Just War Theory

Chrisoula Andreou (University of Utah) Mark Bevir (University of California, Berkeley) Clare Chambers (University of Cambridge) Fabian Freyenhagen (University of Essex) Tim Mulgan (University of St Andrews)

Ian Shapiro (Yale University)

VOLUME 4

Publisher Details

The titles published in this series are listed at brill.com/simp

LEIDEN • BOSTON

2013

The essays have all appeared in the Journal of Moral Philosophy. Library of Congress Cataloging-in-Publication Data

Just war theory / Edited by Thom Brooks. p. cm. -(Studies in moral philosophy, ISSN 2211-2014 ; volume 4) Includes bibliographical references and index.

ISBN 978-90-04-22850-4 (hardback : alk. paper) -ISBN 978-90-04-24202-9 (e-book) 1. Just war doctrine. 2. War--Moral and ethical aspects. I. Brooks, Thom.

U22.J877 2013 172’.42--dc23

2012036486

This publication has been typeset in the multilingual “Brill” typeface. With over 5,100 characters covering Latin, IPA, Greek, and Cyrillic, this typeface is especially suitable for use in the humanities. For more information, please see www.brill.com/brill-typeface.

ISSN 2211-2014

ISBN 978-90-04-22850-4 (hardback)

ISBN 978-90-04-24202-9 (e-book)

Koninklijke Brill NV incorporates the imprints Brill, Global Oriental, Hotei Publishing, IDC Publishers and Martinus Nijhoff Publishers.

All rights reserved. No part of this publication may be reproduced, translated, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission from the publisher.

Authorization to photocopy items for internal or personal use is granted by Koninklijke Brill NV provided that the appropriate fees are paid directly to The Copyright Clearance Center,

222 Rosewood Drive, Suite 910, Danvers, MA 01923, USA. Fees are subject to change.

This book is printed on acid-free paper.

Notes on Previous Publications..................................................... vii

List of Contributors....................................................................... ix

Introduction............................................................................... 1

Thom Brooks

PART I

JUST WAR AND UNJUST WAR

The Permissibility of Aiding and Abetting Unjust Wars 11

Saba Barzargan

Dividing Harm................................................................................ 29

Gerhard Øverland

Consistency in the Armed Enforcement of Human Rights:

A Moral Necessity?...................................................................... 49

Ned Dobos

PART II

JUST WAR AND SELF-DEFENSE

Conditional Threats......................................................................... 69

Gerhard Øverland

Can Wars Be Fought Justly? The Necessity Condition

Put to the Test........................................................................... 81

Daniel Statman

PART III

INNOCENT NON-COMBATANTS AND SELF-DEFENSE

Self-Defence among Innocent People............................................. 101

Gerhard Øverland

Killing in War and Moral Equality................................................. 125

Stephen R. Shalom

Self-Defence and the Principle of Non-Combatant Immunity 143

Helen Frowe

Partiality and Weighing Harm to Non-Combatants 161

David Lefkowitz

PART IV

JUST WAR THEORY AND TERRORISM

Defining Terrorism for Public Policy Purposes: The Group–

Target Definition...................................................................... 183

Eric Reitan

Bibliography................................................................................ 211

Index......................................................................................... 221

Notes on Previous Publications

Saba Bazargan, ‘The Permissibility of Aiding and Abetting Unjust Wars’,

Journal of Moral Philosophy 8(4) (2011): 513–29.

Ned Dobos, ‘Consistence in the Armed Enforcement of Human Rights: A Moral Necessity?’ Journal of Moral Philosophy 8(1) (2011): 92–109.

Helen Frowe, ‘Self-Defence and the Principle of Non-Combatant Immunity’, Journal of Moral Philosophy 8(4) (2011): 530–46.

David Lefkowitz, ‘Partiality and Weighing Harm to Non-Combatants’,

Journal of Moral Philosophy 6(3) (2009): 298–316.

Gerhard Øverland, ‘Conditional Threats’, Journal of Moral Philosophy 7(3) (2010): 334–45.

Gerhard Øverland, ‘Dividing Harm’, Journal of Moral Philosophy 8(4) (2011): 547–66.

Gerhard Øverland, ‘Self-Defence among Innocent People’, Journal of Moral Philosophy 2(2) (2005): 127–46.

Eric Reitan, ‘Defining Terrorism for Public Policy Purposes: The Group– Target Definition’, Journal of Moral Philosophy 7(2) (2010): 253–78.

Stephen R. Shalom, ‘Killing in War and Moral Equality’, Journal of Moral Philosophy 8(4) (2011): 495–512.

Daniel Statman, ‘Can Wars Be Fought Justly? The Necessity Condition Put to the Test’, 8(3) (2011): 435–51.

List of Contributors

Saba Bazargan is a professor at the University of California at San Diego.

Thom Brooks is Reader in Law at Durham University. He is editor and founder of the Journal of Moral Philosophy.

Ned Dobos is a Research Fellow at the Centre for Applied Philosophy and Public Ethics, University of Melbourne and Charles Sturt University.

Helen Frowe is Lecturer in Philosophy at the University of Kent at Canterbury.

David Lefkowitz is an associate professor of philosophy at the University of Richmond.

Gerhard Øverland is a Project Leader at the Centre for the Study of Mind in Nature at the University of Oslo.

Eric Reitan is Professor of Philosophy at Oklahoma State University.

Stephen R. Shalom is Professor of Political Science at the William Paterson University of New Jersey.

Daniel Statman is a professor at the University of Haifa.

Introduction

Thom Brooks

Just War Theory

Just war theory raises some of the most pressing and important philosophical issues of our day. When is a war a just war, if ever? Do all soldiers in war have moral equivalence? What is the difference between combatants and non-combatants? This book brings together some of the most important essays in this area written by leading scholars and offering significant contributions to how we understand just war theory.1 The essays have all appeared in the Journal of Moral Philosophy, an internationally recognized leading philosophy journal. This introduction will introduce the essays included and provide some general background.

Just war theory has a venerable past history with roots in the work of figures such as Augustine and Thomas Aquinas. Contemporary popular interest has grown since the controversial invasion of Iraq in 2003 which brought just war considerations to the centre of debate. A second pivotal moment was the 2009 publication of Jeff McMahan’s Killing in War which further developed and extended his past work in this area to a new level of philosophical sophistication that has won many admirers and contributed to an explosion of new work on just war and related issues (see McMahan 2009).

The idea of a just war is often premised on two normative conditions. The first is the idea of a jus ad bellum, or the justification for engaging in war. This is most often defended in terms of a state’s right to self-defence analogous to individuals. The argument is that individuals are justified i n engaging in self-defence against unjust aggressors. Likewise, states are justified in engaging in self-defence against unjust aggressor states. A standard example is a military intervention against a sovereign state without just cause. If there is no just cause for war, then this normative condition is absent and the military action cannot be considered a ‘just war’ (see McMahan 2005).

1 See also Brooks (2002, 2008, 2011, 2012, 2013).

The second normative condition of a just war is its jus in bello, or the justice arising within war. This relates to the justification of how states engage in war. The fact that a state may possess a just cause for war does not entail that this state would be justified in fighting the war however it might choose. Instead, there are normative constraints on how just wars should be fought if the military action is to be considered a ‘just war’. The standard example is the distinction between combatants and noncombatants where the latter are often considered to be innocent and free from direct targeting during wartime because they are not combatants and, thus, not involved with military operations. Much debate has centred on whether such a distinction is easily applicable in practice. Furthermore, some have wondered how uninvolved non-combatants genuinely are in such operations.

A just war is a military operation where these normative conditions are satisfied. Thus, a just war is an activity that has a just cause and fought within specified normative constraints. A military operation that lacks one or both conditions is not considered a ‘just war’ strictly speaking. The idea of a just war is subject to some revision. For example, classical proponents such as Aquinas believed there were further criteria that should be satisfied, such as the presence of a just government. A just war is one fought with a just cause by a just government in a just way. However, this position has relatively few defenders now. Another view growing in popularity is the idea of a jus post bellum, or justice after the war. This relates to post-war efforts and any duties arising.

The following chapters explore various issues relating to a just cause and self-defence, justice in war and the distinction between combatants and non-combatants, humanitarian interventions, and terrorism. These chapters will now be introduced by section before concluding with some remarks about their selection in this book and this series.

Part I: Just War and Unjust War

The first section contains essays examining the just war and unjust war distinction from different angles. If a war is unjust, then it may seem intuitively obvious that we have a strong moral reason to avoid contributing to it. In the first chapter, Bazargan (2011) argues that this presumption is based on a mistake.2 This is because there may be circumstances where

2 This essay was originally published in a special issue on the topic of ‘Just War’ that appeared in the Journal of Moral Philosophy volume eight, number four in 2011. Papers it may be permissible such an act and this depends on the alternatives available. Furthermore, there is an important difference between the determination of whether a government may engage in war and whether an individual person may promote this war. This issue has some significance because this difference is often manifest in unjust wars relating to humanitarian aims, an increasingly more common variety of military intervention.

A common view is that moral, and not merely causal, responsibility for harm has moral significance. For example, the fact that someone possesses some moral responsibility for harm to another is often considered to be central. We may not believe that our lacking moral responsibility for harm to another has moral significance for us. In the second chapter, Gerhard Øverland (2011) presents a powerful argument against this view in defending ‘the asymmetrical fair share procedure’ whereby innocent aggressors have a duty to take a fair share of harm where dividing shares is possible. Mere contribution has moral significance where it involves harm to others and even where we lack moral responsibility for it.

The third chapter addresses humanitarian intervention. International norms pertaining to human rights are all too often enforced selectively. For example, some oppressive regimes become subjected to military intervention while other similar regimes are not. Ned Dobos (2011) examines the issue of whether a military operation to defend and promote human rights becomes morally illegitimate where an intervening state has failed, is failing or will fail to intervene in similar circumstances elsewhere. Dobos argues that such considerations may lead us to be critical about intervening states and the decisions they determine about when to intervene elsewhere, but they do not undermine the moral legitimacy of the particular intervention. The latter should be assessed on its individual merits.

Part II: Just War and Self-defence

The second part of this book addresses a major topic in just war theory: just war and self-defence. The first chapter considers the moral status of conditional threats. Gerhard Øverland (2010) examines whether it is morally permissible to use potentially lethal defensive force against included in this issue and this book includes Bazargan (2011), Frowe (2011), Øverland (2011), and Shalom (2011). My thanks again to Helen Frowe for her assistance with most of these papers. threats that are merely conditional. This examination involves a novel understanding of self-defence and how it relates to issues of risk and moral responsibility.

In the next chapter, Daniel Statman (2011) offers a powerful critique of the necessity condition. He argues that it is a mistake to believe that it is a mistake to argue that moral considerations relevant to individual selfdefence apply equally to collective self-defence. Statman argues that this view may have some radical unwelcome implications at the level of jus in bello. Instead, he defends a contractualist view about killing in war that is more permissive to killing combatants during wartime. If we want to understand how war might be fought justly, contractualism may have real advantages to available alternatives.

Part III: Innocent Non-combatants and Self-defence

This book’s third part focuses our attention on a specific case pertaining to just war theory and self-defence, namely, the case of innocent noncombatants. In the first essay, Gerhard Øverland (2005) argues that our moral responsibility to render assistance is unaffected by whether or not we contributed to a situation by causing harm. This argument focuses upon the moral significance of need. Severe need trumps considerations of contribution. It may then be unnecessary to argue that we should provide assistance as a negative duty for our contributing to global poverty.3

Suppose a state satisfies just war criteria and, therefore, justified in attacking a military target. A plan is enacted to send a bomber pilot to destroy the target. One likely side-effect is that innocent civilians living nearby will be killed. Do these civilians have a right to self-defence against the bomber pilot? While such a scenario may be justified in McMahan (2009), Stephen Shalom (2011) argues that this is a mistake. The innocent civilians and bomber pilot do not share moral equivalence. If a military attack is supported by just war criteria, then its opponents lack a right of self-defence that might undermine a just war.

In the third chapter, Helen Frowe (2011) presents an important argument in support of reductivism in just war accounts. This position argues that the moral rules pertaining to killing in war are reducible to the moral

3 See Brooks (2007). rules governing killing between individuals. Some argue that this position is problematic. This is because it might have unwelcome potential consequences, such as justifying civilian deaths by terrorist attacks. Frowe argues that a wide conception of reductivism which accepted that innocent people may be permissible killed in self-defence is able to support the prohibition of terrorism if it included the distinction between direct and indirect threats. Frowe argues further that this view of reductivism does not support the blanket immunity of non-combatants to defensive killing. This should compel us to consider how the reductivist account might be amended, but it would be a mistake to reject it.

Finally, David Lefkowitz (2009) considers an argument that has been made by both Thomas Hurka and Frances Myrna Kamm: combatants ought to give greater weight to collateral harm to compatriot noncombatants rather than to enemy non-combatants. Lefkowitz rejects this partiality by use of several illustrations and illuminating analogies.

Part IV: Just War Theory and Terrorism

The final part contains an important contribution to how we understand terrorism and its definition (see Brooks 2010). Eric Reitan argues that we must have a clear definition of terrorism in order to best evaluate public policy decisions aimed at combating terrorism. Reitan considers a wide number of popular understandings of what is terrorism found in the literature. He proposes an alternative understanding based upon the group-target distinction.

Conclusion

A final comment should be made about how these essays were selected for publication. All chapters originally appeared in recent issues of the Journal of Moral Philosophy. I founded the journal with Fabian Freyenhagen in 2003 and it launched the following year. The JMP has been published quarterly since 2009. All submissions were subjected to rigorous anonymous peer review by our international editorial board and referees. We publish a list of our referees in the final issue of each volume. Our standards are high and our acceptance rate is about 5%. We receive submissions from all over the world with the great majority coming from either the United States or Great Britain. Furthermore, we have endeavoured to ensure high quality review standards with swift turnarounds and we are normally able to review about 90% of all submissions in three months and about 85% in two months or less.

The essays in this volume have satisfied our high standards and all have appeared in recent issues of the Journal of Moral Philosophy ensuring their quality and timeliness. Each was selected for this book because of its genuine contribution to the topic of just war theory and related issues. Rather than offer a single narrative, the essays instead may be read in any order and present a number of important perspectives and insights on the topic that should help provide further clarifying illumination on central debates and ideas in this field. Readers interested in learning more about what other essays have been published in the JMP should consult this book’s bibliography towards the end where the full publication details of all articles are listed. Anyone interested in submitting new work for future issues should submit through our online submission system found on our website.4

This book is amongst the first in our new Studies in Moral Philosophy book series published by Brill, the publishers of the JMP. The book series aspires to fulfil the same high standards and quality of its sister journal. We will aim to publish leading work in the areas of moral, political, and legal philosophy. This book helps launch this new series. Prospective authors interested in submitting a proposal for a monograph or edited book for this series should contact me.5

I hope that you will enjoy reading the essays in this book as much as I have. Many thanks are due to Suzanne Mekking at Brill for her strong support for this book and the book series. My thanks also to Liesbeth Hugenholz for her assistance in producing this book. I am also most grateful to Fabian Freyenhagen for his assistance over the years, as well as to our advisory committee for this book series. Finally, my most sincere thanks must be reserved for the authors for choosing the Journal of Moral Philosophy for their work. The journal’s success is primarily not through any particular effort of its editors, but rather a reflection of the high quality of our author’s important contributions. We look forward to producing further volumes on other topics in future.

4 The Journal of Moral Philosophy website may be found here: http://www.brill

.nl/jmp.

5 My contact details are available on my personal website: http://thombrooks.info.

Bibliography

Bazargan, Saba (2011). ‘The Permissibility of Aiding and Abetting Unjust Wars’, Journal of Moral Philosophy 8(4): 513–29.

Brooks, Thom (2002). ‘Cosmopolitanism and Distributing Responsibilities’, Critical Review of International Social and Political Philosophy 5: 92–97.

Brooks, Thom (2007). ‘Punishing States That Cause Global Poverty’, William Mitchell Law Review 33(2) (2007): 519–32.

Brooks, Thom (ed.) (2008). The Global Justice Reader. Oxford: Blackwell. Brooks, Thom (2010). ‘Justifying Terrorism’, Public Affairs Quarterly 24: 189–95.

Brooks, Thom (2011). ‘Rethinking Remedial Responsibilities’, Ethics and Global Politics 4(3): 195–202.

Brooks, Thom (ed.) (2012). Global Justice and International Affairs. Leiden: Brill. Brooks, Thom (2013). Global Justice: An Introduction. Oxford: Blackwell.

Dobos, Ned (2011). ‘Consistence in the Armed Enforcement of Human Rights: A Moral Necessity?’ Journal of Moral Philosophy 8(1): 92–109.

Frowe, Helen (2011). ‘Self-Defence and the Principle of Non-Combatant Immunity’, Journal of Moral Philosophy 8(4): 530–46.

Lefkowitz, David (2009). ‘Partiality and Weighing Harm to Non-Combatants’, Journal of Moral Philosophy 6(3) (2009): 298–316.

McMahan, Jeff (2005). ‘Just Cause for War’, Ethics and International Affairs 19: 1–21. McMahan, Jeff (2009). Killing in War. Oxford: Oxford University Press.

Øverland, Gerhard (2005). ‘Self-Defence among Innocent People’, Journal of Moral Philosophy 2(2): 127–46.

Øverland, Gerhard, ‘Conditional Threats’, Journal of Moral Philosophy 7(3): 334–45. Øverland, Gerhard (2011). ‘Dividing Harm’, Journal of Moral Philosophy 8(4): 547–66. Reitan, Eric (2010). ‘Defining Terrorism for Public Policy Purposes: The Group-Target

Definition’, Journal of Moral Philosophy 7(2): 253–78.

Shalom, Stephen R. (2011). ‘Killing in War and Moral Equality’, Journal of Moral Philosophy

8(4): 495–512.

Statman, Daniel (2011). ‘Can Wars Be Fought Justly? The Necessity Condition Put to the Test’, 8(3): 435–51.

Part One: Just War and Unjust War

The Permissibility of Aiding and Abetting Unjust Wars

Saba Bazargan

1. Introduction

As citizens, we are often in a position to contribute politically to a war fought by our government. More often we are compelled to contribute financially. Some of us are in a position to promote a war more directly, by enlisting for military service. It is no surprise, then, that we are often concerned about the moral permissibility of contributing to wars fought by our government. It is tempting to assume that if a war is unjust, it is morally impermissible to promote that war. But I will argue that the impermissibility of promoting a war cannot be inferred from the fact that the war is unjust.

The argument, put briefly, is as follows. The relevant alternatives available to a government waging a war differ systematically from the relevant alternatives available to civilians, combatants, and other governments, etc., who might contribute to the war. Whether it is permissible to perform an act often depends on the alternatives available to the actor. Hence the conditions determining whether it is permissible for a government to wage a war often differ from the conditions determining whether it is permissible for other actors to promote that war. This difference in conditions is often present in unjust wars that have both just and unjust aims. As a result, it is sometimes permissible for individuals to promote such wars.

Before proceeding, it is important to distinguish two ways in which it might be permissible to promote an unjust war. It might be permissible to do so either for reasons having to do with the aims of the war or for personal or private reasons unconnected with the aims of the war. Suppose that by enlisting in the military, a particular civilian will advance her career, make her parents proud, fund her college education, etc. Or more dramatically, suppose that enlisting will save her life, by making her inaccessible to mafia members intent on murdering her. Even if these facts provide a decisive moral reason to participate in an unjust war, they do not provide the sorts of reasons that I will address. Instead, I will argue that moral and non-moral facts about an unjust war itself—as well as facts regarding an individual’s ability to affect that war—can provide sufficient moral reasons to promote that war.

2. Preliminaries

Before I present the primary argument I will address some preliminaries regarding terminology, the conceptual relationship between aims and wars, as well as the different ways in which aims can be unjust.

Often the word “war” is used analogously to the word “duel”—i.e., to refer to both sides in a particular type of conflict. This is not the sense of ‘war’ that I will be using here. Following the practice of just war theorists, my use of the word “war” will distinguish between sides in a conflict. For example, what we call ‘the Franco-Prussian War’ consisted of France’s war against Prussia and Prussia’s war against France. These were, in one sense, distinct wars. This is how I will use the word “war”. Only a war fought by one side or another can be just or unjust. Understood as that which is fought by all the belligerent parties, a war can be neither just or unjust, since it is itself a conflict between the just and the unjust (or between unjust parties).

Waging a war involves the pursuit of aims through the application of military force. I will make a conceptual distinction between two types of aim. The ultimate aims of a war are those that explain the resort to war. These aims are the motivating reasons for the government’s resort to war. When referring to the aims of a war, I will assume that the aims are ultimate, unless stated otherwise.

Ultimate aims subsume subsidiary aims. These are aims the achievement of which is intended to cause or constitute, either wholly or in part, the achievement of particular ultimate aims. For example, an ultimate aim of a war might be to secure direct access to a foreign oil supply. A subsidiary aim of this ultimate aim might be to neutralize enemy anti-aircraft installations in proximity to the oil fields.

It is, of course, an idealization to claim that governments adopt a particular set of aims for a war. There are varying degrees of commitment towards the pursuit of particular aims, and leaders are often capricious in their commitments. Moreover, the aims adopted might be indeterminate or ill-defined. The aims of a war are usually the result of collective decision-making; this can introduce indeterminacy with respect to the war’s aims, depending on the extent of the disagreement among the decision makers and the decision-making procedure that the collective uses. Sometimes the aims are intentionally left vague in order to reach consensus among members of a gridlocked government, or to facilitate postbellum claims of success; sometimes they are unintentionally vague simply as a result of unreflective leadership. I believe the arguments I will provide can be amended to fit these scenarios. But for the sake of simplicity, the hypothetical wars that I will discuss will have clear and stable ultimate aims.

So far I have discussed only the structural relationships among the aims in a war. Now I turn to the moral evaluation of those aims. The following claims will be highly generalized, so that my ultimate conclusion (that promoting an unjust war is sometimes permissible) will be compatible with a variety of theories of jus ad bellum (i.e., the conditions according to which a resort to war is justified).

There are, broadly construed, three reasons why pursuing an aim can be impermissible. First, pursuing an aim might be unjust, and therefore impermissible, if the aim itself necessarily involves the violation of rights. In such cases there are typically no methods of achieving the aim permissibly. A fortiori, military force is generally impermissible as a means to achieving such an aim. Call such aims ‘intrinsically unjust’. Genocide is an obvious example of an intrinsically unjust aim.

Second, pursuing an aim can be impermissible because it violates constraints of propotionality. Jeff McMahan argues that there are two kinds of proportionality violations.1 Sometimes an agent commits a harm or wrong for which that agent is liable to be harmed. But to kill the agent might exceed the harm to which she is liable—that is, the harm may be disproportionate in relation to the degree of her liability. Hence the agent is not liable to be killed. McMahan calls this a constraint of ‘narrow proportionality’. An example of an aim the pursuit of which would violate narrow proportionality is that of marginally improving the status of women, for which no one bears enough liability to be justifiably killed.2

Alternatively, an aim might violate what McMahan calls a constraint of ‘wide proportionality’, in which the good of the aim being pursued (and perhaps the good side-effects of its pursuit) is weighed against the harms caused to wholly innocent people, usually as a side-effect of pursuing the just aim. For example, collateral damage to civilians during a tactical bombing of a munitions factory must be weighed against the good of destroying the munitions factory for the bombing to satisfy the constraints of wide proportionality.

1 See Jeff McMahan, Killing in War (Oxford University Press 2009).

2 This example belongs to Thomas Hurka “Proportionality in the Morality of War,” Philosophy and Public Affairs 33 (2002): 34-66 at pp. 42. Its explication in terms of liability belongs to Jeff McMahan, supra.

