But unless those
reasons are known to me and unless I can trust that you will vote accordingly, my
distrust of you in the personal realm warrantedly translates into distrust in the
political realm. Granted, the same might perhaps not apply as much to peace nego-
tiators themselves on either side of the wartime divide: it is not as clear in those
cases that I should distrust your willingness to keep to our peace agreement, as my
fellow head of government, simply because I know that you routinely break prom-
ises in your political life. Either way, though, trust can be directed and active (at
particular people for doing particular things) or undirected (as when we generally
assume by default that, e.g., we will not be attacked, the restaurant at which we
dined will not over-charge our credit card, the shop from which we bought a tele-
vision did not intentionally supply us with a defective product, etc.).
Those points apply to trust within and between communities generally, but they
are particularly salient following wars in general, and civil wars in particular. They
suggest that the kind and amount of evidence needed on both sides will vary
depending on the circumstances. In ordinary, stable social and political settings, I
may need relatively little evidence to trust my neighbour not to attack me; nor for
that matter do I need any evidence to think that the political community with
whom we share a border will not invade us. After a war, if my neighbour has been
convicted of participating in a genocide against the ethnic group to which I belong,
or of participating in gang rapes, I will need considerable evidence that he will not
attack me at the first opportunity. If my MP’s political party has led us into the
abyss of a civil war for which he voted, I will need more evidence than I needed in
peacetime to trust him with office. Likewise with our neighbours across the border,
and likewise with our socio-economic associates. Of course, how much evidence
people need in order to develop trust also depends on their upbringing, tempera-
ment, and past experiences of good and bad treatment. The basic point is that the
fact that individuals have lived through a devastating and wide-scale conflict
makes the (re)building of trust both particularly important and particularly
difficult.
260 Reconciliation
What does it take, then, to (re)build trust after conflict? I noted above that this
kind of trust is a property of relationships between moral agents who see one
another as moral equals worthy of concern and respect, and who are responsive to
the principle of fundamental equality in their dealings with one another. War
crimes consist in a denial of fundamental equality. Trust cannot exist unless victims
are given evidence that perpetrators do now see them in that way; this in turn
requires that perpetrators be able to see their victims in that way as well; when they
have all been locked in cycles of mutual attritional violence, it requires that victims
not just see themselves as victims but as worthy of equal respect. In addition, per-
petrators too need to trust victims not to attack them in revenge. Finally, for trust
to emerge, particularly in the aftermath of mass atrocities which by their very
nature dehumanize both perpetrators and victims, all must at the very least try to
learn to recognize their shared humanity. In the next three sections, I scrutinize
three different institutional mechanisms or practices which have figured heavily in
post-war reconciliations and which, I argue, help rebuild political and societal trust
between erstwhile enemies. The first two, namely traditional justice fora such as the
Rwandan gacaca courts on the one hand, and Truth and Reconciliation
Commissions on the other hand, have been explicitly designed as tools for recon-
ciliation after civil conflicts; the last, which encompasses both official expressions
of regret and official apologies, have been used in a less systematic way to those
ends but have nevertheless played a part in them.
9 . 4 . R E S TO R AT I V E J U S T I C E A F T E R WA R :
T H E C A S E O F G AC AC A
In ch. 7, we saw that it would be both unfeasible and morally wrong to punish all
war criminals as we would normally punish non-war criminals, in domestic con-
texts, for similar crimes (killing, raping, torturing, maiming, stealing, etc.) As we
also saw there, blanket amnesties are sometimes justified as a last resort if the estab-
lishment of mechanisms for holding wrongdoers accountable (short of full punish-
ment) would lead to the resumption of war. By implication, preserving some
measure of accountability is the morally right thing to do if wrongdoers are not in
fact in a position to derail the peace process. It also serves to signal to victims that
they have not been forgotten, and that they will be given some form of recognition
as victims. This in turn may well help heal some of their wounds and make it less
likely that they would give in to the temptation of resuming war.
When wholesale punitive justice is not on the table, restorative justice might
provide a good reconciliatory alternative to it. Its advocates have developed and
defended it (or conceptions thereof ) mostly in the realm of domestic criminal
justice systems, but it has garnered growing attention in the literature on reconcil-
iation.17 In a nutshell, it aims to bring together victims, offenders, and other parties
17 Good overviews of restorative justice in general are G. Johnstone and D. W. Van Ness (ed.),
Handbook of Restorative Justice (Portland, Or.: Willan Publishing, 2007); H. Strang and J. Braithwaite
Reconciliation 261
affected by the offence, with a view to helping them address the offense and its
aftermath. Restorative processes afford victims an opportunity directly to explain
to offenders what impact the wrong had on their life (physical, affective, psycho-
logical), and to offenders directly to explain why they acted as they did and to
express remorse and offer apologies; within guidelines, all parties agree on appro-
priate punishment and/or reparations. In domestic contexts, offenders’ and vic-
tims’ immediate family typically take part in those discussions, to support their
relative but also because they too have been affected by the commission of the
crime, and/or as a means to help towards the reintegration of offenders. In
post-conflict contexts, as we shall see, entire communities take part.
So construed, and contrary to what may appear, restorative justice is not to be
seen as second-best and diametrically opposed to punitive justice. On the latter
count, they are both addressed to morally competent offenders, and they both rely
on and instantiate the view that offenders must take responsibility for their actions.
However, unlike punitive justice, restorative justice gives victims a central role in
the process; it also aims, not to punish the offender but to repair the harm he
caused to the victim and wider community, and (under most conceptions) to rein-
tegrate him into the latter.
On the former count, some advocates claim that restorative justice, far from
being a second-best alternative to punitive justice when the latter is malfunction-
ing or morally undesirable, is in fact the best way to deal with wrongdoing. This is
not because state-imposed punishment, and more specifically custodial punish-
ment, has in practice singularly failed to cut down reoffending rates, is dispropor-
tionately harmful to offenders and their families, and often fails to give victims
what they want (some acknowledgment by offenders of the harm they have suf-
fered, help in overcoming the trauma of their increased vulnerability, etc.) Rather,
restorative justice is the only way (according to its advocates) to further the goals
of getting the offender genuinely to acknowledge that he did wrong, of rehabilitat-
ing him into the community, of signaling to victims that they are worthy of con-
cern and respect—and in doing all of those, of repairing the relationships broken
by the commission of the crime. Consequently, punitive state justice should be
seen as a response to the failures of restorative justice (which failure occurs when,
(ed.), Restorative Justice—Philosophy to Practice (Darthmouth: Ashgate, 2000); E. G. Weitekamp and
H.-J. Kerner (ed.), Restorative Justice—Theoretical Foundations (Portland, Or.: Willan Publishing,
2002). For a classic philosophical defence of restorative justice, see J. Braithwaite and P. Pettit, Not Just
Deserts: A Republican Theory of Justice (Princeton, N.J.: Princeton University Press, 2002). For restor-
ative justice as a post-conflict reconciliatory practice, see J. J. Llewellyn and D. Philpott (ed.),
Restorative Justice, Reconciliation, and Peacebuilding (Oxford: Oxford University Press, 2014); Philpott,
Just and Unjust Peace; I. Aertsen, J. Arsovska, H.-C. Rohne, M. Valinas and K. Vanspawen (ed.),
Restoring Justice after Large Scale Conflict (Cullompton, UK: Willan Publishing, 2008). Some argue
that whereas punitive justice focuses on whether the law has been broken, restorative justice focuses
on what harm has been caused. (See, e.g., Mallinder, ‘Amnesties in the Pursuit of Reconciliation,
Peacebuilding and Restorative Justice’, 161.) Not all restorative justice advocates would take that view:
on the contrary, many insist that restorative justice requires that the offender should take the full
measure of the wrong he committed, not just of the harm he caused. For a good overview of similari-
ties and differences between punitive and restorative justice, see K. Daly, ‘Revisiting the Relationship
between Retributive and Restorative Justice’, in H. Strang and J. Braithwaite (ed.), Restorative Justice—
Philosophy to Practice (Dartmouth: Ashgate, 2000).
262 Reconciliation
for example, the offender refuses to repent or the victim insists on disproportionate
punishment), and not the other way around.
Whatever one’s stance on the relative strengths and weaknesses of punitive v.
restorative justice, in the aftermath of mass atrocities, we should regard them as
complementary. Instead of construing restorative justice as compromise justice, we
should think of it as exceptional justice—that is to say, as the kind of justice which
is required by this particular context.18 As per ch. 7, some perpetrators ought to be
punished, but most ought not to all things considered; as per my earlier claims in
this chapter, victims are deeply traumatized in their dozens of thousands. The risk
is that understandably aggrieved victims and perpetrators will remain locked in
deeply antagonistic, peace-undermining relationships. Restorative practices aim to
make it possible for the former to take responsibility for their acts, for the latter to
tell their story—and for both to build enough trust as to consider working with
them towards the establishment of a justifiedATC peace.
The best-known mechanisms for post-conflict restorative justice are the so-called
gacaca courts in Rwanda—though similar fora were set up, with varying degrees of
success, in the Democratic Republic of Congo, Timor-Leste, and Uganda.19 By
2001, seven years after the genocide in which close to one million Tutsis and mod-
erate Hutus died, Rwandan prisons were overflowing with over 100,000 suspected
perpetrators awaiting trial in appalling conditions and at enormous social and eco-
nomic costs. As neither the International Criminal Tribunal for Rwanda nor
Rwanda’s national criminal courts could deal with all those cases, the Rwandan
government passed a Gacaca Law in 2001, harnessing traditional, non-adversarial
justice fora—the gacaca courts—to the task of dealing with imprisoned suspects.
Thousands of gacaca courts were set up throughout the country, with split jurisdic-
tion over so-called category 2 crimes (murder, helping with the genocide, rape) and
category 3 crimes (robbery, theft)—while category 1 crimes (planning the geno-
cide and committing genocidal acts while in the position of authority) remained
under the jurisdiction of national and international courts. Following a three-
months intensive rehabilitation programme (including basic civic education lessons)
inmates were released into the communities where they were suspected of having
committed crimes of types 2 and 3, to face the courts. Their victims, as well as other
community members, were mandated to attend the proceedings under the guid-
ance and oversight of especially trained judges, themselves community members.
18 On which point I agree with Daniel Philpott. See Philpott, Just and Unjust Peace.
19 See, e.g., P. Clark, The Gacca Courts, Post-Genocide Justice and Reconciliation in Rwanda—Justice
without Lawyers (Cambridge: Cambridge University Press, 2010); P. Clark, ‘Ethnicity, Leadership and
Conflict Mediation in Eastern Democratic Republic of Congo: The Case of the Barza Inter-
Communautaire’, Journal of Eastern African Studies 2 (2008): 1–17; T. Hohe, ‘Justice without judici-
ary in East Timor’, Conflict, Security & Development 3 (2003): 335–57; Hayner, Unspeakable Truths,
201ff; Philpott, Just and Unjust Peace, ch. 11; Drumbl, Atrocity, Punishment, and International Law,
85–99. My remarks on Rwanda’s gacaca courts rely heavily on Clark’s account. A good summary of
the process can be found on the United Nations’ website at http://www.un.org/en/preventgenocide/
rwanda/about/bgjustice.shtml (accessed 21/11/2014). For exploration of other traditional modes of
conflict resolution, notably in Albania (following the war in Kosovo), or in Arab culture (in the con-
text of possible reconciliation between Israel and Palestine, see Aertsen, Arsovska, Rohne, Valinas and
Vanspawen (ed.), Restoring Justice after Large Scale Conflict.
Reconciliation 263
Both victims and perpetrators were encouraged to, respectively, tell the story of their
suffering and express remorse; those who had witnessed alleged crimes or were able
to exonerate suspects through their testimony also took an active part in the pro-
cess, following which the community as a whole would decide which punishment
perpetrators should suffer if found guilty, ranging from paying a fine to doing
some community work.20
Seen purely as mechanisms for stepping into the punitive breach left open by
national and international courts, the gacaca courts were extensively criticized by
(e.g.) Amnesty International and other non-Rwandan commentators on the follow-
ing grounds: lawyers were expressly barred from proceedings; crimes committed by
the Tutsi Rwandan Patriotic Front members were not part of their remit; more
generally, Western-style due process standards for criminal justice were not met—
notably with the gathering of evidence, the possibility of professionally conducted
cross-examination and the dangers posed by mass involvement of the population.
But—and this is crucial—not only were gacaca courts seen as a way to deal with
the fact that under those circumstances those standards could not be met (so that
the criticism misses the point). In addition, the courts did not aim merely to pun-
ish (or, as the case may be) to exonerate suspects: they also and in fact mostly aimed
to rebuild relationships (between family members, neighbours in the same village,
people from different villages) which had been destroyed by the genocide. They did
so by seeking to establish the truth about what happened in specific cases, to instil
in their participants a sense of the value of mediation in conflict resolution, to
preserve some form of accountability, and to offer some hitherto disempowered
individuals opportunities to acquire authority within their community (notably
women, who were able to, and did, take an active part in the process). Construed in
that way, holding them to the usual norms of due process in criminal proceedings
is inapposite.
That said, the design of restorative processes of that kind should be shaped by
(universal) norms of both substantive and procedural justice. In particular, to the
extent that communities are involved in deciding what ought to be done with
perpetrators, restorative processes might be shaped by prevailing social norms
which might be deeply inimical to, e.g., women. Moreover, critics of restorative
justice and of its deployment in post-conflict settings articulate genuine worries
about lack of consistency and proportionality in the treatment of suspects; they
also express important concerns about the extent to which victims might be under
pressure to bestow forgiveness and extend sympathy where neither are in order.21
20 The thought that wrongdoers might be required to carry out reconstruction work as a way to
make reparations to their victims might strike some as too close to forced labour for comfort—par-
ticularly when one recalls the appalling conditions under which German and Austrian prisoners of war
were made to work by the Allies in the aftermath of the Second World War. (See, e.g., MacDonogh,
After the Reich.) As I argue elsewhere, however, there is a strong justice-based case for requiring of all
(able) agents that they perform a form of civilian service. (Fabre, Whose Body is it Anyway?, ch. 3.) A
fortiori so, then, if they have committed serious wrongdoing—provided that they are treated with
basic decency, etc.
21 See, e.g., A. Acorn, Compulsory Compassion: A Critique of Restorative Justice (Vancouver:
University of British Columbia Press, 2004); A. Ashworth, ‘Some Doubts over Restorative Justice’,
264 Reconciliation
The United Nations’ Handbook on Restorative Justice Programmes (2006) offers

guidelines for addressing those points.
The gacaca were stood down in 2012, and evidence of their success is mixed.
Admittedly, they have not achieved all of their aims. Nor are they necessarily
appropriate mechanisms for reconciliation when the enemies which they seek to
reconcile do not share the same culture and/or do not reside together. In this
respect, they are better suited for the aftermath of civil wars between culturally
similar enemies than for the aftermath of interstate wars. But in the context of
Rwanda, they have facilitated the emergence of trust at micro-levels in three ways:
by getting perpetrators to begin to take the full measure of what their victims had
endured; by helping victims to begin to see why their former neighbours and
friends may have given in to ethnic hatred; by providing all with opportunities for
(at last) seeing one another as moral equals rather than as rival groups. It would be
strange if that had had no impact on post-genocide reconciliation, and odd to deny
that they might be of some use in relevantly similar cases.
9.5. T RU T H A N D R E C O N C I L I AT I O N C O M M I S S I O N S
Traditional fora such as the gacaca are not the only way in which enemies have
sought to foster reconciliation. So-called Truth and Reconciliation Commissions
(henceforth, TRC), of which there have been forty or so to date, are another and
better known such mechanism—in no small part due to the enormous influence
and fame of the South African TRC. The overwhelming majority of TRCs were
established in the aftermath of civil wars or in periods of transition from dictator-
ships to democracy—for example in El Salvador and Guatemala—and sometimes
via the peace agreement itself—for example, in El Salvador, Liberia, and Sierra
Leone. Interestingly, however, following Indonesia’s bloody departure, East-Timor,
then Timor-Leste, resorted to such a commission under the auspices of its interna-
tional transitional administration (UNTAET). This is the only instance to date of
a post- interstate conflict commission—though there were (abortive) attempts to
do something similar in the former Yugoslavia.22
Typically, truth commissions are touted as an institution of restorative justice.
As Lucy Allais insightfully notes, however, they can help instantiate the key justifi-
catory value of punitive justice, namely holding wrongdoers accountable for their
deeds and honouring the dignity of the victims.23 Either way, what reasons are
there to endorse them? For a start, victims may, indeed often do, want some form
Criminal Law Forum 4 (1993): 277–99; A. Ashworth, ‘Responsibilities, Rights and Restorative
Justice’, British Journal of Criminology 42 (2002): 578–95.
22 For a comprehensive study of truth commissions, see Hayner, Unspeakable Truths. Unless other-
wise stated, the factual claims I make in this section are drawn from her work.
23 L. Allais, ‘Restorative Justice, Retributive Justice, and the South African Truth and Reconciliation
Commission’, Philosophy & Public Affairs 39 (2011): 331–63. For the restorative justice view of TRCs,
see, e.g., E. Kiss, ‘Moral Ambition Within and Beyond Political Constraints’, in R. I. Rotberg and D.
F. Thompson (ed.), Truth v. Justice—The Morality of Truth Commissions (Princeton, N.J.: Princeton
University Press, 2000); J. J. Llewellyn, ‘Truth commissions and restorative justice’, in G. Johnstone
Reconciliation 265
of public acknowledgment of what they endured—an aspiration which (as we saw

in chs. 6 and 7, respectively) neither reparations alone nor war crime trials can
fulfil. Moreover, they often want their suffering to be acknowledged once they have
told their story and thanks to having told it. Perpetrators, for their part, sometimes
want forgiveness. To that, they are not entitled. Sometimes, however, they want
and need their victims, or the latter’s relatives, to understand the pull of deep social
conditioning, and to see, when appropriate, that they too have also been victim-
ized. Think, for example, of a combatant ordered at gunpoint to kill civilians and
whose family’s survival may well have been at stake; of the informer who was tor-
tured by the security services into betraying his side; of the combatant who killed
defenceless civilians after losing his entire family in an aerial bombing.
TRCs serve those ends.24 They afford victims and perpetrators an opportunity
to tell their stories, and the latter a chance to express remorse (even if that is not
built into the design of the institution). By that token, in establishing the truth
about what happened to this or that person, and through the accretion of those
individual testimonies, they function as channels for and repositories of something
like the historical truth about patterns of serious wrongdoings; they also often issue
in recommendations for the prosecution of this or that individual, and/or propos-
als for institutional reforms and reparations programmes.
Whether there are compelling moral reasons to establish a truth commission
thus depends, first, on whether any of those things are necessary for reconciliation;
second, if they are, on the likelihood that the commission can achieve that task;
and third, whether it can do so better than or at least as well as other mechanisms,
at a similar or lesser cost. I very much doubt that one can reach such a strongly
verdictive and a priori judgement on either count. If, in a given context, establish-
ing manifold truths about war wrongdoings is necessary to help war victims live
side by side with perpetrators and work with them towards a justifiedATC peace, this
provides us with a pro tanto reason to endorse them. But it is not clear (at least to
me) that one can say with any degree of confidence, in advance or in the course of
peace negotiations, that peace will not obtain unless such a commission is estab-
lished. The most that one can say, perhaps, is that, under propitious circumstances,
a truth commission is likely to be of such help. How strong the reasons are to
endorse it as a reconciliatory mechanism is contingent on the facts of the matter.
Importantly, it is not contingent on perpetrators being protected by the same
norms of due process in use in criminal trials. Truth commissions have sometimes
been criticized for falling short of those standards even though perpetrators confess
to having committed serious crimes and, in so doing, expose themselves to consid-
erable hardship. This criticism is overstated however. Norms of due process vary
depending on the institutions: they are not the same for criminal courts, civil
courts, and parliamentary or congressional committees. Given that TRCs do not
and D. W. Van Ness (ed.), Handbook of Restorative Justice (Portland, Or.: Willan Publishing, 2007);
Philpott, Just and Unjust Peace.
24 For a good summary, see M. Minow, ‘The Hope for Healing—What Can Truth Commissions
Do?’, in R. I. Rotberg and D. F. Thompson (ed.), Truth v. Justice—The Morality of Truth Commissions
(Princeton, N.J.: Princeton University Press, 2000).
266 Reconciliation
have it in their power to punish and that the consequences of being named by a
TRC are not as bad as the consequences of being found guilty by a war crime tri-
bunal, it is not clear why they should be held to the same standards as war crime
tribunals.25
Suppose then that in a given case, a truth and reconciliation commission would
be the best step towards a justifiedATC peace. To the extent that erstwhile enemies
are under a duty to bring about a justifiedATC peace, they are under a duty to sup-
port the establishment of such a commission. Before committing ourselves to this
conclusion, however, we need to unpack that duty.
A duty to do something—in this instance, a duty to support a particular insti-
tution—is a positive duty. As we saw in s.1.2, some advocates of positive duties of
distributive justice hold that those duties are best discharged through political
institutions such as the state and that all that individuals are under a duty to do at
the bar of justice is to support those institutions. Others, by contrast, insist that
duties of justice are discharged by institutions and by individuals in their private
capacity. On the former approach, I discharge my duty of justice by voting for a
political party which will set up an egalitarian taxation system, and by paying my
taxes once that party is in power. On the latter approach, I discharge my duty by
doing all of that and by making choices, in my private life, which will mitigate
inequalities or alleviate poverty.
Let us bring the distinction to bear on the duty to support a TRC. Most obvi-
ously, the duty includes a duty to support moves by public officials to establish the
commission, for example by voting for political parties which are committed to
this particular part of the peace process. The institutionalist approach can easily
handle that. Suppose then that the commission is established. For it to do its work,
perpetrators, victims, and bystanders must be willing to come forward and testify.
The question now is whether those agents are under a duty to do precisely that.
The institutionalist approach to justice by definition cannot support this particular
duty, at least in so far as it is held, not by public officials, but by individuals in a
private capacity. Only the individualist approach can do so.
Would it make sense to insist that duties of justice are necessarily institutional,
and thus to conclude that there can be no such thing as a duty of justice to testify?
I do not think so. For recall that the institutionalist approach to justice is grounded
in the fact that institutions best achieve both relevant knowledge of what policies
will best bring about justice, and the high degree of cooperation between duty-hold-
ers which discharging those duties requires. Even if the institutionalist approach is
correct in the context of obligations of assistance, it does not undermine the indi-
vidualist approach to testimony as a means to discharge general obligations to
promote peace, for I do not need to coordinate efforts with other people to attend
a truth commission.
Thus, the strengths of the institutionalist approach in the realm of distributive
justice do not particularly count in its favour in the realm of reconciliation—at
25 For exploration of due process concerns and responses along the lines I sketch out here, see, e.g.,
Hayner, Unspeakable Truths, ch. 10; Levinson, ‘Trials, Commissions, and Investigating Committees’.
Reconciliation 267
least as far as testifying to a truth commission goes. Conversely, the strengths of

the individualist approach count in its favour with respect to the duty to testify,
for it accounts for the importance of promoting and living by a peace-friendly
ethos which regulates our interpersonal relationships within the interstices of the
law. It is not enough, for example, that we should vote in favour of the welfare
state and anti-discrimination laws: we must also treat our fellow human beings in
our daily interaction with them in a spirit of solidarity.26 As applied to truth com-
missions, then, it is not enough that we should vote in favour of establishing
them: we may well be under duties to do so by testifying to them if called upon
to do so.
That being said, we should be sensitive to individuals’ standpoint in relation to
the atrocities committed during the conflict. The case of perpetrators is relatively
easy. I claimed in s.7.7 that conditional amnesties compel victims to forego their
right to punish perpetrators in exchange for something to which they are entitled
anyway—for example, that perpetrators should testify to what they did. This is not
particularly controversial morally speaking. If I commit a serious crime—say, a
murder—I am not simply under a duty not to actively obstruct the course of jus-
tice by not lying or not resisting arrest. I am also under a duty to support the course
of justice by confessing to the crime—a duty which I can discharge all the more
easily if I am granted amnesty as a result.
But what about victims? Many believe themselves to have particularly stringent
moral reasons to bear witness. But other victims simply cannot bring themselves to
reliving the past: in fact, there is evidence that many who testified to the South
African TRC were traumatized anew as a result. This is particularly true of rape,
admissions of which usually elicits shame in victims and often leads to further
victimization at the hands of their families and communities. On the reasonable
assumption that there is a limit to the burdens that may be imposed on innocent
individuals for the sake of promoting peace, there is no duty on the part of victims
to testify to what happened to them when the costs of doing so exceed those limits.
By implication, third parties, who do not stand in the same painful relationship to
those events, have more stringent reasons to bear witness on their behalf and thus
ensure that knowledge of those individual sufferings contribute to the construction
of truth about the past.
To conclude, truth commissions are not a necessary condition for reconciliation
irrespective of circumstances, anymore than traditional fora such as gacaca are:
countries ravaged by civil wars have managed to move towards reconciliation with-
out them, as have a fortiori interstate belligerents. Under some circumstances, how-
ever, they may well provide a good alternative to wholesale punishment, offer the
best available mechanism for holding many perpetrators accountable, and enable
victims and perpetrators to reconcile with one another and thereby work towards
a just, or at least justifiedATC peace.
26 See Cohen, Rescuing Justice and Equality. On the importance of ethos in general, see also
J. Wolff, ‘Fairness, Respect and the Egalitarian Ethos’, Philosophy & Public Affairs 27 (1998):
97–122.
268 Reconciliation
9.6. S AY I N G S O R RY: A P O L O G I E S A N D R E G R E T
By common consensus, both gacaca and truth commissions work best if perpetra-
tors apologize for their deeds. But apologies are one instance of a broader practice,
other manifestations of which are also important for reconciliation, namely ‘saying
sorry’. Consider the following examples (among very many). In 1970, German
Chancellor Willy Brandt, who had opposed the Nazis during the 1930s and had
spent most of the Second World War in exile and under a false identity, fell to his
knees, head bowed, at the monument to the victims of the Warsaw Ghetto upris-
ing. In 1985, German President Richard von Weizsacker delivered a speech at the
Bundestag in which he unambiguously and emphatically lay the responsibility for
the horrors of the Second World War at Germany’s feet. That same year, the
Japanese Prime Minister Y. Nakasone apologized for Japan’s role in the Second
World War in a landmark speech delivered at the United Nations’ General
Assembly. In 1996, 900 years after the start of the First Crusade, a few hundreds
of Christians set off from Cologne and, taking it in turns, walked all the way to
Jerusalem following the original route taken by the Crusaders and apologizing for
the atrocities which the latter committed along the way against Jews and Muslims.
In 1998, in a speech delivered at Kigali Airport, President Clinton apologized for
the fact that the United States, the international community, and ‘the nations
within Africa’ had stood by while the 1994 Rwandan genocide went on. In 1999,
rebels in Guatemala apologized for atrocities committed by guerrilla movements
during the country’s decades-long civil war—as did the IRA in 2002 for killing
civilians through sectarian violence. In that same year, Israeli Prime Minister Ehud
Barak expressed regret (but did not apologize) for the suffering which decades of
conflict between his country and Arab nations has caused the latter, ‘including the
Palestinian people’. Finally, at many proceedings of the South African Truth and
Reconciliation Commission, some perpetrators of atrocities against black South
Africans apologized to their victims—notably Eugene de Kock, who in the end
was denied amnesty and was sentenced to a 212-year long prison sentence for
crimes against humanity. Yet, many believe that he should not have been paroled
(as he was in January 2015), and that his apologies are meaningless precisely given
the nature of his crimes. In that particular vein, Albert Speer famously commented
that neither he nor his fellow Nazis could ever apologize for the Holocaust.27
27 On the case of Guatemala, see http://www.latinamericanstudies.org/guatemala/regret.htm

(accessed 03/11/2105). For the IRA, see http://www.theguardian.com/uk/2002/jul/17/northernireland.
northernireland2 (accessed 03/11/2015).
On Ehud Barak’s statement, see, e.g., Bradley, Refugee Repatriation, 220. On the Reconciliation
Walk, see D. Celermajer, The Sins of the Nation and the Ritual of Apologies (Cambridge: Cambridge
University Press, 2009), 35. On von Weizsaker’s speech, see, e.g., A. Lazare, On Apology (Oxford:
Oxford University Press, 2004), 80–1. Generally, the literature on apologies is not as rich as the liter-
ature on forgiveness. The most interesting works are: M. Cunningham, States of Apology (Manchester:
Manchester University Press, 2014); Lazare, On Apology; M. Mihai and M. Thaler (ed.), On the Uses
and Abuses of Political Apologies (Basingstoke: Palgrave, 2014); M. Nobles, The Politics of Official
Apologies (Cambridge: Cambridge University Press, 2008); N. Smith, I Was Wrong—The Meaning of
Apologies (Cambridge: Cambridge University Press, 2008); N. Tavuchis, Mea Culpa—A Sociology of
Apology and Reconciliation (Stanford: Stanford University Press, 1991); E. Barkan and A. Karn (ed.),
Reconciliation 269
From those examples, one can extract several features of regret and apologies for
serious wrongdoings committed in the course of conflicts. First, some apologies are
delivered by individual perpetrators for acts which they themselves committed (De
Kock). Second, many apologies are delivered years, sometimes decades (indeed
centuries!) after the facts, by individuals who speak in an official capacity on behalf
of a community whose predecessors were at fault for the wrongdoings, and who
themselves were not at fault, either because they actively resisted the perpetrators
(Brandt) or because they were born after the war (the Reconciliation Walk).
Sometimes, of course, the actor issuing an official apology was himself partly
responsible for the atrocity or the failure to stop it—though falls short of apologiz-
ing for his own failings (Clinton). Third, official expressions of regret, or apologies,
are issued not only by state actors but also by nongovernmental forces (guerrillas
in Guatemala, the IRA in Northern Ireland.) More rarely, apologies can be issued
by individual members of grass-roots movements for deeds for which they them-
selves share no responsibility whatsoever (Reconciliation Walk). Fourth, those in
front of whom the apology is issued are not always their intended main addressees
(von Weizsacker). Fifth, expressions of regret are often used as a way not to apolo-
gize when an apology really ought to be forthcoming (Barak). Finally, there may
well be grievously wrongful acts for which no apology can ever be issued (Speer).
In what follows, I focus on collective and official cases of saying sorry, for these
are the most relevant for political reconciliation—although my earlier claim that
individuals are under the relevant duties stands here too. As preliminary caveat,
explicit apologies or expressions of regret are not necessary for reconciliation in all
cases: the Franco-German relationship since 1945 is ample testimony to that fact.
At the same time, in other contexts, they might well be necessary, or at any rate
particularly helpful, especially if the prevailing culture is one that values apologies
in general—so that explicitly refusing to apologize may be seen as indicative of a
refusal to reconcile.
9.6.1. Apologies and Regret