There is a third way in which the pursuit of an aim by a particular means is impermissible. A particular means to the accomplishment of an aim can satisfy constraints of wide and narrow proportionality, and still be impermissible to pursue, if there is an even less harmful but equally effective means of accomplishing the same aim (or perhaps a different aim that would make an equal contribution to the achievement of the just cause). This is because there is a constraint on the means to the pursuit of an aim that is independent of the constraints of proportionality, viz. the necessity constraint, which rules out harm that is unnecessary for the achievement of a just aim.

To summarize, I have distinguished several ways in which pursuing an aim can be impermissible. An aim might be intrinsically unjust. Or the particular means of pursuing an aim might violate constraints of (wide or narrow) proportionality. Or the particular means might be unnecessarily harmful. Having drawn these distinctions, I can now argue that it is sometimes permissible to promote unjust wars.

3. Aiding and Abetting an Unjust War

Consider the following three cases.

1) A civilian is contributing to a war by political or economic means, or by enlisting for military service. However, she knows that this war is unjust.

2) A combatant is fighting in a war waged by her government. She has come to believe, correctly, that this war is unjust. She is in a position to voluntarily cease fighting.

3) A government of one country is assisting in a war waged by the government of another country. The former recognizes that the latter is waging an unjust war.

In each case, there is a principal wrongdoer: the government violating jus ad bellum. And in each case there is an accessory aiding or abetting the principal wrongdoer: a civilian, a combatant, and a government, respectively. I will focus on the permissibility of aiding and abetting the principal wrongdoers in each of these three cases. This is best done through an example.

Suppose a government, as a result of civil unrest, embarks on a campaign of atrocities against its own population in order to deter further resistance. The government has its soldiers commit unspeakable acts against the civilian population indiscriminately. Call this country ‘atrocity.’ The government of a bordering country is considering military intervention in order to stop the massacres for humanitarian reasons. Call this country ‘intervene.’

The government of intervene is considering launching a ground assault to neutralize the military units carrying out the massacres, most of which consist of the government’s private guard. Suppose the government of intervene is aware that this would effectively eliminate the means by which the country’s despots maintain their control, allowing the people of atrocity, if they wish, to overthrow the government and replace it with a provisional one of their own choosing. The people of atrocity would welcome intervene’s assistance in stopping the massacres and the government of intervene is aware of this. It is also aware that stopping the massacres will satisfy the constraints of necessity and proportionality (both narrow and wide).

However, intervene is considering another aim, in addition to stopping the massacres. atrocity contains a strip of unpopulated land that runs along its border with intervene. This borderland is strategically ideal for intervene as a buffer between the two countries; it also has valuable deposits of oil and natural gas. Because of this, the government of intervene is considering annexing this borderland, in addition to neutralizing the massacring military units. However, if the government of intervene pursues both aims, it will seize the borderland first and only then stop the massacres.

Suppose that annexing the borderland will not physically harm any civilians. Pursuing both aims—stopping the massacre and annexing the borderland—would be better, both for the civilians of atrocity and for the people of intervene, than pursuing neither aim would be. Suppose further that the government of intervene is aware that the people (though, of course, not the government) of atrocity would rather bear the violation of their rightful sovereignty over the borderland than continue to be subjected to massacres by their government. This is not to say that the people of atrocity do not mind the annexation. It can be predicted that once the provisional government is in place, the people of atrocity will, via this government, protest the annexation. However, they will have neither the military, political, nor economic resources to reclaim the borderland.

The ultimate aim of annexing the borderland is intrinsically unjust. intervene has no right to the territory. Acquiring it is not the sort of aim that can permissibly be pursued through military force, regardless of how few casualties are incurred. And the government of intervene is, by hypothesis, in a position to adopt the aim of stopping the massacres without annexing the borderland. So annexing the borderland is not, for the government, subsidiary to the just aim of stopping the massacres.

But suppose that because the government of intervene does not benefit by pursuing only the aim of stopping the massacres, it would rather do nothing, thereby allowing the massacres to occur, than go to war without pursuing the annexation. Is this a reason to believe that a war with the aims of both stopping the massacres and annexing the borderland is permissible?

It is hard to see how this could be so, given that the aim of annexing territory is not subsidiary to stopping the massacres. The government of intervene is, by hypothesis, free to pursue the aim of stopping the massacres without annexing the borderland. Pursuing a set of aims that includes annexing the borderland is unjust partly because doing so involves freely rejecting an alternative set of aims that does not include annexing the borderland. A recalcitrant disregard for reasons not to perform a certain act typically does not diminish the reasons not to perform that act. So even if the government of intervene will stop the massacres only if it also annexes the borderland, pursuing both aims is unjust. This is so despite the fact that pursuing both aims makes things better overall than they would be if the government of intervene chose to pursue none of the aims.

A war that make things better overall relative to the absence of that war, yet is nonetheless unjust, can be called ‘narrowly unjust.’ Unjust wars that do not make things better overall relative to the absence of that war, I will call ‘broadly unjust.’ These classifications help reveal the moral heterogeneity of the possible aims of unjust wars.

It might be argued that a war is just if and only if going to war has better consequences than not going to war. On this view, a war resulting in an improvement over what would have been the case without that war is just. But this view is absurd. If a war is just if and only if it has better consequences than not going to war, then it is morally permissible for a government to ‘tack on’ gratuitously harmful, self-serving aims when waging otherwise just wars, up to the point at which an additional aim would take the war past the threshold of disproportionality. On this view, if a country is the victim of unjust aggression, the government of that country can permissibly pursue aims that, for example, ignore duties of care, as long as pursuing this aim in combination with pursuing the aim of self-defense has better consequences than not going to war at all. Or if the government of a country (such as intervene) is waging a war with a humanitarian aim, the government can permissibly pursue aims wronging the people requiring assistance, as long as these wrongs are, for its victims, a small price to pay in comparison to losing assistance from the intervening power. But this is not just; it is extortion.

Of course, if achieving the aim of stopping the massacres is costly then intervene might be entitled to compensation. For intervene to be entitled to the borderland, the people of atrocity would have to agree to give it up to intervene as compensation for military assistance. Yet it is unrealistic to presume that the victims of an oppressive regime would have the political voice necessary to explicitly contract with a foreign power. As a result, perhaps it is permissible for intervene to act according to a hypothetical contract; its terms are determined partly by what the people of atrocity would agree to, or what it would be rational or reasonable for them to agree to, if they were to explicitly enter into such a contract. I will assume that the people of atrocity would indeed consent to sacrificing the borderland as the price for intervention. This does not mean, however, that intervene is entitled to the borderland. Individuals often agree to contracts under duress or in extreme conditions. Seeking agreement to a contract under such circumstances can be extortionate, if the price of the service offered is either excessive in relation to the cost of providing it or in relation to the value of the service itself—though this does not imply that any non-extortionate demand is morally permissible. The same might be said of hypothetical contracts in which the hypothetical agreement is made under conditions of duress. We can assume both that the people of atrocity are under duress, and that the value of the borderland is significantly higher than what it costs for intervene to stop the massacres; that is, intervene takes more than what is necessary to compensate for the costs of the war, and thereby profits from it. So, even though the people of atrocity would prefer that the government of intervene pursue both the just and unjust aims rather than pursue neither, and even though they would contract accordingly, it does not necessarily follow that intervene is guilty of no wrong for taking the borderland.

In this section I have distinguished broadly unjust wars from narrowly unjust wars, and presented an example of the latter. In the next section, I use this example to show that it is often permissible for combatants to participate in, and for civilians to contribute to, narrowly unjust wars.

4. Aiding and Abetting Narrowly Unjust Wars

According to contemporary orthodox Just War theory, the moral permissibility of participating in a war does not depend on whether that war is just. This view, which has come to be known as ‘the Independence Thesis,’ has recently been contested.3 But instead of arguing for or against this view, I will argue that even if the Independence Thesis is false, it is still often permissible to fight not only in narrowly unjust wars, but also for the unjust aims of such wars. I will also argue that it is often permissible for civilians to promote narrowly unjust wars as well.

I have claimed that whether a set of aims (and thus a war) is just depends on what alternatives are available to the government. In the example I have presented, the government of intervene has the option of waging a war with only just aims. But unlike a government, individual civilians and combatants typically do not have the power to choose what aims a war will have. Because an agent can be morally required to do only what that agent is capable of doing, a typical individual cannot be morally required to change the aims of an unjust war fought by her government.

But a typical individual does have the power to exert a marginal influence on the aims of a war fought by her government. As a result, there are obligations, permissions, and restrictions that apply to marginally promoting some aims over others. Suppose that an individual can promote one or both of the following aims: a) annex the borderland b) stop the massacres

Where (a) and (b) constitute a narrowly just war, and (b) without (a) constitutes a just war. The combination of not-(a) and not-(b) is the absence of a war waged against atrocity. If intervene is pursuing both (a) and (b), then combatants, by participating in the war, might promote one or both aims, though they are usually not in a position to choose which. It is possible, though typically quite difficult, for a combatant to promote none of the aims by ceasing to participate in the war altogether. Civilians can also marginally promote (a), (b), or neither by voting for the appropriate politicians, joining or working for the appropriate organizations (such as

3 See Jeff McMahan, “The Ethics of Killing in War,” Ethics 114 (2004): 693-733, and “On the Moral Equality of Combatants,” Journal of Political Philosophy 14 (2007): 377-93. See also Rodin, D. and H. Shue, eds Just and Unjust Warriors: The Moral and Legal Status of Soldiers (Oxford University Press, 2008). protest groups or military recruitment centers), donating financially to the appropriate campaigns, writing editorials advocating the relevant position on the war, enlisting for or resisting military service, etc. Unlike combatants, civilians typically have more leeway when it comes to deciding which aims to promote, and unlike combatants, a civilian is typically free not only to refrain from promoting any of the aims, but also to promote not-(a) or not-(b). But the contributions made by a civilian are typically less significant than those made by a combatant.

Which aims can a combatant or a civilian permissibly promote? Consider a combatant participating in the pursuit of the just aim (b) of a narrowly unjust war. It is hard to see how the fact that the war as a whole is unjust makes it impermissible to participate in the pursuit of a just aim. The fact that the war is unjust is compatible with the claim that killing the combatants who are participating in the massacres satisfies the constraints of discrimination, proportionality (narrow and wide), and necessity. (At this point, I am assuming that the combatant contributes solely to the achievement of the just aim in the unjust war.) Contemporary orthodox Just War theory, by classifying wars as either ‘just’ or ‘unjust,’ obscures the moral heterogeneity of the aims of unjust wars. Labeling a war as unjust can misleadingly suggest that it is impermissible to participate in the pursuit of any of the aims of the war.

It is likewise permissible for a civilian to promote a just aim in an unjust war. Suppose a plebiscite is held, in which the electorate of intervene is asked to choose between non-intervention and waging the narrowly unjust war against atrocity. In this case, promoting the just aim comes at the cost of promoting the unjust aim. Does this provide a decisive reason for the civilian to vote in favor of pursuing neither aim? Assuming that the civilian is restricted to choosing between these two options, it is morally permissible for her to vote in favor of the narrowly unjust war. That is, the civilian, in this case, is permitted to promote the narrowly unjust war over non-intervention, since promoting a just war is not an option. Even though the aim of annexing the borderland is not, for the government, subsidiary to the aim of stopping the massacres, promoting the annexation of the borderland is—for the civilian—required in order to promote an end to the massacres. The just and unjust aims are, for the civilian, packaged together.

These claims do not entail the view that when our choices are limited to promoting either of two unjust wars, one of which is worse than the other, it is permissible to choose the lesser evil. Rather, I am claiming that when our choices are limited to promoting a narrowly unjust war over no war at all, it is permissible to choose the former, partly because a narrowly unjust war is better, impartially considered, than no war at all (unlike the lesser evil of two broadly unjust wars). Still, it might be asked: why suppose that what is better overall trumps the injustice involved in intervene’s annexation of the borderland?

Annexing the borderland is a rights violation, but sometimes it is permissible to violate rights if doing so is necessary to avert significantly worse consequences, such as massacres, which is also a far more egregious type of rights violation than the violation of territorial sovereignty. There are two ways in which this is so. First, theft is generally not as wrongful or harmful as murder. Second, though violating territorial integrity wrongs everyone in atrocity, this wrong is not in of itself seriously harmful to any particular individual, unlike the wrong of injuring or murdering someone. The annexation is, instead, a widely dispersed, comparatively minor harm that a great many people will suffer. The right protected by stopping the massacres is the individual right not to be murdered, while the right violated by the annexation is a collective right to territorial integrity. Nonetheless, the annexation is still a serious rights violation; it is impermissible for the government of intervene to annex the borderland because doing so is not necessary to avert the massacres. But for civilians choosing between the narrowly unjust war and no war at all, those rights violations are necessary to stop the massacres. Hence it is permissible to promote the narrowly unjust war when the only other option is to allow the massacres to occur.

A combatant contributing solely to the just aim is in a similar situation. I will discuss this possibility shortly. By contributing to the just aim, he is participating in a narrowly unjust war—but this is permissible since the option of pursuing the just aim without participating in a narrowly unjust war is not available to him. But unlike the civilian voter, this combatant is not promoting the unjust aim of the narrowly unjust war. So, interestingly, the act of contributing to the just aim of a narrowly unjust war is morally better than the act of voting in favor of that war. (Of course, contributions to the just aim might contribute to the unjust aim as well—for example, by fighting in the capital to stop the massacre, a soldier from intervene might be drawing atrocity’s soldiers away from the borderland, making it easier for intervene to capture it. I will discuss such possibilities shortly.) It can also be permissible for combatants to participate solely in the unjust aim of the narrowly unjust war against atrocity. Recall that intervene will not pursue the just aim of stopping the massacres until it achieves the unjust aim of annexing the borderland. Because of this, combatants fighting for intervene can permissibly participate in missions promoting the unjust aim of annexing the borderland—even though this is the aim by virtue of which the war is unjust. Put generally, this is because whether a particular act is necessary for the achievement of a desired end may depend on whether we adopt a first-person or thirdperson perspective with respect to that act. For the government of intervene, annexing the borderland is not subsidiary to the aim of stopping the massacres. Rather, the government has chosen to pursue the latter only if it pursues the former. But things are different for those individuals, such as combatants, uninvolved in the government’s choice of aims. The aims of the war are a matter of choice for the government but are unalterable facts about the world from the point of view of the combatant. So even though the aim of annexing the borderland is not subsidiary to the aim of stopping the massacres, the annexation of the borderland is—for the combatant—required in order to stop the massacres. The combatant can permissibly promote the achievement of the unjust aim for the same reason that the civilian can vote in favor of the narrowly unjust war: the alternative allows serious rights violations to occur, which are of a type far more egregious than the kind involved in the pursuit and achievement of the unjust aim in the narrowly unjust war.

It is permissible for combatants and civilians to promote the annexation when they do not have the option of promoting solely the just aim of stopping the massacres, not merely because the annexation is, for the civilian and the combatant, a means to preventing the massacres. As I noted earlier, the people of atrocity would prefer that the government of intervene wage the unjust war rather than wage no war at all. This does not make it permissible to wage the narrowly unjust war. But it does provide a pro tanto reason to promote that war when it is impossible to promote what the people of atrocity would freely consent to, viz., a just war with the sole aim of stopping the massacres.

It might be not only permissible for a civilian to vote in favor of the narrowly unjust war, but morally required as well. By hypothesis, the government of intervene profits from waging a war in which the borderland is annexed. As I have argued, this fact does not itself make it permissible for intervene to wage such a war. But it might require a civilian to promote that war, if the civilian does not have the alternative of promoting a just war, as is the case in the plebiscite. In such a case, on what grounds could a civilian permissibly refrain from promoting the narrowly unjust war? After all, the war is an impartially better state of affairs that it is permissible to promote and is on balance profitable for intervene. Without other reasons in favor of refraining from promoting the narrowly unjust war, promoting it may be not only permissible for a civilian whose only other option is to refrain from promoting any war, but also obligatory. But it is not necessarily obligatory. While it is true that the narrowly unjust war is a significantly better state of affairs overall than the status quo, and that the narrowly unjust war is profitable to intervene, it might very well be a worse state of affairs for a civilian of intervene who (or who has relatives who) might be drafted in military service if her government goes to war. Such a civilian might be permitted to refrain from promoting the narrowly unjust war for agent-relative reasons.

It may be, on the other hand, that a combatant is not morally required to fight in the narrowly unjust war against atrocity, since promoting a war by fighting in it involves significant personal risk. If there is an obligation for a combatant to fight against atrocity, it is likely that the obligation will arise from commitments the combatant has made qua combatant, rather than a duty to promote a narrowly unjust war at significant personal risk. But since (or so I have argued) it is at least permissible for a combatant fighting for intervene to participate even in the unjust aims of the narrowly unjust war, one might ask: can the defenders of the borderland permissibly fight back in self-defense? Suppose the defenders have not been and will not be assigned to the defense of those committing the massacres. Is it permissible for them to fight back against intervene’s combatants? To determine this, it is necessary to address the issue of liability to defensive attack. According to a common understanding of what has come to be known as the Moral Equality of Combatants, those fighting on each side of a war are morally liable to attack by the other side. For the purposes of this paper, I am assuming that the Moral Equality of Combatants is false (as I believe it to be). I am also assuming the falsity of a claim undergirding the moral equality of combatants. According to this claim, an agent who poses a threat to others thereby loses the right not to be attacked in self-defense regardless of whether the threat is justified. But I will assume that justification precludes liability to attack. This is Jeff McMahan’s view. He notes that “it is hard to see how one’s moral immunity to being killed could be compromised merely by one’s acting in a way that is morally justified.”4 So if a combatant is justified in participating in an unjust war, then that combatant is not liable to be attacked— provided that she will not promote unjust aims in the future for which she

4 See Jeff McMahan, “The Basis of Moral Liability to Defensive Killing,” Nous Supplement 15 (2005): 386-405. at pp 288. might be liable to preventive attack, and that she has not promoted unjust aims in the past for which she might be liable to attack in reprisal.

Because intervene’s combatants are not only permitted to attempt to annex the borderland but are positively justified in doing so, the defenders of the borderland are not permitted to fight back in self-defense. For the defenders of the borderland, just as for intervene’s civilians and combatants, the loss of the borderland to intervene is subsidiary to the end of stopping the massacres. The defenders ought not to oppose what is necessary, as far as they are concerned, to stop the massacres. The annexation is not of course necessary for stopping the massacres where the government of intervene is concerned, but the options that the defenders have are different. So even though the defenders of the borderland have a just cause in defending the borderland, they are not justified in trying to achieve it. They are obliged to allow the wrong to be done, in order to prevent (or to not stop the prevention of) the occurrence of even greater wrongs done by others.

Still, it might seem perverse to claim that the soldiers fighting for their country ought to allow a foreign invader to unjustly annex territory. Even if things go impartially best should the defenders surrender, they still have a pro tanto, agent-relative reason to defend the borderland. It is, after all, part of their country—and they have a special interest in keeping their own country intact. It is this special relation—that of citizenship or residence—that grounds an agent-relative, pro tanto reason to defend the borderland, in addition to the pro tanto, agent-neutral reason to prevent wrongful annexations. As a result, it might be argued, the defenders have a pro tanto moral reason to do, in this case, other than what is impartially best. But it is doubtful that this agent-relative reason, in combination with the pro tanto, agent-neutral reason to defend the wrongful annexation of the borderland, is strong enough to outweigh or override the agent-neutral reason to allow the annexation. Perhaps, if the defenders happen to live on the borderland, contra what I assume, their agent-relative reason to defend their homes is strong enough to justify attacking the invaders. But as it stands, their agent-relative reason is not strong enough. And this is not to mention that, in addition, the defenders might also have an agent-relative reason not to impede efforts to stop the massacres precisely because the massacres are being committed by their government, for whose action they may bear some moral responsibility. This is further reason to think that the soldiers ought to accede to the wrongful demands of intervene. It is, of course, unlikely that soldiers would actually do this—and perhaps their culpability for failing to surrender is significantly mitigated by the fact that they correctly see themselves as defending against unjust aggression. But this is not an issue I address here.

In arguing that it can be permissible for combatants to participate in the unjust aims in narrowly unjust wars, I have argued as if the pursuit of an aim in a war is causally isolated from the achievement of other aims in that war. But if the war-planners are strategically rational, then the war’s ultimate aims—both just and unjust—will likely share subsidiary aims. As a result, promoting one aim is likely to promote another. Alternatively, promoting one aim might inhibit the achievement of another aim, either unavoidably or as a result of strategic incompetence. In any case, the permissibility of participating in the achievement of particular aims of an unjust war is complicated by the fact that the aims of the war might be causally mixed.

For example, suppose that annexing the borderland or stopping the massacres requires defeating atrocity’s army. For interevne to defeat this army, then, would promote both the just aim of stopping the massacres and the unjust aim of annexing the borderland. In such a case, if a combatant participates in combat operations against atrocity, she promotes both just and unjust aims. So the combatant can either promote both aims, or cease participating altogether. Like the civilian voter discussed earlier, the combatant in this case does not have the option of promoting solely the just aim. In these circumstances, it is permissible for the combatant to participate in the aim of defeating atrocity’s private army, even though this aim is, for the combatant, subsidiary to both the just and the unjust aims of the war. It is permissible for the combatant to do so for the same reasons that it is permissible for a civilian in intervene to vote in favor of a narrowly unjust war against atrocity. Neither the civilian nor the combatant has choices that discriminate between promoting only the just aims of the war and promoting all the aims of the war.

The claim that combatants participate impermissibly in a war or in the pursuit of an aim is not meant to imply that they are blameworthy for doing so. Participating impermissibly is consistent with doing so non-culpably. There are various mitigating factors that partially or fully excuse combatants for participating impermissibly. For instance, combatants operating at the behest of a state rarely have a choice regarding what missions to participate in, and thus what aims to promote (though combatants participating in loosely organized military units or guerrilla cells often have more lee-way regarding what missions to pursue). Typically, the only alternatives a combatant has to participating in the missions to which she has been assigned are intentionally failing to promote the mission and withdrawing from military service altogether. Threats of physical and psychological punishment imposed by the military against those who disobey their orders mitigate their culpability for impermissible participation in a war. But they do not affect the permissibility of participation.

There are various other factors that potentially mitigate culpability for impermissible participation. These include non-culpable deficiencies in critical reflection on the moral justifiability of particular aims, nonculpable ignorance regarding morally relevant non-moral facts (including unforeseeable causal mixing among seemingly disparate aims), nonculpable irrationality associated with intense combat, etc. But I am here concerned with the moral permissibility of participating in unjust wars— not with the culpability associated with doing so. The latter, though extremely important, cannot be adequately addressed here.

So far I have argued that it is sometimes permissible for civilians to promote narrowly unjust wars, and that it is sometimes permissible for combatants to fight for the unjust aims of narrowly unjust wars. The same reasoning can be applied to governments that act as third parties in a conflict.

Governments are often in a position to promote an unjust war fought by a foreign government. The most direct method of doing so is to join the war. Recall that a reason why intervene’s war against atrocity is unjust is that the government of intervene is in a position to wage the war without the aim of annexing the borderland. To nonetheless wage the war with that aim is to wage an unjust war. Suppose, however, that intervene does not have the capability of destroying fortified military installations protecting the borderland. This prevents the government of intervene from annexing the borderland, but does not prevent it from stopping the massacres. But if it cannot annex the borderland, the government of intervene will not pursue the aim of stopping the massacres.

Suppose that there is another country, ‘help,’ the government of which would like to intervene in atrocity to stop the massacres. Unlike the government of intervene, the government of help is willing unconditionally to pursue the aim of stopping the massacres. But doing so requires large numbers of troops, which help lacks. help does, however, have the technologically advanced munitions that intervene lacks that are required to destroy the fortified military installations guarding the borderland. Thus the government of help is capable of carrying out the aim of annexing the borderland, but not the aim of stopping the massacres. intervene, on the other hand, possesses the troops required to achieve the aim of stopping the massacres, but lacks the weaponry required to achieve the aim of annexing the borderland.

The government of intervene is willing to pursue the aim of stopping the massacres if help agrees to destroy the fortified military installations that protect the borderland first. After this is done, the government of intervene will send in troops to occupy the borderland and to neutralize the military units carrying out the massacres. The government of help can either promote the annexation of the borderland or choose not to go to war, thereby allowing the massacres to continue. The conspicuously absent choice is that of stopping the massacres without annexing the borderland. Like the civilians and combatants discussed earlier, the government of help is limited in its options, in that it can promote the just aim only by contributing to the unjust aim. Achieving the aim of annexing the borderland is, for the government of help, subsidiary to achieving the aim of stopping the massacres. The government of help is therefore permitted to pursue the aim of annexing the borderland for the same reason that intervene’s combatants are permitted to participate in the pursuit of that aim.

Generalizing, I claim that countries can sometimes permissibly join a war, even if that war is overall unjust. Moreover, countries can sometimes permissibly pursue only the unjust aims of an unjust war.

5. Conclusion

I have argued that whether it is permissible to promote an unjust war sometimes depends on who is doing the promoting—the government, combatants, civilians, other governments, etc. Because these actors often have different options available to them, we cannot rely on the conditions under which the government can permissibly wage a war to determine the conditions under which others can permissibly promote that war.

Different theories of just war might provide different conditions under which a resort to war is morally permissible. But because the conditions under which it is permissible for a government to wage a war come apart from the conditions under which it is permissible for others to promote that war, we require an independent theory of permissible promotion by third parties. Such a theory would take into consideration who is promoting the war, how it is being promoted, as well as what aims are being promoted. I have not presented a full theory of permissible promotion in war; rather, I have argued that we need such a theory, by demonstrating that the permissibility of promotion cannot be derived from the fact that a war is unjust. So if we are interested in knowing whether promoting an unjust war is permissible—as we should be, considering that many of us contribute, albeit marginally, to unjust wars—then developing an independent theory of the conditions of the permissible promotion of a war is imperative.5

References

Brandt, R.B. “Utilitarianism and the Rules of War”. Philosophy and Public Affairs (1971/72), 145.

McMahan, J. “The Ethics of Killing in War”, Ethics 114 (2004): 693.

—— “The Basis of Moral Liability to Defensive Killing”, Nous Supplement 15 (2005): 386. at pp. 288.

—— “On the Moral Equality of Combatants”, Journal of Political Philosophy 14 (2007): 377.

—— “Précis: the Morality and Law of War”, Israel Law Review 40 (2007): 310-323. at pp. 321.

Hurka, T. “Proportionality in the Morality of War”, Philosophy and Public Affairs 33 (2002):

34. at pp. 42.

—— “Liability and Just War”, Ethics & International Affairs 22 (2007) 199. Rodin, D. War & Self-Defense, (Oxford University Press, 2005).

Rodin, D. and H. Shue, eds Just and Unjust Warriors: The Moral and Legal Status of Soldiers

(Oxford University Press, 2008).

Shue, H. The Oxford Handbook of Practical Ethics ch. 28 (LaFollette, Hugh ed., 2003), 731.

Walzer, M “Political Action: The Problem of Dirty Hands”, Philosophy and Public Affairs

(1972/73), 160.

—— Just and Unjust Wars: A Moral Argument with Historical Illustrations. (New York: Basic Books, 3rd ed., 2000).

5 I would like to thank Jeff McMahan for invaluable comments and criticism.

Dividing Harm

Gerhard Øverland

Those arguing against the permissibility of killing innocent aggressors take mere causal contribution to harm to be morally insignificant.1 To be a permissible target for defensive force according to this view, a person must be morally responsible for a threat to others. In this paper I argue that mere causal contribution to (the threat of) harm is morally significant on two counts: a) innocent aggressors have a duty to bear additional costs to help protect their potential victims, as compared to the duty innocent bystanders are expected to bear,2 and correspondingly; b) it is permissible to use more force against innocent aggressors, as used in self-defense and the defense of others, than innocent bystanders.

The paper has two parts. First I aim to demonstrate the intuitive plausibility of this proposal and what I call “the asymmetrical fair share procedure.” According to this procedure, innocent aggressors have a duty to take on a fair share of the harm if dividing it is possible, and a fair share of the risk of being harmed if redistribution of harm is impossible. In the second part, I develop a contractual account explaining why mere contribution is morally significant.

Preliminaries

What it means to contribute to harm remains controversial. I am interested in the simple, paradigmatic cases in which an agent initiates a

1 There is a growing literature on the topic and an increasing number of philosophers seem to support this view; see, for instance, N. Zohar “Collective War and Individualistic Ethics: Against the Conscription of ‘Self-Defense’” Political Theory 21 (1993): 606–622; Michael Otsuka “Killing the Innocent in Self-Defence,” Philosophy and Public Affairs 23 (1994): 74–94; Jeff McMahan “Self-Defense and the Problem of the Innocent Attacker” Ethics 104 (1994); David Rodin War and Self-Defense (Oxford: Clarendon Press 2002), at pp. 70–99; Jeff McMahan “The Basis of Moral Liability to defensive Killing” Philosophical Issues 15 (2005), and; Kimberley Kessler Ferzan “Justifying Self-Defense” Lawand Philosophy 24 (2005): 711–749.

2 As for innocent bystanders, I will assume that they ought to comply with a principle of assistance. For instance, if they can prevent something bad from happening at relatively moderate cost then that is what they ought to do. complete causal process resulting in harm.3 For example, someone fires a gun and hits someone else, or they jump off a bridge to hit the person below, and so forth. One might initiate a harmful causal sequence without being morally at fault. Someone could be an innocent initiator of a harmful process, for instance, when the agent excusably believes that harming another is permissible, perhaps because she has good grounds for believing this, or has been manipulated into doing so.

People who are innocent, but who play an essential part in such complete causal processes can be divided into different categories. Some may have good reason to believe that harming a particular person is a necessary corollary of a just cause (innocent aggressors); a person’s actions may have caused harm accidentally (innocent agents of harm); and some may have done so through no action of their own at all (innocent threats). I will not discuss innocent threats in this paper, only cases involving agents whose actions initiate harmful processes, accidentally or deliberately.4

The type of contributor I am interested in is therefore doingly related to the bad outcome.5 A typical case I shall be discussing is the following, and variations of it:

Push: A cart stands at the top of a hill. Bill pushes it and jumps on board. The cart rolls down the hill and injures Alice at the bottom.6

There is no question here about who did it. Bill injured Alice. And if innocent, Bill would be an innocent doer of harm. Bill could, for instance, have had good reason to believe Alice was about to cause harm, and that the best way of preventing her was to jump on the cart and run her over.

3 According to Philippa Foot, the relevant distinction between doing and allowing harm is primarily one of initiating or sustaining a harmful causal sequence, and allowing a harmful causal sequence to run its course. See Philippa Foot “Killing and Letting Die” in

B. Steinbock and A. Norcross Killing and Letting Die (Fordham University Press: 1994): 280– 289, at p. 281. What it means to initiate and sustain a causal sequence needs to be made more precise. I cannot do that here, but will indicate what I think is an essential part of such an account. It would include the notions of energy transfer and energy release; see David Fair “Causation and the Flow of Energy” Erkenntnis 14 (1979): 219–250. I shall leave those who merely sustain a harmful sequence out of the discussion, as including them would complicate matters.

4 For simplicity, I am lumping together those who become innocent aggressors through misfortune and manipulation. I do not think the distinction is very significant, as exposure to manipulation is basically a form of misfortune, but will not argue for it here.

5 In the following paragraphs I depend on a work in progress, Christian Barry and Gerhard Øverland ‘The Doing, Allowing, and Enabling Distinction.

6 See, for instance, Kadri Vihvelin and Terrance Tomkow’s recent paper “The Dif” The Journal of Philosophy 102 (2005): 183–205, at pp. 192–193.

Or perhaps Bill neither knew, nor could have been expected to know of Alice’s whereabouts.

Innocent doers of harm obviously differ from bystanders who are merely allowingly related to the bad outcome. But the doingly related also stand out from another large class of agents who by their actions enable a bad outcome. These agents could plausibly be called contributors, but unlike those doingly related to the outcome, there is no complete causal process between them and the injuries resulting from their conduct. A typical case would be to enable a harmful process by removing an obstacle; consider:

Kick: A cart is heading towards a rock which would bring it to a halt. Bill kicks the rock out of the way and the cart continues down the hill and injures Alice at the bottom.

Bill may plausibly be held to have injured the victim in this case if he is aware of the situation and leaving the rock in place is not costly to him. But one might ask whether by innocently removing it he injures Alice. Bill could be unaware of the approaching cart, nor plausibly be expected to be aware of it, and the rock could be any ordinary rock lying in a field. In which case it is not clear whether Bill is doingly related to Alice’s injuries. Anyway, it would be wrong to say that Bill initiates a causal process by removing the rock.7 The argument I will present for the significance of innocent contribution does not aim to cover instances like Kick. This is fortunate, because while I think innocent doers of harm are liable to the use of force, the case for using force against agents who innocently enable an ongoing process by removing an obstacle seems weaker.

Before I proceed, let me indicate that I take the permissibility of using force against an aggressor to avoid a harm to be closely related to the aggressor’s duty to accept cost in order to avoid causing the harm in question. I shall take the relation between the duty to bear cost and the permissibility to use force to be such that when a person has a duty to bear cost the fairest or morally most desirable state of affairs, other things being equal, would be for the person to bear such costs even if others had to impose those costs on him. Clearly, efforts to enforce the aggressor’s duty may bring additional costs that would have to be taken into account.8

7 Why is that wrong? Well, ultimately I think it has to do with a lack of a complete causal process between Bill and Alice’s injuries but I cannot go into that here.

8 There could be a reduction in permissible level of enforced cost as compared with maximum required duty to shoulder cost. It could then be permissible to impose less cost on a person than the person would have a duty to bear in the first place. Reasons for this

Generally, however, I shall assume it is permissible for prospective victims and third parties alike to intervene in a manner that imposes a level of cost corresponding to the level of cost the aggressor would have had a duty to shoulder. Just as we expect people to discharge their duties, they should be willing to accept that others impose the particular cost on them when they themselves are unable to perform. If, for instance, an innocent aggressor is required to suffer the loss of one leg to save a person’s life, he should not resist having that loss imposed – although his resistance might be understandable and perhaps even excusable.

A Two-part Theory of Self-defence

In discussions of self-defence, there are two main positions on the use of defensive force against innocent aggressors. Those denying that defensive force can be used against innocent aggressors reason that an innocent aggressor has done nothing to lose his or her right not to be attacked or killed, and should be treated accordingly. Moral responsibility is seen as a precondition of defensive force. The second group, taking the opposite stand, regards the use of defensive force as permissible. A rationale of this view is that a person has a right to defend him or herself against any unjustified aggressor or agent, since by merely initiating a harmful causal process toward another person one loses one’s right not to be significantly harmed.9 We need to take proportionality into account, but the fact that the aggressor is morally innocent would be insignificant for the question of the permissibility of using defensive force since he is causally responsible for a threat to another person. Hence, according to the first position, could be respect for a person’s free choice, or perhaps the mere psychological cost of being forced to do something. So that if 100 were the cost one was obliged to shoulder, the cost others were permitted to impose could be only 100 – q, where the constant q denotes any appropriate number reducing the level of enforcement cost. I shall assume there is no need for a q-moderator.

9 One position that appears to take this form is Judith Jarvis Thomson’s rights-based account of self-defence. It seems to put innocent aggressors and culpable aggressors on a par, as the right to self-defence is based on rights violations which, according to Thomson, can be committed by innocent as well culpable aggressors. According to her, culpability has nothing to do with the right of self-defence, since an agent who exercises this right is not an agent of justice. See Judith Jarvis Thomson “Self-Defence”, Philosophy and Public Affairs 20 (1991): 283–310. For a discussion of Thomson’s position, see Yitzhak Benbaji “Culpable Bystanders, Innocent Threats and the Ethics of Self-Defense” Canadian Journal of Philosophy 35 (2005): 585–622. See also D.R. Mapel “Innocent Attackers and Rights of Self-Defense” Ethics and International Affairs 18 (2004). the permissibility of using force hinges on moral responsibility, while according to the second, moral responsibility plays no role.

Both of these positions are less plausible than an intermediate view, since both innocent contribution and moral responsibility matter in selfand other-defence. For instance, if an innocent aggressor is about to cut off somebody’s limb there are strict proportionality considerations that would limit the means one could permissibly use to prevent them. It might, for instance, not be permissible to kill the innocent aggressor to save the limb, even if that is the only way of doing so. Yet, it seems permissible to do more toward the aggressor than toward an innocent bystander who happens to be in the way, or is on the sidelines but can be used to save the limb.

By contrast, if the agent is a culpable aggressor who enjoys separating people from their limbs, significantly more may be done to him if that is necessary to save the victim’s limb. Even killing a culpable aggressor seems permissible if that is required to save the limb. Proportionality requirements still apply, but are different (and easier to meet) from those in situations where force is used against an innocent aggressor. The degree to which they are easier to meet depends on the moral culpability of the agent. Any aggressor or agent of harm would be liable to the use of force as a consequence of merely initiating harm, but his or her liability to use force would increase with increasing culpability.

While the application of permissible force against innocent aggressors is determined by what I call “the asymmetrical fair share procedure”, permissible means arising from the moral culpability of the threatening aggressor should be seen as increasing from this baseline. For instance, an agent might have been ignorant of his role as the initiator of a harmful causal sequence, though he ought to have realised it. Such negligence would make the aggressor or agent more culpable and extend his or her duty to bear cost. There would accordingly be a higher ceiling on the level of force one could permissibly use against him to protect prospective victims. Recklessness would extend the range and severity of the means still further, and intentional wrongdoing further still.10 For this we need a twopart theory laying out permissible use of force against innocent aggressors and agents of harm and then let moral culpability add significance from

10 Hence, we should normally direct defensive force at the person who is morally responsible for a threat, not the innocent aggressor whenever there is such a choice. No such choice exists in the examples discussed in this paper. this baseline. In this paper, however, I simply aim to determine the baseline.

The Asymmetrical Fair Share Procedure

When discussing the permissibility of defensive force against innocent aggressors and agents, it is common to envisage conflict situations involving an aggressor and a victim, one of whom will be killed. In such cases it is either permissible to kill the innocent aggressor in self-defence or it is not. I think we gain valuable insight into this issue by envisaging situations where the harm may be divided and shared between aggressor and victim.

Bill becomes an innocent doer of harm when he initiates a complete causal sequence that leads to Alice being harmed. Suppose Bill, through no fault of his own, is in such a causal relation to the harm suffered by Alice. We can imagine different levels of harm (e.g. a loss of a finger, a limb or her life).11 Alice may not be in a position to defend herself, but Cathy, who is nearby, could intervene to prevent the incident. An innocent bystander, David, is also present. Since Cathy cannot save everyone, we need to ascertain the rule on which she should act to distribute it.

Two questions seem closely connected: a) whether Bill would have a duty to take on more cost than David to protect Alice from harm; and b) whether Cathy would be permitted to impose more cost on Bill than on David to save Alice from harm. As I said at the outset, I argue for a positive answer to both of these questions.

One could, of course, elect to help Alice by reasoning that people have a right to defend themselves against any unjustified aggression or threat. Alternatively, one could give priority to Bill. To become a permissible target one must be morally responsible for putting someone in danger. Innocent contributors would accordingly be morally indistinguishable from innocent bystanders. And since there are strict constraints on what we can do to innocent bystanders to save ourselves and others from harm, there will be strict limits on what we can permissibly do to Bill as well. Call this view the “on-a-par-with-a-bystander view”.12

11 I shall assume that everything else is equal, and that no particular body part has any special value, like a hand that otherwise might have cured cancer. Yet, if that were the case, it seems plausible to give preference to such hands. This is in line with the contractual argument that I present later.

12 The most explicit supporter of the on-a-par-with-a-bystander view is Otsuka in his “Killing the Innocent in Self-Defence”.

The on-a-par-with-a-bystander view assumes an innocent aggressor or agent has no duty to shoulder additional costs to neutralise the threatening consequences of his or her behaviour. Harming Bill to protect Alice in a situation in which the former is about to harm the latter through no fault of his own would be regarded as on a par with harming the innocent bystander, David, for the same purpose. Most people maintain that not much could permissibly be done to David to save Alice from harm.

A Fair Share

An attractive feature of the on-a-par-with-a-bystander view is its emphasis on something of obvious moral significance, namely moral responsibility. The assumption is that merely initiating a causal sequence is morally insignificant. There is some plausibility to this. If a person is unlucky and causes harm to another, that person is unlucky too, as she ends up as the victim. It seems to make little sense to identify any particular person as the unlucky one to determine who should be given priority. Yet although the direction of the causal link may not tell us where our priorities should lie, it would be a mistake not to recognise that the causal process joining them also distinguishes them from others. An initial proposal would therefore be: as both were unlucky, we should divide the harm between them, keeping everyone else out. In the example at hand, harm should be divided between Bill and Alice, and not imposed on David.

Normally, when we claim that fairness requires that something should be divided between particular individuals, we take it to imply it should be divided equally, unless there is reason to the contrary. When two people are involved and dividing the cost is possible, I shall accordingly divide the cost equally between them. Hence, if Alice was in danger of losing both arms due to Bill’s innocent actions, and Cathy was in a position to intervene and divide the harm between Bill and Alice equally, each losing an arm, then she should do so. To take on a fair share in such cases means roughly to divide the harm into equal parts.

At this point one might wonder why we should restrict our attention to those involved in the causal chain and no one else. Given the alleged moral insignificance of innocently initiating a harmful casual sequence, harm to innocent people should be divided equally among all kinds of innocents, including initiators, victims, and bystanders alike. A fair share procedure should consequently include bystanders in addition to those who initiate the harmful sequence and their respective victims. Some theorists may find this view attractive. After all, the distinction between bystanders and innocent agents of harm is a matter of brute bad luck. Therefore, according to luck egalitarians, in terms of fairness, the equal burdening of all members in a relevant population would be the best solution. Indeed, philosophers like Ronald Dworkin, Gerald Cohen, Larry Temkin and many others would argue that justice requires neutralising the effects of brute bad luck. The rationale of such views is obvious: it is unfair that some people are worse off than others through no fault of their own.

That said, however, it would provoke counter-intuitive results. It would be permissible, for example, to impose the loss of one limb on an innocent bystander if such action reduced the loss of limbs for an innocent aggressor or victim from two to one. This proposal would obviate reasons for thinking innocent bystanders have claims on some special kind of protection. Few seem ready to accept this conclusion. In the second part of this paper I argue against this view on contractual grounds.

Luck Reconsidered

The proposal we are considering is this. The innocent initiator of the harmful causal sequence and the innocent victim have both been unlucky, and ought therefore to divide the burden between them. But though the equal-bad-luck view might be plausible on the face of it, there is an important difference between the two parties. It seems wrong to say that the victim has been unlucky if she had been able to avoid the threat even though it caused harm to the innocent aggressor or agent. The prospective victim’s bad luck is essentially connected to the prospect of being harmed. By contrast, it makes perfect sense to say that the innocent aggressor or agent has been unlucky in harming the victim, even though he should be able to avoid harm to himself – and lucky in that respect.

Clearly, being lucky means different things here. In one situation bad luck refers to being a victim, in another it refers to being a contributor to harm. One might perhaps want to say that Alice – the innocent victim – is also unlucky in having to let harm befall an innocent person in order to save herself. But although the latter might be correct, we shouldn’t look upon this bad luck as the bad luck of being a victim or a contributor. Alice’s bad luck in not being able to save Bill is the same bad luck David would have suffered as an innocent bystander if he was unable to save Bill. And though one might feel sorry for someone for lacking the ability to help someone else, one would probably not treat it as an instance of bad luck in contributing to harm.

I propose at this point that those who innocently initiate a harmful causal process have a duty to shoulder a fair share of the harm. To capture this, I propose an asymmetrical fair share procedure, which requires these innocent people to take on a fair share of the burden. A rationale for this duty may be stated as follows:

Asymmetrical fair share: If a person has innocently initiated a causal sequence the effect of which is to harm or threaten to harm someone else, then, other things being equal, he has a duty as a matter of fairness to shoulder a fair share of the cost of the harm.

Consideration of different cases provides support for this proposal. Envisage innocent Bill again. He has for good reasons pushed and jumped onto a small cart in order to hit Alice who is sitting at the bottom of the hill. He then realises that his reasons was mistaken. He is now an innocent aggressor, endangering both of her legs. Alice cannot escape, but Bill has three options. He can continue in the same direction down the hill and save his own legs but at the cost of Alice’s; he can turn left, which would cost them one leg each; or turn right, this time sacrificing his own legs but saving hers. It would seem plausible to expect the innocent aggressor to accept half the burden, and his appeal to innocence would at most justify his reluctance to bear the lot. Hence, Bill would have a duty to turn, and turning left could be enough to discharge this duty.

Since Bill is now in a position to avoid causing the harm, one might think that he is not an innocent aggressor (or agent) but rather a culpable one. But that is not so; he becomes culpable by failing to comply with his duty. When Bill realises the danger he has put Alice in, his duty is determined by the cost an innocent aggressor would be required to bear. It is first when Bill decides not to comply with his duty that he becomes culpable and liable to increased force. This is like an innocent bystander who realises that she can save another person at little cost. She doesn’t become a culpable bystander by this realisation; rather she becomes culpable by her choice of not complying with her duty.

But there is an asymmetry between what can be required of Bill and of Alice. Envisage therefore another version of the cart example. In this case, innocent Bill can do nothing to avoid the impact but Alice has now three options. She can remain where she is and lose her limbs but save Bill’s; she can roll to the left and save one of hers and one of Bill’s; or she can roll to the right and save her own legs but leave Bill to lose his. If we think the fair share procedure should be symmetrical, Alice would have a duty to roll left-wise, saving one of her own legs and one of Bill’s. But this seems implausible. It would seem perfectly permissible for Alice to jump to the right and save both of her own legs, but at the cost of both of Bill’s.

These examples illustrate the asymmetry between Bill and Alice. One might, however, think the difference has to do with Alice simply having to move aside in the second example, while Bill needs to use Alice’s legs to cushion the impact in the first.13 But that is not so. Bill seems to have a duty to let himself be used to protect Alice by, for instance, putting a leg in front of the left wheel. No such duty applies to Alice. She has no duty to let herself – or parts of herself – be used to cushion Bill’s impact. She is free to leave the whole problem to Bill. Moreover, should Bill fall off the cart, and the cart continued to pose a danger to Alice, he would have a duty to run after it and put a leg in front of its left wheel, thereby sacrificing his leg to save one of Alice’s.14

There is an asymmetry between what Bill and Alice can do to each other as well. First, forcing Alice to take on a fair share of the cost does not seem a permissible option for Bill.15 If Alice is able to move out of the way, Bill may not prevent her. By contrast, Alice seems permitted to force half of the burden onto Bill. If Bill is unable to change the cart’s direction, Alice can protect one of her legs by making Bill lose one of his.16 She can do so in whatever way necessary, by shooting his leg, by rolling to the right, or by using her long-distance-body-mover to ensure Bill gets one of his legs tangled up in the left wheel, thereby saving one of hers. This implies, moreover, that Cathy should not impose any harm on Alice to protect Bill either. She could impose half of the burden on Bill to protect Alice, and thereby leave Alice to shoulder her fair share, but Cathy would have to let Alice escape should she be able to do so.

Now, if Alice can move out of the way and let Bill take the impact alone, is she also permitted to use force to get him to take on the whole burden rather than a fair share? That is less clear. Envisage a new version of Push where Alice has two options. She can fire at the left wheel of the cart, save one of her own legs and one of Bill’s; or she can fire at the right wheel, saving both of her own legs but causing Bill to lose both of his. Perhaps what Alice ought to do is to hit the left wheel, and share the burden with

13 On the issue of ducking harm, see Christopher Boorse and Roy A. Sorensen “Ducking Harm” The Journal of Philosophy 85 (1988): 115–134.