As a preliminary step, we need to distinguish between apologies, agent-relative
regret, and agent-neutral regret. An apology for a wrongful or unjust deed or state
of affairs is appropriate only if one is morally culpable for having brought about x.
Agent-neutral regret about x is appropriate just in so far as x is wrongful or unjust.
Reasons to be regretful about x, in such cases, apply to all moral agents. By con-
trast, agent-relative regret about x is appropriate if one is relevantly connected to
(but not culpable for) x. The following three forms of connectedness can give rise
Taking Wrongs Seriously (Stanford, Cal.: Stanford University Press, 2006). See also J. Thompson, ‘Is
Political Apology a Sorry Affair?’, Social and Legal Studies 21 (2012): 215–24; J. Blustein, Forgiveness
and Remembrance (Oxford: Oxford University Press, 2014), 159–60; S. Winter, ‘Theorising the
Political Apology’, Journal of Political Philosophy (2014): 1–21. For a classic discussion of regret, see
B. A. O. Williams, ‘Moral Luck’, in B. A. O. Williams, Moral Luck: Philosophical Papers, 1973–1980
(Cambridge: Cambridge University Press, 1982). See also J. Wallace, The View from Here—On
Affirmation, Attachment, and the Limits of Regret (Oxford: Oxford University Press, 2013).
270 Reconciliation
to agent-relative regret: the agent is responsible (though not culpable) for a wrong-
ful harm, as in Bernard William’s example of the conscientious driver who una-
voidably kills the child jumping in front of his car; the agent has benefited from a
wrongful harm which she has not herself brought about, as when we, here and
now, can be said to benefit from the institution of slavery; the agent belongs to the
same transgenerational community as the perpetrators of the wrongdoing and
there are reasons to believe that her failure to express regret could signal that she
condones the wrongdoing. In the remainder of this section, when the agent has
brought about the wrongdoing, I shall speak of agential regret; in the other two
cases I shall simply speak of non-agential regret.28 I shall use the word ‘regret’ to
denote all instances of regret.
In paradigmatic cases, regret (in both of its guises) and apologies have the fol-
lowing features. Expressions of regret (in response to wrongdoing) convey both
moral condemnation of and thereby distancing from the deed, as well as sorrow for
the suffering which it occasioned. Apologies also convey moral condemnation,
albeit of what one has done; they too express sorrow for the suffering one has
caused, but they also express remorse. Unlike regret, though, they communicate a
resolve not to reoffend; they also often express hope on the part of the perpetrator
that all of those to whom the apology is addressed might be willing to accept that
he is not reducible as a person to what he did.
Moreover, apologies and regret are addressed not just to victims, but to all of
those to whom wrongdoers are accountable. In the present context, where apolo-
gies are expressed or withheld for war-related human rights violations, they are
addressed to humankind at large since those wrongdoings are of concern to
humankind has a whole. In addition, both ought to be sincere if they are to be
meaningful. Although official regretful and apologetic statements are often used as
a strategic means to achieve a particular diplomatic end, using them in this way
subverts their meaning and fails to treat their addressees with the requisite respect.
In fact, utterances of regret and apologies which are known to be fake may be more
likely than not to foster mutual distrust and contempt and thereby to impair pros-
pects for reconciliation (in both personal and political cases).29
Finally, to accept an expression of regret—as distinct from merely to take note
of it—is to acknowledge that those who utter it do indeed condemn and distance
themselves from the wrongdoing. To accept an apology is to acknowledge that the
wrongdoer is genuinely remorseful about it, and that he has shown himself not to
be wholly reducible to what he did. (This is compatible, it seems to me, with not
fully overcoming at least some of one’s negative feelings about him—in other
28 See Williams, ‘Moral Luck’. The term ‘agent-regret’ was coined by Williams to distinguish
between a morally culpable agent (say, a drunk and reckless lorry driver who kills a child by mounting
his truck on the pavement—my example), and the conscientious lorry driver who simply cannot avoid
hitting a child running across the road. The former owes an apology to the parents (at the very least!),
the latter ‘only’ an expression of agent-regret. I use the term ‘agent-relative regret’ to denote that regret
is owed, not just in such cases, but in other cases where the connection between the agent and the
wrongful harm is not agential. In the discussion that follows, I greatly benefited from insightful com-
ments by Jessica Begon, Alasdair Cochrane, and Ed Hall.
29 Pace Lazare, On Apology, 119 and 157–8, and in agreement with Smith, I Was Wrong, 93.
Reconciliation 271
words, with not forgiving him.) In both cases, to accept the utterance for what it
is, is to accept its agent as a moral interlocutor.
In the context at hand, whether regret (and if so which kind) or apologies are in
order depends on the degree to which the agent making the utterance can be
meaningfully said to share responsibility for the wrongdoing (in which case apolo-
gies or agential regret are appropriate), and how the agent is non-agentially con-
nected to the wrongdoing; it also depends on whether the agent has the standing
to speak on behalf of the group. In this context, as in the context of reparations and
restitutions, the problems raised both by the collective nature of many of the rele-
vant wrongdoings and by the passage of time loom large. Let us address both
points in turn.
9.6.2. The Collective Nature of War-Related Crimes

First, then, the collective nature of war-related crimes, beginning with the prob-
lem of collective responsibility. Only agential regret for what one has brought
about on the one hand, and apologies on the other hand, rely on attributions of
collective responsibility. Recall that on my view, ‘individuals can be said to act
jointly towards a particular goal and to be individually responsible for its occur-
rence when they intentionally contribute to an enterprise of which this goal is a
part; further, this is so even though they need not intend to bring about that goal
itself, provided that they can be reasonably expected to know that other individ-
uals not only also do so but have at least overlapping understandings of what that
larger enterprise consists in’ (s.7.4.1). To the extent that agents are individually
responsible in that sense for the war crimes committed on their behalf by their
leaders, they are under a prima facie duty to apologize or express agential regret for
those war crimes. Accordingly, when a public official with the requisite standing
issues an apology for war-related crimes on behalf of his community, he is in fact
apologizing on behalf only of those individual community members who are
responsible in that sense. On my account, Richard von Weizsacker was not apolo-
gizing in 1985 on behalf of the vast numbers of German citizens who were not
adult, or were not even born, during the war. By implication, when there no
longer will be German citizens who were adults and relevantly responsible for the
evils of Nazism, it will not make sense for a high-ranking German official with the
requisite authority to apologize for the Holocaust on behalf of her compatriots.
But it would not be incoherent for her, as a representative of the German people,
to declare publicly, solemnly, and unambiguously, that the Holocaust was a griev-
ous wrongdoing and that German citizens here and now do not identify with this
part of their collective past—though as we shall see presently, there might also
come a time when such expressions of regret would no longer be required. This
particular kind of regret is agent-relative, in that the Germans have reasons to
express it, via the official, which other agents do not. But it is not agential, since
they themselves are not responsible for the Holocaust.
Collective regret and apologies raise another issue, that of who has the standing
to speak on behalf of those group members who are under the relevant duties. For
272 Reconciliation
leaders of a loosely structured group to speak on behalf of the latter is problem-

atic, for the usual reasons to do with lack of appropriate authorization procedures.
It is not always clear, thus, that guerrilla leaders always have the standing to speak
on behalf of their individual fighters—as was the case in Guatemala. By contrast,
we tend to see little difficulty with entrusting a Head of State with this particular
task. This is partly why Clinton’s apology for the international community’s and
African nations’ inaction during the Rwandan genocide rankles: he clearly did not
have the standing to apologize on anyone’s behalf other than (relevantly responsi-
ble) American citizens. (The other reason why it rankles, of course, is that by
making a generic apology, Clinton clearly hoped to deflect attention from his
own failings, as President of the United States, in respect of the prevention of the
genocide.)
Even when we can identify some agent with the relevant standing, collective
expressions of regret or offers of apology remain problematic. To be meaningful
and successful they must be sincere, sorrowful, and (in the case of apologies)
remorseful. But suppose that the high-ranking official is neither sincere nor sor-
rowful nor remorseful. Are we to infer that he is not, in fact, apologizing or
regretting? And when, conversely, he clearly and visibly appears sincere and sor-
rowful, as Willy Brandt did in Warsaw, are we to infer that the apology or regret
he is thereby offering/expressing is more successful qua official apology? Much
depends, it seems, on our reasons for believing that those on whose behalf he acts
regret or are remorseful about the act. To illustrate: with all due respect, Her
Majesty Queen Elizabeth II is not known for impassioned oratory. But if, follow-
ing a long and by all accounts popular campaign, she were officially to express
regret for the bombing of Dresden, we might well regard that as a proper and
genuine utterance on behalf of enough British people to licence us to say that
this community, save for a minority, regrets the bombing. Conversely, were the
British Prime Minister one day tearfully and passionately to apologize to the
Iraqis for the 2003 invasion of Iraq on behalf of British citizens who are collec-
tively responsible for that war, this would not be enough to convince anyone that
the apology is a proper one, should there be evidence that those citizens still
support the war ex post.
The point, thus, is this: in political regret and apologies, the public official need
not feel guilt, remorse, or sorrow. Emotional display or lack thereof on her part is
not the touchstone of the sincerity with which the apology is expressed. What
matters is whether there are reasons to believe that she really is speaking on behalf
of her compatriots and that her compatriots themselves, at least in some meaning-
ful numbers, have the requisite reactive attitudes: in that sense, official apologies or
expressions of regret are not vicarious utterances. Nor is it necessary (in apology
cases) that she herself should have been implicated in the crimes. Interestingly
though, when the official was not implicated, in fact resisted those wrongdoings at
the time of their commission, and yet sorrowfully apologizes or expresses regret, the
impact is all the more powerful: for if she of all people is willingly publicly to act
and exhibit those reactive attitudes on behalf of her fellow citizens when by right
she stands closer to victims than to perpetrators, this suggests that there is hope yet
Reconciliation 273
for reconciliation. This is why, I think, Willy Brandt’s gesture in Warsaw still reso-
nates four decades later.30
Some readers may object that my endorsement of official apologies is incompat-
ible with my rejection of political forgiveness. I rejected political forgiveness as an
appropriate basis for political reconciliation on the grounds that it is an instance of
vicarious forgiveness, and that vicarious forgiveness is not forgiveness at all. But by
parity of reasoning (those readers might object), vicarious acceptance of apologies
is no acceptance either. Had Wiesenthal or the then-President of the World Jewish
Congress been offered an apology by the dying SS, they would not have had the
standing to accept it on behalf of his (primary and secondary) victims any more
than they had the standing to forgive him on their behalf. But if vicarious accept-
ance of apology is incoherent (the objection goes) then the official acceptance of
political apologies is incoherent as well, and there seems little point therefore in
officially offering them.
Upon closer inspection, however, rejecting vicarious forgiveness and vicarious
acceptance of apology (for the two are on a par) is compatible with there being a
duty to apologize. For in apologizing, one repudiates one’s wrongdoing and com-
municates a hope to be seen as a moral interlocutor by the addressees of one’s apol-
ogy. Although an apology is necessarily a public act (in that one cannot meaningfully
apologize in foro interno), it need not be accepted in order to be made. Suppose
that the then-German Chancellor had apologized in 1925 on behalf of his compa-
triots for the invasion of France and Belgium in 1914, in the presence of both
heads of state. Neither would have had the standing to accept that apology on
behalf of the victims of the invasion; but both would have had the standing to
represent those victims at that ceremony as addressees of that apology. The pres-
ence of officials representing its addressees only serves as a way to publicize the
apology in the right direction, so to speak, and to foster trust in its addresses with
respect to the intentions of the addressors. Whether the apology is accepted is then
up to victims, who may (if they so wish) authorize a public official—for example,
their country’s prime minister or president—to accept the apology on their behalf.
9.6.3. The Passage of Time

So much for the collective nature of official regret and apologies. What about the
passage of time?31 Interpersonal apologies too are subject to it. One may wonder,
for example, whether it is appropriate for an individual to apologize or express
regret for the crimes her grandfather committed against another person. But the
30 On the affective dimension of apologies, see L. Bovens, ‘Apologies’, Proceedings of the Aristotelian
Society 108 (2008): 219–39, esp. 225–8. My concern in this paragraph is with the relationship
between the public official’s apologetic or regretful utterance on the one hand, and her compatriots’
apologetic or regretful dispositions on the other hand. As Meira Levinson pointed out to me in private
correspondence, a public official might be justified in offering an apology or expressing regret, qua
public official, hoping in an act of leadership to incite her compatriots to develop such dispositions.
31 For the sake of brevity I do not discuss the issue of posthumous apologies—issued on behalf of
dead people and/or to dead people. For what it is worth, I do not think that they count as genuine
apologies. For discussion, see Smith, I Was Wrong, ch. 5.
274 Reconciliation
issue is particularly salient in the political context, since regret is expressed and
apologies uttered by agents who belong to a transgenerational community and
who, in virtue of that membership, believe that they are, or at any rate are thought
to be, under an obligation so to act even if the crimes were committed before they
were born. Of all the examples with which I started, the Reconciliation Walk is the
most extreme: 900 years after the fact frankly seems to be pushing it.
In fact, however, the issue is not the fact itself of time passing: it is whether a
long ago injustice has contemporary resonance. We might laugh at the thought
that a handful of Christians should apologize to Jews and Muslims in 1996 for the
massacres committed by Christian Crusaders against Jewish and Muslim commu-
nities nine hundred years previously. As should be clear, I do not think that apolo-
gies are in order in this case anyway, for reasons to do with (lack of ) collective
responsibility. However, in a climate of considerable tension between countries of
an explicitly Christian heritage on the one hand, and Muslims on the other hand,
and in the light of enduring Christian anti-Semitism, an expression of regret might
serve as an important reminder to all parties that not all Christians harbour reli-
gious hatred for devout members of the other two Abrahamic religions.
That said, the fact that people still care, here and now, about a long ago war
crime is not enough to warrant calls for expressions of official regret.32 Apparently,
some French people still resent the English for burning Joan of Arc at the stake in
1431, during the One Hundred Year War. I am not convinced at all that the British
government should express regret for that—not because it was not a war crime (it
was), and not because the English authorities were not responsible for her execu-
tion (they partly were), but because French individuals qua members of the polit-
ical community known as France and English individuals qua members of the
political community known as England have since been locked into a deep, long,
eventful history whose different stages (both good and bad) are much more signif-
icant today than that old crime. I suspect that many will agree (and will counsel
Joan’s devoted supporters to get over it). However, they might find it troubling
that, by implication, there might come a time (in 100 years, 200 years, 500 years,
1000 years?) when German citizens will no longer be under a special duty, qua
German citizens, to express regret via high-ranking officials for the Holocaust.
Well, this is an implication of my argument—and not one so troubling, I think, as
to suggest that the argument should be rejected.
9.6.4. Why Apologize and Express Regret?

So far, so good. But on what grounds must erstwhile enemies express regret or
apologize for wrongdoings committed during the war? For what ought they to be
offered? And are addressees under a duty to accept them?
Not apologizing for one’s wrongs, and a fortiori explicitly refusing to do so, adds
to the first wrong the second wrong of indifference to the predicament in which
32 On which point I disagree with Michael Cunningham’s interesting discussion of this issue.
Cunningham, States of Apology, 49.
Reconciliation 275
victims find themselves as a result of one’s own doing. That second wrong is not
merely an insult, it is an injury in its own right, for it betrays a failure to treat the
victim as a moral interlocutor.
Agent-neutral regret, for its part, ought to be expressed on the deontological
grounds that one cannot be committed to the view that perpetrating those deeds is
morally wrong and yet refuse in principle publicly to denounce them: all of us have
a pro tanto reason to express regret that wrongdoings were committed and suffering
wrongfully inflicted (and in so doing reaffirm one’s condemnation thereof ).
Agential regret, for its part, shares one important feature with apologies.
.
Although the agent is not culpable for the wrongful state of affairs, x, about which
regret is appropriate, he nevertheless is responsible for it. To express no regret at all,
or to express regret as if he were not in any way causally implicated in x, is to fail
properly to account for his part in the victim’s predicament. Such a failure is also a
failure of respect.
Finally, non-agential (agent-relative) regret admits of slightly different justifica-
tions. I said above that benefiting from x makes it appropriate to express regret
about the wrongdoing—as does a particular kind of connectedness with perpetra-
tors of the wrongdoing. With respect to benefiting, here again, to express no
regret at all about x, and/or to do so agent-neutrally, is to fail to account properly
for the fact that one is not situated in relation to x in the same way as a bystander,
and that there is a morally salient asymmetry between one’s lot (having benefited
from x) and the victim’s lot (having wrongfully suffered as a result of x). That
failure, in turn, is not merely a lack of intellectual honesty. It is also a moral
failure—the failure of failing properly to acknowledge one’s moral location in
relation to a particular wrongdoing, and thereby the victim’s moral location in
relation to one’s own.
The second kind of non-agential regret relies on a morally relevant connection
between the perpetrators of the crime and the agent called upon to express regret,
such that regret, understood as distancing oneself from the wrongdoing, is appro-
priate. At the bar of cosmopolitan morality, political membership in a transgener-
ational community cannot alone be one such basis. Sometimes however an open
commitment to one’s community’s heritage may give others, notably victims and
their descendants, good reasons to fear that one positively identifies with the
crimes—unless there is explicit distancing therefrom via the expressing of regret. In
this case, expressing regret is warranted, on the simple grounds that instilling fear
and distress in others is generally morally undesirable.
Thus far, my arguments for apologies and regret have been, for the most, deon-
tological. But as the point about reassurance suggests, those utterances and rele-
vant practices also admit of an instrumental justification. Admittedly, when they
are owed to victims, this is not the main reason why parties should so act: if I
apologize or express regret only because I seek peace, and not solely because I believe
myself to be under a duty to treat my victim as my moral interlocutor and to dis-
tance myself from the wrongdoings, my utterances are not genuine, and thus do
not really count as such. Nevertheless, the process of reconciliation does constitute
an additional reason for so acting.
276 Reconciliation
I say ‘when those utterances are owed to victims’. However, victims are not owed
an apology, and it is debatable whether they are even owed agential and non-agen-
tial regret, when they themselves have grievously wronged those who have wronged
them. For example, I am not persuaded that the Germans owe regret or apologies
for the bombing of Coventry in 1940 to British perpetrators of and beneficiaries
of the 1944 bombing of Dresden, and vice versa. Wars offer many examples of
reciprocal and grievous wrongdoings. That said, in such cases, we can and should
justify joint utterances of (agential) regret, along the lines ‘we/our predecessors
should not have done this to each other’, as indicative of a common resolve to
work towards peace and a means to facilitate reconciliatory processes.33
By that very same token, however, apologies and expressions of regret are con-
strained by the imperative not to jeopardize prospects for a justifiedATC peace. If
insisting on public acts of that kind or performing them are likely to do so—for
example because perpetrators would walk away from the peace negotiations—then
one should not do so. This raises the delicate issue of timing: a long delayed utter-
ance might sometimes be insulting for being too little, too late; conversely, a pre-
mature utterance, particularly a premature apology, might also be insulting for
suggesting that the perpetrators are thereby seeking to block victims’ expressions of
legitimate anger.
Note also, furthermore, that what counts as successful expressions may differ
from culture to culture. In both interpersonal and personal contexts, apologies are
often the outcome of more or less complex negotiations between parties, both of
whom have their own understandings of what they will or will not accept. Those
understandings are partly shaped by their own cultural values.34 It is therefore part
of the duties to express regret and to apologize that they should be discharged in a
way that is sensitive to addressees’ cultural practices—at least up to a point: if
either regret or, perhaps more likely, apologizing requires self-abasement in certain
cultures, we might want to resist doing so.
9.6.5. For What Ought We To Apologize and Express Regret?

For what, then, ought regret be expressed and apologies be offered? As we have
seen, for wrongful states of affairs. In the context of this book, those wrongful
states of affairs are characterized by widespread culpable rights violations, non-cul-
pable rights violations, and justified rights infringements. In s.9.3.1, I argued that
only culpable rights violations call for forgiveness, and that no such violation is
unforgiveable in the sense that no one, including its victim, is never justified in
forgiving its perpetrator for it. In that regard, regret and apologies are different
from forgiveness, and from each other. Regret is an appropriate response to wrong-
ful suffering of all three kinds—not just to culpable rights violations. Apologies, on
the other hand, are an appropriate response only for wrongful suffering for which
one is morally culpable—which rules out non-culpable rights violations as carried
33 For the same point in interpersonal cases, see Bovens, ‘Apologies’, 231.
34 See, e.g., Lazare, On Apology, ch. 10; Smith, I Was Wrong, ch. 4.
Reconciliation 277
out by, e.g., child combatants, as well as justified rights infringements. If a child
combatant, once rehabilitated, were to issue an apology to his victims, it would be
better to regard this as an expression of agential regret conveying his repudiation of
his former self and his condemnation of the (objective) wrongs he committed.
Likewise, if Winston Churchill had issued an apology to collateral victims of the
Allies’ bombing campaigns in the Second World War, it would not have been
appropriate to condemn this gesture on the grounds that this collateral damage
was justified, even if it was in fact justified. Rather, it would have been appropriate
to regard it as an expression of agential regret—an expression, moreover, which in
this particular case and unlike both apologies and regret for non-culpable rights
violations, need not convey the view that one ought not to have done it and will
not do it again.
Finally, there are wrongdoings which, as Albert Speer recognized, do not admit
of apologies. To apologize is not merely explicitly to condemn the wrong one com-
mitted and to repudiate oneself as the perpetrator of that wrong; it is also often to
express the hope that one will not be reduced to the sum total of one’s wrongdo-
ings. But some crimes, of which ordering and planning genocide is one, are such
that their perpetrators can only hope of their fellow human beings that the latter
will not dehumanize them in revenge: they cannot hope, let alone expect, for more,
and it is in that sense that apologizing, to the extent that it would encapsulate such
hope and expectation, would be inappropriate. This is not to say that perpetrators
ought to remain silent. On the contrary, they ought to say, as Speer did, that what
they did is unspeakably evil and that it is so evil that apologies are simply not in
order: in that admission lies a deeper acknowledgment, which they owe to their
victims, of the latter’s humanity. This, I think, is how Richard von Weizsaker’s
1985 speech to the Bundestag is best construed.
Speer’s example is interesting, for it suggests that his explicit non-apology was
addressed not just to victims of the Holocaust but to humankind. If one believes
that humankind at large—that is to say, all of us—is wronged by the commissions
of grievous wrongdoings, then humankind is an appropriate addressee of apolo-
gies. However, as I suggested in s.9.3.1 when discussing vicarious forgiveness and
in ch. 7 when discussing crimes against humanity, I am sceptical. Accordingly,
only victims themselves (including primary, secondary, and tertiary victims) have
claims to apologies. But narrowing the reach of apologies in this way is compati-
ble with insisting that expressions of regret are in order vis-à-vis non-victims.
Recall Japanese Prime Minister Nakasone’s apologetic speech, which was deliv-
ered to the United Nations’ General Assembly. The United Nations did not exist
during the Second World War but has stood since 1945 as a representative of the
world community of states from which Japan was seeking a full and forward-look-
ing acknowledgment of its equal status. As addressed to the world so represented,
the speech is best seen as what I have called an expression of (non-agential) regret.
In so far as it was delivered by and on behalf of Japanese citizens who, though not
responsible for war crimes, were responsible for subsequent wrongful failures to
make reparations to victims, it also and at the same time constituted an apology
to the latter.
278 Reconciliation
9.6.6. On the Duty to Accept Apologies

To offer an apology, or express regret, is one thing. To accept those utterances is
another.35 In ordinary interpersonal contexts, it seems that at least in some cases,
addressees are under a duty to so do. If you do something wrong to another person,
and if you later tell me, sincerely and meaningfully, how much you wish you had
not done it, I ought to take your utterance seriously for what it is—an expression
of regret. If you wrong me in some sense, for example by breaking a promise with-
out a good, defeating reason, in a moment of uncharacteristic unreliability, and
you subsequently apologize giving all appearances of remorse, again, I ought to
accept your apology. I ought not to insist that you utter it again and again. In just
the same way as failing to apologize for one’s wrongdoings, or failing openly to
regret the commission of wrongdoings by others, can constitute a failure to respect
victims as moral interlocutors, failing to accept those utterances for what they are
can constitute a failure to respect wrongdoers and those who express regret as
moral interlocutors as well.
However, serious wrongdoings are somewhat different, at least with respect to
apologies and agential regret, though not with respect to non-agential regret and
agent-neutral regret. For when regret is expressed by agents who are not themselves
responsible for the wrongdoing or representatives thereof, it seems that its address-
ees ought to accept it, for reasons just outlined, even for grievous wrongdoings.
Apologies and agential regret by contrast are addressed by those who are culpable
or relevantly responsible to all agents to whom they are in some sense accountable,
and mostly victims. Given the seriousness of those wrongdoings (murder, rape,
torture, wanton destruction of life-sustaining property, etc.), it might well be gen-
uinely impossible for victims actually to acknowledge that culpable wrongdoers are
remorseful and not reducible, morally speaking, to their crimes. When their rights
have been non-culpably violated (e.g., by child combatants) or justifiably infringed
(e.g., by tactical bomber) it might also be genuinely impossible for them to accept
that the agent, or those speaking on his behalf, does wish to distance himself from
their predicament.
9.6.7. Two Worries