14 Hence, the important point seems to be that Bill has initiated a causal sequence leading to Alice being harmed, and not that he is the current or imminent threat. I cannot investigate this difference here though.

15 Unless, perhaps, if Alice has started to use excessive force against him.

16 Another story comes into play if he chooses not to turn the cart.

Bill? I am uncertain about this. My aim is nevertheless to demonstrate that particular duties apply to those who innocently initiate a harmful causal sequence, and that they should be set apart from innocent bystanders. And what seems clear is that Bill is not permitted to prevent Alice from imposing half of the burden on him.

So far, I have not considered what Alice ought to do if the cost cannot be divided between her and Bill. Envisage again innocent Bill who is hurtling towards Alice in his cart. Alice has only two options. She can stay put and lose both legs, or she can give herself and Bill a fifty percent chance each of avoiding harm. It seems plausible that she may impose a fair share of the risk of being harmed on Bill. Hence, she would be permitted to toss a coin and allow Bill to take all the harm should he be selected by this random procedure.

Let’s take stock. Granted that innocent Bill initiates a complete causal process that would lead to X amount of harm to Alice, Bill would have a duty to take on X/2 to protect Alice from this harm. It would also be permissible to impose X/2 harm on him to protect Alice. Alice, on the other hand, would be permitted to escape altogether and leave Bill to suffer X; as a prospective victim she has no duty to allow herself to suffer harm to protect Bill.17 Alice would also be permitted to use force against Bill to impose X/2 harm on him in order to protect herself.

Contractual Considerations

Situations can arise in which a person constitutes no actual threat to others but is morally at fault for behaving in such a way that it is reasonable to interpret him or her as posing a threat. In such cases, given reasonable evidence, it would be uncontroversial to allocate the cost to the apparent aggressor. But there could be situations in which an agent is not to blame for engaging in a certain activity but which is threatening all the same. In those cases, I have argued in previous work that if one’s interpretation of

17 Two qualifications: First, if the harm to Bill was serious and Alice could prevent it at relatively moderate cost then she ought to do so. Second, some may think that although victims have no duty to take on a fair share, they are (slightly) more duty bound than innocent bystanders who are not under threat. We may, for instance, think you ought not duck harm to save an arm if what is at stake for the other is his or her life, though we might be less inclined to tell you to intervene if doing so would cost you an arm. Anyway, I shall assume here that considerations like this will not bring the victim on a par with an innocent initiator, but may slightly set him or her apart from other bystanders. A proper investigation of this would require more space than I have here. the agent’s behaviour is reasonable, he or she still has to carry the cost.18 In this paper I put apparent aggressors to one side and focus on standard cases involving real aggressors and agents of harm.

I invoke hypothetical contractual reasoning to illuminate how harm should be divided among innocent people, and will consider the asymmetrical fair share procedure from the perspective of a hypothetical contract situation. The first thing to decide is who is to be included in the contract. It seems plausible to assume agreement about the permissibility of giving priority to the innocent at the cost of the culpable, at least when the difference in culpability is above a certain threshold. We can accordingly restrict participation in the contract to innocent people or to people whose culpability is below this threshold.

By appealing to hypothetical consent, the point is to identify who would have reason to give their consent in the contractual situation. People make contracts because they have reason for doing so. Good reasons for entering into a contract are the likely benefits to the contracting parties. In a hypothetical contract we conceal information from the participants. The veil of ignorance I envisage here could be called a “conflict veil”. As the name suggests, it is only a question of hiding the identity of those involved in a particular conflict where harm has to be distributed among innocent people. All other information remains available to the parties. In particular, they have access to the best available knowledge about their respective probabilities of ending up in such conflicts. For simplicity, I will assume the chances of being an innocent victim or innocent initiator of a harmful causal process are equal.

Our central concern is what we know about the participants by virtue of which they would have reason to consent to rules of selfand otherdefence. Clearly, their reasons would depend on their interests. One important interest would be to reduce their risk of being harmed and of dying. I omit other reasons to keep things simple. Reasons for accepting rules for regulating actions of selfand other-defence would accordingly be to avoid harm to the innocent but not to the culpable.

Hypothetical (actual preference) contractualism does not exhaust morality. Against a background of deontological restrictions, its aim is to explain when it might be permissible to violate specific restrictions. The basic idea is that such violations might be permissible provided they are in

18 See Gerhard Øverland “Self-Defence among Innocent People” Journal of Moral Philosophy 2.2 (2005): 127–146, and “Killing Soldiers” Ethics & International Affairs 20:4 (2006): 455–475. the interest of the parties involved. As for deontological restrictions, I take them to be best understood as being in the interests of the individuals they protect. In particular, deontological restrictions are meant to prevent the sacrifice of individuals for the benefit of others. Given such restrictions, hypothetical contractualism aims to demonstrate how it might be in the interest of the implicated parties to sidestep certain restrictions. And when that occurs it would not count as unfair treatment if a particular person ended up as the victim. In the case at hand, the restriction to be circumvented is the alleged impermissibility of harming or killing an innocent person for a comparable benefit.

The notion of fairness here is clearly not the same as when we say it is fair to divide a particular benefit between two individuals equally. Contractual fairness is only meant to indicate that the person bearing the cost – the innocent initiator – has no reason to complain since he would have reason to accept the rule that could allocate the cost to him in a particular case. A simple example indicates what I have in mind. Being a visitor in a South American town, Jim is given a choice by Pedro to kill any one of 20 men who are lined up against a wall.19 They will all be killed unless Jim agrees to kill one of them. Should he do so? Yes, because all 20 would probably agree to an outcome in which one of them was randomly killed by Jim. At the point of Jim’s appearance, the 20 men are faced with the prospect of being shot, and we have every reason to believe they would prefer to see one of their number randomly chosen and killed if it would save the rest.20 The randomly selected person would accordingly have no reasonable complaint against being selected. According to my usage here, it is not unfair to let a person shoulder certain cost provided he had reason to accept the rule that makes him bear it.

Epistemic Uncertainty

We can first observe that being an initiating causal factor will often be accompanied by moral culpability. In a conflict between two apparently innocent people, an agent who has initiated a harmful causal sequence

19 See Bernard Williams and J.J.C. Smart, Utilitarianism, For and Against (Cambridge: Cambridge University Press 1973), pp. 98–99 and Bernard Williams, “Consequentialism and Integrity”, in Samuel Scheffler (ed.), Consequentialism and Its Critics (Oxford: Oxford University Press 1988): 20–50, at p. 34.

20 Of course, if what we know about them give us reason to think otherwise, that should guide our conduct. tends to bear more moral responsibility for the situation than a prospective victim. Even if an agent is not culpable to the fullest extent, they may be guilty of negligence or recklessness. And, since being an initiating agent of a complete causal process is more easily discernable than moral culpability, a rule that would permit sacrificing the initiator is likely to save more innocent people in the long run as we are less likely to get it wrong. Epistemic uncertainty therefore elicits a presumption in favour of having a rule that gives priority to the defending party, i.e. the victim, because it will give (innocent) contracting parties a higher chance of avoiding harm. It is difficult for people under threat to know whether the initiating agents are innocent or not. A rule that gives priority to those who respond to a causal threat will therefore enable people to avoid being harmed on what I call wrong occasions. Wrong occasions are instances in which the defender erroneously believes the aggressors to be innocent and therefore fails to defend herself. Were we to choose a rule that permits self-defence only in cases in which the initiating agents were known to be culpable, or where there was some evidence of their culpability, there would be instances in which innocent people would be harmed while the culpable got off scot-free. To avoid this, the parties have reason to adopt a rule permitting prospective victims to use defensive force against apparent inno- cent initiators of harmful causal sequences.

Moreover, such a rule would help third parties coordinate intervention. Third parties will often lack the time and capacity to coordinate intervention, and would appreciate a simple rule which helped them decide on which side to intervene. Being an initiator to harm is a discernible external feature of conduct that can be used for this purpose.

We might consider an incentives-based argument as well. Adopting a rule according to which those who have initiated a harmful causal sequence have to carry the cost would minimise the likelihood of such situations arising in the first place, since people would take pains to avoid becoming initiators. This goal could perhaps be achieved by deeming those who initiate a threat not fully innocent unless they have taken every reasonable precaution to avoid becoming initiators. Only the fully innocent would be fully protected. This would give people an incentive to take all reasonable precautions against becoming initiators. Only people who had taken such precautions would be treated as innocent. A problem with this is that moral responsibility is not an easily observable feature. It will often be difficult to establish whether the agent has taken the required precautions. A rule giving priority to the defending party might serve the contracting parties better. Of course, any known moral responsibility on the part of the initiators would consolidate priorities further and do so according to the degree of moral culpability, as indicated at the outset of this paper.

Clearly, some aspects of responsibility are observable. We can observe drivers failing to exercise reasonable care, or acting recklessly, for instance. But the point is not whether we can observe moral responsibility, but that when we cannot, initiators nevertheless tend to be more responsible for the threat they pose to others than those at the receiving end. Whenever we can observe moral responsibility on either side of a conflict, the situation would be very different. An agent’s culpability would then normally require him to carry the cost.

Epistemic uncertainty tips the balance in favour of the defending party – how far will depend on estimates of likely culpability. How often is an agent about whom we have no particular knowledge in fact innocent? Without attempting precise estimates, it seems uncontroversial to expect apparent innocent initiating agents to be guilty of negligence or recklessness rather than their victim. Of course, the rule preferring the defending party will sometimes give the wrong result, as what looks like a victim may in fact be an aggressor, or the harm suffered by the defending party may be justified for some other reason. But although this is true, the frequency of such mistakes will hardly be sufficient to give the contracting parties reason not to favour prospective victims.21

Due to the complexity of the causal process, it may be impossible to identify the initiator. In such cases it will be difficult to use the idea suggested here for solving the conflict. If we don’t know who initiated a harmful causal process, we cannot use a rule requiring initiators to shoulder the harm.

Regarding the number of innocent initiators and innocent victims involved in the situation, how prospective victims are treated will depend on the epistemic conditions. It is easy to judge wrongly with regard to innocence. If priority on the basis of slightly greater numbers would not save more innocent people than giving priority to the defending party – because, perhaps, some of the aggressors or agents of harm were culpable – the contracting parties would have reason to support a rule that gave priority to the defending party even when there are minor asymmetries in numbers.22 Hence, the contractual argument would recommend

21 Needless to say, whenever we know that culpability lies on the side of the victim, we ought to give priority to the aggressor.

22 See Øverland 2006, “Killing Soldiers”, at pp. 469–470. giving priority to the defending party in situations with more than one initiating agent, but at some point, the asymmetry in numbers will suggest giving priority to the assumed innocent initiators.

Epistemic Transparency

I have indicated several reasons for selecting a rule that gives priority to the defending party. These arguments traded on epistemic uncertainty. But although contribution is often correlated to culpability, there may be instances in which we know this is not the case.

Division

From the perspective of the parties in the contractual situation, whether an individual is unlucky enough to set off a complete causal process has no bearing on choice of distributive pattern. The parties want to reduce the risk of being harmed. Assuming transparency and full knowledge of the parties’ innocence, there might be no grounds to decide either way.

An equal split would possibly be in their interest. But why should the parties be interested in seeing harm divided? Why allocate a particular duty to innocent initiators? Why not just leave them on a par with bystanders? Assuming knowledge, the uncertainty argument does not apply, although it might be worth mentioning that we seldom have full certainty in real life anyway.

The incentives-based argument still carries some weight. By knowing one will be held accountable for initiating a harmful causal process, one has reason to make a serious effort to avoid doing so.23 Again, the expansive responsibility argument could provide some of the same incentive. But besides the limits alluded to above, a further problem with this argument is that it gives people an incentive to feign innocence since so much is at stake. This incentive is reduced if one requires innocent initiators to take on a fair share. A slight increase in culpability would only add a minor cost-related duty, as one would already be required to shoulder a fair share by merely being an initiator. The latter would, of course, have an incentive to pretend not to be an initiator of a harmful causal process. But that is a more difficult enterprise.

23 We see this at play in civil law, as strict liability, where innocent contributors are held accountable for damages that they cannot be held morally responsible for.

Another reason to adopt a fair share procedure, rather than letting the victim bear the whole cost is because it seems better to take a fifty percent loss than a fifty percent risk of sustaining the whole cost when what is at stake is significant. For instance, it seems better to lose an arm than have a fifty percent risk of losing both or losing none. There is likely to be some disagreement here, but I would guess most people would prefer a defined loss to a fifty percent risk of the double loss. Think of any dividable entity of significant value in your possession, like your money, your property, life span, health, or limbs, and consider whether you would prefer losing fifty percent of any one of them or a fifty percent chance of losing nothing at all, but at a fifty percent risk of losing it all.24

It may pay to observe that there is no reason to give priority to the innocent initiator from the perspective of the contractual position. An alternative to fair share, namely letting harm fall where fate dictates – on the victim normally – may nevertheless merit consideration. According to this proposal, when both parties are innocent neither has a duty to shoulder the cost, and both would be permitted to use force in defence. Third parties would have no reason to intervene. But letting the involved parties resolve the problem by fighting would exacerbate the harm, giving the contractual parties reason to reject this option. Allowing third parties to prevent escalations might mitigate the problem, but it would be unreasonable to expect complete success. Moreover, this kind of procedure would favour certain groups and types of people, giving those who are not members of those groups or types reason to reject it.

Asymmetry

Why attribute moral significance to causal impact? If one thinks luckfactors like being an innocent contributor to harm are irrelevant, one might think the burden at stake should be divided with no regard to causal impact. The contracting parties could likewise let all innocents be on a par in order to reduce harm and include those who initiate a harmful process, victims, and bystanders alike. On the face of it, it would seem to benefit all participants by reducing the risk of being harmed and killed.

I previously appealed to the counterintuitive upshots of such views, which would leave the innocent with no special protection. That will not

24 It might be worth mentioning here that the preference I am alluding to is inconsistent with the expected utility decision procedure beloved by many economists and utilitarians. do under contractual reasoning. Explaining restrictions in a contractual argument entails asking whether the parties in a hypothetical contract situation have prudential reasons to accept them. We cannot assume, for example, that harming innocent bystanders is morally worse than harming those who have innocently initiated a harmful causal sequence. On the contrary, the argument should explain the impermissibility of widening the net with reference to the situation we would end up with if it were permissible to use force against victims and bystanders.

The argument from epistemic uncertainty distinguishes between initiators and victims. While, despite appearances to the contrary, initiating a harmful causal process is often correlated to culpability, victim-hood is often associated with innocence. The latter is true of bystanders as well.25 These observations single out innocent initiators. But when there is no uncertainty, the contracting participants may have no reason to discriminate between innocent initiators, victims and bystanders. However, a version of what I elsewhere have called “the undermining argument”26 explains why it is wise to draw a distinction between those who initiate a harmful process on the one hand, and victims and bystanders on the other. The general point is that if there are no constraints, unfortunate long-term consequences will undermine the antecedent rationality of including everyone in the general lot.

The argument has two steps, of which the first concerns the danger of intervening with force, as recipients are unlikely to accept it without defence. Allowing force to be used against bystanders exposes the enforcing individuals to considerable risk. They are exposed to risk because although there are situations where people might agree to be sacrificed to reduce harm to strangers, this is not what they are likely to do for the most part. Moreover, for there to be such a risk it is unnecessary that everybody resists sacrifice, it is enough that a significant amount does so, which seems to be a plausible assumption. Fear of bystander resistance will accordingly reduce the number of interventions by prospective assistors and hence the impact of allowing use of force against bystanders. Whether anything will be gained by permitting force against bystanders is therefore a moot question, as people would be wary of involving themselves.

25 Of course, they may have failed to assist the victim at moderate cost, and in that sense be morally at fault. I do not discuss such failures here.

26 See Gerhard Øverland “Contractual Killing” Ethics 115 (2005): 692–720, at pp. 713–716.

The second step trades on the previous reduction and says that successful interventions by force might cost more than the achieved level of harm reduction because people will want to avoid being used for the purpose of harm reduction. If people are aware of the danger of being used in a bid to temper overall harm, they will initiate steps to avoid getting involved in the first place, leading to fewer people being saved. Being near accidents, like runaway carts, would be hazardous because one would risk being used in whatever manner to reduce overall harm. Likewise, we would have reasons to avoid houses that are burning or people who are downing; being in their vicinity could result in our being sacrificed to save them. And clearly, if people in general try to avoid proximity to such unfortunate situations, it would have severe implications for the prospect of being saved. The upshot would be that people, who could have been saved by other means than the use of force, will not be saved simply because those who could have performed the rescue will have fled in fear of being sacrificed. Contracting participants have reason therefore to grant some people protection against use of force.

The degree of protection depends on how calculations pan out for innocent initiators, victims and bystanders. Remember the situation discussed initially: Bill innocently harms Alice. Cathy is close by and able to intervene in the causal chain to avert the threatened harm. A fourth person, David, who is an innocent bystander, is also present. Cathy and David are bystanders. There are good reasons to let Cathy prevent Bill from harming Alice, but not interfere with David. If Cathy were permitted to use force against David to reduce the overall harm, David, in turn, would be permitted to use force against her. They are unlikely to accept this kind of risk in order to protect Alice, who is a stranger to them. Generally, this norm would give bystanders a bigger incentive to move out of the way than to help those already involved. These negative consequences will have to be measured against the likely benefits of permitting the use of force in the first place. Some benefits are to be expected. A certain number of people, who would otherwise not be saved from harm would now be saved. However, whether the number thus saved would exceed the number not saved is unlikely because bystanders will look to avoid accidents for fear of being used to reduce overall harm. The contracting parties would therefore have reason to prohibit use of force against bystanders.27

27 That is, above what one can expect bystanders to be willing to shoulder in the first place, i.e. moderate costs. This is a good reason why the principle of assistance only tells us to assist others from severe harm if we can do so at moderate cost.

The same reasoning does not apply to those who initiate harmful causal processes. The positive expected effects of permitting the use of force against them outweigh the expected negative effects. In some situations, it will lead to fewer people being harmed or will reduce the severity of the harm. In other situations it will secure a fair distribution of the harm. There are, of course, unwelcome consequences of using force against innocent initiators as well. It will invite defensive force, giving prospective interveners reason for not getting involved (as we see in real life). Bill, however, is already posing a risk, and to disqualify forceful intervention altogether would leave harming and killing too often at the mercy of these initiators.28

28 I would like to thank Christain Barry, Robert Huseby and Igor Primoratz for their helpful comments.

Consistency in the Armed Enforcement of Human Rights: a Moral Necessity?

Ned Dobos

There is no denying that international human rights norms are enforced selectively. Some oppressive governments become the targets of military intervention, while the political sovereignty of other, equally oppressive regimes is left intact. The moral disapproval that this has been known to evoke takes two distinct forms. For some, the interventions make the noninterventions particularly offensive. When former UN Secretary General Boutros Boutros-Ghali expressed outrage at the contrasting international responses to the crises in Yugoslavia and Somalia in the early 1990s, his objection was not that the Bosniaks were receiving help from the outside world, but that the Somalis were not thought worthy of the same.1 For others, it is the cases of intervention that are morally tainted; selectivity is advanced as a reason to condemn a particular humanitarian operation.2 Not surprisingly, this objection is common amongst the states singled out for intervention, their representatives and their apologists.

My aim in this paper is to determine whether a military operation to defend human rights can possibly be made illegitimate by the fact that the state prosecuting it has failed, is failing or will fail to defend human rights under relevantly similar circumstances elsewhere. I will consider four distinct arguments for an affirmative answer: 1) An intervention that forms part of an inconsistent pattern is to be opposed because selectivity is proof that selfish ulterior motives are in play; 2) The international covenants and treaties which oblige states to honour human rights standards are annulled by selective enforcement; 3) A state’s non-interventions make its interventions “comparatively unjust” (or unfair), and; 4) Intervention in X is rendered illegitimate by non-intervention in Y if the non-intervention

1 Boutros Boutros-Ghali, Unvanquished: A U.S.-U.N. Saga, (New York: Random House, 1999), p. 44, 53, 55, 141, 175.

2 See Lori Fisler Damrosch, “The Inevitability of Selective Response? Principles to Guide Urgent International Action”, in Albrecht Schnabel and Ramesh Thakur, Kosovo and the Challenge of Humanitarian Intervention: Selective Indignation, Collective Action, and International Citizenship, (Tokyo; New York; Paris: United Nations University Press, 2000), pp. 405–06: “Nationalist actors within the former Yugoslavia complained vociferously that they were the victims of unequal treatment”. constitutes a dereliction of duty, and is a direct consequence of the intervention. All four approaches come up short. A non-humanitarian motive does not in itself corrupt the action that stems from it, and does not guarantee an undesirable outcome as is sometimes suggested; the obligation to fulfil human rights is not conditional upon the validity of international contracts; the concept of comparative injustice is dubious, and does not translate into injustice all things considered; and an intervention is highly unlikely to be causally connected to a derelict non-intervention in the way required for the final argument to work.

I

Selectivity in the enforcement of human rights is often taken as evidence of ulterior motives. The explanation for why intervention is occurring here but not there, where the humanitarian emergency is equally or more dire, is that the intervening state is, at least in part, economically or politically driven. A good example is Slater and Nardin’s take on the 1983 US intervention in Grenada. The invasion, they say, was justified (in part) by the Reagan administration as intended to promote democracy and oust a government made up of ‘a murderous bunch of leftwing thugs.’ However, in light of the same administration’s support for far more murderous right-wing thugs throughout Central America, it is evident that the real grounds for intervention were ideological, political, and strategic, not a commitment to respect human rights.3

Noam Chomsky is adamant that the Račak massacre cannot have been NATO’s sole reason for bombing Serbia in 1999. His proof is that the West did nothing to stop Indonesian forces from simultaneously murdering refugees in the Timorese village of Liquiçá.4 The implication seems to be that the selective enforcement of human rights is illegitimate because motivated by something other than pure altruistic concern.

In responding to this line of argument the first thing to note is that, in at least some cases, there is a perfectly coherent moral rationale for selectivity. The fact that Western powers intervened in Yugoslavia and Iraq, but not in Russia and China, some would say, is to be explained in terms of

3 Jerome Slater and Terry Nardin, “Nonintervention and Human Rights”, The Journal of Politics, vol. 48, no. 1, February 1986, p. 94.

4 Noam Chomsky, “In Retrospect: A Review of NATO’s War over Kosovo”, Z Magazine, April 2000, at <http://www.zcommunications.org/zmag/viewArticle/12759>, retrieved on 30/8/2008. their strategic and political interests. However one could just as well say that the West refuses to get involved on behalf of the Chechens and Tibetans for sound moral reasons; intervention here would unleash World War III and cause immeasurably more death and suffering than it would prevent. Thus a state that intervenes selectively can at least sometimes maintain that it is acting purely out of moral motives.5

But even when we do have reason to suspect that ulterior motives are in play, this, on its own, is clearly not enough to make an intervention morally wrong. A billionaire who donates large amounts of money to charity only to make the Forbes Magazine list of most generous philanthropists does a morally good thing, but not for morally good reasons. This may lead us to think less of the philanthropist as a person, but we certainly would not object to his giving to charity. Selfish motives taint the agent, not the actions that stem from them.6

The obvious rejoinder is that ulterior motives are likely to adversely affect the consequences of an intervention, and this can undermine the legitimacy of the act itself. If an intervention is driven by national self-interest, the argument goes, it is likely that the means used to prosecute it will not be conducive to a positive humanitarian outcome. The Kosovo campaign is often cited as a case in point. Ellen Meiksins Wood admits that opponents of this war tended to oscillate between principled opposition on the one hand and practical opposition on the other—or opposition to the war only because it was “bungled”. For Wood, however, “the ‘bungling’ [was] inevitable simply because the war [was] driven by overriding factors quite distinct from, and in opposition to, its stated objectives”.7 The real purpose of the war was to put on display NATO’s military muscle, and to force the Milošević government to capitulate

5 For more on this point, see Chris Brown, “Selective Humanitarianism: In Defense of Inconsistency”, in Deen K. Chatterjee and Don E. Scheid, Ethics and Foreign Intervention, (Cambridge; New York: Cambridge University Press, 2003), pp. 31–50.