Before concluding, let me allay two worries which some might have about the
conclusions defended so far. The first worry is a worry about inflation. On my
account, all war-related rights violations or infringements call for apologies and/or
regret. Those rights include not just rights not to be grievously maimed or killed,
but also rights to assistance (by way of military intervention against other political
communities, or to needed resources against occupying forces). In addition, indi-
viduals, and not just institutional agents acting on their behalf, are themselves
under moral duties to condemn collective wrongdoings for which they share
35 I am grateful to David Miller for raising this question with respect to apologies and to Chris
Bennett for a useful correspondence on this point.
Reconciliation 279
responsibility, to specifically condemn wrongdoings from which they benefit, to

condemn those wrongdoings agent-neutrally, and to support reconciliatory pro-
cesses. Taken together, those claims imply that apologies and/or regret are owed for
a huge number of wrongs—far too huge for our limited cognitive, emotional,
time-related resources. By way of reply: even though we are not able to issue apolo-
gies and/or express regret for every single war wrongdoing (a point which one
cannot but readily grant), it does not follow that we may do nothing. In fact,
official ceremonies and speeches are a way to discharge some of those obligations.
At the same time, when the occasion calls for it, we ought, at least sometimes, in
our individual capacity, explicitly to apologize or express regret to a victim of a war
for which we are partly responsible, from which we benefited, and so on.
The second worry is sometimes (though inaptly) called the Apology Paradox. It
goes like this. In expressing regret for war crimes which happened before I was
conceived, I am likely to express regret for the occurrence of events without which
I would not exist. But how can we regret our own existence, indeed, the existence
of our loved ones? If we are genuinely glad that we and our loved ones are alive—a
feeling most of us seemingly have—then we cannot genuinely regret the occur-
rence of war crimes when we would not exist were it not for them.
In response, some have said that one is not really genuinely regretting those war
crimes: one is simply wishing that one’s existence did not depend on the commis-
sion of those crimes, or one is merely expressing a preference for a world in which
they were not committed and we nevertheless exist. Yet, I do not think that this
response quite works. For it seems that we ought to regret, more strongly, that
those wars happened, even though we thereby regret the occurrence of a world in
which we do exist and conversely wish that a radically different world, a just world,
had come into being, albeit at the cost of our existence. To think otherwise is to
wish it to be the case that millions of people should die and suffer for the sake of
bringing us (you, me, our loved ones) into existence. I struggle to see how one can
hold that view and at the same time claim that we ought to treat one another with
equal concern and respect.36
To conclude, agents who are responsible for war wrongdoings are under a prima
facie duty to apologize or express agential regret to their victims; beneficiaries of
those wrongdoings, as well as agents who share certain ties with perpetrators, are
under a prima facie duty publicly to express non-agential regret, via properly
authorized officials. Under some circumstances, addressees are under a duty to
accept expressions of regret, and to try to accept apologies. The lesson to draw from
this short discussion is that there is rather little scope for collective apologies. That
is to say, there is only as much scope for them as there is for attributions of collective
36 See J. Thompson, ‘The Apology Paradox’, The Philosophical Quarterly 50 (2000): 470–5. The
paradox is inaptly named because if I am right that only those who are causally and morally responsi-
ble for a crime can apologize for it, one cannot apologize for a crime committed before one’s concep-
tion. Thompson’s solution to the paradox is the one I describe and reject here. My own response
follows Saul Smilansky’s, whose paper on this issue helped me clarify my inchoate intuitions. See
S. Smilansky, ‘Morally, should we prefer never to have existed?’, Australasian Journal of Philosophy
91 (2013): 655–66. For the opposite view, see Wallace, The View from Here, esp. chs. 5–6.
280 Reconciliation
responsibility. Given that those attributions make less and less sense over time,
apologies ought to be replaced by expressions of regret. At time goes by, however,
those too lose their meaning as utterances coming from a set of actors relevantly
connected to those war wrongdoings—so much so that all that is left, as far as those
wrongdoings are concerned, is a condemnatory moral judgement issued at the
court of humankind.
9 . 7 . C O N C LU S I O N
I began this chapter by noting that the notion of reconciliation has become a cor-
nerstone of conflict studies. To be sure, peace—that state of affairs in which polit-
ical and communal actors in general, and erstwhile belligerents in particular, fulfil
their obligations to one another without threat of violence—does not always
require that their individual members actually engage in reconciliatory processes
with one another. This is particularly true of large-scale interstate conflicts such as
the two World Wars. In such cases, top-down institution-building such as the
construction of the European Union in the 1950s together with the passage of
time might and indeed often suffice. In civil conflict, however, or in transnational
conflicts involving close neighbouring political communities, a more proactive
approach to peace-building is often required—partly because in such cases, the
understandable grievances created both by the conflict itself and by the impossibil-
ity of a fully just peace are more likely to fan the embers of violence.
Accordingly, in this chapter, I have defended three reconciliatory practices as an
alternative to wholesale punishment and full-blown reparations and restitutions
when the latter ought not to be implemented: Rwanda’s gacaca courts, Truth and
Reconciliation Commissions, and official apologies and utterances of regret. None
of them are always necessary for reconciliation, though on the basis of empirical
evidence, they certainly help foster the conditions under which erstwhile belliger-
ents can learn to trust each other and work towards such a peace.
Those practices require, rely on, and produce testimonies, official utterances,
and public records. In this respect, they contribute to fostering over time a culture
of remembrance. If, as I claimed here, belligerents are (sometimes) under a duty to
foster mutual political reconciliation, the question naturally arises as to whether
they also are under a duty to foster remembrance practices. I shall now argue that
they sometimes are.
10
Remembrance
1 0 . 1 . I N T RO D U C T I O N
War remembrance is a particularly salient issue, of course: as we are marking the

centenary of the First World War, we—descendants of those who lived through
that war and fought and died in it—are repeatedly told that we ought to commem-
orate it and are given instructions on how we should do it. Above and beyond
episodic moments such as centenaries, war remembrance is central to the construc-
tion of political identities.1 When we erect monuments and crosses for combatants
who died in war, lay wreaths in their memory, put up commemorative plaques for
murdered civilians, turn battlefields into places to visit and build military muse-
ums, we typically mean to remember something about ‘us’, either as distinct from
‘them’, our erstwhile enemy, or as transcending communal divisions brought about
by war so as to be a new and better ‘we’.
Decisions about which conflict to commemorate and which to ignore, about
whom to commemorate (‘great’ generals? ordinary combatants? ordinary civil-
ians?), how, and why, are normatively laden. Consider the following two literary
extracts in which we are presented with diametrically opposed ethical stands on
war remembrance. As George Eliot would have it,
We see human heroism broken into units and say, this unit did little—might as well
not have been. But in this way we might break up a great army into units; in this way
we might break the sunlight into fragments, and think that this and the other might
be cheaply parted with. Let us rather raise a monument to the soldiers whose brave
hearts only kept the ranks unbroken, and met death—a monument to the faithful
who were not famous, and who are precious as the continuity of the sunbeams is pre-
cious, though some of them fall unseen and on barrenness.2
For Siegfried Sassoon, however, whose poem On Passing the Menin Gate is a classic
of First World War poetry,
1 The following collection of essays provides a fascinating study of the ways in which different
communities have used acts of remembrance as a way to express and buttress their identity: J. R. Gillis
(ed.), Commemorations—the Politics of National Identity (Princeton, N.J.: Princeton University Press,
1994). In the UK, the two-man silence which is routinely observed on 11 November at 11 a.m. has
become an important part of this country’s annual remembrance ritual. It was not always thus, though.
See J. Winter, Remembering War: The Great War between Memory and History in the Twentieth Century
(New Haven, Conn.: Yale University Press, 2006), 141–3; and A. Gregory, The Silence of Memory—
Armistice Day, 1919–1946 (Oxford: Berg, 1994).
2 G. Eliot, Felix Holt the Radical, ed. D. Thorold (London: Wordsworth, 1997 [1866]), 158.
282 Remembrance
Who will remember, passing through this Gate,
The unheroic Dead who fed the guns?
Who shall absolve the foulness of their fate,—
Those doomed, conscripted, unvictorious ones?
…
Was ever an immolation so belied
As these intolerably nameless names?
Well might the Dead who struggled in the slime
Rise and deride this sepulchre of crime.3
Whereas Eliot enjoins us to raise a monument, literally and metaphorically, to
ordinary combatants, Sassoon denounces that act as a futile and hypocritical
attempt to occlude the horrors of war. Where, then, should we stand?
Those practices have long attracted the attention of historians, sociologists, and
students of conflicts and peace processes. Contrastingly, political philosophers in
general, and just war ethicists in particular, have shown scant and indirect interest
in war remembrance. When they mention it, they do so obliquely. For example,
they discuss it as a condition for reparations for historical injustices (which imply
that one may or should remember wrongful past deeds). Or they construe it as a
means to balance the conflicting imperatives of punishing wrongdoers (which also
implies some form of remembrance) and reconciling communities divided by a
legacy of conflict (which requires some form of forgetting).4
Those arguments justify war remembrance by appealing to the special relation-
ship which binds together members of the same political community, or of politi-
cal communities which were once enemies and who must now learn to live
together. Moreover, it seems that it is precisely in so far as war remembrance is tied
to such special relationships that it makes an especially powerful normative demand
on us.
And therein lies the problem—at least, for a cosmopolitan account of peace
after war. For if cosmopolitans are right that national-cum-political borders do not
in fact have the moral status and importance which war remembrance confers on
them, what space is there for the claim that we have compelling moral reasons to
commemorate wars? It would seem that a cosmopolitan will have very little to say,
qua cosmopolitan, about war remembrance.
At the same time, however, the wars which we (at least in the West) perhaps
feel the most impelled to commemorate are the two World Wars and events
within wars which shock the conscience of humankind. Whereas standard justi-
fications for remembrance duties appeal to the moral features of the particular
relationships which we have with other individuals (whether alive or dead),
3 S. Sassoon, ‘On Passing the Menin Gate’, in The War Poems. ed. R. Hart-Davis (London: Faber
and Faber, 1983 [1928]). Copyright Siegfried Sassoon by kind permission of the Estate of George
Sassoon.
4 I have found only three book-length philosophical treatments of the normative issues raised by
remembrance and commemoration: J. Blustein, The Moral Demands of Memory (Cambridge:
Cambridge University Press, 2008); Blustein, Forgiveness and Remembrance; A. Margalit, The Ethics of
Memory (Cambridge, Mass.: Harvard University Press, 2004). For a brief discussion of reparative
remembrance, see Waldron, ‘Superseding Historic Injustice’.
Remembrance 283
many of our remembrance practices seem also to make sense, intuitively at least,
when addressed through the lenses of universally humanist and individualistic
moral principles.
Or so I shall argue. My aim in this chapter is to offer a justification for war
remembrance which transcends national and political borders and yet is appro-
priately sensitive to the specific historical and personal importance which the
remembered war has for those who commemorate it. In s.10.2, I explain how
I construe war remembrance. In s.10.3, I argue that, notwithstanding their
strengths, standard justifications are not sensitive enough to the interplay
between individual and collective memory, to the diverse fabric of the suppos-
edly homogenous community which engages in remembrance, and to important
moral features of war. In s.10.4, I offer a justification which avoids those pitfalls,
and defend war remembrance both as a way to honor the memory of war victims
and as a vehicle for discharging our general, relationship-independent duty to
bring about peace.
Before I begin, some clarificatory remarks are in order. First, although I have a
few things to say about remembrance in the immediate aftermath of conflict, I am
particularly interested in moral reasons for commemorating wars of the distant
past, whose participants are dead or will be dead before too long—in Remembrance
Sunday or the daily Last Post ceremony at the Menin Gate itself in 2015 rather
than in the mid-1920s. Accordingly, my aim is not to show that remembrance can
assist in reconciliation here and now.5
Second, as we saw in chs. 7 and 9, war crime tribunals and truth and reconcili-
ation commissions have an important role to play in recording the individual
memories which, together, constitute shared memories of violence. Although they
serve remembrance, they do not (or, at any rate, ought not) have it as their primary
aim. My focus in this chapter is on practices which are about remembrance itself,
and which I briefly mentioned at the outset: to name but a few, erecting memori-
als, holding commemorative ceremonies, building museums dedicated to war,
organizing a two-minute silence.
Third, when I write that individuals have compelling moral reasons to com-
memorate wars, I mean that they have compelling moral reasons to engage in those
shared practices. I do not mean to say (e.g.) that they have strong compelling
moral reasons to call into their mind what they know about this or that war once
in a while in the privacy of their home. In so far as the verb ‘to remember’ tends to
denote purely private acts, I refrain from using it and resort instead to the verb ‘to
commemorate’. By that token, to the extent that the practices I seek to justify here
5 To my knowledge, the best account of the role of remembrance for reconciliation is Blustein,
Forgiveness and Remembrance. Incidentally, I find it absolutely fascinating that the decision to establish
the Imperial War Museum in Britain was taken by the British Cabinet in February 1917, at a time
when the war not only was ongoing but was far from being won (the US, remember, only entered the
war in April of that year) and when, one might think, those leaders had more pressing things to worry
about than the way in which future generations would remember the war. I am deeply grateful to
Myfanwy Lloyd for this particular point and, more generally, for enriching discussions of the issues
raised in this chapter.
284 Remembrance
are both shared and public, the noun ‘commemoration’ might be a better fit than
‘remembrance’. In the United Kingdom however ‘remembrance’ is firmly associ-
ated with the commemoration of wars, and I therefore use those two nouns
interchangeably.
Fourth, so far in this book I have used the language of rights and duties. In this
chapter, however, I shall speak of compelling moral reasons (to commemorate
wars). Readers may wonder why. War remembrance, I shall argue below, may well
be one way in which we discharge our duty to bring about peace, and it is to that
extent that we ought to engage in it. At the same time, there might be other,
indeed better, routes to the same end. Moreover, it is not clear to me that the
beneficiaries of our remembrance practices always have a right that we should so
act, even when it is the case that we should. More generally put, the conclusions
I reach here with respect to war remembrance are less verdictive than the conclu-
sions I reached with respect to, e.g., military occupation or reparations. Yet the
language of rights and duties is particularly verdictive—more so than the term
‘compelling moral reasons’: I use the latter, therefore, to denote that war remem-
brance is something that we morally ought to do pending arguments to the con-
trary, and that absent such reasons failure so to act would warrant opprobrium.
That said, I shall on occasion point out that such reasons can sometimes be so
stringent as to ground a duty. Readers who hold that the language of rights and
duties is always appropriate in this particular context can easily replace ‘moral
reasons’ with ‘duties’ as they go along, as the arguments I criticize and those
I construe do not depend for their soundness on the adoption of this, or that,
vocabulary.
10 . 2 . M E M O RY A N D WA R R E M E M B R A N C E
Roughly put, a memory is a thought about a past event, a past utterance, or a per-
son from the past. In order to have a memory of some event E, I need not do any-
thing, whereas in order to remember E, I need actively to call E into my mind. Not
only, thus, is the act of remembering volitional: it also aims to track some feature
of the world as it was. If I claim to remember having a banana for breakfast, and if
it truly is the case that I do remember that fact, then it truly is the case that I had
a banana for breakfast. To the extent that war remembrance is built on memories
of past, it too is truth-tracking. And to that extent, it implies that there is such a
thing as historical knowledge. I shall not defend that assumption here. Let me
simply note that, as Hannah Arendt reminds us, we may well have divergent inter-
pretations about the facts of the past, but we cannot (on pain of jeopardizing the
very possibility of our survival as rational and moral agents) claim that those facts
actually happened when they did not, or that they did not happen when they did.
As she reports it, when former French Premier Clemenceau was asked some years
after the First World War by a German official how future generations would view
the question of who was responsible for the war, he is alleged to have replied, ‘This
I don’t know. But I know for certain that they will not say Belgium invaded
Remembrance 285
Germany’. To the extent that there is such a thing as historical knowledge, one can
speak of the truth-tracking value of remembrance.6
Not only can I remember E only if E happened: further, I cannot be described
as having a memory of E unless I experienced it: since I was not born in 1914, I do
not have a memory of the start of the First World War. However, I might be
described as remembering that the First World War started in 1914, where what
I mean is simply that I know that this is when it started. In ordinary language, one
can remember that E happened without having an experience and therefore a
memory of E happening. This distinction—between episodic and semantic mem-
ories—is crucial. Although I shall on occasion say something about recent wars
whose survivors are numerous in our community, I am mostly interested in
remembrance practices which are built on semantic memories.7
Further, my concern is with shared practices of remembrance. Those practices
are typically carried out by individuals acting on their own initiatives and in a
private capacity, as well as by a public official acting on behalf of the community
to which those individuals belong. This raises two issues. First, to argue as I shall
do presently that individuals qua members of a particular group have a strong
moral reason to engage in shared remembrance of a given war presupposes that
there is what some have called ‘a community of memory’ about that war.8 The
question, then, is who constitutes that community, and whose individual memo-
ries are pressed into the service of constructing and nurturing the community’s
shared memory. Second, to hold that those individuals qua group members have
reasons to do x is to imply that each of those individuals has a reason to act in such
a way that she, together with fellow group members, does x. The question, then, is
that of how those individuals should act. In the present context, the claim that
individuals qua group members have reasons to engage in war remembrance might
be read in the following ways: that they merely have reasons to support institu-
tional commemorative practices by not disrupting them, or by voting for parties
which promise to allocate budgetary resources to their conduct, and so on; or that
they have reasons to do all of that and in addition actively to engage in acts of
remembrance in a private capacity, for example by wearing a poppy and attending
commemorative ceremonies, or observing a two-minute silence at 11 a.m. on 11
November. Just as there are institutionalist and individualist approaches to justice,
there are institutionalist and individualist approaches to remembrance. Here, as
when discussing reconciliatory practices, I shall lean towards the individualist
approach.
Finally, war remembrance has emotional and moral valence: its constitutive
practices are usually meant to induce us to mourn those who died and, particularly
though not only in the case of combatants, feel gratitude for their sacrifice; some-
times they are also meant to induce us to condemn much of what was done in the
6 See H. Arendt, ‘Truth and Politics’, The New Yorker, 25 February 1967. I am grateful to Jeremy
Waldron for reminding me of that passage.
7 For the distinction between episodic and semantic memories, see, e.g., Blustein, Forgiveness and
Remembrance, 75.
8 Margalit, The Ethics of Memory.
286 Remembrance
prosecution of a war, or on the contrary to celebrate it. As we shall see, this moral
and emotional expectation is sometimes problematic. Nevertheless, when I speak
of remembrance, or use the phrase ‘to commemorate’, I shall mean the active
recalling to one’s mind of wars of the past through shared practices, accompanied
by the requisite emotional and moral attitude. Although ‘commemorate’ and
‘commemoration’ are usually endowed with positive valence (or so I am told by
many native English speakers)—such that one commemorates the end of the First
World War but not its beginning—I shall use those words in a value-neutral way.
1 0 . 3 . PA S T A N D P R E S E N T:
T WO P RO B L E M AT I C A RG U M E N T S
In this section, I examine two arguments for the view that we ought to remember
wars. The first holds that acts of remembrance are a way properly to acknowledge
what our ancestors did. The second argument claims that such acts are a means to
cement our political and cultural relationship with fellow community members,
which gives us a compelling moral reasons to perform them.
10.3.1. Giving the Past its Due

The claim that we have strong moral reasons to commemorate our predecessors’
wrongdoings is familiar. Less familiar (at least in the academic literature) is the
claim that we have strong moral reasons to honour and express gratitude to those
who benefited us which we can also discharge in this way. As we shall see, both
arguments are problematic.
Past Wrongdoings
Let us begin with the claim the descendants of war criminals have particularly
compelling reasons qua such actors to honor victims of those wrongdoings and
their descendants via acts of public remembrance—whether they are descendants
through kinship, political membership, or both. The example which is the most
often discussed in this context is that of the Holocaust, so we might as well use it
for argumentative purposes. On this view, contemporary Germans have strong
moral reasons—stringent enough in fact that they generate duties—to remember
the Holocaust, for even if they themselves are clearly not complicitous by deed or
omission in it, the crime was committed by their forebears understood as their
predecessors in the transgenerational and national community known as Germany.
The difficulty with the argument under scrutiny is that it relies on an overly
simplistic notion of a national community. Current German citizens are not all
descendants of the wrongdoers whose crimes are to be remembered, either because
their parents or grandparents immigrated to Germany after 1945, or because their
forebears were victims rather than wrongdoers. It is not clear, thus, why a sec-
ond-generation Turkish immigrant who was granted German citizenship, let alone
Remembrance 287
the granddaughter of a German Jew who died at Bergen Belsen, should have com-
pelling moral reasons to commemorate the Holocaust qua German citizens.
Nor does the fact alone that someone’s grandfather was an SS guard at Auschwitz
give that particular individual here and now stronger moral reasons to participate
and support Germany’s acts of shared remembrance, than an American citizen
whose grandfather did not take part in the genocide would have. As we saw in chs.
5 and 6, either responsibility for wrongdoing or benefiting from it are necessary for
reparative obligations. The same applies to moral reasons in respect of remem-
brance. This individual can plausibly be deemed to have compelling moral reasons
to support and take part in the remembrance of the Holocaust only under the fol-
lowing conditions: if his polity—Germany—here and now, is still engaged in acts
of wrongdoings towards victims of the Holocaust and if he himself can plausibly be
regarded as complicitous in this particular wrongdoing, or if Germany here and
now is still benefiting from the Holocaust (which would impose reparative obliga-
tions on him), and if private and public remembrance of the Holocaust is plausibly
construed as a way to rectify those ongoing wrongdoings. Under those conditions,
however, what grounds those reasons is the fact that as a citizen he commits a par-
ticular wrongdoing or that he derives an unjust benefit from the Holocaust—and
neither the fact alone that his grandfather served at Auschwitz nor the fact that he
is German. If the point holds with respect to the grandson of an SS guard, it also
applies, a fortiori, to the grandson of an ordinary German citizen who voted for the
Nazi party in December 1932 and failed to resist its rule in the twelve and a half
years that followed. Note, furthermore, that the point also holds of immigrants at
the time at which they arrive into Germany: if at that time German citizens are still
under collective reparative duties to victims of Nazism and if remembrance is one
way to discharge those duties, immigrants acquire a compelling moral reason to
contribute to it at the same time as they acquire rights to the benefits attendant on
residence. Likewise with those immigrants’ descendants.
Here is another argument in favour of this particular kind of remembrance.
Namely: if we want to take pride in the good aspects of our national culture and
its past, we have to accept that we ought to feel shame at its bad aspects. This in
turn provides us with a compelling moral reason to distance ourselves from those
bad aspects—a reason which is specific to us precisely because it is our culture and
we in some way identify with it. Engaging in appropriately conducted remem-
brance practices is one way in which we can do that. Moreover, even if we do not
feel shame at our culture’s morally troubling past any more than we feel pride at its
achievement, we still have a particularistic moral reason to distance ourselves from
it. For if we fail to do so, others, and particularly surviving victims of our predeces-
sors’ wrongdoings and their descendants may reasonably presume that we do in
fact endorse that past. To the extent that they might feel threatened or disrespected
by our silence, it behoves on us to reassure them—for example, by engaging in the
aforementioned practices.9
9 I owe this objection to Patrick Tomlin, whose comments on this section, together with David
Miller’s, were of great help. We encountered the argument that agents have strong moral reasons to
288 Remembrance
This is a powerful argument. Its strength lies in the surely plausible thought
that if my behaviour understandably leads others to believe that I endorse (e.g.)
racist views, then I ought to reassure those who would feel victimized by those
views. Moreover, the argument has considerable bite if the enactment of remem-
brance itself would shortly follow the commission of the wrongdoings under
remembrance. However, the argument does not work if I do not act in such a
way as to lead others warrantedly to form those beliefs. If I plant the Confederate
flag in my front garden, my neighbours and passers-by can reasonably form the
belief that I do not regard slavery as egregiously wrongful to the point that its
abolition is a just cause for war. But if I choose to live in Mississippi and not do
anything at all to signify any kind of allegiance to the Confederacy, it would be
unreasonable on their part to impute to me feelings of leniency towards slavery.
Treating people with the respect they deserve does involve not thinking the
worse about them absent relevant evidence to the contrary. Were I to encounter
a German national who would express pride at the fact that Germany (or, rather,
its predecessor state) gave J. S. Bach, Beethoven, and Schiller to the world, my
first thought would be that it is rather misplaced pride: after all, his connection
to those geniuses, and theirs to him, are based on the utterly contingent fact of
their birth. But I would not assume that he might be sympathetic to Hitler. The
more time has elapsed, the less plausible, indeed the more disrespectful it is to
presume of others that they endorse their community’s murky past.
In sum, the notions of collective shame and collective identification strike me
as problematic bases for moral reasons in support of remembrance. The under-
lying rationale for rejecting them is the humanist and individualist point that
individuals are not liable to the imposition of reparative duties just in virtue of
their membership in this or that community, but only in virtue either of what
they did, or of the benefits they accrued from unjust acts. To those who believe
that political borders and political membership alone are irrelevant to the con-
ferral of fundamental rights and the imposition of correlative duties, that point
will seem obvious. But it does pay to notice what it implies in the present con-
text—namely that the grandson of an Auschwitz guard does not have a more
compelling moral reason to support and take part in the public remembrance of
the Holocaust than does the granddaughter of the Jewish female detainee whom
his grandfather murdered. To be sure, those two individuals’ emotional atti-
tudes to the Holocaust are likely to differ. In particular, the former might feel
shame at what his grandfather did, so deeply rooted is the view, at least in our
culture, that our parents’ and grandparents’ deeds somehow taint us by associa-
tion. But even if those feelings are not irrational, it is a far cry from concluding
that this individual has particularly compelling moral reasons for remembrance
in virtue of his kinship with his grandfather the SS guard. It is that conclusion
which I reject.
distance themselves from wrongdoings committed by their contemporaries or immediate forebears in

s.9.6 when discussing apologies and regret as reconciliatory practices.
Remembrance 289
Past Benefactors, Present Benefits

So far, I have focused on remembering past war wrongdoings. Often, however, we
remember wars and honour their dead in gratitude for the latter’s sacrifice.
Presumably, the argument ought to extend to all those who took part in the war
and who thereby benefited us—not just combatants, but also civilians who con-
tributed to the war effort at some cost to them. But there are at least two difficulties
with that argument. First, the longer time lapses, the less recognizable a polity is as
that polity by which and on whose behalf the war was fought. Britain has changed
considerably since 1918—culturally, politically, demographically. Thus, in just the
same way as it is not clear why a young German of Turkish descent should, qua
German, remember the Holocaust in sorrow, it is not clear why a young Briton of
Pakistani descent should, qua Briton, remember the British dead of the First World
War in gratitude.
An obvious rejoinder to this is that Britons today would not enjoy the benefits
of living in a peaceful, liberal, democratic society were it not for the fact that
Britain went to war against Germany in 1914. But there is an equally obvious
difficulty with this rejoinder: it relies on the possibility of making counterfactual
claims about the past, which the more distant that past is the more dubious those
claims are. In this case, we simply do not know what would have happened had
Britain not regarded the violation of Belgium’s neutrality by Germany in 1914 as
its ‘red line’. To be sure, it could be that Britons today in fact benefit from the First
World War, which if true would yield the conclusion that they have strong, objec-
tive compelling moral reasons to be grateful to the combatants who fought for
Britain. But given that we do not know for sure, that we have little evidence either
way, and that sustaining and especially financing remembrance practices is both
burdensome to those who engage in them and costly to those who would benefit
from the resources we could instead devote to them, considerations of gratitude
towards the dead seem too weak to generate compelling moral reasons so to act
vis-à-vis wars of the increasingly distant past.
Second, benefiting today from our predecessors’ war-related deeds and sacri-
fices is not enough to generate compelling reasons to experience and express
one’s gratitude.10 For if it were enough, we would have compelling moral reasons
to remember combatants in gratitude for their having died in defence of our
community even if that defensive war was in fact unjust—to be clear, even if they
killed enemy combatants and civilians unjustly. But this cannot be so. Killing is
one of the worst things human beings do to one another, and wrongful killing is
thus one of the most grievous wrongdoings they commit (only torturing some-
one to death, I think, surpasses it). If we are to commit ourselves, as we should,
to respecting the nearly absolute moral imperative ‘do not deliberately kill
another unwarrantedly’, then it is hard to see on what grounds we not only may,
but more strongly still, ought to feel grateful for the fact that we benefited from
10 I am particularly grateful to David Birks and Patrick Tomlin for pressing me hard on the issues
raised in this paragraph.
290 Remembrance
systematic breaches thereof. The point is particularly salient in the case of

the First World War. Suppose for the sake of argument that British leaders were
justified in believing that the neutrality of Belgium, the importance of the
Anglo-French Alliance, and the imperative of containing German expansionist
ambitions in sub-Saharan Africa were just causes for war. Even so, the scale and
manner of the destruction suggest that the war itself was fought unjustly, and in
fact that continuing with the war instead of suing for peace was unjust—to be
clear, that continuing to kill enemy combatants was unjust. I see very little there
to be grateful about. By the same token, incidentally, in some cases we ought to
remember publicly those who refused to fight an unjust war—as is the case in
Germany where war memorials have been erected in memory and honour of
Second World War deserters.
But perhaps this is too quick. Perhaps they are reasons for thinking otherwise.
I can think of two. First, one might hold that, in so far as we and those whom we
love would not exist were it not for the fact that combatants fought those ex hypoth-
esi unjust wars, we do have a pro tanto reason to be grateful. As we saw in s.9.6,
however, we do have overwhelming reasons to regret the occurrence of those griev-
ously unjust deeds thanks to which we and our loved ones now exist. By implica-
tion, we have no such reason to feel grateful to the authors of those deeds. Moreover,
if we all of us together, as fellow members of a community who would not exist but
for those wars, ought to express our gratitude through acts of remembrance, we
ought to do so at least as much vis-à-vis unjust belligerents who started those wars
as vis-à-vis those on our side. This might not be a conclusion which proponents of
the argument under consideration would welcome.
Second, one might hold that combatants were not aware that they were killing
wrongfully and that they incurred considerable physical, psychological, and moral
burdens for our sake, for which we do have compelling reasons to be grateful even
if they in fact acted wrongly. However, whether we ought to experience and express
gratitude to them for what they did, irrespective of the moral features of their war
is doubtful, even if we benefited from it.
Admittedly, perhaps we should be grateful for the fact that they meant to benefit
us, their successors, by fighting that war. To be grateful for that, however, is not the
same as to be grateful for what they did. Analogously, suppose that my mother
gives me a medicine which she thinks will cure me but in fact makes me very ill.
I can be grateful to her for wanting to help me without being grateful to her for
giving me the medicine. Does this move rescue the gratitude argument for remem-
brance? Only partially so. For even if we ought to be grateful for what agents
intended to do and not what they actually did, the gratitude argument would
support only the commemoration of those combatants who did intend to benefit
us (assuming that we can ascertain their motives). It would not support the com-
memoration of those who did not so intend. Nor, finally, would it support the
commemoration of those wars’ victims. In fact, there is a risk that those victims
will be deliberately occluded from our remembrance practices, so focused as we are
to honor our dead. The (architecturally beautiful) Vietnam Veterans Memorial in
Washington perfectly illustrates the point: it honours members of the American
Remembrance 291
Armed Forces who fought in a war that was by all accounts profoundly unjust
whatever they themselves believed at the time: there is not a single mention of their
Vietnamese victims.11
The foregoing remarks suggest that when considering whether to remember
combatants in gratitude, we need to distinguish those who justly fought on our
behalf from those who did so unjustly; we also need to distinguish combatants
who rightly refused to fight at the cost of their life and in so doing kept a moral
spark alive in the midst of horror, from those who continued to fight, also at the
cost of their lives, in furtherance of injustice. It seems that we have compelling
moral reasons to remember the former two, but not the latter, in each pair. Given
the enormous suffering which so many endured, this might seem unacceptably
harsh. But my point is not that we have no compelling moral reasons to remember
unjust combatants—at least, those ordinary combatants who were conscripted
into unjust wars and chose to fight instead of being condemned to death for refus-
ing to face the enemy. In fact, I shall offer some such reasons in s.10.4, and argue
that the requisite emotional attitude when remembering them is that of at the very
least profound regret, indeed sorrow. My point here is that gratitude is not the right
kind of reason in those cases.
To conclude, gratitude for past deeds only provides a limited (and on some
construals problematic) justification for commemorating wars and their actors—as
do appeals to past injustice. As I shall now show, appeals to the importance of
cementing our present bonds with community members are also problematic.
10.3.2. Strengthening Community Bonds in the Present