6 This insight can be attributed to John Stuart Mill, who draws an important distinction between an agent’s intention and his motive for acting. The intention is “what the agent wills to do” or the act contemplated. By contrast the motive is the further goal that the agent wishes to accomplish with the intended act. Mill argues that “the morality of the action depends entirely upon the intention—that is, what the agent wills to do” and not the underlying motive for the action. Thus whether I save someone from drowning out of compassion or out of a desire for applause, the intention is still to rescue, and therefore the moral status of the action (though perhaps not that of the agent) remains the same. See John Stuart Mill, Utilitarianism, Roger Crisp (ed.), (Oxford: Oxford University Press, 1998), p. 65, n. 2.

7 Ellen Meiksins Wood, “Kosovo and the New Imperialism”, in Tariq Ali (ed.), Masters of the Universe? NATO’s Balkan Crusade, (London; New York: Verso, 2000), p. 196. with NATO unconditionally. It is not surprising, then, that NATO forces adopted a mode of warfare that best served these ends: high-tech, highaltitude and highly destructive bombing. This, however, proved disastrous for the very people that the war was ostensibly waged to protect. On 23rd March 1999, the day prior to the commencement of the campaign, the number of people that had been forced out of their homes was estimated at 230,000. By the end of the war 1.4 million were displaced, and of these 860,000 had fled the province entirely for Albania or Macedonia. Prior to the intervention, 2,500 people had been killed in the civil war between the Serbs and the KLA. During the intervention, approximately10,000 were killed, most of them Albanian civilians killed by the Serbs.8

Peter Coghlan argues that American interests have had a similar effect on the Iraq war. He identifies three distinct motives behind the decision to invade. The first was punitive: to take revenge against the terrorist cells that attacked New York and Washington on 11 September. The second was to show off the military strength of the US, and with this to send a message to potential threats. The third was to liberate the Iraqi people. The first and second reasons, says Coghlan, have “undermined” and “corrupted” the third. US forces wanted to help the Iraqi people, but have largely failed in this objective because they were so determined to “kick some ass” along the way.9

With this in the background, the first argument from inconsistency can be reformulated as follows: an intervention that forms part of a selective pattern is probably motivated by something other than pure altruism, and for this reason it will not produce a positive humanitarian outcome at an acceptable cost.

While there is no denying that ulterior motives can have an adverse affect on consequences, I am not convinced that the influence of national self-interest should be simply presumed inimical to good results. In fact, there are a number of ways in which national self-interest can improve the prospects of a positive humanitarian outcome.

States are often reluctant to sustain casualties where they have no national interests at stake, and this can spell disaster for the intended beneficiaries of a disinterested humanitarian intervention. NATO’s campaign over Kosovo only exacerbated the forced expulsion of the ethnic Albanians,

8 Michael Mandelbaum, “A Perfect Failure: NATOs War Against Yugoslavia”, Foreign Affairs, 78, 5 (Sept/Oct 1999), p. 3.

9 Peter Coghlan, “War and Liberation” in Raimond Gaita (ed.), Why the War Was Wrong,

Melbourne: Text Publishing, 2003, pp. 143–44. at least in the short term. Wood’s explanation for why the intervention was limited to high-altitude bombing is that this best served the selfish ulterior motives of NATO’s member states. But an equally plausible explanation is that this method of war was adopted precisely because NATO states had no strategic or economic interests at stake. This made a ground assault politically unfeasible, since democratic citizens have shown themselves unwilling to tolerate military casualties purely for the sake of defending foreign nationals (see Somalia). The presence of national interests can make up for this “commitment shortfall”.10

Indeed this is what political realists have been telling us for years. According to the “high-priest of post-war realism”, Hans J. Morgenthau, states lack both the will and the resources to pursue moral objectives to completion.11 “Moralistic” doctrines are only ever implemented in “fits and starts… here half-heartedly and with insufficient means, there with all-out military commitments, there not at all…”.12 The result is negative net utility overall. The national interest is sacrificed, few outside the state are made significantly better off and many are likely to be made worse off. As Morgenthau puts it, the intrusion of “sentimentalism” into foreign policy is to blame for the fact that “political success has been sacrificed without appreciable gain in universal morality”.13 Arguably, national self-interest alone can generate and sustain the political will needed to carry out a successful intervention.

Aside from this, national interest can help keep the ruthlessness of war in check. Disinterested humanitarians may be particularly disposed to a dangerous sense of self-righteousness.14 The more soldiers see themselves as the defenders of justice, good and civilisation, (as opposed to the national interest), the more likely they are to “simplify the moral boundaries of the conflict” and to perceive the enemy as the embodiment of

10 Michael Wesley, “Towards a Realist Ethics of Intervention”, Ethics and International Affairs, vol. 19, no. 2, 2005, p. 70.

11 Tim Dunne and Brian T. Schmidt, “Realism” in John Baylis and Steve Smith (eds.), The Globalization of World Politics, 2nd edition, (New York: Oxford University Press, 2001), p. 147.

12 Hans J. Morgenthau, In Defense of the National Interest: A Critical Examination of American Foreign Policy, (New York: Knopf, 1951), p. 119.

13 Ibid., p. 114. This is not to say that realists are opposed to all humanitarian interventions. The realist is opposed only to those lacking a coincident national interest rationale. See for instance Charles Krauthammer, “When to Intervene: What’s Worth Fighting For?”, The New Republic, 192, 6 May 1985, pp. 10–13.

14 See Anthony Coates, “Humanitarian Intervention: A Conflict of Traditions” in Terry Nardin and Melissa S. Williams (eds.), NOMOS XLVII: Humanitarian Intervention, (New York and London: New York University Press, 2006), pp. 73–79. injustice, evil and barbarism; as an opponent that is not deserving of the respect that is usually owed to the enemy.15

This can potentially turn what Plato referred to as mere “discord” into all out war. Where armed conflict breaks out between enemies who are “by nature friends”—where Greeks are pitted against other Greeks—the goal of the war is to resolve the dispute that triggered it and to restore the peace, says Plato. Such wars are fought in a limited way: “they quarrel as those who intend some day to be reconciled.”16 As a result, Plato tells us:

They will not devastate Hellas, nor will they burn houses, nor ever suppose that the whole population of a city – men, women, and children – are equally their enemies, for they know that the guilt of a war is always confined to a few persons and that the many are their friends.17

Discord between combatants who are by nature friends is contrasted with war between natural enemies.. Plato cites the conflict between Greeks and Barbarians as the clearest example. The aim of such wars is not specific redress and a restoration of the peace, but annihilation and total defeat. Anthony Coates explains that between natural enemies, “a standing cause of war exists – a cause that owes nothing to the specific threats made or injuries received, but one that is rooted in the perception of the enemy’s fundamental Otherness.”18 As a result, war is conducted with greater intensity, savagery and ruthlessness than discord, and those engaged in it tend to pay less regard to the distinction between combatants and noncombatants insofar as both share in the “fundamental Otherness” to which Coates refers. In 1139, the Lateran Council sanctioned the use of crossbows against infidels but not against fellow Christians. In the 1800s, soft-nosed dum-dum bullets, “designed not just to penetrate the body but to tear it apart,” were developed by the British for use in colonial wars but not European wars. In World War II the Germans adopted radically different rules of engagement on the Eastern and Western Fronts. They recognised the common humanity of the English and Americans, but not of the Slavic Russians. Similarly, the Americans tended to treat German solders in accordance with the rules of war, while the bodies of Japanese combatants were regularly mutilated.19

15 Ibid., p. 76.

16 Anthony Coates, “Culture, the Enemy and the Moral Restraint of War” in Richard Sorabji and David Rodin (eds.), The Ethics of War: Shared Problems in Different Traditions, (Aldershot: Ashgate, 2006), p. 212.

17 Quoted in Ibid., p. 212. 18 Ibid., p. 213.

19 Ibid., pp. 213–14.

If soldiers are aware that they are fighting for mundane national interests, rather than lofty moral principles, they are arguably less likely to demonise and dehumanise the enemy. An acute awareness of the selfserving nature of an intervention can help to prevent the target state and its supporters from being cast as fundamentally Other. With this, national interest can arrest the degeneration of discord into war.

While we may have some reason to think that a politically or economically driven intervention will have bad consequences for its intended beneficiaries, then, we also have reason to be cautious of unadulterated altruism in international relations. Hence even where selectivity can be taken as proof of ulterior motives, this, in itself, does not seem to be a good enough reason to oppose an otherwise justified humanitarian intervention.

II

The second argument that we need to look at says that selective enforcement annuls the obligation to honour human rights, and thus undercuts the justification for enforcing them. We start with the premise that states are duty-bound to live up to international norms—including human rights norms—because they have freely entered into international treaties and covenants which require them to do so. These obligations, in other words, are the products of a state’s voluntary transactions. But each state that enters into some such contract does so on the reasonable expectation that its terms will be enforced without prejudice against all parties. No state would rationally consent to observe these constraints on its conduct were it aware that some signatories would be allowed to violate human rights with impunity, while others, itself included, would be subjected to military intervention. From this it is inferred that the selective enforcement of human rights covenants renders them null, undercutting the justification for humanitarian intervention.20

Again there are several problems with this line of argument. First, human rights abuses are sometimes tolerated for morally legitimate practical and prudential reasons, and this cannot sensibly be said to void international human rights law. No state can reasonably expect that the terms of its international agreements will be enforced even where the

20 Lea Brilmayer briefly discusses this line of argument in “What’s the Matter With Selective Intervention”, Arizona Law Review, vol. 37, 1995, pp. 962–63. practicalities of the situation do not allow for it or where the consequences would be disastrous.

But let us grant for the sake of argument that the International Covenant on Civil and Political Rights, The International Covenant on Economic, Social and Cultural Rights, and all other such agreements have indeed been invalidated by selective enforcement. On one way of understanding human rights this may be enough to undermine the justification for humanitarian intervention. The position I have in mind is that which is sometimes attributed to legal positivists. It says that the only rights are legal rights or rights that originate within a legal system. There can be no rights that exist prior to, and independently of, legal codification. It follows that if the covenants that enacted human rights into international law are null and void, then human rights simply do not exist. And if there are no human rights, the imperative to protect them cannot be invoked to justify the use of military force.

But this is entirely wrongheaded. We condemn a government that deliberately starves its own people without regard to whether that government is bound by an international covenant proscribing deliberate starvation. And this is precisely because we recognise its citizens as having a pre-institutional moral right not to be treated this way. The legal obligation to honour this right simply reinforces or supplements the moral obligation. Accordingly, the annulment of international human rights treaties can at most be said to remove the source of the additional, supplementary obligation. But each state’s natural duty to respect the human rights of its people remains intact.

In this connection, humanitarian intervention in the absence of valid international human rights agreements can be contrasted with the prevention and punishment of violent crime at the domestic level in the absence of political obligation. Political obligation refers to the moral duty of citizens to obey the law simply because it is the law. This obligation stems, at least on one prominent theory, from a citizen’s consent, (tacit or explicit, actual or hypothetical) to his government’s authority. I consent to obey, and therefore disobedience constitutes a breach of contract on my part, which is a transgression of the rights of my government and compatriots.

Now in some societies the conditions necessary for free and voluntary consent to be registered do not obtain. Here, citizens are not under a political obligation, and thus do not owe it to their government to comply with its directives. Someone that breaks a law prohibiting assault in this setting cannot be said to wrong his state. But he obviously wrongs his victim just the same. And importantly, we still feel that he can be justifiably punished and prevented from re-offending by agents of the state. Few would maintain that the Nazi government had no right to coerce rapists because it had no claim to their obedience. It had this right because everybody has this right. No special authority is needed. Likewise, in the absence of valid international human rights agreements each state still owes it to its people to respect their human rights; it simply does not owe this duty to other states in the international arena. But this is just not necessary to justify humanitarian intervention.

The claim that selectivity renders human rights treaties void, then, needs to be qualified. Inconsistency due to principled and practical considerations cannot support any such claim. But even if we set this aside the argument is fatally flawed. From the fact that human rights covenants are invalid it does not follow that human rights do not exist or that they cannot be legitimately enforced.

III

Two kinds of injustice can be distinguished: non-comparative and comparative. The former involves failure to give an individual his due, where this is not determined or affected by how others are treated, but only by his rights and deserts. Whether an individual has been treated properly according to standards of comparative justice, on the other hand, depends on how his treatment compares to that of others.21 One suffers comparative injustice if he is arbitrarily discriminated against, or treated worse than other people that are “relevantly similar,” even if he is treated in accordance with the demands of non-comparative justice (he is given his due, not burdened undeservedly and so on).

The selective enforcement of human rights norms would seem to involve comparative injustice against those regimes that are singled out. Hence a state’s decision not to intervene in some cases does affect the moral status of its actions in other cases; the non-interventions make the interventions comparatively unjust.

There are two questions that need to be addressed here. First, we need to consider whether “comparative injustice” does in fact constitute injustice, properly speaking. Is it really the case that someone who gets exactly

21 Joel Feinberg, “Noncomparative Justice”, The Philosophical Review, vol. 83, no. 3, July 1974, p. 299. what he deserves is nevertheless wronged if other people in relevantly similar circumstances are not treated in a like manner? Whether comparative injustice can ever suffice to make an intervention impermissible all things considered is the follow-up question. Let us consider each in turn.

Four combinations of comparative and non-comparative justice are conceivable. An action can be just in both senses, or unjust in both senses, but the two prongs can also yield conflicting demands. Joel Feinberg offers as an example a society in which illegal parking is punished by beheading, but where this punishment is administered “uniformly and without discrimination to all offenders”.22 Here comparative justice is done, since like cases are treated alike, but insofar as people (presumably) do not deserve to be beheaded for traffic offences, this practice is non-comparatively unjust. Conversely, where illegal parking is punished with a small fine but the sanction is applied arbitrarily to some but not others, we have noncomparative justice together with comparative injustice. It is the last of these four combinations that we need to take a closer look at. Our question is: can a course of action that satisfies the non-comparative standard properly be considered unjust in any sense simply because it falls short of the comparative standard?

Joshua Hoffman is sceptical. He uses Feinberg’s example of the Augustinian theory of salvation to illustrate why. On Augustine’s account, no human being deserves to be saved from hellfire and brimstone, but God, out of mercy, chooses to grant salvation to a select few. Those of us who do not receive God’s grace cannot complain of non-comparative injustice, since we all thoroughly deserve damnation. Yet Feinberg suggests that we suffer comparative injustice insofar as other people, who are no more deserving of it, are granted everlasting life. To call this injustice, Hoffman argues, is tantamount to saying that a random act of charity wrongs all those who are equally needy but receive nothing, and this is clearly counter-intuitive. If X does not owe anything to Y, he does not treat Y unjustly by failing to assist him financially. Should X then choose to donate to Z, nothing changes. From this Hoffman draws the conclusion that as long as an individual is not treated contrary to his rights and deserts, he is not treated unjustly in any sense simply by virtue of being treated unequally.23

22 Ibid., p. 311.

23 Joshua Hoffman, “A New Theory of Comparative and Noncomparative Justice”,

Philosophical Studies, vol. 70, 1993, pp. 174–75.

But let us make yet another concession for the sake of argument, and admit that there is such a thing as comparative injustice. It follows that an intervention which forms part of an inconsistent pattern is morally tainted. But is this ever enough to make an otherwise legitimate humanitarian intervention impermissible all things considered? The answer to this I think is clearly no.

Feinberg admits that the right to be given one’s due is more important than the right not to be discriminated against.24 He goes so far as to say that: the superiority of the claims of non-comparative to comparative justice is in some cases so striking that one might well raise the question whether comparative justice, in those cases, makes any claims at all.25

If this applies anywhere, it is in cases where humanitarian intervention is called for. A state that violates the fundamental rights of its people is guilty of non-comparative injustice of the highest order. The imperative to prevent this from occurring or continuing well and truly buries whatever claim to equal treatment its perpetrators might have. Indeed, the fact that a tyrant or dictator will be treated unfairly hardly seems worth mentioning. Let us move on.

IV

The fact that a state chooses to defend human rights here but not there might make its actions comparatively unjust, annul the international treaties which bind states to honour human rights, and expose ulterior motives. But none of this, I have suggested, can suffice to make a humanitarian intervention morally impermissible. Suppose, however, that the human rights abuses occurring there are far more serious than the oppression occurring here, so much so that intervention there is morally obligatory, while intervention here is merely permissible. Suppose further that the reason for the non-intervention is that the intervention has exhausted the discretionary resources of the acting state. In this scenario, a state is rendered incapable of fulfilling its moral obligation to rescue the subjects of one country by prosecuting a humanitarian intervention in another, to whose citizens it does not owe any such obligation. This could plausibly be

24 Feinberg, “Noncomparative Justice”, p. 317. 25 Ibid., p. 317. said to make the intervention wrongful. (A father may normally be entitled to spend his extra money on his hobby, but should his daughter become ill and need the money for treatment he loses this prerogative. What is normally permissible then becomes impermissible insofar as it leaves the father incapable of discharging his parental responsibilities.)

This argument presupposes that humanitarian intervention is sometimes a right and sometimes a duty. But this position is fraught with difficulties. Up to this point I have simply assumed that defending the human rights of foreign nationals is (at most) the prerogative of every state, but the responsibility of none. This needs to be rebutted if the argument under discussion is to even get off the ground. But providing a persuasive argument for a duty of military intervention is no easy feat, and even if the case can be made, another problem immediately arises. The few ethicists who do accept that humanitarian intervention can rise to the level of duty seem to think that all permissible humanitarian wars are also obligatory. This is equally unconducive to the objection spelled out above.

Two distinct arguments in support of a duty of humanitarian intervention can be found in the literature. The first rests on the premise that human rights impose positive as well as negative obligations. If I have a right to life, this enjoins a right not to be murdered (obviously), but also a prima facie right to be protected against murder. If nobody has an obligation to come to my defence when my life is threatened, then I do not have a right to life in any practically meaningful sense.26 From this a number of ethicists have attempted to derive a duty of humanitarian intervention. But this is not as straightforward an inference as it might at first appear.

Any duty to defend human rights must be qualified by a high-cost proviso. One cannot be obliged to render assistance to others if this would involve sustaining too high a cost or taking too great a risk. Now while states—especially affluent states—may be in a position to render assistance without significant sacrifice of their vital interests, the soldiers that actually do the fighting are not. Humanitarian war does, or at least can, put members of the armed forces in grave danger. And even where the risk to soldiers is minimal, they are still exposed to severe psychological damage as a result of their involvement. In the course of a typical humanitarian intervention a soldier can expect to witness killing and maiming.27

26 Henry Shue, “Conditional Sovereignty”, Res Publica, vol. 8, no. 1, 1999, pp. 1–7.

27 As many as 85% of former peacekeepers reported that they witnessed shootings and 47% had seen dead or wounded people. See I. Bramsen, A.J.E. Dirkzwager and H.M. Van der Ploeg, “Predominant Personality Traits and Exposure to Trauma as Predictors of

He will often feel helpless to defend the people that he has been sent to rescue. On occasion he may be called upon to clear or recover dead bodies. He must exercise restraint in the face of danger, but also resort to deadly force when necessary. The psychological scars that such experiences can potentially leave should not be overlooked.

As Dave Grossman and Bruce Siddle point out, “combat, and the killing that lies at the heart of combat, is an extraordinarily traumatic and psychologically costly endeavour that profoundly impacts all who participate in it.” The natural resistance to killing one’s own kind unleashes a “psychological backlash” when contravened, often resulting in depression, severe cases of Post-Traumatic Stress Disorder and even suicide.28 Merely witnessing traumatic death can also have long-lasting effects. In fact, even the anticipation of exposure to traumatic death has been known to cause significant distress among military personnel.29 Moreover, where the primary aim of a military operation is to protect civilians and keep the peace soldiers are expected to exercise greater restraint than would usually be required, since this is crucial for effective peacekeeping.30 But a recent study has suggested that this can cause anxiety and frustration among combat-trained soldiers, and lead to the development of mental disorders such as chronic hostility later in life.31

Thus even where a soldier’s life is not in grave danger, he may be exposed to the kind of psychological damage that would render him incapable of living a flourishing life upon returning from duty. This brings us to an important question. If participating in a humanitarian intervention is potentially so costly for soldiers, are they not entitled to invoke the aforementioned high-cost proviso? If they are, this presents a problem for the duty of intervention. How can a state be obliged to initiate an intervention if its soldiers are not obliged to execute it? At most, a state might have a

Posttraumatic Stress Symptoms: A Prospective Study of Former Peacekeepers”, American Journal of Psychiatry, vol. 157, no. 7, 2000, pp. 1115–1119.

28 As of 1996, three times as many Vietnam veterans had died from suicide after the war than died from enemy attack during the war. See Dave Grossman and Bruce Siddle, “Psychological Effects of Combat”, Killology Research Group, retrieved on 30/8/2008 from <http://www.killology.com/print/print_psychological.htm>.

29 James E. McCarroll, Robert J. Ursano, Carol S. Fullerton and Allan Lundy, “Traumatic Stress of a Wartime Mortuary: Anticipation of Exposure to Mass Death”, Journal of Nervous and Mental Disease, vol. 181, no. 9 September, 1993.

30 K. Allard, Somalia Operations: Lessons Learned, (Washington DC: National Defense University Press, 1995).

31 Brett T. Litz, Lynda A. King and Daniel W. King, “Warriors as Peacekeepers: Features of the Somalia Experience and PTSD”, Journal of Consulting and Clinical Psychology, vol. 65, no. 6, 1997, p. 1008. duty to try and assemble a special expeditionary force made up of volunteers, and to intervene using these forces.

The second argument found in the literature runs into similar problems. The claim is that refusing to defend people against abuses of their fundamental human rights is inconsistent with the Kantian dictum to “treat humanity always as an ends, never solely as a means”.32 Again, however, this injunction needs to be qualified. To say that an individual is obliged to defend others no matter the personal cost is essentially to reduce that individual to a means; a means to other people’s well-being.

The obvious response is that by enlisting in the armed forces, a soldier freely consents to a role which involves being exposed to death and injury (including psychological injury), and that the high-cost proviso is therefore not available to military personnel. But this does not apply to conscripts, andas Ihavediscussedelsewhere, theissueof whattheprofessional soldier “signs up for” is a point of contention.33 On Martin Cook’s account of the implicit military contract soldiers enlist to defend their nation’s vital interests, not the human rights of foreigners. Therefore the high-cost proviso is unavailable only with respect to wars of national self-defence. Although I think we have reason to be sceptical of Cook’s picture, if there is to be a persuasive argument for a duty of humanitarian intervention, it needs to be dealt a fatal blow. I am not claiming that this task is insurmountable, only that the duty to treat people as ends in themselves or to defend their human rights cannot be said to straightforwardly entail a duty of armed intervention.

But getting over this hurdle is only the first step. Once we acknowledge that humanitarian intervention can rise to the level of duty, we still need to ask: are all permissible humanitarian wars obligatory, or only some? KokChor Tan maintains that the conditions which justify the transgression of sovereignty also suffice to make intervention obligatory.34 Raimond Gaita concurs:

32 See Carla Bagnoli, “Humanitarian Intervention as a Perfect Duty: A Kantian Argument” in Terry Nardin and Melissa S. Williams (eds.), NOMOS XLVII: Humanitarian Intervention, (New York and London: New York University Press, 2006), pp. 117–42.

33 See Ned Dobos, “On Altruistic War and National Responsibility: Justifying Humanitarian Intervention to Soldiers and Taxpayers”, Ethical Theory and Moral Practice, vol. 13, no. 1, 2010.