In separate works, Jeffrey Blustein and Avishai Margalit both give the following
as compelling—in fact, particularly stringent—moral reasons for remembrance
in general, which I apply here to the specific case of war.12 It is indisputable that
a community’s culture, whether that community is a family, a cultural or ethnic
group within a state, or a nation state, is shaped over time not just by historical
events but by the fact that, and the multifaceted ways in which, those events are
explicitly remembered either through well-orchestrated ceremonies or through
the production of mnemonic artefacts (monuments, plaques, crosses, museums,
and so on). In publicly remembering those events, we acknowledge that the
bond that unites us to our fellow members is rooted in the history which we
share with them—in a way not dissimilar to the interpersonal bonds of kinship
which are weaved and strengthened over time. In just the same way as valuing
our kin and the transgenerational family which they constitute requires remem-
bering what previous, now-dead, family members did that made the family
what it is, so does valuing our relationship to fellow nationals. To illustrate:
11 For an interesting discussion of the symbolism of this memorial, see Griswold, Forgiveness, ch. 5.
12 Blustein, The Moral Demands of Memory, 222–8; Margalit, The Ethics of Memory, 104–5;
Blustein, Forgiveness and Remembrance, 261–78. For an illuminating account of the community bonds
on which this kind of argument rests, see T. Hurka, ‘The Justification of National Partiality’, in
R. McKim and J. McMahan (ed.), The Morality of Nationalism (Oxford: Oxford University Press, 1997).
292 Remembrance
assume that Britain here and now would not exist in its present form but for
what First World War combatants endured; accept that British citizens here and
now value the community to which they together belong and wish to ensure
that it endures; suppose that it will not endure but for remembering those par-
ticular contributions. Taken together, those points support the claim that the
British have strong moral reasons to remember what First World War combat-
ants did for them.
However, the argument strikes me as problematic for several reasons. For a start,
and to reiterate a point made in the previous section, it works only to the extent
that the communal relationships which are sustained through acts of remembrance
are morally valuable. If those relationships are founded in and perpetuate morally
abhorrent ideals, in so far as there can be no moral reason to sustain them, there
can be no moral reason to perform those acts. Blustein accepts the point but it pays
to note how restrictive it is of remembrance practices—simply because the over-
whelming majority of political communities were for a long time organized around
abhorrent ideals, and in fact (for the most) continue to be so. To give but two
examples, the year 1919 marks the beginning of decades of remembrance of the
Great War, as well as the treaty-based carving up by the victorious belligerents of
their enemies’ former colonial empires—over which the war was in part fought.
Colonialism, whose racist underpinnings are unquestionable, was a dominant ide-
ology in both France and Britain until the Second World War. In a similar vein, the
Soviet authorities engaged in yearly commemorative exercises in honour of the
combatants who died fighting the Nazis—culminating in 1965 with the erection
of the Tomb of the Unknown Soldier in Moscow; at the same time they used the
memories, both individual, common, and shared, of the Great Patriotic War as a
means to cement one of the most grotesquely abusive regimes there ever was. Once
again, if one thinks that there nevertheless are reasons to commemorate those wars,
one will have to look elsewere.13
Moreover, appeals to the notion of a national transgenerational community are
doubly exclusionary. On the one hand, they only give us compelling moral reasons
to remember our communal predecessors’ contributions to who we are, here and
now. As a dual French and British citizen, I only have reasons to remember, and
perhaps even mourn, French and British First World War veterans, and not their
German and Austrian counterparts. Of course, perhaps it matters, not that I
should be French and British but, rather, that I should be a citizen of two countries
which together with Germany and Austria now form the European Union. But if
this is what matters, then as a French citizen, it is not clear whether it is more
important that I should attend and take part in my native country’s national
remembrance of the war—which remembrance pits the French against the
German—or if I should on the contrary identify with President Mitterrand’s and
13 For the extent to which conflicts over colonial possessions (notably Morocco) heightened
pan-European tensions in the 1900s and were a contributing factor to the outbreak of the war, see,
e.g., Stevenson, 1914–1918, 18ff; H. Strachan, The First World War, vol. 1—To Arms (Oxford: Oxford
University Press, 2001), 18–35. On Soviet commemorations, see C. Merridale, Night of Stone—Death
and Memory in Twentieth Century Russia (London: Penguin Books, 2000).
Remembrance 293
Chancellor Kohl’s joint act of remembrance on the battlefield of Verdun in 1984.14

It is of course open to Margalit and Blustein to stress that I should do both—
honour the dead of Verdun qua French national in recognition of the battle’s
importance in contemporary French collective memory, and honour them with
my German contemporaries in recognition of the special relationship, which is at
the heart of the European Union, between our two countries. But this of course
would leave out—problematically so in the light of their sacrifice—American and
Russian veterans, whose descendants are not part of the EU. To justify remember-
ing those combatants, one would have to look to the gratitude argument, suitably
constrained by considerations about the moral status of the war.
On the other hand, appealing to the transgenerational community does not
properly cater for the fact that such a community may well be made of a plurality
of groups not all of which can be expected to identify with the dominant historical
narrative. The difficulty is compounded by the fact that the events which we have
supposedly compelling moral reasons to remember take on the function of a
myth—such that remembering those events is valuable not because and to the
extent that the practice of remembrance tracks some truth about the past, but
because it contributes, as myths do, to sustaining the community here and how.
Myths, however, are inevitably distorting, as are the shared memories which take
on this function. Unfortunately, this particular kind of distortion—which takes
the form of failures to mention truthful facts about the past or, worse, of outright
lies about it—is often crucial to strengthening the dominant community of mem-
ory within the nation, at the cost of further marginalizing minority communities
of memory which do not and cannot be expected to share in the myth-like narra-
tive of remembrance.
Again, Blustein acknowledges the point—who could not?—and stresses that
remembrance can strengthen marginalized groups’ demands for recognition.
Further, he insists that the truth-tracking function of remembrance should best be
reserved for occasions when we engage in remembrance practices not as members
of particularized groups such as the nation state, but of humanity as a whole.15
I shall return to the latter point in s.10.4. With respect to the former, my worry is
twofold. First, there is a tension between the mnemonic demands of minority
recognition, which more often than not require that unpalatable truths be told
about the past, and the mnemonic demands of patriotism, which more often than
not require that those truths be altogether forgotten. For example, consider the
battle of Gallipoli of 1915, in which the Allies incurred dozens of thousands of
casualties yet failed to take the Gallipoli Peninsula from the Turks. Of those casu-
alties, over ten thousand belonged to the Australian and New Zealand Army Corps
14 See ‘Mitterrand and Kohl honor Dead of Verdun’, New York Times. http://www.nytimes.
com/1984/09/23/world/mitterrand-and-kohl-honor-dead-of-verdun.html (4/10/2013).
15 Blustein, The Moral Demands of Memory, ch. 4. At the same time, he notes that for a community
to take responsibility for its collective past is a moral imperative in so far as its current identity is
shaped by failures to recognize the wrongs which it committed in the past. Blustein, The Moral
Demands of Memory, 144–58. In so far as taking responsibility for one’s collective past requires telling
and acknowledging the truth about the past, there is a tension in his overall argument.
294 Remembrance
(ANZAC). It is hard not to regard Gallipoli as a complete military and political

disaster. Yet, it has become the crucible in which the Australian nation was
formed—a byword for courage and heroism, an event which, to this day, is key to
all Australian war commemorations. Many Australians believe themselves to be
under stringent moral reasons to commemorate Gallipoli—on so-called Anzac
Day—as a means to help sustain a transgenerational community which had begun
to take political shape in the years preceding the war and which continues to this
day. Yet, focusing on Gallipoli risks occluding the wars which were waged by set-
tlers and the then-authorities against the Aborigines and which resulted in the
destruction of those communities as self-sustaining organizations. In a similar
vein, monuments to the dead of the (American) Civil War extol Southerners’ fight
for states’ sovereignty and praise Northerners’ defence of the Union—and in so
doing link the erstwhile enemies into a common celebration of the relatively
unproblematic notion of sovereignty; but at the same time, they erase the central
and morally egregious issue of the war, namely slavery, from its commemorative
architecture. The question, then, is which of those conflicting demands should be
honoured, and thus which remembrance practices we have the most compelling
reason to embrace. The justification under scrutiny cannot on its own answer this
particular question.16
Second, to say that we, citizens of a particular polity, have compelling moral
reasons as members of this polity to remember a particular war on the grounds that
by remembering our combatants who sacrificed their lives for us we strengthen the
bond which unites us to one another now, is to suppose that there is a ‘we’ which
transcends deep differences between them, and us. However, the further back one
goes in time, the less current generations have in common with their predeces-
sors—so much so that it is not clear at all that there is such a ‘we’. The analogy with
friendship and other interpersonal relationships, which is often meant to provide
support for this justification for remembrance duties, will not work here, precisely
because we do not stand in the same interpersonal relationship with our fellow
citizens, particularly over decades, as we do with our friends and kin. To be sure, in
the immediate aftermath of a war, kinship mediates our political relationship with
our fellow citizens. As Jay Winter argues in his fascinating accounts of the remem-
brance of the First World War, in the 1920s and 1930s mourning combatant citi-
zens who had died for the homeland was a way to grieve for one’s lost son, father,
or husband.17 But when there are no longer any surviving participants in that war
16 See K. Savage, ‘The Politics of Memory: Black Emancipation and the Civil War Monument’, in
J. R. Gillis (ed.), Commemorations—the Politics of National Identity (Princeton, N.J.: Princeton
University Press, 1994). The point I am making about Australia also applies to the 4th of July and
commemoration of the War of Independence. There is no federal day of remembrance for the wars
waged against Native Americans. In Australia, the week from the first to the second Sunday of July
starts with National Aborigines and Islanders Day and celebrates the culture and achievements of
Aboriginal and Torres Straits Islander people—which is a way (albeit indirectly) to remind Australians
of what their predecessors did to them. However, whereas Anzac Day is a national public holiday,
National Aborigines and Islanders Day is not.
17 See Winter, Remembering War, ch. 6; J. Winter, Sites of Memory, Sites of Mourning: The Great War
in European Cultural History (Cambridge: Cambridge University Press, 1995).
Remembrance 295
(as is the case with the First World War, and as soon will be the case with the
Second World War), and in fact when there no longer is anyone alive who has had
a meaningful relationship (as parent, daughter, spouse, sibling, friend, grandchild)
with participants, the bonds of co-citizenship which unites the current generation
of ‘rememberers’ to the war’s participants, witnesses, and victims are no longer
mediated by personal attachments. What moral reasons, then, do we have to
remember?
Finally, Blustein’s and Margalit’s defence of remembrance presupposes that there
is a duty to be patriotic, at least to the extent that our relationship to our compa-
triots qua compatriots is intrinsically morally valuable. But many are sceptical that
there is such a duty—in fact, it is a central assumption of this book that there is
not. If that is right, the claim that we have compelling moral reasons to remember
war, and particularly wars of the distant past, cannot be justified primarily by
invoking the intrinsic goodness of those relationships.
1 0 . 4 . R E M E M B R A N C E A N D T H E D U T Y
TO P RO M OT E P E A C E
10.4.1. Justifying Remembrance

Let us take stock. Appealing to past wrongdoings as a moral reason for war remem-
brance will only work for those cases where we, here and now, either are morally
responsible for the unjust war or atrocities at issue, or are still benefiting from it.
The gratitude argument will only work for those cases—of which there are very
few—where gratitude is in order. Appeals to cementing current communal rela-
tionships will only work if those relationships are morally acceptable. All three
leave many remembrance practices without defence. Moreover, assuming that
those arguments are in fact sound, they all appeal to particularistic morality—or at
any rate to the morality of special relationships—and thus do not deliver a cosmo-
politan case for war remembrance. At this juncture, cosmopolitans have two
options. Either they simply reject the intuition that war remembrance is warranted
whenever the particularistic arguments leave it undefended, and accept that they
have nothing to say qua cosmopolitans about it; or they attempt to construe a
cosmopolitan argument for it which covers both cases which particularistic argu-
ments can accept and cases which particularistic arguments ought to reject. In the
remainder of this chapter, I go down the second route.
There is an important lesson to be learnt from the foregoing discussion, namely
that moral reasons for war remembrance must be sensitive to the following consid-
erations: (a) the moral features of the relationships which are nurtured by acts of
remembrance; (b) the culturally and historically diverse features of putatively
national communities of memory; (c) the moral features of the wars under remem-
brance. With those points in mind, we are in a position to construct a three-
pronged argument in support of war remembrance. The first prong appeals to the
inherent value of all human beings wherever they are; the second prong appeals to
296 Remembrance
instrumentally particularistic means for promoting peace, while the third prong is
thoroughly universal.
In chs. 7 and 9, I offered a gloss on John Donne’s assertion that the death of any
human being diminishes us all, in the sense that it is of concern to all of us irre-
spective of borders. This is why we all have punitive jurisdiction (via appropriate
institutions) over the commission of crimes of war; this is also why we all are under
a duty to support reconciliation processes wherever they occur. Valuing all human
beings wherever they are, and in particular attending to the plight of those who
suffer from and through war, provides us with a reason to ensure as much as we can
that they are not forgotten. Not doing anything publicly to remember them is a
failure to acknowledge the seriousness of the wrongdoings to which they were
subject and, thereby, a failure of respect.18 Their unique and specific value partly
explains why (I think) it is important to remember them by name if we can, as we
do on memorials to combatants. The Australian War Memorial nicely illustrates
the point: every day, at 5 p.m., its staff conduct a public Last Post ceremony and
tell the story of one of the thousands of Australian dead combatants whose name
figures on the wall. Above and beyond the politically and culturally specific details
of those biographies, those daily ceremonials do have universal moral valence. Of
course, the more time passes, the less it seems to matter in respect of those individ-
uals each taken on their own that we should seek to preserve a record of what they
did and/or went through—if only because more recent victims are likely to make
greater demands on our commemorative resources. If we can do so though, it
seems that we do have compelling moral reasons to honour their memory.
To this victim-centered argument for war remembrance, a cosmopolitan can
add two different kinds of considerations: the role of remembrance in fostering
peace-conducive political relationships on the one hand, and the universal moral
significance of the wars under commemoration on the other hand.
Take the first consideration. As we saw throughout s.10.3, individuals who are
called upon to remember a past war in which they did not take part often have
little in common with the combatants and civilians who are remembered through
acts of shared remembrance. There are many reasons for that: they are relatively
recent immigrants, or their direct ancestors acted or were treated in the war in ways
which markedly differ from the dominant, quasi-official narratives (viz., the
deserter v. the heroic combatant who held the rank), or the cultural, social, and
political fabric of the community has changed to such an extent that they here and
now are closer to their contemporaries from different political communities than
to their political, indeed familial, ancestors. Across those differences, however,
those individuals have the following crucial features in common: they all have
the capacity for rational and moral agency; they all are under a duty to avoid
18 Blustein makes a similar point. But he does not couch it in cosmopolitan terms. In fact, his
concern is with societies which need to come to terms with internal conflicts, and with remembrance
practices within those societies. What he calls international duties of remembrance are duties to assist
societies in transition with their own remembrance practices—on which point I obviously agree.
Likewise, I am in agreement with much of what he says on the symbolic and moral value of remem-
brance in general. See Blustein, Forgiveness and Remembrance, chs. 5–7.
Remembrance 297
unwarrantedly killing, torturing, raping, or harming one another; they all are
under duties to try and ensure as far as possible that those acts not be carried out
on their behalf. Against the background of their shared humanity and its concom-
itant duties, their relationship to their compatriots can help them do two things:
meaningfully affirm their commitment to the moral imperative not to participate
in unjust wars on the one hand, and act by that imperative on the other hand. And
it does so in so far as their communal values are consonant with or instantiate that
imperative. (‘We, the British, Americans, etc., do not do that.’) If commemorating
those wars the legacy of which is an enduring part of their political culture enables
members of a given political community more strongly to identify with one
another qua fellow members, and if they are thereby better equipped to affirm and
comply with that imperative, then they have compelling moral reasons to do so.
The point applies to all of those who, one way or another, take part in and are liable
to the burdens and privileges of the political relationship which this political cul-
ture underpins: it does not just apply to those whose ancestors were members
thereof at the time at which the war in question took place. To the extent that a
German citizen of Turkish descent whose parents immigrated to Germany in 1970
has a reason to engage in and support public remembrances of the Holocaust, he
has it not qua German who, it is mistakenly said, ought to experience collective
shame, but rather as a participant in a political relationship (that which unites
contemporary residents of Germany) the sustenance of which enables all Germans
to fulfil their general obligations to bring about peace for all.
On the view I have just defended, remembrance is justified both by appeal to its
expressive dimension and by the role it might play in deterring us from prosecut-
ing unjust wars. I take the first point to be relatively uncontroversial: a true, genu-
ine commitment to human rights and, more deeply, to the inherent value of all
human beings, is not to be affirmed merely in foro interno: it is to be affirmed
openly, explicitly, and particularly when it is under threat.19
On the second point, however, establishing a causal relationship between war
remembrance and the promotion of peace is epistemically arduous. For example, it
is not clear that yearly Remembrance Sunday ceremonies, let alone a one-off cente-
nary, helps us ensure that Britain will not wage unjust wars (or, at least, not do so as
often as it otherwise would). That said, there might be a chance that properly con-
ducted commemorative acts of that kind help maintain this country’s social fabric
and in so doing improve prospects for a durable peace. More plausibly, they might
well contribute to consolidating the view, which does seem deeply rooted in British
public culture, that we ought not to go to war without being able to at least make a
strong moral case for it. What we can say, thus, is that if the remembrance of wars
which play a particular part in the construction of our political culture enables us to
abide by the universal moral imperative to bring about peace, then we have strong
moral reasons to engage in and support the relevant practices.
Admittedly, some might object that this case for war remembrance is too weak
to generate moral reasons to engage in potentially expensive commemorative
19 Blustein, The Moral Demands of Memory, esp. 35–7 and 266–8.

298 Remembrance
ractices. The budget for the First World War centenary in Britain alone totals £50
p
million, and we might be better inspired to allocate this sum to help rebuild war-
torn communities (or so it might be claimed).20 I myself raised that worry against
the gratitude argument. Still, the cases are importantly different. There, my point
was that gratitude to dead combatants could not bear the weight of war remem-
brance. Here, my point is that we owe it to our contemporaries and successors to
strive for peace, failing which they might die. Granted, if war remembrance does
not in fact lead to peace, we might be morally compelled to divert the resources it
normally commands towards practices and institutions which do. But because the
beneficiaries of our remembrance practices are not already dead and might be
exposed to the horrors of war as a result of our commemorative failures, the justifi-
catory threshold for abandoning those practices is higher at the bar of my cosmo-
politan argument than it is at the bar of the gratitude argument.
Pending further objections, then, the case for war remembrance based on
the importance of fostering peace-promoting political relationships stands.
Furthermore, it is immune from the main criticism I levelled in s.10.3.2 against
the view that remembrance sustains our national transgenerational community.
There I argued that appealing to such a notion excludes both outsiders and insid-
ers who do not share in the dominant discourse. The case for remembrance I have
just sketched out does not have this weakness, for two reasons. It implies that
(e.g.) the British here and now have compelling moral reasons to commemorate
not just what their communal ancestors did or suffered at the hands of German
combatants during the First World War but also what British combatants suffered
at the hands of their own civilian and military leaders, and what the Germans
suffered at the hands both of the British and of their own leaders. It also implies
that the British, here and now, have reasons to remember how colonial troops
under British command were treated during the war and how colonial popula-
tions under British administration were treated in the post-war settlement, many
of whose descendants are now living in Britain as marginalized second- or
third-generations immigrants.
The reason in turn why my argument has this two-pronged implication is this.
The British have compelling moral reasons to commemorate the First World War
vis-à-vis one another not qua British citizens, but as a means to bring about
peace for all: this implies that the particular relationship which unites the British
to their fellow residents must be constructed in a way that does not pit some of
its participants—‘us’, the British of British descent—against ‘them’, namely their
erstwhile enemies (in this instance, the Germans) or their erstwhile subordinates
(formerly colonial populations). The same considerations apply to French and
German commemorations of that war. Interestingly, the last two months of
2014 saw two very different ways of remembering it. In London, two artists
20 See N. Clark, ‘Communities Secretary Eric Pickles says 2014’s £50m World War I commemo-
rations must not turn into “anti-German festival”’, The Independent. http://www.independent.co.uk/
arts-entertainment/art/news/communities-secretary-eric-pickles-says-2014s-50m-world-war-i-
commemorations-must-not-turn-into-antigerman-festival-8652405.html (accessed on 16/01/2016).
Remembrance 299
filled the Tower of London’s moat with 888,246 ceramic red poppies, each rep-
resenting a British military fatality of that war. By contrast, on 11 November, the
French President, British Prime Minister, and German Chancellor gathered in
Arras to inaugurate an International Memorial—or ‘Ring of Remembrance’—to
all military personnel who died in Artois and South Flanders in the Great War,
irrespective of nationality and rank, and listed in alphabetical order. The Tower
of London display is at odds with cosmopolitan remembrance, while the Ring of
Remembrance best instantiates it.
So far, I have sought to justify war remembrance by appealing to the instrumental
value, for our ability to affirm a commitment to fundamental human rights and to
bring about universal peace, of particular political relationships. In so doing, I
focused on the remembrance of those wars whose legacy still endures in the political
culture which underpins those relationships. However, if the ideal of peace ulti-
mately provides justificatory force to calls for war remembrance, it also gives us
compelling moral reasons to remember those wars not in so far as they have that
cultural historical significance for us as members of a given political community,
but in so far as they have the deepest and most universal moral significance to us qua
human beings. By extension, we have similarly strong reasons, qua human beings,
to remember those wars which have universal moral significance, and to which we
are not connected as participants or descendants thereof.
Let me elaborate. Uncontroversially, we all are under a duty to affirm cosmo-
politan commitments; we are also under a duty to ensure as far as we can that
unjust wars in which millions of people die should not take place, that no one
should be a victim of genocide, that atrocities which combatants commit against
civilians should be punished. Moreover, we all have a stake in bringing about
peace, not just in the obvious sense that we each benefit from a peaceful world,
but in the sense that there always is a risk, if not that we will ourselves be victims
or perpetrators of grievous war violence, at least that those deeds will be ordered
and committed on our behalf. George Santayana famously claims that ‘those
who cannot remember the past are condemned to repeat it’ and Edmund Burke
no less famously writes that ‘people will not look forward to posterity, who never
look backwards to their ancestors’. Both points buttress the view that fears about
repeating our predecessors’ mistakes—and in particular their moral mistakes—
provide a powerful justification for remembrance duties.21 If commemorating
the First World War serves as a reminder of the horrors of war and helps us to
keep in view the moral requirement not to senselessly sacrifice lives, then of
course we should do so—not as French, British, or indeed German citizens, but
as human beings.
This argument provides strong reasons vis-à-vis our contemporaries and our
successors to remember the two global wars taken in toto, as they and their legacy
transcend national borders. But it also justifies the remembrance of specific events
21 E. Burke, Reflections on the Revolution in France. ed. C. O’ Brien (Harmondsworth: Penguin
Classics, 1986 [1790]); G. Santayana, The Life of Reason (Gutenberg Project, 1905), at http://www.
gutenberg.org/files/15000/15000-h/15000-htm (accessed on 16/ 01/2016).
300 Remembrance
within those wars as well as non-global particular wars, in so far as they display the
very worst and the very best of which human beings are capable towards one
another qua human beings. We all have strong moral reasons, qua human beings,
to support and take part in the remembrance of genocides, wherever and against
whomever they have been committed. We all have strong moral reasons, qua
human beings, to support the remembrance of atrocities against civilians, wherever
and against whomever they have been committed. We all have strong moral rea-
sons, qua human beings, to support the remembrance of those combatants whose
lives were squandered in vain, precisely because the wrongdoing which was com-
mitted against them is one that they suffered as human beings: a French or a
British combatant were not differentially brutalized, disrespected, shamed, and
humiliated by trench warfare than were German combatants. Finally, and by the
same token, we have strong moral reasons qua human beings to commemorate the
acts of (justified) heroism, kindness, and compassion which combatants and civil-
ians carried out in wars, precisely in so far as those acts displayed the best of which
we are capable as human beings.
This argument is less vulnerable to the passing of time than backward-looking
and community-strengthening justifications for war remembrance. In s.10.3.1
I noted that those arguments tend to occlude the fact that, the more distant in the
past the war under remembrance has occurred, the less plausible it is to postulate a
homogeneous community of ‘rememberers’. On my account, however, we may
well have strong moral reasons to remember wars of the distant past, if remember-
ing those particular wars does (for whatever reason) enable us to discharge our
general obligation to promote peace. To the question, ‘will our successors have
strong moral reasons to remember the Holocaust two hundred years from now?’,
the answer might well be ‘yes’, if remembering the Holocaust is one way to affirm
rejection of its evilness and to minimize the risk that they will allow yet another
genocide to be carried out.
Before I rehearse objections to my cosmopolitan arguments for war remem-
brance, I should like to address two further issues. First, I have argued throughout
this book that we all are under a duty to support processes for a justifiedATC peace.
That duty can be decomposed in duties on the part of all of us qua citizens to sup-
port the relevant just institutions, in duties on the part of public officials to act as
required by a justifiedATC peace, and in duties on the part of all of us so to act in
our private, daily life. When discussing truth and reconciliation commissions in
s.9.5, I argued that the duty to work towards peace was a duty not just to support
the work of those commissions by, e.g., not interfering with their activities, or by
supporting the allocation of required material resources, but also a (prima facie)
duty on the part of perpetrators, bystanders, and sometimes victims themselves, to
testify. As applied to moral reasons for war remembrance, the question, then, is
whether those moral reasons are merely reasons to support institutions of remem-
brance which in turn supply officials with role-based reasons for remembering on
our behalf (as when the Head of State attends relevant ceremonies on 11 November),
or whether we have compelling moral reasons to engage in those practice ourselves.
I believe the latter, contra the institutionalist approach to duties of justice. For even
Remembrance 301
if that approach is correct with respect to the distribution of resources, it does not
undermine the individualist approach to remembrance as a means to discharge
general obligations to promote peace: after all, I do not need to coordinate efforts
with other people to attend remembrance ceremonies, wear a poppy, remain silent,
at the very least remain aware of those practices. Furthermore, in s.9.5, I argued for
the importance of promoting and living by a justice-friendly ethos which regulates
our interpersonal relationships within the interstices of the law. In this particular
context, then, it is not enough that we should vote in favour of erecting war memo-
rials or against scrapping remembrance ceremonies: there are compelling moral
reasons for at least on occasion engaging in remembrance practices ourselves.
The second issue is this. As we saw in s.10.2, war remembrance has moral and
emotional valence. Accordingly, reasons to commemorate wars are also reasons to
do so with the appropriate emotional and moral attitudes. This raises the question
of the extent to which public officials who perform commemorative acts on our
behalf must feel the requisite emotions for the acts to be appropriately commem-
orative. When dealing with this question in the context of public expressions of
regret and apologies in s.9.6, I claimed that emotional display on the part of the
official is not a prerequisite of a valid expression. The same applies to commemo-
rative acts, mutatis mutandis.
The moral and emotional valence of war remembrance raises yet another ques-
tion. As I also noted in s.10.2, the events which we are typically invited to remem-
ber are usually painful and more often than not egregiously wrongful. Appropriately
commemorating war does not simply consist in openly marking its beginning and
its end: it also requires that we should commemorate individual actors who typi-
cally suffered in it: the combatants who fought and killed; those who refused to
fight and were punished with death; those who lost control of their bowels while
going over the top and fell screaming in agony; civilians who were subject to the
horrors of gang rape, torture, and summary execution by occupying and/or alleg-
edly liberating combatants; those who collaborated with those combatants and
were complicitous in the latter’s wrongdoings out of greed, ideological commit-
ment, or the simple need to survive; those who lost loved ones to its horrors. Not
only, thus, does it behove all of us, and not just the British, to (e.g.) remember
George Eliot’s ‘faithful who are not famous’ or Siegfried Sassoon’s ‘unheroic Dead’:
it is incumbent on all of us on the one hand to remember them with sorrow and
on the other hand, if sending them to their death really was a crime, to remember
them as victims thereof. The Menin Gate memorial is a symbol of moral bank-
ruptcy, Sassoon tells us, is precisely because it embodies a failure to remember the
lost combatants of Passchendaele as they should be remembered—as victims.
Likewise, mutatis mutandis, with any morally and emotionally distorting act of
remembrance.
That said, although the dominant emotion which we are supposed to experience
in war remembrance is sadness, we might also be invited to express joy that the war
did end after all. For a few years following the end of the First World War, a fairly
substantial number of veterans insisted that Armistice Day should be seen as an
opportunity to rejoice and celebrate, not to mourn—an emotional attitude which
302 Remembrance
the bereaved often found difficult to accept.22 I do not think that one can be overly
prescriptive here, if only because the range of emotions one may appropriately feel
partly depends on one’s own individual experience of the war. That said, unalloyed
joy that the war is over with no sensitivity to the suffering it caused, unalloyed
sadness with no tolerance for the fact that some actors in war may feel emotionally
ambivalent at its ending, unalloyed and vengeful delight at the thought that the
enemy are dead, indifference to the plight of combatants and civilians as one
stands, mechanically and unthinkingly, at the tomb of the Unknown Combatant—
all of these do strike me as morally problematic. At the very least, and however
ambivalent one might feel in relation to war in general and specific wars and their
actors in particular, one should remember them with some regret, if not sorrow.
This applies even to those who fought unjust wars and suggests that they (at least
sometimes) ought to be remembered qua human beings, and no longer, or at least
not just qua wrongdoers: there comes a time when the perception that one’s ances-
tors were wronged by the enemy and (conversely) when the recognition that our
ancestors did wrong, must give way to profound regret at what was done (as dis-
tinct from anger for what they did to us and guilt at what we did to them). This,
in effect, is exactly what President Mitterrand and Chancellor Kohl were doing at
the battlefield of Verdun in 1984.
The point is controversial of course. For surely (some might object) it simply
cannot apply to Osama Bin Laden, or to the committed members of the Nazi
paramilitary Einsatzgruppen who perpetrated atrocities against Jews in the wake of
the Wehrmacht’s campaigns and were deeply committed to what they were doing.
Yet I think that it does apply, for there is always a sense in which the death of
another person is bad—even if we are justified in killing him, and even if, in fact,
all things considered it is better that he died. To be sure, in some cases, of which
the Einsatzgruppen combatant and Bin Laden are paradigmatic examples, the
sense in which a death is always bad is very tenuous: but to my mind at least, it is
there and it is important not to lose sight of it.23
As regard moral attitude, I am implying that it is appropriate to bear a moral
judgement on what our predecessors did or failed to do, even though their own
moral views were vastly different from our own. There are deeds which do shock
the conscience of humankind and it is those deeds which above all warrant com-
memorating publicly with the requisite degree of moral condemnation. (In less
grievously wrong cases, such condemnation can take the form of, on the one hand,
denying that combatants of the past were justified in acting as they did, and on the
other hand, admitting that they had an excuse for so doing.) Still, not all wars, or
phases of war, admit of clear-cut moral judgements. Many—most?—are extremely
22 Gregory, The Silence of Memory.

23 I am particularly grateful to John McLaughlin (to whom I owe the example of the Einsatzgruppen)
and Felix Pinkert for helping me clarify my own thoughts on this delicate issue. It is worth noting,
incidentally, that the emotional dimension of remembrance is important not just in the sense that it
tracks the reactive attitude which war suffering itself should elicit in us, but also in the sense that the
past is more likely to endure in our mind if we remember it affectively. See Blustein, Forgiveness and
Remembrance, 205.
Remembrance 303
controversial morally speaking. In those cases, if we choose to engage in a com-

memorative act, we should do so in explicit acknowledgment of that controversy.
Thus, many regard the decision by the British Cabinet to authorize the systematic
and deliberate targeting of German civilian populations during the Second World
War (Hamburg, Cologne, Dresden) as a morally legitimate response to the Nazis’
bombing of British cities and a morally legitimate means to defeat an evil oppo-
nent; but others regard it as an egregious wrongdoing, indeed a war crime.
Unsurprisingly, the erection of monuments in memory of Sir Arthur ‘Bomber’
Harris and the airmen who took part in those bombings under his command has
proved extremely controversial, not least because commemorative buildings do not
generally reflect how divisive that policy is—indeed, was at the time. As a com-
mentator insists, remember if you must, but at least do it with moral honesty.24
If erecting monuments will not do it, do it some other way.
10.4.2. Three Objections

The view that war remembrance is best justified by appeal to the moral imperative
of bringing about universal peace might seem objectionable on three counts.
According to the undue optimism objection, most political actors, indeed citizens
themselves, simply do not have the honesty and self-restraint to remember wars as
they should be remembered, and in fact use the war and remembrance thereof as a
means to fan the flames of violence—so much so that in many cases, bringing
about peace requires that we forget about war, so divisive the latter was. Good
examples of commemorative acts which undermine peace are the marching parades
in Northern Ireland and ceremonies commemorating Croatia’s war against Serbia
in the 1990s.25 We encountered a similar point in s.7.7 when discussing amnesties
and amnesia as alternatives to punishment. By way of reply: the worry is particu-
larly acute, of course, the less time has elapsed between the end of the war and the
commemorative events. Either way, if the imperative of bringing about peace
requires some form of forgetting, or at any rate, not actively remembering, then
the moral reasons I have provided in favour of war remembrance should be put to
the side. At any rate, they should give way to peace-based reasons for renouncing,
at least for a while, public, official acts of remembrance. However, we should not
destroy the possibility, in the future, for remembering—and thus should ensure
that records are kept, albeit sealed for some time; nor should we prevent witnesses
or survivors from speaking out, e.g., through censorship laws.
Suppose, however, that remembrance will help bring about peace. According to
the over-demandingness objection, my view implies that we all have compelling
moral reasons to remember all the genocides, all the atrocities, all the senseless wars
of which we know. However, our resources are scarce—be it the financial resources
24 See J. Jones, ‘The Artistic Jingoism of the Bomber Command Memorial’, The Guardian. http://
www.theguardian.com/commentisfree/2012/jun/29/bomber-command-memorial-artistic-jingoism
25 I am grateful to Margaret Moore and Cheney Ryan for pressing me on this. For discussion, see
Blustein, Forgiveness and Remembrance, 216–17.
304 Remembrance
which we need to erect memorials and conduct commemorative ceremonies,

the time resources which we spend when attending commemorative events, and
perhaps more importantly still the cognitive and emotional resources which
we should deploy whenever we take part in those practices. Given that we can-
not possibly remember all the wars ever fought, the objection goes, we simply
lack compelling moral reasons to do so. As I was once asked by a US Air Force
cadet, who, really, apart from the English and possibly the people of Normandy,
has any interest in commemorating the battle of Hastings, let alone a moral
reason to do so?
This interesting and important objection invites two responses. First, one of our
remembrance practices can be institutionalized globally. In just the same way as the
UNESCO maintains through the World Heritage a list of sites whose natural or
cultural significance is such as to warrant special conservationist measures, it should
do so for sites of war memories which have significance above and beyond borders,
such as the battlefields of the two World Wars. As it happens, the UNESCO already
does so for the Holocaust: the complex of Auschwitz-Birkenau is a protected site, as
‘the symbol of humanity’s cruelty to its fellow human beings in the 20th century’.
I see no reason not to extend the remit of such a list to sites of war.26
Second, the objection is similar to familiar concerns raised by opponents to
cosmopolitan obligations of justice, and can be met in the same way: as many
cosmopolitans aver, the best way to discharge those obligations is to allocate the
task of helping individuals to the political community of which they are members.
By parity of reasoning, the best way to discharge the universal imperative to pro-
mote peace through acts of remembrance is to allocate those acts among actors
who, by dint of their common history and geographical location, will find it easier
to remember this, or that, war or atrocity.27 This point draws strength from the fact
that publicly to remember a war and its victims implies actively recalling it, and
them, to your mind, with the requisite reactive attitude. It is easier to do so when
the legacy of that war is visible in the landscape of your childhood and discernible
in the fabric of your own individual history than it is when it is inscribed in far-
flung corners of the world. Thus, the labour of remembering genocides in general
can be appropriately divided into the moral tasks of remembering the Armenian
Genocide, the Holocaust, and the Rwandan genocide—with each task befalling
individuals who are connected to perpetrators and victims through their political
membership in relevant communities.
26 See http://whc.unesco.org/en/list/31 (accessed on 16/01/2016).

27 For a similar argument, see, e.g., Blustein, The Moral Demands of Memory; Margalit, The Ethics
of Memory. A crucial difference between their account of remembrance duties and mine is that they
endorse a two-tiered view of morality, whereby we have moral obligations to all human beings and
ethical obligations to compatriots (as well as close friends and relatives). Both include obligations of
remembrance, but on their acccount, ethical obligations are justified by appeal to the intrinsic value
of the relationship of co-citizenship or co-nationality (which I disputed in s.10.3.2). Moreover,
Blustein and Margalit restrict the reach of the imperative to track truth in remembrance to (universal)
moral obligations, on the grounds that historical remembrance may appropriately be turned into
nation-building myth. I reject that view too. I am grateful to Michael Growden for helpful comments
on the points raised in this and the next five paragraphs. For scepticism on the notion of universal
remembrance (as it were), see S. Linfield, ‘Memory’s Lair’, Boston Review, 01/06/2003.
Remembrance 305
At this juncture, an objection of a rather different kind might seem to emerge—

which we might call the over-universalism objection. On my view, the wars and
their actors which we have compelling moral reasons to commemorate are seem-
ingly understood as tokens for the broad phenomenon ‘war’ and the broad catego-
ries of ‘combatants’, ‘victims of atrocities’, and so on. Thus, it would seem that the
Holocaust is to be remembered as an example of a genocide—and not for the
specific event that it was. And this might seem problematic on two grounds. First,
we risk unhinging wars from their distinct historical context and thereby overlook-
ing crucial and morally relevant differences between our predecessors (broadly
understood as those who lived before us, irrespective of community membership)
and us (similarly broadly understood). As a result, we risk going to war unjustly on
the grounds that our predecessors’ failure to wage a particular war in the past
ensued in a worse state of affairs than if they had dared to fight. Conversely, we risk
not going to war on the grounds that our predecessors’ decision to do it proved
disastrous, when we in fact morally ought to act. To illustrate, it is not far-fetched
to suppose that repeated public remembrance of the horrors of trench warfare in
post–First World War France partly explains why the French adopted a resolutely
defensive military strategy in the 1930s and were reluctant unilaterally and coer-
cively to block Hitler’s repeated violations of the Versailles Treaty at a time when
the German Army would not have been able to repel a French attack.
There is something to that concern, which can only be countered by unflinching
scrutiny of both the facts of the wars at issue and conflicting interpretations thereof,
and in awareness of the psychological biases to which we are prone when deciding
whether or not to go to war. To the extent that perceptions of past defeats and
victories shape our contemporary attitudes to war, we must remain attentive to the
degree to which public remembrance itself informs those perceptions.28
The second problem with over-universalism is this: it seems that, with respect to
commemorating wars and events which have the deepest and most universal moral
significance to us qua human beings, any war or war atrocity will do—particularly
when there no longer are victims. In other words, in a century from now, why
remember the combatants who endured trench warfare in the Battle of the
Somme in 1916 rather than those who endured appalling suffering at the battle of
Stalingrad in 1942–43? After all, the dehumanization of combatants remains such,
wherever it happens. Similarly, why ensure that what remains of Treblinka not rot
to oblivion, so long as Auschwitz remains standing? After all, any reminder of the
Holocaust will do. If that is an implication of my argument, then it will (I suspect)
be troubling to many, in so far as it seemingly rides roughshod over the deeply
rooted intuition that there is something uniquely valuable and thus worth preserv-
ing about each, or at least most, of the events and sites which together are part and
parcel of a war.
I see the force of the intuition. What makes an event or a site uniquely distinct
is an interesting question which I cannot tackle at length. Irreplaceability has
something to do with it, at least when it comes to material objects, as does the
28 On the influence of perceptions of defeat and victory on leaders’ decision to wage war or to
desist from it, see Johnson and Tierney, Failing to Win.
306 Remembrance
notion of historical significance for both sites and events.29 Let us assume, then,
that Auschwitz and Treblinka are each unique and to be valued for their unique-
ness as well as in virtue of the fact that they are sites where dozens of thousands of
people were gassed to death mostly for being Jewish. And let us make a relevantly
similar claim about the horrors of both trench warfare and war in the depth of the
Russian winter. Still, all that the claim of unique value shows is that our successors
may well have a reason to remember both wars and maintain all of their sites—but
not that they will have a compelling moral reason to do so at the bar of the univer-
sal requirement to promote peace, such that failing to comply with that reason
would elicit moral opprobrium. Absent instrumental peace-promoting reasons for
preserving both the battlefield of Stalingrad and that of the Somme, our successors
will be situated in relation to those wars in exactly the same way as we now stand
in relation to the Crimean War or indeed Napoleonic wars—all of which did see
their fair share of horrors. In one sense, my point is the often made one that there
is a time limit to what we ought to do with respect to the past. In another sense, it
is also that there may well be a limit to the number of loci for discharging our uni-
versal duty to our contemporaries and immediate descendants to promote peace.
1 0 . 5 . C O N C LU S I O N
In this chapter, I have defended the view that there are compelling moral reasons
for war remembrance which do not flow from our special relationships to others
(be it the relationship of co-citizenship, that of wrongdoers and victims or descend-
ants thereof ). Those reasons apply to us as human beings who, more deeply, owe it
to one another to bring about peace. None of the above implies that war remem-
brance cannot be justified on grounds other than those offered here. In particular,
acts of remembrance such as Memorial Day in the US or Remembrance Sunday in
the UK have considerable expressive, and in this respect, intrinsic, value above and
beyond the fact that they might (if properly conducted) help citizens of those two
polities meet their general obligations to humanity. My concern, however, was
with providing moral reasons for war remembrance such that failing to act on
those reasons would leave us vulnerable to moral opprobrium. Unless one takes the
maximizing consequentialist view that one is vulnerable to moral opprobrium for
failing to maximize the good, I doubt that particularistic expressivist considera-
tions can, in general and in this particular case, ground moral reasons of that kind.
Finally, as I also argued, although we ought to side with Eliot, we should neverthe-
less heed Sassoon’s worry that the call to emotionally laden remembrance is but a
cloak for mindless, self-congratulatory, and deceitful political myth-making. That
particular kind of remembrance, we can, indeed ought to dispense with.
29 For a recent discussion, see Gregory, The Silence of Memory; E. H. Matthes, ‘History, Value, and
Irreplaceability’, Ethics 124 (2013): 35–64.
Conclusion
I have sought to offer cosmopolitan principles for the regulation of belligerents’

conduct as they consider suing for peace and settle on peace terms. The just war
tradition in its Vitorian incarnation prescribes the punishment of the enemy, the
restitution by unjust belligerents of what they wrongfully took during the war, and
the removal of the enemy regime by the victorious party once the war is over if
failing to do so would endanger the latter’s community. Following Grotius, and
with the exception of Kant, it also regards peace treaties as important instruments
for delineating and enforcing belligerents’ post bellum rights and obligations. To
these constitutive components of a just peace, and modulo some qualifications,
we can add reparations for wrongful harms and remembrance practices; to peace
treaties, we can add the deployment of international peacekeeping forces, the insti-
tution of international governance mechanisms, and the establishment of reconcil-
iatory processes aimed at fostering trust amongst enemies.
In this concluding chapter, I will not rehearse my arguments and conclusions as
they have unfolded chapter by chapter. Instead, I will revisit the four questions
which have recurrently emerged throughout the book:
(1) the dependence v. independence question—to wit: the extent to which the moral
status of a belligerent’s war has a bearing on its rights and obligations post-war.
(2) the individual v. collective question—to wit: the extent to which an individu-
alist theory of peace after war can account for the collective dimension of
both war and peacemaking.
(3) the just peace v. justifiedATC peace question—to wit: the extent to which it may
not be possible to implement a just peace, and thus to which it may be nec-
essary to compromise in favour of justified peace all things considered (whose
constitutive rights and obligations also need ascertaining).
(4) the cosmopolitanism question—to wit: the extent to which my theory of
peace after war is genuinely cosmopolitan, given that its central principles in
the main apply, not to the world at large, but to erstwhile belligerents, and
thus seem better supported at first sight by particularistic considerations
than by universal norms.
T H E D E P E N D E N C E V. I N D E P E N D E N C E Q U E S T I O N
Let me start with the first question. It issues from contemporary just war theory,
whose dominant preoccupation over the last fifteen years has been the relationship
308 Conclusion
between the jus ad bellum and the jus in bello. Followers of Michael Walzer argue
that soldiers are entitled to kill enemy soldiers once the war has started and irre-
spective of its justness. Others, notably Jeff McMahan, reject that very point and
insist on tying the permissibility of war killings to the permissibility of the recourse
to war. On the latter account, which I also endorse, individuals’ wartime rights and
obligations are dependent both on the moral status of their contribution to the war
and on the moral status of the war itself.
Against this normative background, it makes sense to ask whether individuals’
ex bello and post bellum rights and obligations are also dependent on those factors.
As we saw, we must distinguish between procedure and substance. Substantive
rights and obligations set out the terms under which belligerents must behave
towards one other, for example with respect to suing for peace, reparations, and
punishment. Procedural rights and obligations pertain to belligerents’ standing to
sue for peace, and to negotiate, endorse, and enforce peace terms. Clearly, an
unjust belligerent does have standing to sue for peace. The question of its standing
with respect to peace negotiations and peace enforcement is more interesting. On
my account, a belligerent’s standing as a peace negotiator is partly shaped by the
moral status of its war, which in turn has a bearing on the procedural justness, or
lack thereof, of the resulting peace settlement and, by implication, on its binding-
ness. I say ‘partly’, and not ‘wholly’. Sometimes, an unjust belligerent regime has
acted so grievously unjustly that it has lost its standing as a peace negotiator. It does
not follow however that the just, or less unjust victorious belligerent may unilater-
ally impose terms on the defeated populations. Sometimes, alternatives to that
regime, for example in the form of resistance movements, might acquire such
standing if given the chance; in addition, the endorsement of peace terms by the
defeated population is sometimes required for those terms to be binding.
Furthermore, and in any event, to insist that having conducted a just war is a nec-
essary condition for negotiating peace terms (as opposed to having those terms
imposed by victorious belligerents, à la Versailles) is a profound mistake: for it may
well be that an unjust belligerent would, thanks to negotiations, provide for both
its members and outsiders better conditions for the protection and enjoyment of
their moral rights and duties than they would have otherwise.
Turning now to the enforcement of peace terms, as well as to the enforcement
of short-term measures such as ceasefires while negotiations are ongoing, the status
of one’s war is less dispositive of one’s rights and obligations than might first appear.
For example, it is a key condition of the success of peacekeeping operations, whose
task is to oversee and enforce aforementioned short-term measures, that peace-
keepers should be seen as impartial. This dictates in favour of multilateralism,
whereby international institutions deploy international forces funded by member
states. On that view, merely being a belligerent in that particular conflict disquali-
fies an actor from enforcing ceasefires and providing basic humanitarian relief on
the ground: the moral status of one’s participation in that conflict is neither here
nor there. In the context of punishment for war crimes, by contrast, whether a
belligerent’s forces have committed war-related crimes in the course of a given con-
flict can have a decisive bearing on its standing as a punishing party. It undermines
Conclusion 309
it if those crimes were committed or condoned in such a way as to give objective

grounds for suspecting that the perpetrators will be treated too leniently; it
strengthens it if by punishing the perpetrators appropriately, that belligerent would
reaffirm and strengthen its commitment to fighting just wars in the future.
What about substantive rights and obligations? Again, the picture is somewhat
nuanced. For example, it does not follow from the fact that one’s war is just at t1
that one has the right to carry on with it at t2—any more than it follows from the
fact that it is unjust at t1 that one must sue for peace at t2. Likewise, one may well
have conducted an unjust war, yet nevertheless acquire occupation rights over the
territory of one’s enemy. Similarly, one may well have conducted an unjust war, yet
without being under a restitutive obligation to return the military equipment or
even the territory which one has wrongfully seized; conversely, one may well have
conducted a just war yet still be under a prima facie obligation to give back that
which one has rightfully taken. In all those cases, the moral status of the war is
severed from one’s entitlements and obligations post-war through changes in cir-
cumstances, the availability of new information, and our enemy’s behaviour: for
example, developments on the battlefield would make continuing with the war
unjust; if we return to our enemy matériel which we wrongfully took from its
forces, it will use it to unjust ends, etc. By contrast, reparative and punitive rights
and obligations seem more firmly tied to the moral status of one’s war. Leaders who
have wrongfully ordered their armed forces into unjust wars, soldiers who have
killed in those unjust wars, and citizens who have wrongfully contributed to an
unjust war, are prima facie liable to reparative and punitive burdens, whereas just
belligerents (be it the leaders, soldiers, or citizens) are not.
I say ‘seem more firmly tied’, because on closer inspection, and through the
lenses of the kind of moral individualism which informs this book, one must dis-
aggregate war into its individual constitutive acts of violence and contributions
thereof. To know whether a particular agent is liable to punitive or reparative harm
post-war, we need to know whether she herself violated or justifiably infringed the
rights of another person in the course of the war. Even if the war which she and her
associates waged was just overall, some of her contributions to the conflict may
have been profoundly unjust—and vice versa. In a curious twist, then, individuals’
rights and obligations post bellum do not rest on an assessment of their war in toto
as just or unjust. Rather, an individual’s fundamental rights and obligations in
general and in this particular context do not depend on the mere fact of her mem-
bership in this, or that group which happen to conduct this, or that, collective
enterprise. They depend on what she does or did as a moral agent, on the unjust
harms which she incurred, and on the benefits which she unjustly derived.
T H E I N D I V I D U A L V. C O L L E C T I V E Q U E S T I O N
This point leads us straightaway to the individual v. collective question, to wit, how
to account for the inherently collective dimension of war, and of peace after war.
War is of course fought by individual combatants; the suffering it occasions is felt
310 Conclusion
by individuals. To the extent that individual acts of maiming and experiences of

suffering are what ultimately needs justifying, it is appropriate to approach war
from the standpoint of moral individualism. At the same time, individual combat-
ants do not act on their own, but as part and at the behest of a collective—usually,
their political community; individual non-combatants who participate in war also
do so as part and at the behest of a collective, and are maimed and killed by dint
of the fact that they belong to such a collective. The collective dimension of war
raises a number of difficult issues for an individualist normative account of peace
after war. First, it raises the issue of the imposition of post-war burdens. That
issue—of harm imposition for participation in collective enterprise—also arises in
war itself. Admittedly, when a combatant attempts to kill another soldier in prose-
cution of an unjust cause, one can say without too much difficulty that he is liable
to be killed. However, many participants in war each contribute in marginal ways
to the collective enterprise (think about the munitions factory worker, the tax-
payer, the soldier who never fires a shot). The question then is whether they are
liable to be killed (on the grounds that they are participating in a lethal enterprise
even if they themselves do not kill), or whether they are liable to lesser harm (on
the grounds, precisely, that they themselves do not kill). Post-war, the question is
whether the reparative and punitive burdens to which participants in an unjust
war/phase of war are liable ought to be calibrated to those participants’ individual
contributions (or, in the reparative context, to the unjust benefits they individually
derive from the war); or whether they ought to be calibrated to the wrongful harms
occasioned by the war collectively understood as the sum total of those individual
contributions. On the sum-total view, the risk is that many of those participants
will be unjustifiably overburdened; on the individual contribution view, the risk is
that they will be left off the moral hook, at the expense of their victims. In answer-
ing this difficult question, I have argued with respect to reparations that we must
eschew tracking all but the most grievously and obviously wrongful individual
contributions and instances of wrongful benefiting, and settle in all other cases for
a Reconstruction Principle. Its constitutive duties fall on rights violators, justified
rights infringers and unjust beneficiaries, as well as on bystanders, who do not fall into
either category. In the case of bystanders, the reconstruction duties are justified
both at the bar of an ethics of assistance and on the grounds that non-participants
and non-beneficiaries must work on the realistic assumption that, sooner or later,
they too will be embroiled in unjust wars, either as participants or victims. With
respect to punishment, I have argued against collective punishment, on the
grounds that to risk letting wrongdoers go unpunished, except in the most egre-
gious cases, is better than to risk punishing the innocent. However, I have also
argued that an agent who takes direct part in the commission of a collective war-
related crime such as genocide is liable to punitive harm for that collective crime,
even if he did not know or intend to take part in a genocide: it is enough that he
could reasonably have been expected to know that his contribution was a contri-
bution to this collective crime.
The second issue raised by the individual v. collective question pertains to the
ways in which belligerents sue for peace and negotiate and enforce peace terms.
Conclusion 311
As a matter of fact, these are collective acts, in the sense that some public officials
(typically the head of state or government, or the guerrilla leaders) make the rele-
vant declarations, announcements, and offers on behalf of the political community
on whose behalf they claim to act. As a matter of principle, this is how it should
be. For just as I cannot, as a private agent or indeed as a citizen acting on my own,
declare that my country is now at war with yours, I cannot declare that it is no
longer at war and that it accepts these or those peace terms. This second issue yields
several interesting lines of inquiry. In particular, it invites us to think about the
relationship between individuals’ and public officials’ rights. It is a key assumption
of this book (and in line with my argument in Cosmopolitan War) that public offi-
cials’ rights to govern supervene on individuals’ pre-institutional rights; moreover,
public officials have the right to govern if and only if, through the laws for which
they vote and enforce and the executive decisions which they make on the basis of
those laws, they respect the human rights of both state members and outsiders. In
the context of this book, for example, public officials’ right to punish war criminals
supervenes on the right which victims and all other human beings have to see to it
that war crimes not go unpunished; they have that right, moreover, only if they
better enable victims and all other human beings to exercise their pre-institutional
right to punish and fulfil its concomitant duties (such as the duty not to punish
more harshly than is warranted, out of a thirst for revenge). Similarly, a public
official’s right to extract reparation payments supervenes on claimants’ pre-institu-
tional right to do the same, etc.
Another line of inquiry pertains to the importance of ensuring that a public
official is authorized so to act on behalf of individual agents to whom her deci-
sions will apply. When, in virtue of occupying a public role, she can better enable
those agents to fulfil their general duties—for example, their duty not to take
part in a wrongful invasion, or not to commit atrocities, or to accept the pres-
ence of peacekeeping forces—that alone is enough to render her decisions bind-
ing. Put differently, the consent of agents to whom her decisions apply is not a
necessary condition for her decisions to be binding. In the language I used when
discussing peace agreements, declarative clauses, and by implication declarative
decisions, derive their authority solely from the fact that they instantiate or act
as a vehicle for discharging the duties which correlate with human rights. Often,
however, individuals will not regard themselves as so bound unless they consent
to that official’s decisions: while they are mistaken, the risks that they might
resume war, or at any rate fail to support the peace process, might be great
enough that their consent should be sought all things considered. Moreover,
general duties need specifying, and can often be specified in different and mor-
ally equivalent ways. In such cases, respect for individuals’ interest in shaping the
conditions under which they conduct their life dictates that they should have a
say in the negotiation of peace terms; it thereby requires both that they authorize
public officials to act on their behalf, for example through elections, and that
they endorse the resulting peace terms. The more wide-ranging peace terms are,
the more important it is that individuals each have a say over those terms, via
proper authorization procedures.
312 Conclusion
The latter point occludes two different difficulties. The first difficulty is this.
Assume that we can plausibly aver that the citizenries of two countries recently at
war authorized their respective governments to settle on peace terms at t1, and
endorsed those terms at t2. Are those terms still binding several decades later, when
no individual member of those citizenries as they existed at t1 and t2 is still alive?
Yes, to the extent that those terms instantiate general, independently justified, and
time-independent rights and obligations. When those terms create new rights and
obligations, or specify general rights and obligations, the answer to that question is
also ‘yes’, so long as each cohort of citizens does not in its majority signal to the
other and contemporary citizenry that they do not consider themselves so bound.
The second difficulty arises from the fact that some of our post-war obliga-
tions—namely our obligations to apologize and express regret for war-related
crimes, and more generally to remember war itself—have affective valence. When
the Queen stands at the Cenotaph on Remembrance Sunday, she represents all
British citizens in that commemorative act. Yet, British citizens may not in fact all
feel the same way about the wars in which the United Kingdom has been involved,
and it seems overly ambitious to say ‘the British together feel that…’—more ambi-
tious indeed than saying ‘the British have decided to do x’ (for how on earth can
we know what anyone, let alone several people, really feels?) Moreover, even assum-
ing a collective affective agency (as it were), the Queen’s affective attitude may not
itself match ours. Those remarks raise the problems of ascribing agency and inten-
tion to a collective of individual agents, the related problem of minorities, and the
problem of mismatch between citizens’ commitments and officials’ decisions.
These are familiar problems in democratic theory, but they arise in this context
too—and if nothing else are a salutary reminder of the importance of drawing
on various areas of political philosophy when thinking about what is required of
belligerents after war.
T H E J U S T P E A C E V. J U S T I F I E D ATC P E A C E Q U E S T I O N
So what, then, is required of belligerents? It is tempting to think at first sight that

just as belligerents must strive to wage and fight a just war, they must try to seek
and achieve a just peace simpliciter. Yet it becomes clear on closer inspection that
belligerents may not be able to do that. Early on in the book, I characterized a just
peace as follows (verbatim): a state of affairs where all individuals actually enjoy
their human rights to the freedoms and resources they need to lead a flourishing
life. By implication, a just peace does not obtain only between belligerents; it also
obtains with respect to all other parties who may have been wrongfully harmed by
the war though were not themselves participants in it (e.g., bystander communities
and their members whose infrastructure was damaged by mis-targeted bombs); it
also obtains, finally, between parties who were neither harmed by the war nor con-
tributed to it but whose prospects for a flourishing life partly depend on the peace
settlement. For there to be a just peace, thus, it is not enough that belligerents no
longer have a justified grievance against each other: it must also be the case that the
Conclusion 313
rights of all human beings, wherever they are, are not violated in the aftermath of
the war.
As we saw throughout the book, however, the achievement of a just peace faces
three feasibility constraints: scarcity of resources; epistemic uncertainty; and agents’
non-compliance. In so far as material and human resources are scarce, and in the
light of the sheer numbers of war victims and the sheer scale of the destruction
wrought by war, it simply is not possible to meet all reparative and restitutive
demands; nor is it possible to meet all punitive claims—for example by putting all
war actors deemed guilty of wrongful war killings in prison. Even if it were possible
to do all of that, it would not be possible at the same time to honour the non-war
related distributive demands of individuals who do not have prospects for a flour-
ishing life.
Suppose that I am wrong—that as a matter of fact, the resources are there to do
that. Even so, we face insuperable epistemic constraints. Remember that on my
account of peace after war, individuals each on their own are liable to the relevant
burdens for the wrongs they committed in the war and/or the unjust benefits they
derived from it; individuals each on their own have restitutive, reparative, and puni-
tive claims on the basis of the wrongs to which they were subjected. But we are not
and can never be in a position to know of every single participant in and victim of
war how they contributed to the war and/or the extent to which they suffered from
it. We therefore cannot hope to achieve a just peace simpliciter, since such a peace
requires that we implement restitutive, reparative, and punitive principles in full.
Those two kinds of constraints—resource scarcity and inability to garner the
required information—are practical constraints: we might one day have the
required resources; we might also one day be able to get that information. If so, we
would be able to achieve a just peace. But they are not merely practical constraints:
they are morally inflected, in the sense that they compel us to make decisions
which have considerable moral costs (for example: which reparative claimants to
satisfy at the expense of other such claims, which suspected war criminals to let go
at the expense of his victims, etc.). The third constraint, by contrast, is directly
moral, in that it results from a conflict between the demands of the just peace on
the one hand, and the demand of satisfying the preferences of those who are under
a duty to comply with those demands. In the present context, those preferences are
indicative of individuals’ moral failure to comply with the demands of a just peace.
For example, individuals strongly suspected of war crimes go on the run, or
threaten to derail the peace process unless they are granted an amnesty; taxpayers
whose democratically elected regime waged an unjust war with their support refuse
to consent to the tax rise needed to pay up for reconstruction; combatants of
defeated communities refuse to comply with the directives of a justified peacekeep-
ing or occupation force—and so on.
Faced with those various constraints, we may have to accept that a just peace
might not be achievable. The question, then, is what we must strive for at the very
least, and what we may justifiably accept as a compromise—put differently, what a
justified peace all things considered consists in. It consists, at a minimum, in a state
of affairs where individuals do not enjoy all of their non-basic rights as a result of
314 Conclusion
either one or more of those constraints (which is why it is not just tout court), but
where they nevertheless enjoy their basic human rights (failing which it would not
be justified). It permits the following (inter alia): more flexible territorial and reset-
tlement solutions (as a compromise on the right of refugees to return to and govern
over the exact places from which they were expelled); general duties of reconstruc-
tions (partly in lieu of the payment of reparations by all perpetrators of war-related
crimes to all victims of such crimes); the granting of selective amnesties and the
adoption of restorative processes (as a compromise on punishing all war criminals);
and a moratorium on public practices of remembrance (as a compromise on
unflinching scrutiny of the past lest it should fan the embers of war). When a just
peace is impossible and war not permissible, a justifiedATC peace so construed is
what we must strive for.
T H E C O S M O P O L I TA N I S M Q U E S T I O N
Let me turn, finally, to the cosmopolitanism question—namely, to the cosmopolitan