34 Kok-Chor Tan, “The Duty to Protect” in Terry Nardin and Melissa S. Williams (eds.), NOMOS XLVII: Humanitarian Intervention, (New York and London: New York University Press, 2006), p. 85. See also Darrel Moellendorf, Cosmopolitan Justice, Boulder: Westview Press, 2002, pp. 122-23, and John Lango, “Is Armed Humanitarian Intervention to Stop Mass Killing Morally Obligatory”, Public Affairs Quarterly, vol. 13, no. 3, 2001, pp. 173–92.

Justice permits us to go to war for the sake of those who suffer criminal injustices at the hands of their governments only when, all things considered, we are obliged to do so rather than merely when great good is likely to result from it.35

Both John Rawls and Michael Walzer also seem committed to this view, though they do not commit themselves quite so explicitly. In his Law of Peoples, Rawls argues that only “well-ordered” states are morally immune from foreign intervention. A state must meet three basic conditions to count as well-ordered, one of them being respect for “human rights proper”, including the right to personal security, to the means of subsistence, liberty from slavery, personal property and to equal treatment under the law. But a well-ordered regime need not fulfil all of the rights that a liberal government guarantees.36 Undemocratic states that deny freedom of religion and free speech can still be well-ordered, and therefore entitled to the protection that sovereignty affords.

Now the tenor of Rawls’ discussion suggests that intervention is not purely optional where it is legitimate. Within the class of societies that are not “well-ordered”, Rawls draws a further distinction, setting “outlaw states” apart from “burdened societies”. Outlaw regimes choose not to honour human rights, while burdened societies “lack the political and cultural traditions, the human capital and know-how, and, often, the material and technological resources needed to be well-ordered”.37 Only in relation to burdened states does Rawls explicitly talk about a “duty of assistance”; liberal states are obliged to ameliorate the unfavourable circumstances that obstruct their development into well-ordered societies. But some such duty must also be owed to the citizens of abusive “outlaw” regimes. If people are entitled to have the “burdens” which prevent them from enjoying well-ordered political arrangements removed, why should it matter if these burdens are deliberately imposed? A purely optional ethic of intervention against outlaws does not sit comfortably with the duty to assist burdened societies and the reasons given to support it.38 And insofar as only outlaw states are legitimate targets of intervention on Rawls’ account, it turns out that all legitimate intervention is prima facie obligatory.

35 Raimond Gaita, “A Last Resort” in Raimond Gaita (ed.), Why the War was Wrong,

(Melbourne: Text Publishing, 2003), p. 100.

36 John Rawls, The Law of Peoples; with The Idea of Public Reason Revisited, (Cambridge Mass: Harvard University Press, 1999), p. 79.

37 Ibid., p. 106.

38 Ibid., p. 81.

Walzer’s body of work tells a similar story. Humanitarian intervention is permissible only when the absence of a morally significant union or “fit” between the rulers and the ruled is made “radically apparent”, according to Walzer. And this is only in cases of massacre, enslavement or ethnic cleansing. At the time that Walzer expounded this position it was offered as an account of when humanitarian intervention should be considered permissible. But in a recent paper Walzer argues that the rights against massacre, enslavement and forcible expulsion are supported by a right to be rescued from these crimes.39 Like Rawls, then, Walzer seems to think that the conditions which lift the ban on intervention also make it a duty. To see why this position is fatal to the line of argument under consideration, let us return to the scenario sketched at the beginning of this section. In our example, state ‘X’ is incapable of fulfilling its moral obligation to rescue the citizens of ‘A’ from severe human rights violations, say in the form of genocidal assault. The reason that ‘X’ is incapable of intervening in ‘A’ is that it has exhausted its discretionary resources by intervening in ‘B’, where the humanitarian situation was serious enough to justify intervention, but not to make it obligatory. Here we might say that ‘X’ did the wrong thing by getting involved in ‘B’. But what if the intervention in ‘B’ was also prima facie obligatory? It makes little sense to object to an intervention on the grounds that it causes a state to default on its duties, if nonintervention would have equally constituted a dereliction of duty. For the argument to work, intervention must occur where it would otherwise be purely discretionary, and this must foreseeably result in inaction where it is obligatory. If all permissible humanitarian wars are morally required, however, it is impossible for this scenario to even arise.

But even if we set all this aside the argument still seems to have little purchase on reality. Usually, a state’s failure to intervene in one place will not be directly attributable to its humanitarian commitments elsewhere. The US and Europe were fiercely criticised for failing to respond to the Rwandan genocide in the same way that they had responded to the atrocities taking place in the former Yugoslavia in the early 1990s. But the inaction in Rwanda was not because of the action in Bosnia; the latter was not causally related to the former in the way that the argument requires. The most likely explanation for the inconsistency is that the Western powers

39 Michael Walzer, “Beyond Humanitarian Intervention: Human Rights in Global Society” in Thinking Politically: Essays in Political Theory, ed. by David Miller, (New Haven and London: Yale University Press, 2007). simply had no strategic interests in Rwanda. If there had been important national interests at stake, it is safe to say that there also would have been an intervention.

Admittedly there may be exceptions. The ongoing civil war in Sudan has so far killed tens of thousands of civilians and made refugees out of more than 2 million. Women and children have been enslaved, prisoners have been tortured and Christian pastors have reportedly been crucified. The international community is yet to respond with force. Alex Bellamy suggests that the inaction in the case of Sudan can be attributed, at least in part, to the ill-fated “Operation Iraqi Freedom”.

Due to its commitments in Iraq, the US cannot afford to go it alone in Sudan; it has reached the point of military “overstretch”. Thus a collective humanitarian action via the UN is the only feasible option, says Bellamy. However the standing of the US and UK as leaders and “norm carriers” has been severely damaged by the Iraq war, which was widely seen as an abuse of the emerging norm of humanitarian intervention. As a result, “it has become harder for these states to persuade others to act decisively in humanitarian emergencies at precisely the moment when those states themselves are less able to bear the costs of acting outside of the world’s institutional framework”.40 The international community has failed to reach a consensus on intervention in Sudan, and the diminished credibility of the US and UK means that an aggressive diplomatic push for coercive measures on their part is likely to be counter-productive.41 If Bellamy is right, then the US and UK are now unable to initiate either unilateral or multilateral action in Sudan, and this is, at least in part, because of their actions in Iraq.

Now if true this certainly makes the Iraq intervention all the more regrettable, but is it grounds for saying that the decision to depose Saddam was immoral ? The answer obviously depends on whether the US and UK could have reasonably foreseen that their actions would have this consequence. Arguably, they could have predicted a reduction in their warmaking capacity for the duration of the Iraq conflict, and a temporary loss of credibility in the international arena, given the overwhelmingly negative response to the war’s proposal. But could President Bush and Prime Minister Blair have foreseen that the rights abuses in Darfur would reach

40 Alex J. Bellamy, “Responsibility to Protect or Trojan Horse? The Crisis in Darfur and Humanitarian Intervention after Iraq”, Ethics and International Affairs, vol. 19, no. 2, 2005, p. 33.

41 Ibid., p. 51. genocidal proportions? Could they have foreseen that Operation Iraqi Freedom would cost trillions of dollars and lead to military overstretch? Could they have foreseen that conditions in Iraq would deteriorate to such an extent as to preclude any recovery of the US and UK’s moral credibility as “norm carriers”? Could they have foreseen that no other state or coalition would be able and willing to address the worsening crisis in Sudan in the absence of their leadership? To condemn the coalition for its decision to invade Iraq on the grounds that it led to non-intervention in Sudan, an affirmative answer to each of these questions is required. Intervention in X can be rendered illegitimate by non-intervention in Y only if the nonintervention constitutes a dereliction of duty, and this is a foreseeable consequence of the intervention.

V

I have surveyed what I take to be the strongest arguments for the claim that a humanitarian intervention can be rendered illegitimate by the fact that the state prosecuting it has failed, is failing or will fail to defend human rights in relevantly similar circumstances elsewhere. Each of the arguments has come up short.42 The first succeeds, at most, in showing that inconsistency gives us reason to be critical of the intervener, not the intervention. The second rests on the mistaken premise that human rights do not exist independently of international covenants and treaties. The third overstates the place of comparative justice in an all things considered moral judgment. And the fourth is contingent upon a contentious account of the ethics of intervention, and would apply only in circumstances that seem rather unlikely to arise in the real world. We must conclude, then, that an intervention cannot be made illegitimate by the mere fact that it occupies a place in an inconsistent or selective pattern.43

42 Although my discussion has been restricted to armed humanitarian intervention, my conclusion applies equally to inconsistency in the criticism of tyrants. Again, selectivity might make the non-criticisms worse, morally, than they would otherwise have been, but the criticisms that do occur are still likely to be desirable and justified, all things considered.

43 I would like to thank Igor Primoratz, Andrew Alexandra, and two anonymous reviewers for the Journal of Moral Philosophy for their helpful comments and suggestions on an earlier version of this paper.

Part Two: Just War and Self-defense

Conditional Threats

Gerhard Øverland

A person might want your money and be willing to kill you for it. But instead of just shooting you and making off with the dough, she might offer the infamous choice, your money or your life! Situations like this land us with difficult choices. Not, of course, whether you should hand over the money and live, or decline and die. Yet, if you were presented with the option of killing the threat to save both money and life, whether it would be permissible to proceed is not clear.

David Rodin uses the notion of ‘conditional threat’ to describe people who, instead of trying to kill you outright, threaten to kill you unless you give them what they want.1 The problem of conditional threats comes about when the threat posed by an aggressing party is an immediate threat only to something of a lesser value, and merely conditionally a threat to life. The term may also be applicable to invading forces.2 Invading forces will often not kill anybody unless people put up a show of resistance. Since one might believe that responding to conditional threats with lethal force is impermissible unless what is extorted is of sufficient merit to justify so doing, one might conclude that it would be impermissible for an invaded country to mobilize its armed forces to protect something of only a lesser value.3

In this paper I ponder the moral status of conditional threats, in particular the extent to which a threatened party would be permitted to use (lethal) defensive force. I first investigate the mugger case before turning briefly to the more complicated issue of national defence in the face of an invading army. One should not, I argue, exaggerate the level of protection people under threat owe their conditioned killers simply because what is

1 See David Rodin, War and Self-Defense (Oxford: Oxford University Press, 2002), pp. 133–38.

2 Richard Norman has, for instance, argued that the Falklands War was a classic example of a situation where defensive force should not have been applied (see Richard Norman, Ethics, Killing and War [Cambridge: Cambridge University Press, 1995], pp. 156–58).

3 Norman holds that ‘the fact that unjust demands are backed up by lethal threats does not by itself justify killing in response; it depends on the importance of what is defended’ (Ethics, Killing and War, p. 133). extorted is of little value. After all, either the conditional threat is willing to kill for something she has no claim on at all, or, if she has a just claim on it – or good reason to believe that she has – she is willing to kill to (re) acquire something of not much worth. In both cases the moral culpability of the conditional threat could be seen as reducing her claim on protection from harm.

Risk

Jeff McMahan has little sympathy with conditional threats, and holds that if a thief threatens to kill you if you resist, then ‘one is permitted to create the conditions of one’s own lethal defense. For the thief’s threat does not nullify one’s right to resist’.4 McMahan’s idea is that if someone is trying to steal money from you, you are permitted to use proportional means to avert the crime, for example, by pushing the assailant away. He then maintains that this option cannot become impermissible by being presented with a conditional threat.5 And then, the argument goes, since trying to push someone who’s waving a gun at you is not always the most advisable formula, the defendant may bypass that step and activate the lethal force alternative forthwith.

But this move merely seems to imply that a person is killed for an unjustified cause, namely an individual’s right to resist a mugging. We therefore need to know why it is permissible to set in motion events that will lead to the death of the aggressor when what is being protected is of little value. What is it about the conditional threat that makes lethal force permissible despite the fact that the value at stake does not make it so?

Initially Rodin’s view is that ‘it is never legitimate to respond to conditional threats with lethal force unless what is extorted is itself of sufficient value to justify the use of lethal force’.6 That sounds counterintuitive. It neither takes into account the culpability of the agent, nor the danger to which the person under threat has been exposed. After having discussed the views of John Locke, Rodin agrees to let the risk imposed on the victim play a role. Whether there is a right to kill in response to conditional

4 Jeff McMahan, ‘Innocence, Self-Defence and Killing in War’, Journal of Political Philosophy 2 (1994): 193–221, at p. 196.

5 Observe nevertheless that if the conditional threat affects a third party, and pushing the aggressor out of the way exposes the third party to a higher risk, it could be impermissible to take that course of action.

6 Rodin, War and Self-Defense, p. 134. threats, he now maintains, turns on ‘the nature and value of what the aggressor demands and the level of risk assumed’.7

But if there is a significant risk one might question the importance of being a conditional threat, as the risk itself might justify lethal defensive force. If a person for no good reason poses a significant risk to another, the latter should be morally free to forcibly defend himself. Take Russian roulette with an unwilling player as a case in point. Although there is only a one in six risk of death, shooting and killing the player would seem entirely permissible if that were the least intrusive way of preventing the player from pulling the trigger.

Anyway, the risk factor itself may not change things very much with regard to a trustworthy conditional threat, i.e. a person who will only cause harm if she doesn’t get what she wants. In those cases, it could be argued, the level of risk depends mainly on the response of the victim. The victim will only be killed if he defends himself forcibly – provided we assume with Rodin that it is wrong to use defensive force in a situation in which one is not exposed to significant risk, and that trustworthy conditional threats do not expose us to a significant risk. But the latter assumption is not true; a victim is exposed to risk whether or not he responds by acting ‘wrongly’ such as trying to kill the conditional threat. For example, the victim may falsely believe that he will be able to run for cover, which means that he will be shot dead if he tries. And clearly, trying to escape is not morally wrong, no matter how stupid. Hence, not even a trustworthy conditional threat is in overall control of the risk she poses. Risk caused by the unwise behaviour of the victim, for instance, is something the conditional threat is willing to impose upon other people. That counts against her.

Moral Responsibility

The mere fact that what is at stake is of minor value speaks against permitting killing conditional threats. If you are sitting in a café, and an unarmed pickpocket tries to relieve you of a few dollars, you would not be permitted to resist if killing the thief was your only recourse. Yet, this simple observation does not take us very far. There is a big difference between ordinary thieves and conditional threats.

7 Rodin, War and Self-Defense, p. 137.

To illustrate the difference, imagine an escalated situation between a petty thief and a victim. The victim resorts to lethal defensive force to protect his money. The thief, on her side, happens to have access to a weapon and can use it to protect herself against the victim’s use of lethal force. However, she neither intended to use any weapon to obtain the money nor was it reasonable to construe her behaviour as threatening. We have to take immediate and possibly lethal action to save either the victim, who is trying to protect a few dollars, or the petty thief, who out of the blue has procured a gun from somewhere. It would at this point seem permissible to save the thief who was not a threat to anyone and rather kill the person who knowingly resorted to lethal force to defend something of as little value as a few dollars (assuming here that the owner of the money is not poor or in any particular urgent need). If that is true, it must be because the wrong committed by resorting to lethal force to deter a petty thief is a significantly greater moral wrong than the thief’s unsuccessful unarmed attempt to purloin the few dollars.

The same does not seem to be true when the thief is a trustworthy conditional threat. If a person not only wants the victim’s money but also is willing to kill for it, as revealed by holding him at gunpoint while offering the choice between money and life, things take on a different hue. If the victim now resorted to defensive force, and we had to select whom to help, it would be wrong to save the conditional threat at the cost of the other. We may want to say that responding with lethal force was wrong, given that what was at stake was something of minor value. Yet, we nevertheless seem obliged to give priority to the person defending himself against the conditional threat.

This observation cannot be explained by the fact that victims of conditional threats are exposed to risk, and that we therefore can see that they have a reason for using lethal force, although what is at stake is not of much worth. At the time of the intervention there is no difference in risk between the escalated situation involving the petty thief and the one with the conditional threat. It must therefore be explained by the serious moral wrong it is to be a conditional threat in the first place. It is the threat’s moral responsibility for waiving a loaded gun that would require third parties to assist the defendant whenever there is a question of saving either of them. The aggressor’s willingness to use deadly force to obtain something of lesser value makes him morally culpable, and this culpability affects the amount of force that it would be proportional for the victim and others to use. Hence, the means we permissibly may invoke to frustrate conditional threats is neither a simple question about whether what is extorted is itself of sufficient value to justify the use of lethal force, nor a conjoined evaluation of the risk to the defending party and the value at stake, but a conjoined evaluation of the two former and the moral culpability of the agent.

Generally, moral culpability matters in selfand other-defence. For instance, if an innocent aggressor is about to cut off somebody’s limb there are strict proportionality considerations that would limit the means that could be used to prevent it. It would, for instance, not be permissible to kill the innocent aggressor to save the limb. By contrast, if the agent is a culpable contributor, who enjoys splitting people from their limbs, it seems uncontroversial to assume that significantly more may be done toward the wrongdoer if that is necessary to save a limb of the victim.8 Even killing a culpable aggressor seems permissible when need be to save a limb. Proportionality considerations still apply, but are more relaxed compared to situations where force is used against an innocent contributor, and the degree to which they are more relaxed depends on the degree to which the agent is morally responsible for his or her actions.9 It is therefore not surprising that the moral responsibility on the part of a conditional threat should likewise impinge on the permissibility of using force against her.

Asymmetry

Having seen that moral culpability matters, there is a certain peculiarity to a conditional threat that might lead one to suggest that she would be more culpable when less is at stake. The rationale would be the following. First, if what is at stake is something of minor value that will reduce the permissibility of killing in its defence. But if so, we should also expect that the moral culpability of the person who is willing to take life in order to obtain something of minor value would be increased. This will then reduce her claim on protection from harm. Granted the threat is trustworthy, she is now willing to kill for almost nothing. That is a serious wrong on her part.

8 Judith Jarvis Thomson observes this in ‘Self-Defence’, Philosophy and Public Affairs 20 (1991): 283–310, at p. 289. See also Yitzhak Benbaji, ‘Culpable Bystanders, Innocent Threats and the Ethics of Self-Defense’, Canadian Journal of Philosophy 35 (2005): 585–622, at pp. 597–99.

9 This proposal does not imply that I am committed to a punitive account of defensive force. A punitive account seems to imply that the person deserves that something bad happens to her; my proposal is compatible with the weaker claim that she deserves less protection in a conflict between persons.

It is therefore not obvious that a reduction in the value at stake will narrow the permissible range of defence options. One might even be tempted to say that it is the other way around, as a willingness to kill for very little may look worse than killing for something of significant value.

But as it stands, that proposal cannot be correct. It doesn’t seem to apply to the acquisition of values you have no claim on in the first place. Whether or not the conditional threat attempts to get hold of a few dollars or millions doesn’t seem to alter our judgments about her moral culpability. It is not OK to be a conditional threat as soon as the value at stake is ‘worth’ killing for.

A plausible option is to say that when an agent is willing to kill for something she has no just claim on, her moral culpability is immune to changes in its value. It is equally wrong to be a conditional threat for the sake of a few dollars, as it is for the sake of a million. And then the interest at stake will work as an independent factor making use of lethal force increasingly proportional. For instance, if the person threatens to kill you unless you sign off a significant part of your wealth, killing the threat could be justified.

This proposal implies that it might be morally permissible to use lethal force in order to protect a particular interest from a conditional threat and impermissible in order to protect it from a simple thief. For instance, you may have to let a thief take off with your car, but be permitted to defend it with lethal force if you are threatened with deadly consequences. Moreover, the significance of the moral culpability for being a trustworthy conditional threat ensures that we should always side with the victim and never the conditional threat, as the moral culpability associated with the latter trumps the moral wrong of resorting to lethal force to defend something of minor value.10

Things look different when the conditional threat attempts to extort things to which she has a just claim. In this case, her moral culpability will be determined by its value, but in inverse proportion. Even though a person correctly believes that her neighbour has stolen something from her, a

10 Clearly, part of the explanation why it seems wrong to shoot a conditional threat when only a few dollars are at stake might in the ordinary case be that we do not really believe the person is actually prepared to kill us if we refuse to submit to her demand. I am therefore alluding to trustworthy conditional threats, that is, people who will not kill unless they are met with resistance, but they will if need be. If we alter the second part of this clause, things take on a different hue. If the agent will not kill for money, and is merely putting up a show to increase her chances of success, she would not be exposing anyone to risk – except herself, as she may be taken to be trustworthy and therefore killed – and her moral culpability would be considerably smaller. hundred dollars say, she better know that killing to obtain it is disproportional. Hence, trustworthy threatening her neighbour with lethal consequences if he doesn’t hand back the money would be morally wrong. By contrast, if the neighbour had stolen something significant, being a conditional threat may very well be a permissible option.

The value at stake bears differently on the culpability of conditional threats. For a conditional threat with no just claim, it simply has no impact; she is fully culpable no matter the value of what she is after. For a conditional threat with a just claim, the interest at stake will impact inversely on her moral culpability; she becomes fully culpable as the value at stake approaches zero. This implies that a conditional threat with a just claim on something of insignificant value is (almost) as culpable for waiving a gun as any conditional threat with no claim. A difference between them is nevertheless observable in that the fully culpable conditional threat without a just claim may in addition try to extort something of value, making defensive force easier to justify.

It may at this point pay to observe how the value at stake and the moral culpability of the conditional threat with a just claim interact. If the value is significant – perhaps almost worth killing for – it is not very wrong to be willing to kill for it. Moreover, the conditional threat has now a just claim on something valuable, and since the neighbour’s initial theft was of this item, it was a significant wrong on his part. Third parties would therefore have to side with the conditional threat. By contrast, if it is something of little value, it is indeed very wrong to be willing to kill for it. And now the interest of the conditional threat would not carry much weight either. This leaves her with very little to support her case. And although the neighbour’s initial theft was wrong, it would not be significant enough to outweigh the wrong on her part. Hence, third parties should side with the neighbour (the simple thief) in this case.11

War

A surprising event during Germany’s invasion of Norway, early on 9 April 1940, was the sinking of the German Navy’s newest ship, the heavy cruiser

11 If we vary the neighbour’s moral culpability by assuming that he has an excuse in the sense that he had good reason to believe that the object at stake was his to take in the first place, things become more difficult. In that case you defend something of significant value, but yet not worth killing for, leaving some culpability on your part, while your neighbour is innocent in the relevant sense.

Blücher, by the 47-year-old 28cm guns of Oscarsborg fortress. A total of 830 German soldiers died that morning either from the blast or by drowning in the cold water. As it turned out, the Germans persecuted the Jewish community in Norway as well, which, had one known at the time, would have justified lethal resistance against the invading forces.12 But when sailing up the Oslo fjord, the Germans had no intention of killing anybody; they only wanted to take over a country.

The asymmetry between the permissibility to use defensive force against simple thieves and conditional threats has gone mostly unnoticed, and that is particularly unfortunate with regard to the question of defensive wars. In such cases, the aggressor typically claims to have some kind of just cause, but fails to note that the value they are about to fight for is something of a lesser value not worth killing for. They may even have a just claim on it, but they ought to know that killing to obtain it is disproportional. Their culpability will therefore be set inversely to the value at stake, securing permissibility to defend with lethal means that which it otherwise would not have been permissible to kill for.

Invading soldiers may, of course, have a just claim – or a just cause – and they may therefore be permitted to kill the defending soldiers. Certain instances of humanitarian interventions would be examples. I will not discuss such cases here. My interest is primarily with situations where the invading soldiers have no just claim, but have good reason to believe that they have, and therefore perhaps are fully excused for their mistake in this regard. It is at least possible that the soldiers on the deck of Blücher fell within this category.