underpinnings of my theory of the just peace. According to cosmopolitans, all
human beings wherever they reside have rights to the resources and freedoms
which they need in order to live a flourishing life, against all other human beings
wherever they reside (subject to ability to meet the relevant duties). Political-cum-
national borders, in other words, are irrelevant to the conferral of those rights and
the imposition of the correlative duties. More generally, cosmopolitans are wary of
special relationships based on communal membership and/or shared history, in so
far as those relationships are exclusionary vis-à-vis outsiders. Yet relationships
between erstwhile belligerents, as the latter prepare to end war and negotiate peace,
are inherently exclusionary. What, then, can a cosmopolitan qua cosmopolitan say
about the special rights and obligations which those relationships generate?
She can make at least three points. First, cosmopolitans can articulate and bring
our attention to rights and obligations which belligerents have vis-à-vis one
another, not qua erstwhile belligerents, but qua human beings (via the relevant
institutions). Post bellum, those general rights and obligations can and indeed often
are enshrined in what I have called the declarative clauses of a peace agreement.
Second, cosmopolitan norms underpin and/or constrain those rights and obli-
gations in several ways. Most obviously, restitutive and reparative principles correct
for rights violations and rights infringements, and are thus parasitic on non-correc-
tive principles of justice. Since those non-corrective principles are cosmopolitan
principles, cosmopolitan norms underpin restitutive and reparative rights and
obligations. Furthermore, a belligerent may not give preferential treatment to its
own members on the mere grounds of co-membership—for example by granting
an amnesty to its own soldiers and not to enemy soldiers when those agents have
all taken part in the same crime. In this case, there is no conflict of obligations,
which makes the expression of patriotic partiality particularly egregious. But even
in conflict cases, there are considerable limits to partiality on grounds of commu-
nal membership and of the shared history of war. Thus, when a belligerent is not
Conclusion 315
in a position to meet both its post-war obligations to distant strangers and its post-
war obligations to compatriots, it may not privilege the latter on grounds of patri-
otic partiality. When it is not in a position to meet both its post-war special
obligations to the victims of its war (whether those victims are compatriots or
members of the enemy) and its general obligations to non-war victims, it may not
prioritize the former merely on the grounds of the shared history of the war. Finally,
when receipt of wrongful benefits itself generates reparative obligations, it does so
irrespective of borders.
Third, and most importantly, the imperative to promote and support peace is a
cosmopolitan imperative, to which all human beings wherever they are must com-
mit themselves. This yields a universalist interpretation of the right to punish war
criminals, a strong duty to provide assistance to war victims wherever they are by
helping with the reconstruction of their community and supporting reconciliatory
processes, and a global ethics of war remembrance. When the question arises of
how to institutionalize those rights and obligations, cosmopolitans are committed
to endorsing the principle of universal jurisdiction for the punishment of war
crimes and the principle of multilateralism for the deployment of peacekeeping
and/or occupying forces.
Non-cosmopolitans will find much to worry about in those pages. They will
object that my cosmopolitan account of peace after war leaves too little space for
political self-determination and communal values. They will point in particular to
my defence of transitional foreign administrations and of international war crimes
tribunals; they will also complain (I suspect) that cosmopolitan war remembrance
is overly universal and thereby too divorced from ordinary practices to have much
normative force. I have tried to respond to those worries throughout the book. Let
me simply restate here that there is space for political self-determination when it
comes to specifying belligerents’ rights and obligations, whether those rights and
obligations are general or, on the contrary, issue from their shared history as ene-
mies in war. There is also space for political self-determination even under the
auspices of international peacekeeping, occupying, and governance forces, whose
mandate is precisely to help provide the conditions under which members of war-
torn communities can come to govern themselves. Moreover, although human-
kind via international punitive institutions has primary jurisdiction over war-related
crimes, states have jurisdiction over those crimes when international judicial insti-
tutions do not or cannot exercise the right to punish or when it is better from the
point of view of justice after war that they should not do so. In addition, while
I do think that there are strong, universal, and morally compelling reasons to
remember some wars of the distant past, there are also culturally and communally
specific reasons to do so. That those reasons are morally weaker does nothing to
undermine the cultural and political value of the commemorative practices which
they support.
All of that being said, cosmopolitan peace admittedly offers only limited scope
in the end for patriotic partiality and political self-determination. If this is the
political and moral price of a commitment to the inherent value of every human
being wherever he or she is, I for one am more than happy to pay it.
Works Cited
Acorn, A., Compulsory Compassion: A Critique of Restorative Justice (Vancouver: University

of British Columbia Press, 2004).
Aertsen, I., Arsovska, J., Rohne, H.-C., Valinas, M. and Vanspawen, K. (ed.), Restoring
Justice after Large Scale Conflict (Cullompton: Willan Publishing, 2008).
Afflerbach, H. and Strachan, H. (ed.), How Fighting Ends: A History of Surrender (Oxford:
Allais, L., ‘Wiping the Slate Clean: The Heart of Forgiveness’, Philosophy & Public Affairs
36 (2008): 33–68.
Allais, L., ‘Restorative Justice, Retributive Justice, and the South African Truth and
Reconciliation Commission’, Philosophy & Public Affairs 39 (2011): 331–63.
Allawi, A. A., The Occupation of Iraq—Winning the War, Losing the Peace (New Haven,
Conn.: Yale University Press, 2007).
Altman, A. and Wellman, C., A Liberal Theory of International Justice (Oxford: Oxford
Amery, J., At the Mind’s Limits—Contemplations by a Survivor on Auschwitz and its Realities
(London: Granta Books, 1999).
Angstrom, J. and Duyvesteyn, I. (ed.), Understanding Victory and Defeat in Contemporary
War (London: Routledge, 2007).
Applbaum, A. I., ‘Forcing a People to Be Free’, Philosophy & Public Affairs 35 (2007):
359–400.
Applebaum, A., Gulag—A History (London: Penguin Books, 2004).
Arendt, H., ‘Truth and Politics’, The New Yorker, 25 February 1967.
Arendt, H., Eichmann in Jerusalem: A Report on the Banality of Evil, rev. and enl. edn.
Ashworth, A., ‘Punishment and Compensation: Victims, Offenders and the State’, Oxford
Journal of Legal Studies 6 (1986): 86–122.
Ashworth, A., ‘Some Doubts over Restorative Justice’, Criminal Law Forum 4 (1993):
277–99.
Ashworth, A., ‘Responsibilities, Rights and Restorative Justice’, British Journal of
Criminology 42 (2002): 578–95.
Ashworth, A., Principles of Criminal Law, 5th edn. (Oxford: Oxford University Press,
2006).
Ashworth, A., ‘Ignorance of the Criminal Law, and Duties to Avoid it’, The Modern Law
Review 74 (2011): 1–26.
Atwood Lawrence, M., The Vietnam War—A Concise International History (Oxford: Oxford
Augustine, The City of God against the Pagans. ed. R. W. Dyson (Cambridge: Cambridge
University Press, 1998 [426]).
Authers, J., ‘Making Good Again: German Compensation for Forced and Slave Laborers’, in
P. De Greiff (ed.), The Handbook of Reparations (Oxford: Oxford University Press, 2006).
Baier, A., ‘Trust and Antitrust’, Ethics 96 (1986): 231–60.
Barkan, E., The Guilt of Nations: Restitution and Negotiating Historic Injustices (New York:
Thornton, 2000).
318 Works Cited

Barkan, E. and Karn, A. (ed.), Taking Wrongs Seriously (Stanford, Cal.: Stanford University
Press, 2006).
Bass, G.J., ‘Jus Post Bellum’, Philosophy & Public Affairs 32 (2004): 384–412.
Bass, G.J., Stay the Hand of Vengeance: The Politics of War Crimes Tribunals (Princeton, N.J.:
Princeton University Press, 2002).
Bazargan, S., ‘The Permissibility of Aiding and Abetting Unjust Wars’, Journal of Moral
Philosophy 8 (2011): 513–29.
Bazargan, S., ‘Proportionality, Territorial Occupation, and Enabled Terrorism’, Law and
Philosophy 32 (2013): 435–57.
Bederman, D. J., International Law in Antiquity (Cambridge: Cambridge University Press,
2001).
Beevor, A., The Fall of Berlin 1945 (London: Penguin, 2002).
Beilin, Y., ‘Just Peace: A Dangerous Objective’, in P. Allan and A. Keller (ed.), What is a Just
Peace? (Oxford: Oxford University Press, 2008).
Bell, C., On the Law of Peace: Peace Agreements and the Lex Pacificatoria (Oxford: Oxford
Bell, D., ‘John Stuart Mill on Colonies’, Political Theory 38 (2010): 34–64.
Bellamy, A. J., Griffin, S. and Williams, P., Understanding Peacekeeping (Cambridge: Polity,
2003).
Bennett, C., The Apology Ritual (Cambridge: Cambridge University Press, 2008).
Benvenisti, E., The International Law of Occupation (Princeton, N.J.: Princeton University
Press, 2004).
Bhatia, M., ‘Postconflict Profit: The Political Economy of Intervention’, Global Governance
11 (2005): 205–24.
Bhuta, N., ‘The Antinomies of Transformative Occupation’, European Journal of International
Law 16 (2005): 721–40.
Biggar, N., In Defence of War (Oxford: Oxford University Press, 2013).
Blocq, D., ‘The Fog of UN Peacekeeping: Ethical Issues regarding the use of Force to pro-
tect Civilians in UN Operations’, Journal of Military Ethics 5 (2006): 201–14.
Blustein, J., The Moral Demands of Memory (Cambridge: Cambridge University Press, 2008).
Blustein, J., Forgiveness and Remembrance (Oxford: Oxford University Press, 2014).
Boraine, A., ‘Truth and Reconciliation in South Africa’, in R. I. Rotberg, and D. F.
Thompson (ed.), Truth v. Justice—The Morality of Truth Commissions (Princeton, N.J.:
Bovens, L., ‘Apologies’, Proceedings of the Aristotelian Society 108 (2008): 219–39.
Bradley, M., Refugee Repatriation—Justice, Responsibility and Redress (Cambridge:
Braithwaite, J. and Pettit, P., Not Just Deserts: A Republican Theory of Justice (Princeton, N.J.:
Brock, D. W., ‘Defending Moral Options’, Philosophy and Phenomenological Research 51
(1991): 909–13.
Brooks, R. E., ‘The New Imperialism: Violence, Norms, and the “Rule of Law”’, Michigan
Law Review 101 (2003): 2275–340.
Broomhall, B., International Justice and the International Criminal Court: Between Sovereignty
and the Rule of Law (Oxford: Oxford University Press, 2004).
Brownlie, I., Principles of Public International Law, 6th edn. (Oxford: Oxford University
Press, 2003).
Buchanan, A. E., Secession: The Morality of Political Divorce from Fort Sumter to Lithuania
and Quebec (Boulder, Col.: Westview Press, 1991).
Works Cited 319

Bunting, M., The Model Occupation—The Channel Islands under German Rule, 2nd edn.
(London: Pimlico, 2004).
Burke, E., Reflections on the Revolution in France. ed. C. O’ Brien (Harmondsworth: Penguin
Classics, 1986 [1790]).
Burke-White, W. W., ‘A Community of Courts—Towards a System of International
Criminal Law Enforcement’, Michigan Journal of International Law 24 (2002–2003):
1–101.
Burke-White, W. W., ‘Proactive Complementarity: The International Criminal Court and
National Courts in the Rome System of International Justice’, Harvard International
Law Journal 49 (2008): 53–108.
Butler, J., ‘Sermon VIII. Upon Resentment and Forgiveness of Injuries’, in J. Butler, Fifteen
Sermons Preached at the Rolls (London: J. and J. Knapton, 1726).
Butt, D., Rectifying International Injustice—Principles of Compensation and Restitution
Between Nations (Oxford: Oxford University Press, 2009).
Butt, D., ‘Colonialism and Postcolonialism’, in International Encyclopedia of Ethics (Oxford:
Blackwell Publishing Ltd, 2013).
Butt, D., ‘“A Doctrine Quite New and Altogether Untenable”: Defending the Beneficiary
Pays Principle’, Journal of Applied Philosophy 31 (2014): 336–48.
Caplan, R., International Governance of War-Torn Territories: Rule and Reconstruction
Carens, J. H., The Ethics of Immigration (Oxford: Oxford University Press, 2013).
Cassese, A., International Criminal Law, 2nd edn. (Oxford: Oxford University Press,
2008).
Celermajer, D., The Sins of the Nation and the Ritual of Apologies (Cambridge: Cambridge
Chandler, D., Empire in Denial—The Politics of State Building (London: Pluto, 2006).
Chehtman, A., The Philosophical Foundations of Extraterritorial Punishment (Oxford:
Chehtman, A., ‘Developing Local Capacity for War Crimes Trials: Insights from BiH,
Sierra Leone, and Colombia’, Stanford Journal of International Law 49 (2013):
297–329.
Chesterman, S., ‘Occupation as Liberation: International Humanitarian Law and Regime
Change’, Ethics & International Affairs 18 (2004): 51–64.
Chesterman, S., You, The People—The United Nations, Transitional Administration, and
State-Building (Oxford: Oxford University Press, 2004).
Chiu, Y., ‘Liberal Lustration’, Journal of Political Philosophy 19 (2011): 440–64.
Christensen, J., ‘Weapons, Security, and Oppression: A Normative Study of International
Arms Transfers’, Journal of Political Philosophy 23 (2015): 23–39.
Christman, J., The Myth of Property: Towards an Egalitarian Theory of Ownership (Oxford:
Chwang, E., ‘On Coerced Promises’, in Sheinman, H. (ed.), Promises and Agreements
Cicero, On Duties. ed. E. M. Atkins (Cambridge: Cambridge University Pres, 1991 [44
BC]).
Clark, N., ‘Communities Secretary Eric Pickles says 2014’s £50m World War I
Commemorations Must Not Turn Into “Anti-German Festival” ’, The Independent.
http://www.independent.co.uk/arts-entertainment/art/news/communities-secretary-eric-pickles-
says-2014s-50m-world-war-i-commemorations-must-not-turn-into-antigerman-festival-
8652405.html
320 Works Cited

Clark, P., ‘Ethnicity, Leadership and Conflict Mediation in Eastern Democratic Republic
of Congo: The Case of the Barza Inter-Communautaire’, Journal of Eastern African
Studies 2 (2008): 1–17.
Clark, P., The Gacca Courts, Post-Genocide Justice and Reconciliation in Rwanda—Justice
without Lawyers (Cambridge: Cambridge University Press, 2010).
Coady, C. A. J., ‘Escaping from the Bomb: Immoral Deterrence and the Problem of
Extrication’, in Shue, H. (ed.), Nuclear Deterrence and Moral Restraint (Cambridge:
Coady, C. A. J., Morality and Political Violence (Cambridge: Cambridge University Press, 2008).
Cockburn, P., The Occupation—War and Resistance in Iraq (London: Verso, 2007).
Cohen, G. A., ‘On the Currency of Egalitarian Justice’, Ethics 99 (1989): 916–44.
Cohen, G. A., Rescuing Justice and Equality (Cambridge, Mass.: Harvard University Press,
2008).
Coleman, J. L., Risks and Wrongs (Oxford: Oxford University Press, 2002).
Coleman, P. T., Deutsch, M. and Marcus, E. C. (ed.), The Handbook of Conflict Resolution—
Theory and Practice, 3rd edn. (San Francisco, Ca.: Jossey Bass, 2006).
Colonomos, A. and Armstrong, A., ‘German Reparations to the Jews after World War II:
A Turning Point in the History of Reparations’, in P. De Greiff (ed.), The Handbook of
Reparations (Oxford: Oxford University Press, 2006).
Conquest, R., The Harvest of Sorrow (London: Hutchinson, 1986).
Cottingham, J., ‘Varieties of Retribution’, The Philosophical Quarterly 29 (1979): 238–46.
Cousens, E. M., ‘From Missed Opportunities to Overcompensastion: Implementing the
Dayton Agreement on Bosnia’, in S. J. Stedman, D. Rotchild, and E. M. Cousens (ed.),
Ending Civil Wars—The Implementation of Peace Agreements (Boulder, Col., and London:
Lynne Rienner Publishers, 2002).
Crawford, J. and Brownlie, I., Brownlie’s Principles of Public International Law, 8th edn.
Cronin, A. K., How Terrorism Ends—Understanding the Decline and Demise of Terrorist
Campaigns (Princeton: Princeton University Press, 2009).
Cunningham, M., States of Apology (Manchester: Manchester University Press, 2014).
Daly, K., ‘Revisiting the Relationship between Retributive and Restorative Justice’, in H.
Strang, and J. Braithwaite (ed.), Restorative Justice—Philosophy to Practice (Dartmouth:
Ashgate, 2000).
Darwin, J., Unfinished Empire: The Global Expansion of Britain (London: Penguin Books,
2012).
Davis, N., ‘Rights, Permission, and Compensation’, Philosophy & Public Affairs 14 (1985):
374–84.
De Greiff, P., ‘Justice and Reparations’, in P. De Greiff (ed.), The Handbook of Reparations
De Greiff, P. (ed.), The Handbook of Reparations (Oxford: Oxford University Press, 2008).
Deak, I., Gross, J. T. and Judt, T. (ed.), The Politics of Retribution in Europe—World War II
and its Aftermath (Princeton, N.J.: Princeton University Press, 2000).
Deigh, J. ‘Promises under Fire’, Ethics 112 (2002): 483–506.
Diamond, L., ‘What Went Wrong in Iraq’, Foreign Affairs 83 (2004): 34–56.
Dinstein, Y., War, Aggression and Self-defence, 3rd edn. (Cambridge: Cambridge University
Press, 2001).
Dinstein, Y., The International Law of Belligerent Occupation (Cambridge: Cambridge
Dolinko, D., ‘Some Thoughts About Retributivism’, Ethics 101 (1991): 537–59.
Works Cited 321

Doyle, M. W., ‘Kant, Liberal Legacies, and Foreign Affairs’, Philosophy & Public Affairs 12
(1983): 205–35.
Doyle, M. W., ‘Kant, Liberal Legacies, and Foreign Affairs, Part 2’, Philosophy & Public
Affairs 12 (1983): 323–53.
Doyle, M. W., ‘Liberalism and World Politics’, The American Political Science Review 80
(1986): 1151–69.
Doyle, M. W., Ways of War and Peace: Realism, Liberalism, and Socialism (New York;
London: Norton, 1997).
Doyle, M. W., and Sambanis, Nicholas, Making War and Building Peace: United Nations
Peace Operations (Princeton: Princeton University Press, 2006).
Drumbl, M. A., Atrocity, Punishment, and International Law (Cambridge: Cambridge
Duff, R. A., Punishment, Communication, and Community (Oxford: Oxford University
Press, 2001).
Dunn, J., ‘Trust and Political Agency’, in D. Gambetta (ed.), Trust: Making and Breaking
Cooperative Relations (Oxford: Blackwell, 1988).
Easterday, J. S., ‘Peace Agreements as a Framework for Jus Post Bellum’, in J. S. Easterday,
J. Iverson, and C. Stahn (ed.), Jus Post Bellum (Oxford: Oxford University Press, 2014).
Edelstein, D. M., Occupational Hazards—Success and Failure in Military Occupation (Ithaca,
N.Y.: Cornell University Press, 2008).
Eisikovits, N., ‘Forget Forgiveness: On the Benefits of Sympathy for Political Reconciliation’,
Theoria 52 (2004): 31–63.
G. Eliot, Felix Holt the Radical, ed. D. Thorold (London: Wordsworth, 1997 [1866]).
Elster, J., Closing the Books: Transitional Justice in Historical Perspective (Cambridge:
Enoch, D., ‘Whose Right Is It? Reflections on Harel’s Reflections on Palestinians’ Interest
in Return’, Theoretical Inquiries in Law 5 (2004): 367–78.
Estlund, D., ‘Utopophobia’, Philosophy & Public Affairs 42 (2014): 113–34.
Evans, M., ‘Balancing Peace, Justice and Sovereignty in Jus Post Bellum: The Case of ‘Just
Occupation’’, Millenium—Journal of International Studies 36 (2008): 533–54.
Evans, M., ‘Moral Responsibilities and the Conflicting Demands of Jus Post Bellum’, Ethics
& International Affairs 23 (2009): 147–64.
Evans, M., ‘At War’s End’, in J. S. Easterday, J. Iverson, and C. Stahn (ed.), Just Post Bellum
Fabre, C., Whose Body is it Anyway? Justice and the Integrity of the Person (Oxford: Oxford
Fabre, C., Justice in a Changing World (Cambridge: Polity Press, 2007).
Fabre, C., ‘Mandatory Rescue Killings’, Journal of Political Philosophy 15 (2007): 363–84.
Fabre, C., ‘Guns, Food, and Liability to Attack’, Ethics 120 (2009): 36–63.
Fabre, C., Cosmopolitan War (Oxford: Oxford University Press, 2012).
Fabre, C., ‘Internecine War Killings’, Utilitas 24 (2012): 214–36.
Fabre, C., ‘Cosmopolitanism and Wars of Self-Defence’, in C. Fabre and S. Lazar (ed.), The
Morality of Defensive War (Oxford: Oxford University Press, 2014).
Fabre, C., ‘Rights, Justice and War’, Law and Philosophy 33 (2014): 391–425.
Farah, D., ‘Papers Show U.S. Role in Guatemalan Abuses’, Washington Post. http://www.
washingtonpost.com/wp-srv/inatl/daily/march99/guatemala11.htm (7/02/2014).
Farmer, S., ‘Postwar Justice in France: Bordeaux 1953’, in I. Deak, J. T. Gross, and J. Judt
(ed.), The Politics of Retribution in Europe: World War II and its Aftermath (Princeton,
322 Works Cited

Feinberg, J., ‘The Expressive Function of Punishment’, The Monist 49 (1965): 397–423.
Feinberg, J., ‘Voluntary Euthanasia and the Inalienable Right to Life’, Philosophy & Public
Affairs 7 (1978): 93–123.
Feldman, L. G., ‘The Principle and Practice of “Reconciliation” in German Foreign Policy:
Relations with France, Israel, Poland and the Czech Republic’, International Affairs
(Royal Institute of International Affairs 1944–) 75 (1999): 333–56.
Feldman, N., What We Owe Iraq—War and the Ethics of Nation Building (Princeton, N.J.:
Fiddian-Quasmiyeh, E., Loescher, G. Milner, J., Long, K. and Sigona, N. (ed.), The Oxford
Handbook of Refugee and Forced Migration Studies (Oxford: Oxford University Press,
2014).
Findlay, T., The Use of Force in UN Peace Operations (Oxford: Oxford University Press,
2002).
Firth, J. and Quong, J., ‘Necessity, Moral Liability, and Defensive Harm’, Law and Philosophy
31 (2012): 673–701.
Fishkin, J. S., ‘Justice Between Generations: Compensation, Identity, and Group
Membership’, in J. W. Chapman (ed.), Compensatory Justice (New York: New York
Flikschuh, K., ‘On the Cogency of Human Rights’, Jurisprudence 2 (2011): 17–36.
Fortescue, W., The Third Republic in France, 1870–1940: Conflicts and Continuities
(London: Routledge, 2000).
Fortna, V. P., Does Peacekeeping Work? (Princeton, N.J.: Princeton University Press, 2008).
Fox, G. H., ‘The Occupation of Iraq’, Georgetown Journal of International Law 36 (2005):
195–297.
Fox, G. H., Humanitarian Occupation (Cambridge: Cambridge University Press, 2008).
Freeman, M., Necessary Evils—Amnesties and the Search for Justice (Cambridge: Cambridge
Freire, M. R. and Lopes, P. D., ‘Peacebuilding in Timor-Leste: Finding a Way between
External Intervention and Local Dynamics’, International Peacekeeping 20 (2013):
204–18.
Fried, C., Contract as Promise: A Theory of Contractual Obligation (Cambridge, Mass.:
Harvard University Press, 1981).
Frowe, H., Defensive Killing (Oxford University Press: Oxford, 2014).
Fullinwider, R., ‘Preferential Hiring and Compensation’, Social Theory and Practice 3
(1975): 307–20.
Galtung, J., Essays in Peace Research (Copenhagen: Ejlers, 1975).
Gans, C., The Limits of Nationalism (Oxford: Oxford University Press, 2003).
Gans, C., ‘The Palestinian Right of Return and the Justice of Zionism’, Theoretical Inquiries
in Law 5 (2004): 269–304.
Gaus, G. F., ‘Does Compensation Restore Equality?’, in J. W. Chapman (ed.), Compensatory
Justice (New York: New York University Press, 1991).
Gerson, A., ‘Peace Building: The Private Sector’s Role’, The American Journal of International
Law 95 (2001): 102–19.
Gilabert, P. and Lawford-Smith, H., ‘Political Feasibility: A Conceptual Exploration’,
Political Studies 60 (2012): 809–25.
Gildea, R., Marianne in Chains—Daily Life in the Heart of France during the German
Occupation (New York: Picador, 2002).
Goldman, A. H., ‘The Paradox of Punishment’, Philosophy & Public Affairs 9 (1979):
42–58.
Works Cited 323

Goodin, R. E., ‘What Is So Special about Our Fellow Countrymen?’, Ethics 98 (1988):
663–86.
Goodin, R. E., ‘Theories of Compensation’, Oxford Journal of Legal Studies 9 (1989): 56–75.
Goodin, R. E. and Barry, C., ‘Benefiting from the Wrongdoing of Others’, Journal of
Applied Philosophy 31 (2014): 363–76.
Goodin, R. E. and Lepora, C., On Complicity and Compromise (Oxford: Oxford University
Press, 2013).
Govier, T., ‘Forgiveness and the Unforgivable’, American Philosophical Quarterly 36 (1999):
59–75.
Govier, T., Forgiveness and Revenge (London: Routledge, 2002).
Greenawalt, K., ‘Amnesty’s Justice’, in R. I. Rotberg, and D. F. Thompson (ed.), Truth v.
Justice—The Morality of Truth Commissions (Princeton, N.J.: Princeton University Press,
2000).
Gregory, A., The Silence of Memory—Armistice Day, 1919–1946 (Oxford: Berg, 1994).
Griswold, C. L., Forgiveness—A Philosophical Exploration (Cambridge: Cambridge
Grotius, H., The Rights of War and Peace. ed. R. Tuck (Indianapolis, IN: Liberty Fund, 2005
[1625]).
Grotius, H., Commentary on the Law of Prize and Booty. ed. M. J. van Ittersum (Indianapolis:
Liberty Fund, 2006 [1606]).
Hampton, J., ‘The Moral Education Theory of Punishment’, Philosophy & Public Affairs 13
(1984): 208–38.
Hampton, J. and Murphy, J. G., Forgiveness and Mercy (Cambridge: Cambridge University
Press, 1988).
Haque, A. A., ‘Law and Morality at War’, Criminal Law and Philosophy 8 (2014): 79–97.
Hardin, R., Trust and Trustworthiness (New York: Russell Sage Foundation, 2002).
Harel, A., ‘Whose Home Is It? Reflections on the Palestinians’ Interest in Return’, Theoretical
Inquiries in Law 5 (2004): 333–66.
Hart, H. L. A., ‘Are There Any Natural Rights?’, The Philosophical Review 64 (1955):
175–91.
Haydar, B. and Øverland, G., ‘The Normative Implications of Benefiting from Injustice’,
Journal of Applied Philosophy 31 (2014): 349–62.
Hayner, P. B., Unspeakable Truths: Transitional Justice and the Challenge of Truth Commissions,
2nd edn. (London: Routledge, 2011).
He, Y., The Search for Reconciliation—Sino-Japanese and German-Polish Relations since
WWII (Cambridge: Cambridge University Press, 2009).
Hellmueller, S., ‘The Power of Perceptions: Localizing International Peacebuilding
Approaches’, International Peacekeeping 20 (2013): 219–32.
Herman, B., ‘Contigency in Obligation’, in B. Herman, Moral Literacy (Cambridge, Mass.:
Harvard University Press, 2007).
Hermanson, C., ‘Duties in the Wake of Atrocity: A Normative Analysis of Post-Atrocity
Peacebuilding’ (D.Phil thesis, University of Oxford, 2012).
Hieronymi, P., ‘Articulating an Uncompromising Forgiveness’, Philosophy and Phenomenological
Research 62 (2001): 529–55.
Hobbes, T., Leviathan, ed. N. Malcolm (Oxford: Oxford University Press, 2012 [1651]).
Hohe, T., ‘Justice without judiciary in East Timor’, Conflict, Security & Development
3 (2003): 335–57.
Honore, T., ‘Ownership’, in A. G. Guest (ed.), Oxford Essays in Jurisprudence (Oxford:
324 Works Cited

Howard, M. L., UN Peacekeeping in Civil Wars (Cambridge: Cambridge University Press,
2008).
Human Rights Watch, ‘Israel/Occupied Territories: Jenin War Crimes Investigation
Needed’, Israel/Occupied Territories: Jenin War Crimes Investigation Needed (London:
Human Rights Watch, 2002).
Hume, D., A Treatise of Human Nature. ed. P. H. Nidditch (Oxford: Clarendon Press, 1976
[1740]).
Hurka, T., ‘The Justification of National Partiality’, in R. McKim and J. McMahan (ed.),
The Morality of Nationalism (Oxford: Oxford University Press, 1997).
Hurka, T., ‘Proportionality in the Morality of War’, Philosophy & Public Affairs 33 (2005):
34–66.
Ignatieff, M., Empire Lite—Nation-Building in Bosnia, Kosovo and Afghanistan (London:
Vintage, 2003).
Iklé, F. C., Every War Must End, 2nd revised edn. (New York: Columbia University Press,
2005).
Iverson, J., ‘Contrasting the Normative and Historical Foundations of Transitional Justice
and Jus Post Bellum’, in J. S. Easterday, J. Iverson, and C. Stahn (ed.), Jus Post Bellum
Jensen, M., ‘The Limits of Practical Possibility’, Journal of Political Philosophy 17 (2009):
168–84.
Johnson, D. P. and Tierney, D., Failing to Win: Perceptions of Victory and Defeat in
International Politics (Cambridge, Mass.: Harvard University Press, 2006).
Johnson, J. T., Ideology, Reason, and the Limitation of War: Religious and Secular Concepts,
1200–1740 (Princeton, N.J.: Princeton University Press, 1975).
Johnson, J. T., Just War Tradition and the Restraint of War: A Moral and Historical Inquiry
(Princeton, N.J.: Princeton University Press, 1981).
Johnstone, G. and Van Ness, D. W. (ed.), Handbook of Restorative Justice (Portland, Or.:
Willan Publishing, 2007).
Jones, J., ‘The Artistic Jingoism of the Bomber Command Memorial’, The Guardian. http://
www.theguardian.com/commentisfree/2012/jun/29/bomber-command-memorial-
artistic-jingoism.
Jones, K., ‘Trust as an Affective Attitude’, Ethics 107 (1996): 4–25.
Judt, T., Postwar—A History of Europe since 1945 (London: Pimlico, 2007).
Kagan, S., ‘Causation and Responsibility’, American Philosophical Quarterly 25 (1988):
293–302.
Kaldor, M., New & Old Wars, 2nd edn. (Cambridge: Polity, 2006).
Kamm, F. M., The Moral Target: Aiming at Right Conduct in War and Other Conflicts
Kant, I., ‘Perpetual Peace: A Philosophical Sketch’, in H. Reiss (ed.), Kant—Political
Writings (Cambridge: Cambridge University Press, 1991 [1795]).
Kant, I., ‘Metaphysics of Morals’, in H. Reiss (ed.), Kant—Political Writings (Cambridge:
Cambridge University Press, 1991 [1797]).
Kant, I., ‘On the Common Saying: That May be Correct in Theory but It Is No Use in
Practice’, in M. J. McGregor and A. W. Wood (ed.), Practical Philosophy (Cambridge:
Cambridge University Press, 1996 [1793]).
Kecskemeti, P., Strategic Surrender: The Politics of Victory and Defeat (Stanford: Stanford
Keller, A., ‘Justice, Peace and History: A Reappraisal’, in P. Allan and A. Keller (ed.), What
is a Just Peace? (Oxford: Oxford University Press, 2008).
Works Cited 325

Kelman, H. C., ‘Building Trust Among Enemies: The Central Challenge for International
Conflict Resolution’, International Journal of Intercultural Relations 29 (2005): 629–50.
Kershaw, I., The End: Hitler’s Germany, 1944–45 (London: Allen Lane, 2011).
Kimmel, D., From Promise to Contract—Towards a Liberal Theory of Contract (Oxford: Hart
Publishing, 2003).
Kiss, E., ‘Moral Ambition Within and Beyond Political Constraints’, in R. I. Rotberg and
D. F. Thompson (ed.), Truth v. Justice—The Morality of Truth Commissions (Princeton,
Kohn, M., ‘Colonialism’, Stanford Encyclopedia of Philosophy (2012), available at http://
plato.stanford.edu/archives/sum2012/entries/colonialism%3E
Kolodny, N. and Wallace, J., ‘Promises and Practices Revisited’, Philosophy & Public Affairs
31 (2003): 120–54.
Kumar, R., ‘Why Reparations?’, in J. Oberdiek (ed.), Philosophical Foundations of the Law
of Torts (Oxford: Oxford University Press, 2014).
Kunz, J. L., ‘Ending the War with Germany’, The American Journal of International Law 46
(1952): 114–19.
Kutz, C., Complicity: Ethics and Law for a Collective Age (Cambridge: Cambridge University
Press, 2000).
Kutz, C., ‘Justice in Reparations: The Cost of Memory and the Value of Talk’, Philosophy &
Public Affairs 32 (2004): 277–312.
Lacey, N. and Pickard, H., ‘From the Consulting Room to the Court Room? Taking the
Clinical Model of Responsibility Without Blame into the Legal Realm’, Oxford Journal
of Legal Studies 33 (2013): 1–29.
Lacey, N. and Pickard, H., ‘To Blame or to Forgive? Reconciling Punishment and
Forgiveness in Criminal Justice’, Oxford Journal of Legal Studies 35 (2015): 665–96.
Lawand, K., ‘The Right to Return of Palestinians in International Law’, International
Journal of Refugee Law 8 (1996): 532–68.
Lawford-Smith, H., ‘Understanding Political Feasibility’, Journal of Political Philosophy 21
(2013): 243–59.
Lazar, S., ‘Responsibility, Risk, and Killing in Self-Defense’, Ethics 119 (2009): 699–728.
Lazar, S., ‘Scepticism about Jus Post Bellum’, in L. May, and A. T. Forcehimes (ed.), Morality,
Jus Post Bellum, and International Law (Cambridge: Cambridge University Press, 2012).
Lazar, S., Sparing Civilians (Oxford: Oxford University Press, 2015).
Lazare, A., On Apology (Oxford: Oxford University Press, 2004).
Levinson, S., ‘“Responsibility for Crimes of War”’, Philosophy & Public Affairs 2 (1973):
244–73.
Levinson, S., ‘Trials, Commissions, and Investigating Committees—The Elusive Search for
Norms of Due Process’, in R. I. Rotberg and D. F. Thompson (ed.), Truth v. Justice—The
Morality of Truth Commissions (Princeton, N.J.: Princeton University Press, 2000).
Linfield, S., ‘Memory’s Lair’, Boston Review, 1 June 2003.
List, C. and Pettit, P., Group Agency—The Possibility, Design, and Status of Corporate Agents
Llewellyn, J. J., ‘Truth commissions and restorative justice’, in G. Johnstone and D. W. Van
Ness (ed.), Handbook of Restorative Justice (Portland, Or.: Willan Publishing, 2007).
Llewellyn, J. J. and Philpott, D., ‘Restorative Justice and Reconciliation: Twin Frameworks
for Peacebuilding’, in J. J. Llewellyn and D. Philpott (ed.), Restorative Justice,
Reconciliation, and Peacebuilding (Oxford: Oxford University Press, 2014).
Llewellyn, J. J. and Philpott, D. (ed.), Restorative Justice, Reconciliation, and Peacebuilding
326 Works Cited

Long, K., The Point of No Return: Refugees, Rights, and Repatriation (Oxford: Oxford
Lowe, A. V., International Law (Oxford: Oxford University Press, 2007).
Luban, D., ‘A Theory of Crimes against Humanity’, Yale Journal of International Law 85
(2004): 85–168.
Luban, D., ‘Fairness to Rightness: Jurisdiction, Legality, and the Legitimacy of International
Criminal Law’, in J. Tasioulas and S. Besson (ed.), The Philosophy of International Law
Luban, D., ‘War as Punishment’, Philosophy & Public Affairs 39 (2011): 299–330.
MacDonogh, G., After the Reich—From the Liberation of Vienna to the Berlin Airlift
(London: John Murray Publishers, 2007).
Macmillan, M., Peacemakers—Six Months that Changed the World (London: John Murray,
2001).
MacQueen, N., Peacekeeping and the International System (London: Routledge, 2006).
Mallinder, L., ‘Amnesties in the Pursuit of Reconciliation, Peacebuilding and Restorative
Justice’, in J. J. Llewellyn and D. Philpott (ed.), Restorative Justice, Reconciliation, and
Peacebuilding (Oxford: Oxford University Press, 2014).
Mandel, R., ‘Reassessing Victory in Warfare’, Armed Forces and Society 33 (2007):
461–96.
Margalit, A., The Ethics of Memory (Cambridge, Mass.: Harvard University Press, 2004).
Marten, K. Z., Enforcing the Peace: Learning from the Imperial Past (New York: Columbia
Matthes, E. H., ‘History, Value, and Irreplaceability’, Ethics 124 (2013): 35–64.
May, L., Crimes against Humanity: A Normative Account (Cambridge: Cambridge University
Press, 2005).
May, L., War Crimes and Just War (Cambridge: Cambridge University Press, 2007).
May, L., Aggression and Crimes Against Peace (Cambridge: Cambridge University Press,
2008).
May, L., After War Ends—A Philosophical Perspective (Cambridge: Cambridge University
Press, 2012).
May, L., ‘Jus Post Bellum, Grotius, and Meionexia’, in J. S. Easterday, J. Iversonand C. Stahn
(ed.), Jus Post Bellum (Oxford: Oxford University Press, 2014).
Mayeur, J.-M. and Reberioux, M., The Third Republic in France, 1870–1940: Conflicts and
Continuities (Cambridge: Cambridge University Press, 1988).
McCready, D., ‘Ending the War Right: Jus Post Bellum and the Just War Tradition’, Journal
of Military Ethics 8 (2009): 66–78.
McMahan, J., ‘Self-Defense and the Problem of the Innocent Attacker’, Ethics 104 (1994):
252–90.
McMahan, J., ‘War as Self-Defense’, Ethics & International Affairs 18 (2004): 75–80.
McMahan, J., ‘Just Cause for War’, Ethics & International Affairs 19 (2005): 1–21.
McMahan, J., ‘The Morality of War and the Law of War’, in D. Rodin and H. Shue (ed.),
Just and Unjust Warriors (Oxford: Oxford University Press, 2008).
McMahan, J., Killing in War (Oxford: Oxford University Press, 2009).
McMahan, J., ‘The Morality of Military Occupation’, Loyola International and Comparative
Law Review 31 (2009): 101–23.
McMahan, J., ‘The Just Distribution of Harm Between Combatants and Noncombatants’,
Philosophy & Public Affairs 38 (2010): 342–79.
McMahan, J., ‘Proportionality and Time’, Ethics 125 (2015): 696–719.
Works Cited 327

Megret, F., ‘Should Rebels be Amnestied?’, in J. S. Easterday, J. Iverson and C. Stahn (ed.),
Jus Post Bellum (Oxford: Oxford University Press, 2014).
Meierhenrich, J., ‘The Ethics of Lustration’, Ethics & International Affairs 20 (2006):
99–120.
Meisels, T., Territorial Rights (Dordrecht: Springer, 2005).
Melvern, L., Conspiracy to Murder: The Rwandan Genocide (London: Verso, 2006).
Merridale, C., Night of Stone—Death and Memory in Twentieth Century Russia (London:
Penguin Books, 2000).
Mihai, M. and Thaler, M. (ed.), On the Uses and Abuses of Political Apologies (Basingstoke:
Palgrave, 2014).
Millar, G., van der Lijn, J. Ø. and Verkoren, W., ‘Peacebuilding Plans and Local
Reconfigurations: Frictions between Imported Processes and Indigenous Practices’,
International Peacekeeping 20 (2013): 137–43.
Miller, D., ‘Political Philosophy for Earthlings’, in D. Leopold M. Stears (ed.), Political
Theory: Methods and Approaches (Oxford: Oxford University Press, 2008).
Miller, R. W., ‘The Ethics of America’s Afghan War’, Ethics & International Affairs 25
(2011): 103–32.
Minear, R. H., Victor’s Justice—The Tokyo War Crimes Trial (Princeton, N.J.: Princeton
Minow, M., Between Vengeance and Forgiveness—Facing History after Genocide and Mass
Violence (Boston: Beacon Press, 1998).
Minow, M., ‘The Hope for Healing—What Can Truth Commissions Do?’, in R. I. Rotberg
and D. F. Thompson (ed.), Truth v. Justice—The Morality of Truth Commissions (Princeton,
Moellendorf, D., ‘Reconciliation as a Political Value’, Journal of Social Philosophy 38 (2007):
205–21.
Moellendorf, D., ‘Jus ex Bello’, Journal of Political Philosophy 16 (2008): 123–36.
Moellendorf, D., ‘Two Doctrines of Jus Ex Bello’, Ethics 125 (2015): 653–73.
Montague, P., ‘Rights and Duties of Compensation’, Philosophy & Public Affairs 13 (1984):
79–88.
Moore, M., A Political Theory of Territory (Oxford: Oxford University Press, 2015).
Moore, M. (ed.), National Self-Determination and Secession (Oxford: Oxford University
Press, 1998).
Moore, M. S., A Theory of the Criminal Law (Oxford: Oxford University Press, 2010).
Morris, B., The Birth of the Palestinian Refugee Problem Revisited, 2nd edn. (Cambridge:
Morris, H., ‘Persons and Punishment’, The Monist 52 (1968): 475–501.
Mulgan, T., ‘Two Conceptions of Benevolence’, Philosophy & Public Affairs 26 (1997):
62–79.
Munzer, S. R., A Theory of Property (New York: Cambridge University Press, 1990).
Murphy, C., A Moral Theory of Political Reconciliation (Cambridge: Cambridge University
Press, 2010).
Murphy, J. G., Getting Even—Forgiveness and its Limits (Oxford: Oxford University Press,
2003).
Murphy, J. G., Punishment and the Moral Emotions—Essays in Law, Morality, and Religion
Murphy, L. B., ‘The Demands of Beneficence’, Philosophy & Public Affairs 22 (1993):
267–92.
328 Works Cited

Murphy, R., UN Peacekeeping in Lebanon, Somalia and Kosovo—Operational and Legal
Issues in Practice (Cambridge: Cambridge University Press, 2007).
Nabulsi, K., Traditions of War: Occupation, Resistance, and the Law (Oxford: Oxford
Nagel, T., ‘What Makes a Political Theory Utopian?’, Social Research 56 (1989): 903–20.
Neff, S. C., War and the Law of Nations: A General History (Cambridge: Cambridge
Neff, S. C., ‘Conflict Termination and Peace-making in the Law of Nations: A Historical
Perspective’, in C. Stahn and J. K. Kleffner (ed.), Jus Post Bellum—Towards a Law of
Transition from Conflict to Peace (The Hague: T.M.C. Asser Press, 2008).
Nobles, M., The Politics of Official Apologies (Cambridge: Cambridge University Press,
2008).
North, J., ‘Wrongdoing and Forgiveness’, Philosophy 62 (1987): 499–508.
Nozick, R., Anarchy, State, and Utopia (New York: Basic Books, 1974).
Offe, C., ‘Disqualification, Retribution, Restitution: Dilemmas of Justice in Post-
Communist Transitions’, Journal of Political Philosophy 1 (1993): 17–44.
Orend, B., ‘Jus Post Bellum’, Journal of Social Philosophy 31 (2000): 117–37.
Orend, B., ‘Justice after War’, Ethics & International Affairs 16 (2002): 43–56.
Orend, B., The Morality of War (Peterborough, On.: Broadview Press, 2006).
Osborne, R., Herbert von Karajan—A Life in Music (London: Random House, 1998).
Pappé, I., The Forgotten Palestinians: A History of the Palestinians in Israel (New Haven, CT:
Yale University Press, 2011).
Parfit, D., Reasons and Persons (Oxford: Clarendon, 1984).
Parfit, D., On What Matters, vol. 1 (Oxford: Oxford University Press, 2011).
Paris, R., At War’s End—Building Peace After Civil Conflict (Cambridge: Cambridge
Pasternak, A., ‘Voluntary Benefits from Wrongdoing’, Journal of Applied Philosophy 31
(2014): 377–91.
Patten, A., ‘Democratic Secession from a Multinational State’, Ethics 112 (2002): 558–86.
Pattison, J., ‘Jus Post Bellum and the Responsibility to Rebuild’, British Journal of Political
Science 45 (2013): 635–61.
Paul, E. F., ‘Set-asides, Reparations, and Compensatory Justice’, in J. W. Chapman (ed.),
Compensatory Justice (New York: New York University Press, 1991).
Peled, Y. and Rouhana Nadim, N., ‘Transitional Justice and the Right of Return of the
Palestinian Refugees’, Theoretical Inquiries in Law 5 (2004): 317–32.
Pendas, D. O., The Frankfurt Auschwitz Trial 1963–65: Genocide, History, and the Limits of
Law (Cambridge: Cambridge University Press, 2006).
Pettit, P., ‘The Cunning of Trust’, Philosophy & Public Affairs 24 (1995): 202–25.
Phillips, A. L., ‘The Politics of Reconciliation: Germany in Central-east Europe’, German
Politics 7 (1998): 64–85.
Philpott, D., Just and Unjust Peace: An Ethic of Political Reconciliation (Oxford: Oxford
Pitts, J., A Turn to Empire: The Rise of Imperial Liberalism in Britain and France (Princeton,
Pufendorf, On the Duty of Man and Citizen. ed. J. Tully (Cambridge: Cambridge University
Press, 1991 [1673]).
Quinn, W., ‘The Right to Threaten and the Right to Punish’, Philosophy & Public Affairs
14 (1985): 327–73.
Quong, J., ‘Liability to Defensive Harm’, Philosophy & Public Affairs 40 (2012): 45–77.
Works Cited 329

Radley, K. R., ‘The Palestinian Refugees: The Right to Return in International Law’,
American Journal of International Law 72 (1978): 586–614.
Räikkä, J., ‘The Feasibility Condition in Political Theory’, Journal of Political Philosophy 6
(1998): 27–40.
Rawls, J., A Theory of Justice (Harvard: Harvard University Press, 1971).
Rawls, J., The Law of Peoples (Cambridge, Mass.: Harvard University Press, 1999).
Rawls, J., Justice as Fairness: A Restatement. ed. E. Kelly (Cambridge, Mass.: Harvard
Reinhard, W., A Short History of Colonialism (Manchester: Manchester University Press,
2011).
Reiter, A. G., ‘Examining the Use of Amnesties and Pardons as a Response to Internal
Armed Conflict’, Israel Law Review 47 (2014): 133–47.
Renzo, M., ‘Crimes Against Humanity and the Limits of International Criminal Law’, Law
and Philosophy 31 (2012): 443–76.
Reydams, L., Universal Jurisdiction: International and Municipal Legal Perspectives (Oxford:
Rigby, A., Justice and Reconciliation After Violence (Boulder, Col.: Lynne Riener: 2001).
Rings, W., Living with the Enemy—Collaboration and Resistance in Hitler’s Europe 1939–
1945 (London: Weidenfeld and Nicolson, 1982).
Roberts, A., ‘What Is a Military Occupation?’, British Yearbook of International Law 55
(1984): 249–305.
Roberts, A., ‘Transformative Military Occupation: Applying the Laws of War and Human
Rights’, The American Journal of International Law 100 (2006): 580–622.
Roberts, A. and Guelff, R. (ed.), Documents on the Laws of War, 3rd edn. (Oxford: Oxford
Rocheleau, J., ‘From Aggression to Just Occupation? The Temporal Application of Jus Ad
Bellum Principles and the Case of Iraq’, Journal of Military Ethics 9 (2010): 123–38.
Rodin, D., War and Self-Defense (Oxford: Clarendon Press, 2002).
Rodin, D., ‘Two Emerging Issues of Jus Post Bellum: War Termination and the Liability of
Soldiers for Crimes of Aggression’, in C. Stahn and J. K. Kleffner (ed.), Jus Post Bellum—
Towards a Law of Transition from Conflict to Peace (The Hague: T.M.C. Asser Press,
2008).
Rodin, D., ‘Ending Wars’, Ethics & International Affairs 25 (2011): 359–67.
Rodin, D., ‘The Reciprocity Theory of Rights’, Law and Philosophy 33 (2014): 281–308.
Rodin, D., ‘The War Trap: Dilemmas of jus terminatio’, Ethics 125 (2015): 674–95.
Rodin, D. and Shue, H. (ed.), Just and Unjust Warriors (Oxford: Oxford University Press,
2008).
Ronen, Y., ‘Post Occupation Law’, in J. S. Easterday, J. Iverson,and C. Stahn (ed.), Jus Post
Bellum (Oxford: Oxford University Press, 2014).
Rosato, S., ‘The Flawed Logic of Democratic Peace Theory’, American Political Science
Review 97 (2003): 585–602.
Rosenberg, T., ‘Overcoming the Legacies of Dictatorship’, Foreign Affairs 74 (1995):
134–53.
Russell, F. H., The Just War in the Middle Ages (Cambridge: Cambridge University Press,
1975).
Ryan, C. C., ‘The Normative Concept of Coercion’, Mind 89 (1980): 481–98.
Santayana, G., The Life of Reason (Gutenberg Project, 1905).
Sassòli, M., ‘Legislation and Maintenance of Public Order and Civil Life by Occupying
Powers’, European Journal of International Law 16 (2005): 661–94.
330 Works Cited

Sassoon, S., ‘On Passing the Menin Gate’, in The War Poems. ed. R. Hart-Davis (London:
Faber and Faber, 1983 [1928]).
Savage, K., ‘The Politics of Memory: Black Emancipation and the Civil War Monument’,
in J. R. Gillis (ed.), Commemorations—the Politics of National Identity (Princeton, N.J.:
Scanlon, T., What We Owe to Each Other (Cambridge, Mass.: Harvard University Press,
1998).
Scanlon, T. M., ‘Reasons, Responsibility, and Reliance: Replies to Wallace, Dworkin, and
Deigh’, Ethics 112 (2002): 507–28.
Schabas, W., Unimaginable Atrocities: Justice, Politics, and Rights at the War Crimes Tribunals
Scheffer, D., All the Missing Souls—A Personal History of the War Crimes Tribunals (Princeton,
Scheffer, D. J., ‘Beyond Occupation Law’, The American Journal of International Law 97
(2003): 842–60.
Schwartz, D., ‘The Justice of Peace Treaties’, The Journal of Political Philosophy 20 (2012):
273–92.
Shafer-Landau, R., ‘The Failure of Retributivism’, Philosophical Studies: An International
Journal for Philosophy in the Analytic Tradition 82 (1996): 289–316.
Sher, G., ‘Transgenerational Compensation’, Philosophy & Public Affairs 33 (2005):
181–200.
Shlaim, A., ‘The Rise and Fall of the Oslo Peace Process’, in L. Fawcett (ed.), International
Relations of the Middle East (Oxford: Oxford University Press, 2013).
Shue, H., Basic Rights: Subsistance, Affluence, and U.S. Foreign Policy (Princeton, N.J.:
Shue, H., ‘Mediating Duties’, Ethics 98 (1988): 687–704.
Shue, H., ‘Do We Need a “Morality of War”?’, in D. Rodin and H. Shue (ed.), Just and
Unjust Warriors (Oxford: Oxford University Press, 2008).
Simmons, A. J., ‘Locke and the Right to Punish’, Philosophy & Public Affairs 20 (1991):
311–49.
Simpson, G. J., Law, War and Crime: War Crimes Trials and the Reinvention of International
Law (Cambridge: Polity, 2007).
Smilansky, S., ‘Morally, should we prefer never to have existed?’, Australasian Journal of
Philosophy 91 (2013): 655–66.
Smith, D. L., Constitutional Royalism and the Search for Settlement (Cambridge: Cambridge
Smith, N., I Was Wrong—The Meaning of Apologies (Cambridge: Cambridge University
Press, 2008).
Statman, D., ‘Fabre’s Crusade for Justice: Why We Should Not Join’, Law and Philosophy
33 (2014): 337–60.
Stevenson, D., 1914–1918: The History of the First World War (London: Allen Lane, 2004).
Stiglitz, J. and Bilmes, L., The Three Trillion Dollar War: The True Cost of the Iraq Conflict
Stilz, A., ‘Authority, Self-Determination, and Community in Cosmopolitan War’, Law and
Philosophy (2013): 1–27.
Stilz, A., ‘Occupancy Rights and the Wrong of Removal’, Philosophy & Public Affairs 41
(2013): 324–56.
Strachan, H., The First World War, vol. 1 —To Arms (Oxford: Oxford University Press,
2001).
Works Cited 331

Strang, H. and Braithwaite, J. (ed.), Restorative Justice—Philosophy to Practice (Darthmouth:
Ashgate, 2000).
Suarez, F., ‘De Bello’, in J. B. Scott (ed.), Selections from Three Works of Francisco Suarez, S.J: De
Triplici Virtute Theologica, Fide, Spe, et Charitate (Oxford: Clarendon Press, 1944 [1621]).
Tadros, V., The Ends of Harm: The Moral Foundations of Criminal Law (Oxford: Oxford
Tansey, O., Regime-Building: Democratization and International Administration (Oxford:
Tavuchis, N., Mea Culpa—A Sociology of Apology and Reconciliation (Stanford: Stanford
Teitel, R., Transitional Justice (Oxford: Oxford University Press, 2000).
Thompson, J., ‘The Apology Paradox’, The Philosophical Quarterly 50 (2000): 470–5.
Thompson, J., Taking Responsibility for the Past: Reparation and Historical Justice (Cambridge:
Polity, 2002).
Thompson, J., ‘Is Political Apology a Sorry Affair?’, Social and Legal Studies 21 (2012):
215–24.
Thomson, J. J., ‘Preferential Hiring’, Philosophy & Public Affairs 2 (1973): 364–84.
Thomson, J. J., ‘Rights and Compensation’, Noûs 14 (1980): 3–15.
Thomson, J. J., The Realm of Rights (Cambridge, Mass.: Harvard University Press, 1990).
Tuck, R., The Rights of War and Peace: Political Thought and the International Order from
Grotius to Kant (New York: Oxford University Press, 1999).
Tusa, A. and Tusa, J., The Nuremberg Trial (New York: Skyhorse Publishing, 2010).
Tutu, D., No Future without Forgiveness (London: Rider Books, 1999).
U. N. ‘Report of the Panel on United Nations Peacekeeping.’ 2000.
Vattel, E. d., Le droit des gens, ou, Principes de la loi naturelle appliques à la conduite et aux
affaires des nations et des souverains. ed. A. Geouffre de Lapradelle (Washington, D.C.:
Carnegie Institute, 1916 [1758]).
Vernon, R., ‘What is a Crime Against Humanity?’, Journal of Political Philosophy 10 (2002):
231–49.
Vitoria, F., ‘On the Law of War’, in A. Pagden and J. Lawrance (ed.), Political Writings
(Cambridge: Cambridge University Press, 1991 [1539]).
Voltaire, Philosophical Dictionary. ed. J. Morley and T. Smollett (New York: E.R. DuMont,
1901 [1764]).
von Hirsch, A., Censure and Sanctions (Oxford: Oxford University Press, 1993).
Waldron, J., The Right to Private Property (Oxford: Oxford University Press, 1988).
Waldron, J., ‘Superseding Historic Injustice’, Ethics 103 (1992): 4–28.
Waldron, J., ‘Redressing Historic Injustice’, The University of Toronto Law Journal 52
(2002): 135–60.
Waldron, J. ‘Settlement, Return, and the Supersession Thesis’, Theoretical Inquiries in Law
(2004): 237.
Walker, M., Moral Repair—Reconstructing Moral Relations after Wrongdoing (Cambridge:
Wallace, J., The View from Here—On Affirmation, Attachment, and the Limits of Regret
Walzer, M., ‘Prisoners of War: Does the Fight Continue after the Battle?’, The American
Political Science Review 63 (1969): 777–86.
Walzer, M., ‘Just and Unjust Occupations’, Dissent, Winter 2004.
Walzer, M., Just and Unjust Wars—A Moral Argument with Historical Illustrations, 4th edn.
(New York: Basic Books, 2006).
332 Works Cited