There are important differences between selfand other-defence in civil life and war. Let me here point to two considerations that seem to pull in different directions, the question of soldiers’ innocence and the likelihood of collateral damages. Many think that soldiers’ participation in unjust aggression can often be morally excused. The fact that soldiers tend to be young, uneducated, and are strongly loyal to their country are some reasons for this. And if indeed the belligerent soldiers are innocent, the defending party appears to have a duty to endure higher risk and expose their own forces to greater harm so as to reduce the harm done to the aggressing soldiers. I cannot here discuss whether soldiers normally

12 It might also arguably have helped the Norwegian royal family, parliament and cabinet to escape to London. have a moral excuse for participating in their country’s unjust war, but let me say a few words about the claim that defending soldiers ought to endure higher risks in order to avoid harming excused soldiers.

One might perhaps wonder why innocent defending soldiers, who are not acting wrongly, should bear greater costs to avoid harming innocent aggressing soldiers engaged in wrongdoing. Note, however, that the claim is only that they should bear greater cost in this case than if the aggressors were morally culpable. I have already indicated that if an innocent aggressor is about to cut off somebody’s limb the proportionality considerations limiting the means that could be used to prevent it would be stricter than if the aggressor were morally culpable. It is easy to envisage situations involving numerous aggressors where similar considerations apply. Moreover, I am not suggesting that the innocence of the aggressors would make defensive force impermissible. Simply being an innocent contributor to harm has severe implications for the amount of defensive force that may be used. Yet, moral culpability expands the range of permissible defensive means. Innocent aggressors are not on par with innocent bystanders as has been suggested by quite a few philosophers recently,13 but they are not on par with culpable aggressors either, as I discuss elsewhere.14

As for a conditional threat, the idea is that the risk one is required to take in order to avoid harming this person decreases as a function of her moral culpability. But then observe that it is not clear whether invading soldiers who pose only a conditional threat to life should be regarded as innocent, even though they may be excused for reasons of loyalty etc. After all, soldiers are typically geared up to kill if ordered to do so. We can therefore assume that an invading army is trustworthy, and that its soldiers are ready to kill to advance their objective. But making a conditional

13 See N. Zohar, ‘Collective War and Individualistic Ethics: Against the Conscription of “Self-Defense”’, Political Theory 21 (1993): 606–622; Michael Otsuka, ‘Killing the Innocent in Self-Defence’, Philosophy and Public Affairs 23 (1994): 74–94; Jeff McMahan, ‘Self-Defense and the Problem of the Innocent Attacker’, Ethics 104 (1994): 252–290; David Rodin, War and Self-Defense (Oxford: Clarendon Press, 2002), pp. 70–99; Jeff McMahan, ‘The Basis of Moral Liability to Defensive Killing’, Philosophical Issues 15 (2005): 386–405; and Kimberley Kessler Ferzan, ‘Justifying Self-Defense’, Law and Philosophy 24 (2005): 711–49.

14 A first attempt is found in ‘Forced Assistance’, Law and Philosophy 28 (2009): 203–232, where I propose an asymmetrical fair share procedure, according to which innocent contributors have a duty to take on a fair share of the harm if dividing it is possible, and a fair share of the risk of being harmed if redistribution is impossible. I develop the account further in ‘Dividing Harm’, a work in progress. threat to life in order to obtain something of a lesser value reveals a moral fault on part of the threat. Hence, although soldiers may (falsely) believe their country is fighting in order to reclaim something they have a just claim on, they ought to know it is wrong to kill in the service of the objective for which they are invading the other country. Compare with the mugger. While she might have good reason to believe that the few dollars belong to her, she should know that she ought not to kill in order to obtain them – provided she is not starving or finds herself in a particularly desperate situation. Similarly, we may partly excuse the forces of a belligerent state engaged in unjust aggression, as they may, for instance, have good reason to believe that the land they are about to occupy was unjustly stolen from them some years back. What cannot be excused is their willingness to resort to lethal means to secure their prize. Hence, though they are excused for having wrong beliefs about who are entitled to what, there is no excuse for their decision to take up arms, and they are morally culpable in that respect.

Now, these lines of thought might seem to entail that whilst soldiers who are killing in the name of an unjust war should be afforded protection by those they aim to kill, soldiers who pose only a conditional threat in the name of an unjust war are not innocent, and thus are fair game for the defending forces. That is not the proposal, of course. First of all, if it is morally wrong to be a conditional threat because the value at stake is of minor importance, it is wrong to carry out the killing as well. Moreover, merely being an innocent contributor is significant, but so is moral culpability. I take no stand here as to what is most significant, and, needless to say, that would very much depend on the degree of moral culpability. It might be worth observing, though, that in a choice between harming a conditional threat and an innocent aggressor, it is not at all obvious that you ought to harm the latter. Suppose that you can survive by way of killing a person on your left or on your right. On the left there is an innocent aggressor coming against you; on the right there is a conditional threat that will kill you unless you hand her some money. Your morally preferred option in this case seems to me that you kill the conditional threat.

Of course, the aggressing soldiers may have been deluded about the true cause of the war as well. They might, for example, believe they are on a liberating mission, when they are really engaged in an attempt to gain control over some natural resources. That being the case they should be considered innocent and the defending party should accept a higher risk of harm to avoid killing them. It is, however, not clear what one could have told the soldiers on Blücher’s deck early that morning which gave them reason to believe they were justified in killing people in a democratic state like Norway in order to gain control over their country.15

While there clearly may be instances in which what is at stake is something of a lesser value and not worth killing for, it seems that Rodin and others overstate the instances in which that should have implications for conditional threats. They seem to think that a state that launches a defensive war against an invading army that’s only seeking to take away things like collective self-determination, land and valuable resources, or a people’s ‘common way of life’, fails to meet the requirement of proportionality because the value of such factors is not high enough to warrant the taking of life.16 I think it is perfectly permissible to kill if protecting such values requires it, provided the conditional threats are trustworthy and are aware of their mission’s real aims. By knowingly engaging in such threats, one demonstrates serious disrespect of others. They may have reason to believe that they have a just claim on what they are ready to fight for, but they have seriously misjudged the extent to which it is permissible to take up arms to claim it. Hence, resorting to lethal defensive force might not be permissible simply because one wants to protect something of a lesser value, nor perhaps simply because one is exposed to a particular risk – though that may be the case if the risk is significant – but because the aggressors are willing to kill for these objectives.

Turning now to the obverse case, namely the likelihood of death of non-threatening innocent people, or collateral damage. While it often is permissible to kill well-informed trustworthy conditional threats to protect something that in itself does not merit lethal defence, this observation does not imply that waging a defensive war in order to protect such objectives always is permissible. The reason is simply that more than wellinformed trustworthy conditional threats might be killed in the war.17

15 One might perhaps also like to argue that the value at stake should be evaluated differently from the two viewpoints. To gain control over others may have some value, but to maintain control over oneself seems to have higher value. Hence, the Germans’ case looks even worse. Their allegedly just claim would clearly not be worth killing for, making their culpability quite severe, while the Norwegians’ interest in maintaining self-control strengthen their case of taking up arms.

16 Rodin, War and Self-Defense, p. 134. See also David Rodin, ‘Beyond National Defense’, Ethics and International Affairs 18.1 (2004): 95. Richard Norman holds essentially the same view; see his Ethics, Killing and War (Cambridge: Cambridge University Press, 1995), pp. 120–56.

17 McMahan argues that defensive war waged by a victim state can be proportionate given that other nations will be dissuaded from committing aggression as a result of waging the defensive war. Jeff McMahan, ‘War as Self Defense’, Ethics and International Affairs

18.1 (2004): 79. While this looks like a plausible consideration when it is talk of merely

Since an essential justification of resorting to lethal means against conditional threats is based on the moral culpability of the threat, this type of justification will not suffice for actions that cause the death of people outside the ranks of the belligerent soldiers.18 killing trustworthy well-informed conditional threats, it seems much less weighty in order to justify killing innocent non-threatening people, collaterally or otherwise.

18 I would like to thank an anonymous referee of this journal for helpful and challenging comments.

Can Wars Be Fought Justly? The Necessity Condition Put to the Test

Daniel Statman*

After years of disregard, questions regarding the meaning of the proportionality condition and its application to wars are finally attracting some philosophical attention.1 Yet, as David Mellow noted a few years ago, “there continues to be a significant lack of clarity, precision and agreement about how the proportionality condition should be defined and applied in practice.”2 The situation is even worse with regard to the necessity condition,3 especially insofar as it concerns (if indeed it does) jus in bello. This paper is an attempt to fill this gap. I seek to show that, properly understood, the necessity condition leads to very demanding constraints on the use of force in war, constraints which cast doubt on the possibility that wars can be fought justly. I then argue that this conclusion lends support to a recent interpretation of just war theory, one that emphasizes the contractual nature of the war convention.

for Helpful Comments on Earlier Drafts, I Am Greatly Indebted to Yitzhak Benbaji, Yuval Eylon, Michael Gross, and Saul Smilansky.

1 See especially Thomas Hurka, “Proportionality in the Morality of War,” Philosophy &

Public Affairs 33 (2005): 34–66; Idem, “Proportionality and Necessity” in Larry May (ed.), War: Essays in Political Philosophy (New York: Cambridge University Press, 2008), 127–144; Jeff McMahan, “The Ethics of Killing in War,” Ethics 114 (2004): 693–733; David Mellow, “Counterfactuals and the Proportionality Criterion,” Ethics and International Affairs 20 (2006): 439–454.

2 Mellow, ibid., 440. See also Michael Gross, “The Second Lebanon War: The Question of Proportionality and the Prospect of Non-Lethal Warfare,” Journal of Military Ethics 7 (2008), 2 (“Although the principle of proportionality enjoys widespread support, it is generally ill defined, misunderstood, wrongly applied, and without any feasible operational definition”). The legal notion of proportionality in international law fares no better, as shown by Abraham Bell, “Optimal Proportionality in the Law of War” (available at http://www.law

.northwestern.edu/colloquium/international/documents/Bell.pdf), p. 3 (“Proportionality means different things in different contexts within the law of war, and the scope and application of the concept is far from clear in many cases”).

3 See George Fletcher and Jens David Ohlin, Defending Humanity: When Force is Justified and Why (New York: Oxford University Press, 2008), 93 (“Very few scholars pay careful attention to the requirement of necessity, even though it is a universal requirement of selfdefense… The literature of international law speaks of the requirement of necessity as well as of proportionality, but no real sustained attention is paid to either of these criteria”).

I. Necessity and Proportionality: Preliminaries

Let me start by exploring the meaning of the necessity condition and the proportionality condition in individual self-defense. Some writers seem to assume that the necessity condition is no different to the proportionality condition because choosing the least harmful course of action (which is what the necessity condition requires) is the same as choosing the proportionate one. This view, however, seems wrong. Think of an aggressor, A, posing a threat to a potential victim, V. Even if, as a matter of fact, the only way V can stop A from carrying out an unjust attack upon her is to carry out a self-defensive act, D, which would otherwise be immoral (or illegal), carrying out D might still be disproportionate to the unjust attack (=cause much more harm than the unjust attack would) and hence be ruled out by the proportionality condition. The distinction between the necessity condition and the proportionality condition can thus be expressed as follows: The necessity condition concerns the relation between some defensive act, D, and all alternative defensive acts D1, D2… Dn. It requires that D be less harmful than all these alternative acts. The relevant comparison is independent of the threat posed to V by A. Whatever its nature and magnitude, V is required to take the least harmful measure to block it. In contrast, the proportionality condition contains no reference to alternative defensive acts, only to the relation between D and the unjust threat. If D is disproportionate to the threat, then it is ruled out by the proportionality condition regardless of the existence or the nature of alternative defensive actions that might prevent the threat from materialization.

To these two conditions, we should add a third which is usually overlooked, namely, the Success Condition.4 The first question that V must ask herself is which actions are available to her that could reasonably be expected to block the threat. If the answer is none, then V has no right of self-defense against A (that is, no right to carry out otherwise immoral acts against A). If only one act, D, passes this primary test, then the necessity condition is satisfied by way of default. If more than one act passes the primary test, the act that is least harmful to A must be selected. D must then face the proportionality test. If D passes it, V is justified in carrying out the act. If not, V is doomed. She will have to suffer the aggression with no moral means of self-protection.

4 See my “The Success Condition for Legitimate Self-Defense,” Ethics 118 (2008): 659–686.

The order of conditions is, then, success->necessity->proportionality. V must first establish that D can block the unjust attack, then that it is the least harmful way to do so, and finally that it is proportionate to the attack. This order reflects an ascent in the severity of the moral constraint imposed on V. The success condition is the weakest constraint, because if some act is not expected to provide any defense at all, surely it cannot enjoy the protection of the right to self-defense. The necessity condition is more demanding, because it prohibits V from taking a course of action that is expected to block an unjust aggression – simply because there is a less harmful one which could do the same job. The proportionality condition is the most demanding constraint because it prohibits V from doing D in order to block an unjust attack, even when D is expected to block it, and even when it is expected to do so in the least harmful way.5 That the proportionality condition poses a more demanding constraint on self-defense than the necessity condition explains the fact that while no modern legal system omits the necessity condition from the list of conditions for legitimate self-defense, some omit the proportionality condition or reserve it only for extreme cases. But it is not only the demanding nature of the proportionality condition that explains the lack of consensus surrounding it, but the apparent unfairness involved. It seems unfair that when a selfdefensive act, D, is necessary but disproportionate, it is V who is doomed to suffer rather than A.

II. The Necessity Condition and Warfare

The analysis of the previous section helps us to see how the application of the necessity condition leads to opposing conclusions regarding the morality of war – to weaker restrictions, on the one hand, and to stronger restrictions, on the other. Or so it seems.

5 Here and in the rest of the paper, I’ll be assuming a subjective reading of the conditions under discussion. What matters for the legitimacy of D is not whether D is in fact effective, necessary, or proportional, but that it is reasonably believed to be so by the agent. For the sake of simplicity, I shall often use objective language (e.g. “D is disproportionate to the threat”), but I’ll always be assuming a subjective one. On this point I am in disagreement with Mellow, who explicitly accepts an objectivist view of the justification for selfdefense. See David Mellow, “Iraq: A Morally Justified Resort to War,” Journal of Applied Philosophy 23 (2006), 299–300 and note 18. Because he holds this view, he believes that the justification for war is essentially retroactive, which leads him to justify the war in Iraq by its actual results as he judges them. This retroactive viewpoint explains his worry about counterfactuals when it comes to the application of the proportionality condition; see Mellow, “Counterfactuals and the Proportionality Criterion.”

Why might one think that correct application of the necessity condition should relax the constraints on warfare? The answer has to do with the essential future-oriented nature of self-defense. For an otherwise immoral/illegal act, D, to be justified under the right of self-defense, it must be reasonably expected to provide defense from some future aggression. If D cannot be thought to provide such a defense but is instead justified by reference to some past aggression, then though it might be justified on other grounds, such as retribution, it cannot find solace under the right to self-defense. Hurka shows that what follows with regard to the proportionality condition is that the proportionality sought is not between the self-defensive act and some past act of aggression, but between the selfdefensive act and some potential aggression in the future.6

Similarly, I should add, with the success and the necessity conditions. Insofar as acts of warfare are conceived as responses to enemy attacks in the past, there is a temptation to judge them according to their success or necessity in appropriately responding to such attacks. This means that if the original enemy attack was not that severe, the counter-attack ought not to be that severe either, a view assumingly in line with the necessity condition. The temptation exists in ordinary wars too, but it especially applies to low-intensity ones, which look like a series of attacks, counterattacks, and counter-counter-attacks, each allegedly deriving its justification from the need to respond to the one preceding it. Once this temptation is resisted, as it should be, states might be allowed to respond to a perceived unjust attack against them with an attack which is much harsher and more destructive than the one they themselves suffered if they have a reasonable basis for believing that such a response is necessary to block future attacks.

One might object by saying that harsh responses to unjust attacks presume a kind of foreknowledge that we humans don’t have. Even retroactively we can only rarely determine whether such a response was necessary, but prospectively we almost never can. Hence, we must adjust the severity of our responses to that of the attacks against us, and refrain from striking back in a harsh manner just because such a strike is envisioned to prevent some attack, or series of attacks, in the future. However, such epistemic modesty regarding our capacity to assess what some enemy might do to us in case we don’t act in a certain way leads to a prohibition against a less harsh response too. After all, the justification for this less harsh

6 Hurka, “Proportionality in the Morality of War.” response must also be the prevention of future attacks, but to make such a claim, one would have to be reasonably sure (a) that the less harsh attack would deliver the goods, and (b) that there is no other way of doing so. Moreover, condition (a) seems actually harder to satisfy in the less harsh response than in the harsher one, for the simple reason that a less harsh response would usually have a weaker chance of either deterring the enemy or significantly undermining the enemy infrastructure. Let me explain why. Regarding infrastructure, it seems that a limited response would have a merely negligible effect in preventing future attacks. The point about deterrence is more complex. Deterrence functions differently in different contexts. Particularly relevant to the present discussion is the distinction between the measures that might deter potential wrongdoers who are motivated by sheer self-interest and those that might deter wrongdoers who are ideologically motivated. For the former, a moderately higher cost than benefit would usually be sufficient to deter them. For the latter, a much higher cost would be required. The role of deterrence in international relations tends to be closer to its role in ideologically motivated wrongdoing. If the response which group A gets for its attack on group B is a moderate one, the response will hurt, but usually will not convince members of group A to refrain from further attacks in the future. It might even increase their motivation to do so by adding an element of revenge to their considerations. To deter them from further attacks, the response of group B must often be so painful that further attacks on the part of group A would be regarded as out of the question.7 All the more so with strongly ideological organizations like Hamas or Al-Qaeda who are ready to sacrifice a lot in pursuit of their goals and who glorify martyrdom, extolling the virtues of those who die (or kill themselves) in the war against the enemy

(shaheeds).8

7 How exactly deterrence works on the international level is a matter of dispute. See, for instance, Jeffrey Berejikian, “A Cognitive Theory of Deterrence,” Journal of Peace Research 39 (2002), 165–183. In Berejikian’s own view, the most effective stance is a “firm-but-fair” one because excessive bullying diminishes deterrence. That sounds logical, but note that a “firm” response would often be “disproportionate” in the regular use of the term.

8 With regard to terrorists there is a further reason why only especially harsh measures might achieve the required defense, namely, that terrorists are “unwilling to enter into trusting relationships with their enemies” (Matthew Noah Smith, “Terrorism, Shared Rules and Trust,” Journal of Political Philosophy 18 [2008] 216). Their contempt for the shared rules of war makes them “utterly untrustworthy” (ibid.), which means that the terrorists cannot be trusted to “learn the lesson” in the way other people, who are willing “to share a society with other human beings” (ibid. 215), would learn it. Hence, the lesson must be forced upon the terrorists—drilled forcefully into their minds, so to say—for them to be deterred from future acts of violence. This would involve sending an unequivocally clear message, one which can only be conveyed through the use of harsh measures.

However, some forms of deterrence are nevertheless available even visà-vis organizations like these. A well-known statement by the Hizbullah leader, Hassan Nasrallah, can serve as an example. In August 2006, shortly after the war with Israel had ended, Nasrallah announced that “he would not have ordered the capture of two Israeli soldiers had he known it would lead to such a response.”9 It is hard to determine how sincere Nasrallah was in this declaration, and it is premature to predict whether, to what extent, and how long he will indeed be more restrained in the future.10 (People tend to forget or to underestimate the losses and pain they suffered from their adversaries in the past and also to overestimate their ability to defeat them in the present without suffering similar losses.) It does, however, seem reasonable to assume that a less harsh response on the part of Israel would have failed to bring about such restraint.11

The reader should not get the impression that, in my view, as individuals and as groups, we must respond with full force to any unjust attack against us. Often a moderate “reasonable” response would suffice by virtue of the clear message it sends to the attacker, “Hey there, don’t mess with me.” Signaling to our rivals and enemies that we are “serious about it” and that we won’t be trodden upon is sometimes enough to break the cycle of violence. But, of course, deterrence of this kind can only work provided that when my enemy does strike again, I respond in a much harsher way; otherwise I’ll make a fool of myself and make myself easy prey after having proven that I can only bark, not bite.12 To this one should add that countries and organizations fighting against liberal democracies often perceive them as spoiled, weak, and hedonistic13 and hence short of the mental determination and fervor required to get involved in a real war. Against this background, restrained and (typically) low-risk attacks from the air or the sea might not only fail to deter such organizations, but in fact bring about the opposite effect by confirming the above perception.

9 Jerusalem Post August 28, 2006.

10 At least in the short run, deterrence seems to have worked in the case of Hizbullah. I refer to its refraining from firing rockets against Israel during the Israeli campaign in Gaza in January 2009.

11 That odd attacks are inefficient against international terror has been pointed out by Michael Bonafede, “Here, There, And Everywhere: Assessing the Proportionality Doctrine and the US Uses of Force in Response to Terrorism After the September 11 Attacks,” Cornell Law Review 88 (2002), 212.

12 For a similar approach, see Bell, “Optimal Proportionality in the Law of War,” Part II, who argues that the common understanding of the proportionality condition makes it counterproductive. To achieve deterrence, states must convey a clear message “that sufficiently high penalties will be imposed on aggressors” (p. 23, italics added).

13 See Ian Buruma and Avishai Margalit, Occidentalism (New York: Penguin, 2004).

All this seems to imply that the necessity condition imposes weaker constraints on self-defense than commonly thought. However, as indicated above, the application of the necessity condition to warfare also seems to lead to stronger restrictions. The reason for this is the following. If killing human beings in war is justified in terms of self-defense, then all the standard conditions for self-defense must serve as constraints on waging war too. This is most obvious for those who subscribe to what I shall call ‘The Individualistic View of Justified Killing in War [IJW]’, according to which to fight justly is to be able to explain, as it were, to each enemy individual who is about to be killed in war, how his or her fate is justified according to the standard conditions of self-defense. This is so because, as McMahan puts it,

[T]he morality of defense in war is continuous with the morality of individual self-defense. Indeed, justified warfare just is the collective of individual rights of selfand other-defense in a coordinated manner against a common threat.14

On IJW, the same moral limits on killing that apply to the coordinated defense of three people against one apply also to three against three, or to 10,000 vs. 10,000. If the killing of Bob cannot be grounded in the right to selfdefense when Bob alone poses an illegitimate threat, it cannot be so grounded just because he poses such a threat together with 10,000 other people, i.e. an entire army. Thus, as far as “the deep morality of war” is concerned, to borrow McMahan’s expression,15 the necessity condition must be fulfilled in each case of killing in war. The same holds true for the success condition because, as we saw above, the success condition precedes the necessity condition in order of justification, and must be satisfied before moving on to the necessity condition. An act of killing in war that makes no contribution to blocking the threats imposed by the enemy (to the collective as a whole or to individual members of it) cannot find solace in the right to self-defense.16

The problem is that if the necessity condition and the success condition apply to jus in bello, we seem to be on the safe road to pacifism17 because

14 McMahan, “The Ethics of Killing in War,” 717, who also refers (in fn. 40) to his “War as Self-Defense,” Ethics & International Affairs 18 (2004): 75–80.

15 McMahan, “The Ethics of Killing in War,” 730.

16 For the purpose of the present argument, I bracket the paradox raised by the success condition on which I elaborate in “The Success Condition for Legitimate Self-Defense” (supra note 4).

17 I use ‘pacifism’ in a weak sense, by which I mean a view that perceives most wars in the actual world as morally indefensible, not in the strong sense of an in-principle clearly much of the killing taking place in war is either non-effective or unnecessary for victory. This means that wars inevitably involve impermissible killing and hence cannot be morally justified.

To further clarify this point, consider a domestic analogy. Suppose I refuse to pay protection money to the local mafia. The mafia members threaten to burn down my business, a common tactic against stubborn non-payers. Everybody knows that these guys are serious. Suppose further that the local police are at best impotent and at worst controlled by the mafia, so that I can expect no help there. In these circumstances, I have a right to protect myself and my property from mafia aggression, even if this involves the use of lethal measures. However, I don’t have a right to go around and kill any member of the mafia in the country (or even in the county) just because I want to defend myself from the above threat. I must make sure that such killing can reasonably be thought to be both effective and necessary in providing the required defense; otherwise it could not be justified under the right of self-defense. Similarly in war, the fact that an unjust enemy threatens me does not in itself justify the killing of each and all members of the enemy side. I must first make sure that each act of killing is both effective and necessary.