Walzer, M., ‘The Aftermath of War’, in E. Patterson (ed.), Ethics Beyond War’s End (Cambridge:
Wang, D., China’s Unequal Treaties: Narrating National History (Lanham, Md.: Rowman &
Littlefield, 2005).
Weinberg, G. L., A World at Arms: A Global History of World War II (Cambridge: Cambridge
Weinstock, D., ‘Building Trust in Divided Societies’, Journal of Political Philosophy 7 (1999):
287–307.
Weitekamp, E. G. and Kerner, H.-J. (ed.), Restorative Justice—Theoretical Foundations
(Portland, Or.: Willan Publishing, 2002).
Wellman, C., ‘A Defence of Secession and Political Self-Determination’, Philosophy &
Public Affairs 24 (1995).
Wellman, C., ‘Rights and State Punishment’, The Journal of Philosophy CVI (2009): 419–39.
Wellman, C. H., ‘The Rights Forfeiture Theory of Punishment’, Ethics 122 (2012): 371–93.
Wenar, L., ‘Reparations for the Future’, Journal of Social Philosophy 37 (2006): 396–405.
Wertheimer, A., Coercion (Princeton, N.J.: Princeton University Press, 1987).
Whitman, J. Q., The Verdict of Battle—The Law of Victory and the Making of Modern War
(Cambridge, Mass.: Harvard University Press, 2012).
Wiesenthal, S., The Sunflower (London: W. H. Allen, 1970).
Wilde, R., International Territorial Administration: How Trusteeship and the Civilizing
Mission Never Went Away (Oxford: Oxford University Press, 2008).
Williams, B. A. O., ‘Moral Luck’, in B. A. O. Williams, Moral Luck: Philosophical Papers,
1973–1980 (Cambridge: Cambridge University Press, 1982).
Williams, R. E. and Caldwell, D., ‘Jus Post Bellum: Just War Theory and the Principles of
Just Peace’, International Studies Perspectives 7 (2006): 309–20.
Winter, J., Sites of Memory, Sites of Mourning: The Great War in European Cultural History
(Cambridge: Cambridge University Press, 1995).
Winter, J., Remembering War: The Great War between Memory and History in the Twentieth
Century (New Haven, CT: Yale University Press, 2006).
Winter, S., ‘Theorising the Political Apology’, Journal of Political Philosophy (2014): 1–21.
Wittmann, R., Beyond Justice: The Auschwitz Trial (Cambridge, Mass.: Harvard University
Press, 2005).
Wolff, C., Jus gentium methodo scientifica pertractatum. ed. J. H. Drake (Oxford: Clarendon
Press, 1934 [1750]).
Wolff, J., ‘Fairness, Respect and the Egalitarian Ethos’, Philosophy & Public Affairs 27
(1998): 97–122.
Yannis, A., ‘The Concept of Suspended Sovereignty in International Law and Its Implications
in International Politics’, European Journal of International Law 13 (2002): 1037–52.
Yee, S., ‘The Tu Quoque Argument as a Defence to International Crimes, Prosecution, or
Punishment’, Chinese Journal of International Law 3 (2004): 87–134.
Ypi, L., ‘What is Wrong with Colonialism’, Philosophy & Public Affairs 41 (2013): 158–91.
Zaum, D., The Sovereignty Paradox: The Norms and Politics of International Statebuilding
Zaum, D., ‘The Norms and Politics of Exit: Ending Postconflict Transitional Administrations’,
Ethics & International Affairs 23 (2009): 189–208.
Zimmerman, M. J., ‘Intervening Agents and Moral Responsibility’, The Philosophical
Quarterly 35 (1985): 347–58.
Zimmerman, M. J., ‘Rights, Compensation, and Culpability’, Law and Philosophy 13 (1994):
419–50.
Index
Authors are listed only if and when they appear in the main text. Specific peace treaties are listed in
alphabetical order under the entry ‘treaty’. Specific peace agreements and accords are listed under the
letter corresponding to their name (for example, the Dayton Agreement appears under D.) Wars and
genocides are listed alphabetically under the entries ‘war’ and ‘genocide’, respectively (this applies to
the Holocaust.)
acts and omissionsâ•‡ 51 and suing for peaceâ•‡ 28–32

actus reus and surrenderâ•‡ 28–32, 46–7
and punishment in generalâ•‡ 178 and transitional foreign administrationsâ•‡ 237
and war-related crimesâ•‡ 185–8, 194 see also consent
Afghanistanâ•‡ 26n1, 42n21, 63, 219, 238n28
African Unionâ•‡ 54 Barak, Prime Minister Ehudâ•‡ 268–9
Algeriaâ•‡ 26, 35, 41n20, 241–2 Barre, President Siadâ•‡ 57
Allais, Lucyâ•‡ 264 Belgiumâ•‡ 91, 92, 108, 168, 206, 222, 273, 284,
Allende, President Salvadorâ•‡ 215 289, 290
amnestiesâ•‡ Bell, Christineâ•‡ 90
v. amnesiaâ•‡ 216 Bin Laden, Osamaâ•‡ 302
conditional v. blanketâ•‡ 210 Bismarck, Otto vonâ•‡ 111–12
full v. partialâ•‡ 210 Blackman, Alexanderâ•‡ 203, 204n38
justifyingâ•‡211–17 Blustein, Jeffreyâ•‡ 292–5
v. not punishingâ•‡ 210 Bomber Harrisâ•‡ 303
and patriotic partialityâ•‡ 214–15 Bosnia-Herzegovinaâ•‡ 24, 107, 157, 167, 225
annexationâ•‡ 131–2, 222, 241 Brandt, Chancellor Willyâ•‡ 268–9, 272–3
Anzac Dayâ•‡ 294 Brazilâ•‡ 209–10, 216
Apartheidâ•‡ 180, 211–12, 253 Bremner, Consul Paulâ•‡ 77
apologies Burke, Edmundâ•‡ 299
Apology Paradoxâ•‡ 279
and collective responsibilityâ•‡ 271–3 Cameroonâ•‡222
duty to accept apologiesâ•‡ 278 causal contribution to warâ•‡ 6
and emotionsâ•‡ 272 Chileâ•‡215–16
and inflationâ•‡ 278–9 Chinaâ•‡ 86, 94–5, 102, 114, 238
justifying apologiesâ•‡ 274–6 Churchill, Prime Minister Winstonâ•‡ 109, 277
and the passage of timeâ•‡ 273–4 Cicero, Marcus Tulliusâ•‡ 14
and sincerityâ•‡ 272 Clémenceau, Prime Minister Georgesâ•‡ 284–5
personal v. officialâ•‡ 269 Clinton, President William Jeffersonâ•‡
v. regretâ•‡ 269–70 268–9, 272
vicarious apologiesâ•‡ 273 Cohen, G. A.â•‡ 18–19
see also authority; regret collateral damageâ•‡ 37, 42, 44, 51, 68, 150, 153,
al-Assad, President Basharâ•‡ 169 154, 277
Arafat, Chairman Yasserâ•‡ 101 colonialismâ•‡
Arendt, Hannahâ•‡ 284–5 definedâ•‡238
Argentinaâ•‡ 29n4, 216 functional objection to colonialismâ•‡ 239–42
Aristotleâ•‡252 justificatory objection to colonialismâ•‡ 242–4
Augustine, Stâ•‡ 14 commemoration, see remembrance of wars
Austriaâ•‡ 99n10, 136, 163, 207, 292 Congoâ•‡ 54, 236n25, 241n34, 262
Auschwitzâ•‡ 204, 287, 288, 304–6 consent
authority and distributive justiceâ•‡ 160–1
and military occupationâ•‡ 74–7, 80–1, 228 to military occupationâ•‡ 69–70, 75, 80–2
and official apologiesâ•‡ 271–3 to peace agreementsâ•‡ 98, 105–6, 110–11
and peace agreementsâ•‡ 103–12 to peacekeepingâ•‡ 54–8
and punishmentâ•‡ 178–9 sexual consentâ•‡ 208
and resort to warâ•‡ 14–15, 28–9 to territorial restitutionâ•‡ 132–3
334 Index
consent (cont.) resource constraints 19
to transitional foreign administrations see also punishment; reparations; restitution
231–3, 243–4 Feinberg, Joel 149–50
to war 29, 48 First Gulf War 25, 147n1
see also authority forgiveness, defined 253–4
Correlates of War Project 10, 53 interpersonal v. political 255–6
cosmopolitanism, key tenets 2–5; see also and reconciliation 253–7
justice, cosmopolitan unforgiveable crimes 254–5
crimes vicarious 256–7
of aggression 183, 186–7 France 24, 25, 54, 68, 80, 91–2, 108, 111–12,
against humanity 180–3 147n1, 155, 169, 215, 220, 222–3,
against peace 91, 191 241, 248, 250n4, 273–4, 292, 305
of war (war-related crimes) 172–3, 180–5 Franco, General Francisco 209
see also actus reus; mens rea Fritzl, Josef 207
Czechoslovakia 136 Fürtwangler, Wilhelm 196n31
Danzig 222 gacaca
Dayton Agreement (1995) 24, 54, 107, 112, and due process 263
225–6, 227, 233 justifying 262–3
discrimination Gaza 74, 101, 135, 139, 140, 142, 184
between combatants and non-combatants 10 Geneva Agreement on Vietnam (1954) 92
and ending wars 40–1 Geneva Conventions and Protocols Additional
and military occupation 66 (1949 and 1977) 64, 77, 213
and peacekeeping 55 genocide 29, 41, 43, 51, 109, 149, 151, 187, 202,
disenfranchisement 134 213, 259, 277, 299–300, 303–4, 310
doctrine of double-effect, see military Armenian 172, 180, 216, 304
occupation Holocaust 136, 177, 251, 257, 268, 271,
Dönitz, Admiral Karl 30, 206n41 274, 277, 286–8, 289, 297, 300, 304–5
Donne, John 202, 257, 296 Rwandan 109, 149, 206, 249, 254, 262–3,
Dresden bombing 272, 276, 303 268, 272, 304
duties, see obligations Germany 24, 25, 27, 30, 33–4, 42, 45, 47, 50,
64, 73, 76, 80, 82–4, 91–2, 99n10,
East Timor (Timor Leste) 167, 221, 104, 108, 112, 136–7, 147n1, 151,
224–7, 234, 240, 242, 251, 155, 168, 180n9, 204–5, 222, 228–9,
262, 264 248, 250n4, 251, 257, 268, 284–90,
Egypt 54 291–2, 297
Eichmann, Adolf 177 Good Friday Agreement (1998) 91
Eisenhower, General Dwight 157 Govier, Trudy 256
El Salvador 209, 264 Gratian 14
Eliot, George 281–2, 301, 306 Gregory IX Pope 14
emotions, and morality 252 Grotius, Hugo 15–16, 113, 307
England 208, 274 Guatemala 136, 139, 157–9, 248, 251, 264,
enslavement 3, 41, 43, 51, 180–1 268, 269, 272
equality
and complying with promises and Hague Regulations (1907) 64, 77
agreements 95 Hammarskjöld, UN Secretary-General Dag 55
between individuals 3, 98, 163, 260 Herman, Barbara 212
between regimes 109 Hirohito, Emperor 30
between states 16 Hiroshima 205
of welfare 18–19 Hitler, Adolf 30, 34, 109, 288, 305
Erdemovic, Dragan 190 Hobbes, Thomas 113
Estonia 84 Hong Kong 102
European Union 25, 248, 280, 292–3 Hume, David 97
exile 214 Hungary 136, 163
Hussein, Saddam 26, 166, 196, 228
feasibility, defined 18
epistemic constraints 19 impartiality
hard v. soft constraints 19 in peacekeeping 61–2, 73, 76–7
non-compliance constraints 19 in punishment 177, 179, 200, 203, 205–6
practical v. moral constraints 18–19 and reconciliation 251
Index 335
imprisonment 72, 171, 191, 192–5, 198, institutionalist v. individualist approaches 4,
207, 217 169, 266–7, 300–1
independence v. dependence thesis 21, 308–9 procedural v. substantive 20–1, 89–90, 95,
and peace agreements 106–12 102–3, 263, 308
India 54, 63n12, 238, 241–3 restorative 261–4
Indonesia 221, 224, 251, 264 transitional 1–2
insurgency see also reparations; restitution; sufficientism;
and military occupation 78, 79n30, 80, punishment
130, 233
and war continuation 45, 34 Kagan, Shelly 152–3
International Criminal Court (ICC) 23, 172, Kamm, F. M. 18, 36–7, 152, 212–13
180, 183n13, 199, 204 Kant, Immanuel 1, 16–17, 18, 307
International Criminal Tribunal for the Former Karajan, Herbert von 196n31
Yugoslavia (ICTY) 23, 172, 187, 190, Karadzic, Radovan 210, 225
195, 199, 202 Kecskemeti, Paul 32–3
International Criminal Tribunal for Rwanda killing
(ICTR) 172, 199–200 and enforcing peace agreements 113–14
International Monetary Fund 240 and military occupation 80–2
Iraq 1, 26, 45, 63, 64, 70–1, 77, 147, 162, and peacekeeping 60
166, 184, 196, 203, 219, 228–30, and promises 95
240n33, 272 punitive 38
Israel 54, 63, 74, 94, 101, 103, 104, 114, 135, redemptive 38
136, 137, 139, 141–3, 151, 159, 184, and remembrance 289–90, 296–7
248–50, 268 and termination costs 42–3
in war and self-defence 6–9, 11–12,
Jelisic, Goran 187 51, 178
Jerusalem 104, 135, 139, 177, 182, see also crimes, of war; discrimination;
223, 268 liability, to be killed; liability, to be
Joan of Arc 274 punished; proportionality
Jodl, Marshall Alfred 30 Kock, Eugene de 268–9
Joint Declaration between the USSR and Japan, Kohl, Chancellor Helmut 292–3, 302
Joint Declaration (1956) 27 Kosovo 167, 226–7, 232, 240
jus ad bellum and jus in bello, their Kutz, Christopher 187
relationship 20
jus habitationis 122, 134 Lacey, Nicola 176
jus sanguinis 121–2, 140 League of Nations 222–3
jus solis 121–2 Lebanon 54
just cause for war (and lack thereof ) 10–12, 16, lesser-evil arguments 46, 68, 124
17–18, 29 levée en masse 28
for defensive harm 7 Liberia 264
for First World War 290 liability
and military occupation 65–6, 69–71, 78 and causal contribution to/participation in
and peacekeeping 55–6 war 6, 23, 150–1, 155, 158–9,
and punishment 183, 184–5 185–8, 308
and reparations 150, 154 and instrumentality 154n10
and restitution 131–2, 137 to be killed 6–8, 42, 43, 50, 60, 185,
and transitional foreign 237, 310
administrations 228–30 to punishment 174–6, 184–90
and war continuation 30, 35–7, 38, 42, 45, to reparations 23, 156, 309
47–51 see also responsibility
just war principles, listed 10, 55, 183, 235; Luxembourg Agreement (1952) 104, 147n1
see also discrimination; just cause for
war; necessity; proportionality; success Mandates system 222–4
requirement Mandela, Nelson 254
justice Margalit, Avishai 291–5
corrective 117–18, 120, 126, 145 Marshall Plan 167–8
cosmopolitan 2–5, 8, 23, 44, 93, 118, 123, MacArthur, General Douglas 27
126, 150, 163, 220, 229, 233–6, 249–51 May, Larry 168–9
distributive 4, 23, 117, 121, 144, 146, 151, McMahan, Jeff 11, 12, 38, 67, 308
156, 168–9, 266–7 Memories, episodic v. semantic 285
336 Index
mens reaâ•‡ Nuremberg Trialsâ•‡ 23, 25, 172, 177, 180, 195,
and punishment in generalâ•‡ 178 203, 205, 206
and war-related crimesâ•‡ 188–9
military occupationâ•‡ obligations
of Baltic States (by the USSR)â•‡ 82–4 conflictsâ•‡
of China (by Japan)â•‡ 86 between reparative and distributive
and civil warsâ•‡ 70–2 obligationsâ•‡ 148, 155–6
and collaboration/cooperationâ•‡ 84–8 between reparative obligationsâ•‡ 148, 155–6
and the doctrine of double effectâ•‡ 67–9 cosmopolitan, of justiceâ•‡ 5
duty to comply with occupiers’ diachronicâ•‡134–5
directivesâ•‡80–4 promissoryâ•‡ 91, 95–102
and exitâ•‡ 73–4 special v. generalâ•‡ 92
of France (by Germany)â•‡ 80, 86, 87 occupancy, justification for occupancy
of Gaza/ West Bank (by Israel)â•‡ 63 rightsâ•‡ 120–1, 122, 134, 137, 138–9;
of Germany (by the Allies)â•‡ 64, 228 see also right(s)
and humanitarian interventionâ•‡ 75 Orend, Brianâ•‡ 18
of Iraq (by the US-led coalition)â•‡ 63, 64, Oslo Accords (1993)â•‡ 94, 101, 104, 107, 114
70–1, 228 Ottoman Empireâ•‡ 222–3
just principles forâ•‡ 66 ownership
laws ofâ•‡ 64, 77–9 incidents ofâ•‡ 119–20
v. peacekeepingâ•‡ 53 justifications for rightsâ•‡ 121–2, 124–5
and reparationsâ•‡ 150–1 private v. publicâ•‡ 120–1
resistance toâ•‡ 80–4 see also right(s)
and restitutionsâ•‡ 131–2
transformativeâ•‡228–30 Pakistanâ•‡ 54, 289
see also authority; consent; discrimination just Pal, Judge Radhabinodâ•‡ 205–6
cause for war; moral hazard; necessity; Palestineâ•‡ 101, 104, 135, 139–40, 142, 222,
proportionality; success requirement 249–50
Mill, John Stuartâ•‡ 243 Parfit, Derekâ•‡ 9
Miller, Davidâ•‡ 18–19 Paris Peace Accords (1973)â•‡ 92
Milosevic, Slobodanâ•‡ 210, 225–6, 232 patriotic partialityâ•‡
Mitchell, Georgeâ•‡ 112 and amnestiesâ•‡ 215
Mitterrand, Francoisâ•‡ 292–3, 302 and cosmopolitan justiceâ•‡ 5
Moellendorf, Darrellâ•‡ 35 and military occupationâ•‡ 86
moral hazard and war endingsâ•‡ 44
and military occupationâ•‡ 71–2 personal prerogativeâ•‡ 4–5, 9, 44, 70, 84, 125
and the right to returnâ•‡ 137–8 peace
and war continuationâ•‡ 49 agreement/treaty
multilateralismâ•‡ and coercionâ•‡ 113–14
and military occupationâ•‡ 76–7 and complianceâ•‡ 96–102
and peacekeepingâ•‡ 61–2 and corruptionâ•‡ 113
murderâ•‡ 3, 28, 48, 54, 80 declarative clauses (defined)â•‡ 92
and crimes against humanityâ•‡ 180–2 de novo clauses (defined)â•‡ 94–5
v. war killingsâ•‡ 186, 189, 192 and fraudâ•‡ 113
procedural v. substantive conditionsâ•‡
Nagasakiâ•‡205 89–90
Nakasone, Yasuhiroâ•‡ 268, 277 specifying clauses (defined)â•‡ 92–3
Napoleon IIIâ•‡ 111 see also authority; consent
NATOâ•‡ 225–6, 232 just v. justifiedATC (defined)â•‡ 17–20,
necessityâ•‡10 312–14
and military occupationâ•‡ 66 as a just war principleâ•‡ 12–13
and peacekeepingâ•‡ 59 liberal thesisâ•‡ 219
no-undue sacrifice provisoâ•‡ 3, 13, 18, 20, 31, negative v. positiveâ•‡ 12
61, 84, 133, 150 peacebuildingâ•‡ 54, 55
non-identity problemâ•‡ 125–7 peacekeeping
Northern Irelandâ•‡ 91, 112, 269, 303 duty to deploy peacekeeping troopsâ•‡ 60–1
Nozick, Robertâ•‡ 160 first-generation v. second generation
Nuremberg defence against criminal peacekeepingâ•‡54–5
liabilityâ•‡189–90 and humanitarian interventionâ•‡ 56, 61
Index 337
just peacekeeping principles 56, 58–60 interpersonal v. political 247–9
mandate of peacekeeping forces 55–6 see also forgiveness; gacaca; truth and
and neutrality 54 reconciliation commissions; trust
see also authority; consent; discrimination; Reconciliation Walk 269, 274
just cause; success requirement reconstruction
Picasso, Pablo 128 and conditional aid 165–6
Pickard, Hanna 176 duties 166–9
Pinochet, Augusto 215–16 and feasibility constraints 169–70
piracy 200 Reconstruction Principle 146, 163–5
Pol Pot 257 redemption (and war continuation) 38
Poland 82, 136, 204n38, 222, 250n4 regime change
Portugal 223–4 in the just war tradition 15–16
Potsdam Agreement (1945) 136–7 and transitional foreign administration
power, Hohfeldian 221, 230
to declare war 29, 32 regret
to govern 5, 33, 109, 132, 133, 228, 231 agent-relative v. agent-neutral 269–70
to negotiate and endorse peace treaties 107 agential v. non-agential 270
to punish 178, 201–2, 215 and collective responsibility 271–2
to sue for peace 32 and emotions 272–3
proportionality 10 justifying 274–6
and military occupation 66 and the passage of time 273–4
narrow v. wide 11–12 and sincerity 272
and peacekeeping 59, 60 see also apologies; authority; emotions
and punishment 193–5 remembrance of wars
and sunk costs 38–40 and collective shame 286–8
and transitional foreign administrations 235 and demands for recognition 292–3
and war endings 37–40 and emotions 285–6, 301–2
Pufendorf, Samuel 16 and gratitude 289–91
punishment 176, 179, 181 individualist v. institutionalist
and the confiscation of property 198 approaches 300–1
and duress 190–1 justifying remembrance 295
duty to punish 178, 202 over-demandingness objection to
epistemic defence 191–2 cosmopolitan 305–6
expressivist justifications for 175–6 over-universalism objection to
and feasibility constraints 194 cosmopolitan 303–4
fittingness v. proportionality 192–3 and the transgenerational community 291–5
and humanitarian intervention 183, 184 and truth 284–5, 303
and lustration 195–7 undue-optimism objection to
Nuremberg defence 189–90 cosmopolitan 303
protective (deterrence) justifications for 175 reparations
retributivist justifications for 174–5 and the fault principle 152–8, 162–3
and sanctions 197–8 and feasibility constraints 147–8, 152
tu quoque defence 206n41 and humanitarian intervention 151–2, 154
see also actus reus; authority; crimes, of war; and the no-unjust benefit principle 158–63
liability to punishment; mens rea; Reparative Principle 146
proportionality; universal jurisdiction see also liability; right(s), inheritance
responsibility
Rabin, Prime Minister Ytzhak 101 individual v. collective 187–8, 271–2,
Rambouillet Accords over Kosovo (1999) 278–9
226, 232 for wrongful threats 8, 42, 48, 66, 75, 155,
Ramses II 13 157, 164, 170, 211–12
rape, and crimes against humanity 180–2 restitution
Rawls, John 12, 18, 97, 115 and feasibility constraints 122, 129, 135
reconciliation justifying 123–5
after intrastate v. interstate conflicts 249–50 and requisitioned/stolen military
definition 247 matériel 129–30
duty to reconcile 251–2 and stolen artwork 127–9
individualist v. institutionalist see also right(s), inheritance; right(s),
approaches 251–2 ownership
338 Index
right(s) termination costs 42–4, 50–1
to assistance/resources 3, 94, 170 Thomson, Judith Jarvis 149–50, 178
basic v. non-basic 3–4 Tocqueville, Alexis de 242
civil 67, 82, 84, 195 Tokyo Trials 23, 172, 205
general v. special 92, 118, 251 torture, as a crime against humanity 180–1
human 3 transitional foreign administrations
individual v. jointly-held 67–8 duty to establish a TFA 233–4
inheritance 125, 128 and humanitarian intervention 233–4
occupancy 120–1, 123, 137–8 and the impact of treaty agreements 230–3
ownership 119–21, 124–5, 138, 149, 161 just principles for TFAs 234–5
political 33, 82, 84, 195, 228, 229 justifying TFAs 227–30
of return 135–43 and military occupation 218, 227–30
sovereignty 5, 65–70, 119, 123, 132–4, and the use of force 236–7
138–40, 241 see also authority; colonialism; just cause
territorial 121–3, 134 Treaties between the United Kindgom and
violations v. justified infringements 8–9 China (1842‒1898) 94–5, 102
see also occupancy; ownership; power, Treaty of Frankfurt (1871) 112
Hohfeldian Treaty of Kadesh (c.1259 bc) 13
Roosevelt, President Franklin Delano Treaty of St Germain (1919) 163
105, 109 Treaty of San Francisco (1952) 27
Rwanda 54, 62, 109, 136, 149, 152, 169, 172, Treaty of Trianon (1919) 163
194, 199–200, 206, 222, 248, 249, Treaty of Versailles (1919) 91–2, 94, 108, 172,
254, 260–4, 268, 272, 304; see also 305, 308
genocide, Rwandan; gacaca Treaty of Vienna (1815) 1
trust
Santayana, George 299 horizontal v. vertical 259–60
Sassoon, Siegfried 301, 306 and negotiating peace agreements 110–11
Scanlon, T. M. 97–101 political 258–9
Schabas, William 177 v. reliance 258
secession 132–3 Trusteeship system 223, 235
Shue, Henry 3 truth and reconciliation commissions (TRC)
Sierra Leone 209–10, 216n52, 227, 264 and due process 265–6
Simpson, Gerry 177 duty to establish and support a TRC 266–7
slavery individualist v. institutionalist
and crimes against humanity 180–1 approaches 266–7
as a just cause for continuing war 36–7 justifying TRCs 265–6
Somalia 56, 227 see also El Salvador; Guatemala; Sierra
South Africa 211–12, 213, 222, 253, 264, Leone; South Africa
267–8 Tudjman, Franco 225
Soviet Union (USSR) 1, 27, 82–4, 109, 112, Turkey 208, 216, 293
131, 137, 182 Tutu, Archbishop Desmond 253–5, 257
Spain 209–10, 215, 223
Speer, Albert 268–9, 277 Uganda 210, 262
Srebrenica 59, 190 Ukraine 205
Stalin, Joseph 33, 83, 109 famine 12
Stalingrad 305–6 United Kingdom 54, 91, 102, 105, 137,
state legitimacy 5, 33–4, 54, 79, 103, 200, 228; 284, 312
see also authority United Nations 25, 54, 59, 62, 64, 112, 167,
Stimson Doctrine 23n34 211, 213, 264, 268, 277
success requirement 10, 11 UN Charter 55, 64, 113, 223, 235
and military occupation 66 UN Compensation Commission 25, 147n1
and peacekeeping 56, 58 UN Convention on the Law of the Sea
and war endings 35–6 (1982) 200
Suez Crisis 54 and transitional administrations 22–7
sufficientism 3–4, 125, 160, 164 United States 31, 91, 92, 108, 137, 157–8,
surrender 166, 168, 215, 219, 268, 272
v. suing for peace 31–2, 41–4 universal jurisdiction
unconditional v. conditional 45–7 and the complementarity principle
see also authority 199, 208
Index 339
defined 199 Iraq War 1, 26, 45, 64, 70–1, 219, 230, 272
justifying universal jurisdiction 200–6 Opium Wars 94
and states’ competence to punish 207–9 punitive war 17
Uruguay 209–10 Second World War 27, 29, 33–4, 42, 45, 47,
64, 70, 76, 86, 99n10, 109, 128, 135,
Vattel, Emerich de 1, 16, 17, 34 136, 158, 184, 223, 248, 249, 263n20,
Vienna Convention on the Law of Treaties 91, 268, 277, 290, 292, 295, 303
103n11, 111n19, 112 Six-Day War 142
Vietnam 26, 42, 92 Vietnam War 92
Vitoria, Francisco 1, 14–16, 307 Weizsacker, President Richard von 268–9, 271
West Bank 63, 101, 104, 135, 139, 142, 184
Walzer, Michael 1, 8, 80, 82, 308 West New Guinea 221
war Wiesenthal, Simon 255–6, 273
American Civil War 294 Wilson, President George 222
defined 10 Winter, Jay 294
Cold War 54 World Bank 1
English Civil War 198 World Jewish Congress 104, 256, 273
First Gulf War 25, 147n1 Wolff, Christian 1, 16–17
First World War 37, 38, 108, 172, 222, wrongdoing
223, 281–6, 289–92, 294–5, 298, 299, fact-relative, v. evidence-relative, v. belief-relative
301, 305 accounts 9, 110–11
Franco-Algerian War 26, 35, 41n20
Franco-Prussian War 24, 111–12 Yalta Conference (1945) 109
individual v. collective dimensions 5–6, Yugoslavia 23, 172, 190, 199, 200, 225, 226,
309–12 248, 264

Fabre, Cécile - Cosmopolitan Peace-Oxford University Press (2016)

Uploaded by

Copyright:

Available Formats

Fabre, Cécile - Cosmopolitan Peace-Oxford University Press (2016)

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Fabre, Cécile - Cosmopolitan Peace-Oxford University Press (2016)

Uploaded by

Copyright:

Available Formats

OUP CORRECTED PROOF – FINAL, 28/05/16, SPi

1. Cosmopolitanism and War 1

3. Peacekeeping and Military Occupation 53

6. Reparations, Distribution, and Reconstruction 145

8. Transitional Foreign Administrations 218

10. Remembrance 281

Works Cited 317

2 Cosmopolitanism and War

Europe after the Napoleonic Wars, with no consideration of reconstruction, repa­

Cosmopolitanism and War 3

4 Cosmopolitanism and War

Cosmopolitanism and War 5

6 Cosmopolitanism and War

Cosmopolitanism and War 7

So much, then, for the individualist framework of this book. Admittedly,

8 Cosmopolitanism and War

Cosmopolitanism and War 9

10 Cosmopolitanism and War

13 See http://www.correlatesofwar.org/ (accessed on 13/01/2016).

Cosmopolitanism and War 11

12 Cosmopolitanism and War

Cosmopolitanism and War 13

1.4. J us P ost B ellum i n t he J u s t Wa r T r a d i t i o n

14 Cosmopolitanism and War

Cosmopolitanism and War 15

23 Vitoria, ‘On the Law of War’, Question 3, arts 5–9.

16 Cosmopolitanism and War

Cosmopolitanism and War 17

1.5 . Thi n ki n g A b o u t P eace A f t e r Wa r

1.5.1. Just Peace v. JustifiedATC Peace

18 Cosmopolitanism and War

Cosmopolitanism and War 19

20 Cosmopolitanism and War

consists in. It consists, at a minimum, in a state of affairs where individuals do not

1.5.2. Jus Ad Bellum, Jus In Bello, Jus Ex Bello, and

Cosmopolitanism and War 21

22 Cosmopolitanism and War

In the remaining nine chapters, I offer a cosmopolitan account of just transitions

Cosmopolitanism and War 23

justness, or lack thereof, of the resulting peace settlement and, by implication, on

24 Cosmopolitanism and War

Cosmopolitanism and War 25

s.2.3, I argue that a belligerent whose war is just at t1 is nevertheless sometimes

or at the very least institutional unresponsiveness—are not propitious to formal

which a whole community is deemed no longer to be engaged in hostilities.

2.3.1. Just Cause and Reasonable Chance of Success

2.3.4. How to End an Initially Just War: Suing

this particular, limited objective within the constraints of the requirement of

2.3.5. To Whom is the Duty Owed?

2.4.1. Ending an Unjust War: Conditional v. Unconditional Surrender

2.4.2. Acquiring Justifications for Continuing with an

goes to war in breach of one or more of the other requirements, it is justified in

distinctions I prescind from addressing cases where the group of beneficiaries

intractable is precisely that war, notwithstanding the destruction it causes, has

54 Peacekeeping and Military Occupation

The (vast) literature on peacekeeping standardly distinguishes between peacekeep-

Peacekeeping and Military Occupation 55

consent—neutrality—limited force in favour of more robust—some would say,

56 Peacekeeping and Military Occupation

A peacekeeping mission takes place as belligerents are (seemingly) willing to

1. Cosmopolitanism and War 1

3. Peacekeeping and Military Occupation 53

6. Reparations, Distribution, and Reconstruction 145

8. Transitional Foreign Administrations 218

10. Remembrance 281

Works Cited 317

Europe after the Napoleonic Wars, with no consideration of reconstruction, repa