To this I should add that the constraint posed by the necessity condition is stronger the less culpable the aggressor is for the threat he poses. It is bad enough to kill somebody unnecessarily when the person killed is fully culpable for an unjust attack. It is much worse to kill him unnecessarily if he is less than fully culpable – which is precisely the case with most soldiers, especially conscripts, who are at most only partially culpable for the threat they pose. In these circumstances, asking the fighting parties to make sure that each act of warfare is both effective and necessary makes a lot of sense – yet, in practice, it is a request that is almost impossible to fulfill.

How could friends of IJW respond to this challenge?

The first response would be to emphasize the subjective nature of the conditions under discussion. In introducing the problem under discussion, I said that the problem for IJW is that much of the killing taking place in war is either non-effective or unnecessary for victory. However, as I myself noted earlier,18 these constraints on self-defense should be understood subjectively. This means that what matters is not the ex post objection to all wars (typically based on an assumed unconditional prohibition to kill human beings).

18 Supra note 5. estimate that much of the killing (and wounding and harm) in wars is non-effective or unnecessary, but the ex ante estimate about the effectiveness and necessity of individual acts of warfare. And once this shift to the subjective point of view is made, it becomes clear – or so this response suggests – that, in most cases, the killing, wounding and harm involved in war do meet the success and the necessity conditions. In most cases, when soldiers open fire on their enemies, attack their bases, and bomb their posts, they act under the reasonable belief that doing so will make a contribution to the defeat of the enemy and that this contribution is necessary.

Let me strengthen this response by connecting it to Kim Fezan’s account of self-defense.19 According to Ferzan, the right to kill in self-defense requires the existence of an objective “triggering condition,” namely that A culpably intends to kill V, as well as a subjective condition to the effect that V predicts that harm will occur if she doesn’t act. If the triggering condition is not met, V can, at most, have an excuse for carrying out D, but no justification. If it is met, then “any likelihood of harm should be sufficient for self-defense.”20 Though Ferzan’s emphasis is on the likelihood of harm, the same should apply to the likelihood of the effectiveness and of the necessity of D in blocking it. According to Ferzan, if A culpably intends to harm V, then even if the means he uses are only 1% likely to succeed, V has a right to carry out D against him in order to prevent the threat from materializing. The same seems to hold if the probability of the threat materializing is very high (close to 100%) but the probability of D blocking it is only 1%. Ferzan says that if V knows that defensive force is not necessary, she may not kill A,21 which seems to imply that so long as she doesn’t know this, and furthermore holds the contrary belief, then even if the belief (that D is necessary) is a weak one, she still has a right to kill A.

This, however, seems to make the necessity condition mute as a constraint on self-defense. If any likelihood of success and necessity is enough, then there would hardly be any self-defensive acts that would be ruled out by the conditions under discussion. This would mean that almost all killing in war would meet these conditions – but so would almost all killing of mafia members in the above example, and, more generally, the killing of almost anybody connected to V’s attackers in a way that gives her some

19 Kimberly Kessler Ferzan, “Justifying Self-Defense,” Law and Philosophy 24 (2005): 711–749.

20 Ibid., 744, italics added. 21 Ibid., 748. reason to think that the planned aggression would be blocked as a result. In other words, the response just outlined does not really show how IJW can accommodate the necessity condition, but rather how in practice this condition poses no real constraint on self-defense. This seems too big a price to pay to salvage IJW. This high price is especially troubling given the less-than-full culpability of most soldiers.

Note that although the ex post judgment cannot by itself undermine the moral status of the self-defensive act, it does shift the burden of proof onto the shoulders of soldiers in the ex ante situation in the following sense: given the experience in past wars, soldiers have a good reason to doubt whether their attacks will be both effective and necessary. Hence they may not carry them out before first establishing that, notwithstanding this lesson from past wars, some particular attacks can nonetheless be reasonably thought to succeed in defeating the enemy and to do so in the least harmful manner. It is pretty hard to establish that some particular act of warfare – a sniper killing some individual enemy soldier, an aircraft bombing some distant military base – can succeed in making a significant contribution to the defeat of the enemy. It is even harder to establish that there is no less harmful way of doing so.

Finally, if the success and the necessity conditions can be satisfied by “any likelihood” (once the triggering condition has been satisfied), this should apply to jus ad bellum too, with the implication that waging war might be justified even if the probabilities of success and of necessity were very low indeed. The two conditions (“last resort” and “reasonable hope of success”) would then be satisfied in almost all wars and hence, in practice, stop serving as constraints. One might be tempted to say that the moral status of states is different from that of individuals; therefore the conditions for legitimate self-defense function differently when the defense of states is concerned. But this route is unavailable to friends of IJW because for them, “warfare just is the collective of individual rights of selfand other-defense in a coordinated manner against a common threat;”22 hence, if the above interpretation of these conditions is accepted for individual self-defense, it must apply to wars too. Thus, the price of avoiding an ethic of war which is too demanding at the jus in bello level is to end up with an ethic of war which is too permissive at the jus ad bellum level.

To further clarify this point, let’s make the same argument the other way round, from jus ad bellum to jus in bello. According to the accepted

22 See text near note 14 above. understanding of the conditions for jus ad bellum, launching war might be justified only after making sure that the war has a reasonable likelihood of success and only after all alternatives to war – diplomatic measures, economic measures, etc. – have been exhausted. Applying this reading of the success and necessity conditions to jus in bello implies that each individual act of warfare might be justified only after making sure that it has a reasonable likelihood of success and only after all alternatives to it have been exhausted. Yet conducting warfare within such constraints would be virtually impossible.

In response, one might suggest that the necessity condition should be interpreted differently in jus ad bellum and in jus in bello. But this move is ad hoc and – once again – runs against the fundamental idea of IJW which assumes that the same conditions guide all forms of self-defense, individual and national alike.

Let me examine two other suggestions to avoid the all too restrictive implications of the necessity condition for warfare. The first is to rely on the doctrine of double effect [‘Double Effect’] and permit the unnecessary killing of enemy combatants under the title of collateral damage. This suggestion concedes that many enemy soldiers are killed unnecessarily, but says that since many are killed necessarily and since the killing of the former is not intended, it is covered by Double Effect. But I doubt whether Double Effect can do the required justificatory work. The enemy soldiers whose killing is unnecessary for victory are not killed incidentally, as a side-effect, but are directly targeted and killed – by bullets, bombs, and rockets – together with those soldiers whose deaths are necessary. One cannot seriously claim that in bombing the enemy headquarters, one intends to kill only those soldiers the killing of whom meets the requirements of the success condition and the necessity condition, while all the other soldiers are killed incidentally.23 This seems more like a reduction of Double Effect than like a helpful reliance on it.

The second suggestion is that the necessity condition does not require that each individual act of killing in war be necessary, only that the total

23 For a similar reason, Double Effect cannot solve the problem of how it might be justified to kill enemy soldiers given that many (probably most) of them are not morally culpable for the war in which they are participating. One cannot seriously say that the nonculpable enemy soldiers may be killed as a side-effect of killing the culpable ones. See recently Seth Lazar, “The Responsibility Dilemma for Killing in War: A Review Essay,” Philosophy & Public Affairs 38 (2010), pp. 210–211 who argues that since modern warfare has no way to sort out the enemy soldiers that are liable to attack from those who are not, in practice “killing in war… will involve widespread and serious rights violations” (211). number of enemy soldiers who are killed be necessary. From the point of view of the right to self-defense, given that all enemy soldiers are legitimate targets, it makes no moral difference which soldier exactly is killed. Therefore, if the attacked side needs to kill no less than, say, 10% of the enemy force in order to defend itself, it is allowed to do so without having to establish that each individual act of killing is necessary. If, to defeat the unjust side, the just side has to kill approximately one thousand soldiers out of soldiers S1, S2 … Sn, then it satisfies the necessity condition if it kills any of (S1 v S2 … v Sn), provided that the number does not exceed one thousand. (Nor should the number go too much below one thousand; otherwise it would fail to meet the success condition.) And while establishing the necessity for each individual killing is a mission impossible, establishing the necessity for some number of killings which is necessary for victory is not.

This is a tempting solution to our puzzle, but I doubt if it will work. It relies on two dubious premises: (a) That we have a fair idea about the number of enemy soldiers who must be killed in all wars (or in some specific war), and (b) that, in terms of defeating the enemy, it makes no difference which enemy soldiers are targeted (hence the above disjunction). But there is no way of making serious and responsible estimates in advance regarding the number (or the percentage) of enemy soldiers the killing of whom would be necessary for victory, and there is no basis for thinking that killing one soldier (a senior officer, for example) is just as effective as killing another (a military truck driver) in order to defeat the enemy.

If none of these solutions is convincing, we are left with the puzzle with which we started, namely, that the necessity condition can’t be satisfied at the level of jus in bello, which implies that wars can’t be fought justly. And if wars can’t be fought justly, they must be morally prohibited, which means that pacifism is the only decent response to wars.24

That a consistent application of IJW leads to pacifism or to a position very close to it is already not a new idea. Benbaji makes the same point in his critique of McMahan, the central figure at the IJW camp,25 and I try to

24 This conclusion does not preclude the possibility that states, or individual soldiers fighting on their behalf, might be excused for participating in war (and violating the necessity condition), though one would have to show what the excuse is based on. But just war theory as well as commonsense morality assumes that some wars are not only excusable but just.

25 See Yitzhak Benbaji, “The Responsibility of Soldiers and the Ethics of Killing in War,” Philosophical Quarterly 57 (2007): 558–73 in which he shows that McMahan’s justice-based conception of self-defense yields a nearly pacifistic ethics of killing in war. show that Rodin, another figure in that camp, is much more committed to pacifism than he himself realizes.26 The argument of this section further substantiates this connection between IJW and pacifism by suggesting a certain understanding of the necessity condition and by showing the implications of applying it to jus in bello.

Some readers might be more convinced than I am by one of the above attempts to show that IJW is compatible with the necessity condition for justified self-defense. To such readers, I plead that they at least be open to the possibility that some other account of just war theory can do a better job than IJW in explaining the nature and justification of the limits posed by it. In the next section I point to such an account.

II. The Necessity Condition and Non-Individualist Justifications for War

In Section II, I argued that since the individualistic view of justified killing in war [IJW] requires the consistent application of the standard conditions for individual self-defense to the domain of war, and since a consistent application of the necessity condition in this domain renders much of the killing in war impermissible, IJW leads to the conclusion that much of the killing in war is not permissible. And this means that IJW must oppose most wars, or, in other words, it must accept pacifism.

Two routes to dodge this result suggest themselves. The first, still within the framework of self-defense, replaces the individualistic view of justified killing in war (IJW) with a collectivist one (The Collectivist View of Justified Killing in War [CJW]). The second relies on a different normative framework altogether, viewing the war convention as a tacitly accepted social norm. It is beyond the scope of the present paper to examine the details of these alternatives to IJW. I will limit myself to examining how they account for the necessity condition and whether they do a better job than IJW in blocking the route from this condition to pacifism.

A. The Necessity Condition and CJW

According to CJW, part of the justification for killing people in war lies in their very membership of the aggressive collective. A war is a conflict

26 Daniel Statman, “Moral Tragedies, Supreme Emergencies and National-Defense,”

Journal of Applied Philosophy 23 (2006): 311–322. between collectives which are metaphysically distinct from individuals,27 and hence warrant a morality of their own – a collective morality – which is different from and irreducible to the morality that governs the relations between individuals. Within such a collectivist morality, soldiers lose their individual faces and become, as far as the war is concerned, just part of the collective.

As indicated at the end of the previous section, CJW, like IJW, functions within the framework of self-defense, which means that the standard conditions for legitimate self-defense apply to it too. Why, then, might one think that the necessity condition poses a weaker constraint on conducting war according to CJW than it does according to IJW? Here is one possible answer. On CJW, when two collectives are engaged in war against one another, the individuals fighting in the war lose their identity qua individuals and are completely swallowed up by their respective collectives. It is as if when one collective fights against another, all it sees in front of it is the enemy collective, not individual members of it. This seems to imply that the constraints on self-defense apply only at the level of collectives, which in turn seems to imply that the necessity condition applies only to the measures taken against the enemy collective, not to the measures taken against individual soldiers. And this – finally – seems to be the same as saying that the necessity condition applies only at the level of jus ad bellum, not at the level of jus in bello. What the necessity condition requires is that the war against the enemy collective be necessary, not that each attack against any enemy soldier or enemy post be necessary. What the necessity condition requires is that the war be a last resort (as required by the standard formulation of just war theory), not that individual acts of warfare be so.

To further substantiate this line of defense, consider the following analogy. When an individual is attacked by a culpably unjust aggressor, she is allowed to use force in order to defend herself from the attack. But this surely doesn’t mean that she needs to show that each individual hit or kick is necessary to block the attack. In a similar way, one might argue that when a collective – which, under CJW, is a kind of agent irreducible to the individuals it comprises – is fighting to defend itself, it does not need to justify each act of warfare it engages in, or each case of killing or wounding

27 See Fletcher and Ohlin, Defending Humanity, 178 (“When the collective acts, the whole is greater than the sum of its parts. The nation and its army enjoy an existence of their own”). This formulation stands in stark contrast to McMahan’s view cited above near note 14. an enemy soldier. It needs to care about necessity only at the level of jus ad bellum, not at the level of jus in bello.

However, this reading of CJW takes the analogy between individuals and collective entities too far. It is one thing to say that wars have an irreducible collective dimension. It is a different thing to say that the collective dimension is all that exists, that the individual faces are completely lost in them. Zohar’s suggestion that the morality of war is a dual one, comprising both collectivist and individualist dimensions, sounds much more plausible.28 But within such a morality, there is no reason why the necessity condition should apply only to defensive acts against the enemy collective, at the level of jus ad bellum, and not to defensive acts against enemy individuals, at the level of jus in bello.

A different reading of CJW would see it as a theory about the conditions for being liable to self-defensive attack, not as a theory about the constraints on mounting such attacks. While on IJW, a necessary condition for such liability is posing a threat to a potential victim, namely, being an aggressor, on CJW, membership in the aggressive collective is sufficient. But on this reading, CJW does not affect the standard constraints on selfdefense, namely, the success condition, the necessity condition, and the proportionality condition. It explains how individuals might be targeted even though they are not aggressors (nor threats) in the usual sense of the word. It fails to explain how they might be targeted even if doing so is not necessary to provide the desired defense from the threat posed by the enemy collective.

B. The Necessity Condition and CONTRACT

The challenges posed by IJW to traditional just war theory, especially those developed by McMahan, have led some thinkers to suggest a different ground for this theory, according to which concepts other than selfdefense do the main work in explaining the moral status of killing in war. Central to this endeavor is the reciprocal29 or the contractual understanding of the morality of war. Let’s call this ‘The Contractual View of the War Convention’ or ‘CONTRACT.’30 Benbaji explains it as follows: states have a

28 See Noam Zohar, “Collective War and Individualistic Ethics,” Political Theory 21 (1993): 606–622.

29 See Fletcher and Ohlin, Defending Humanity, chapter 8, who argue that “the principle of reciprocity governs the entire law of armed conflict” (180).

30 In “Proportionality in the Morality of War,” Hurka seems to hold a view close to, if not identical with IJW. He says that “the structure of just war theory closely parallels that of the strong interest in being able to defend themselves from unjust attacks against them; since they cannot be sure that they will always be the stronger party in such conflicts, they have an interest in limited, rule-governed wars over unlimited, “total” ones; since deciding on such rules would be impossible once a war breaks out, they have an interest in agreeing on such rules ex ante. Such an agreement would be guided by a desire to minimize the harm of war, on the one hand, without making it impractical to win it, on the other. In Benbaji’s view, then:

[T]he objective of the contracting parties is minimizing the harm inflicted on morally innocent people within wars, without limiting the rights states have to use force in protecting their just claims.31

How would the necessity condition figure in this view? The objective of minimizing the harm inflicted by war while still being able to fight seems to be precisely what the necessity condition requires – bringing about no more harm than necessary. But following the argument in Section II, if the necessity condition required that each individual act of warfare be shown to be necessary, that would make war practically impossible. Hence, this understanding of the necessity condition is inconsistent with CONTRACT, according to which the conventions of war must be such that they do not preclude effective self-defense.

This is all the more so with regard to the proportionality condition, insofar as it is interpreted, like in theories of self-defense, as a constraint on the measures that might be taken against an attacker (rather than against third parties or innocent bystanders). Thus understood, the proportionality condition is even more demanding than the necessity condition; hence if CONTRACT finds the necessity condition too demanding morality of self-defense” (38) and later says in terms reminiscent of McMahan (see citation near note 14 above) that in a legitimate military action “a large group of individuals act collectively, through their political institutions to protect the rights of another large group of individuals, who may be themselves.” In “Proportionality and Necessity,” 136, however, he seems to shift to a contractual view of war, at least insofar as the moral status of soldiers is concerned. He uses the analogy of boxing to illustrate how people can permit each other to do to them what would be forbidden outside of the ring. See also his Thomas Hurka, “Liability and Just Cause,” Ethics and International Affairs 21 (2007), 210. The boxing analogy is popular among contractualists, see e.g. Benbaji, “A Defense of the Traditional War Convention,” 487. It is rejected by IJW supporters, see Jeff McMahan, “On the Moral Equality of Combatants,” Journal of Political Philosophy 14 (2006), 381.

31 Yitzhak Benbaji, “The War Convention and the Moral Division of Labor,” Philosophical Quarterly 59 (2009): 593–618. See also idem., “A Defense of the Traditional War Convention,” Ethics 118 (2008), Section III.

(again, in jus in bello), then surely it would find the proportionality condition too demanding as well. A war convention that included the proportionality condition would impose too strong a constraint on the ability of states to use force in protecting their just claims.

If the loosening of these two conditions, necessity and proportionality, on the conduct of warfare sounds too permissive, they are no more permissive than other features of CONTRACT, mainly the permission they grant to unjust combatants (=combatants of the unjust side) to kill just combatants (=combatants of the just side) who, by definition, are morally innocent in the relevant sense. If, under CONTRACT, it is permissible to kill combatants in war even if they pose no illegitimate threat, then it is not unreasonable to assume that it might be permissible to kill combatants even if such killing is not necessary (in the sense prevalent in moral and legal theories of individual self-defense) to achieve the goals of the war.

The difference between IJW and CONTRACT with respect to the necessity condition has to do with the different role of self-defense in each of these theories. In IJW, the ultimate justification for all killing in war lies in the right to self-defense, which means that each act of killing in war must ultimately meet all the conditions for legitimate self-defense. By contrast, in CONTRACT, the ultimate justification lies in the assumed contract between the fighting parties, self-defense coming into play only as the rationale that underlies the parties’ decision to undertake the contract.

To sum up, according to The Individualist View of Justified Killing in War, the same constraints that limit the use of otherwise immoral measures in individual self-defense apply to collective self-defense too, which means that the necessity condition (as well as the other conditions for justified self-defense) must be met in all cases of killing in war. I expressed doubt as to whether this condition could be met, and I argued that, if it cannot, we are on a road – a dangerous one, in my view – to pacifism. I then suggested that the contractual view of killing in war offers a better way of dealing with this issue, a way that, on the one hand, grants the necessity condition a role in moral reasoning about war, while, on the other, does not lead to pacifism. At least in this respect, CONTRACT has an advantage over other candidates in explaining how wars might be fought justly.

Part Three: Innocent Non-combatants and Self-defense

Self-defence Among Innocent People

Gerhard Øverland*

In this paper I propose a principle by which we can decide which party is justified in using defensive force when unfortunate circumstances have brought innocent parties in a deadlock in which it appears that if one innocent party does not kill the other the latter will kill the former. Typical situations may be when what appears like an aggressor is not an aggressor at all, but someone doing something making him or her look like one (the case of the apparent aggressor); or when circumstances outside a person’s control has made his or her into a threat for another person (the case of an innocent threat); or when a person mistakenly has good reasons to believe that there is a justifying cause for aggression and when a person lacks the required mental capacity to evaluate his or her actions in order to be held morally responsible for them (cases of an innocent aggressor).

While we may believe the grounds for acting of the typical innocent aggressor to be reasonable, it might also be the case that the innocent aggressor lacks the capacity to evaluate his or her grounds for acting. In neither case could the agent be held morally responsible for his or her actions. An innocent threat is a person who is causally involved in a threat of harm to another person but not through his or her own agency. An innocent threat is not doing anything; it is merely something that happens to the person that makes him or her into a threat. Apparent aggressors are people who look like aggressors but who in fact are not engaged in aggressive action. Hence a person using defensive force against an apparent aggressor may be an example of an innocent aggressor.

Although it is widely accepted that use of defensive force is justified in cases when the aggressor is a culpable person with an intention and ability to kill, what its minimal requirements should be is contested.

* Much inspiration to write this paper stems from reading chapter 6 of Arthur Ripstein’s book equality, responsibility, and the law (Cambridge: Cambridge university press, 1999), and I am grateful for his encouraging words. An earlier version of this paper was presented at the 21st IVR (Internationale Vereinigung Für Rechtsund Sozialphilosophie) world congress in Lund, Sweden, and later at a seminar held by the ethics programme, university of oslo. I am indebted to Antony Duff for helpful comments on an earlier version of this paper. I would also like to thank Kai Dramer and John Richard Sageng for stimulating discussions, and Jakob Elster for written comments.

My proposal is that in order for one innocent person to be justified in using defensive force against another innocent person, this other person must have initiated the apparent threatening behaviour, but the defendant’s interpretation of that behaviour as being threatening must be reasonable. This view implies that although reasonable beliefs do not always justify, a reasonable interpretation of an initial behaviour by another party as being threatening does.

Preliminaries

It is common to differentiate between two main usages of the term ‘justification’.1 In one version, ‘justification’ is used to designate an objective and impartial perspective, which implies that a person is justified when he or she acts in accordance with what it would have been permissible to do. The circumstances that make an agent’s act permissible are the same circumstances that make it a candidate for being justified. However, in order to be justified the agent must know about those circumstantial factors.2 This observation implies that all permissible acts need not be justified simply because the agent may not know about the justifying circumstances,3 but whenever the agent does, and only then, will he or she be justified in doing so and so. One attraction of such a true belief view of justification is that saying that a conduct is justified has implications for what we take to follow from it, namely that the conduct may be supported and not prevented.4

In a second usage, what determines whether or not an action is justified is evaluated from the perspective of the agent. According to this view a person might be justified in acting in a particular manner if he or she acts on his or her beliefs about a situation and those beliefs are reasonable.

1 See for example, Suzanne Uniacke, Permissible Killing (Cambridge: Cambridge University Press, 1994), pp. 9–26.

2 See John Gardner, ‘Justifications and Reasons’, in A.P. Simester and A.T.H. Smith (eds.),

Harm and Culpability (Oxford: Oxford University Press, 1996), p. 105.

3 One could, of course, include all permissible action in those that are justified, and thereby deny the knowledge requirement. Such a pure objectivist idea of justification would be coextensive with what is permissible, but is not, I think, a plausible candidate for how we should use the term.

4 See Kent Greenawalt, ‘The Perplexing Borders of Justification’, in Michael Louis Corrado (ed.), Justification and Excuse in the Criminal Law (New York: Garland Publishing, 1994), p. 363. See also George Fletcher, Rethinking Criminal Law (New York: Oxford University Press, 2000), p. 762.

One attraction of such a reasonable belief view of justification is that it matches up well with instances in which we think it appropriate to exempt a person from moral blame. One of the difficulties of this view, however, is that it must deny the existence of an important relationship between saying that an action is justified and whether or not it may be prevented. Moreover, it may very well allow for two parties being justified in using defensive force against each other.

Confronted with a moral wrong, we can assess it with regard to two parameters. We might think it is permissible to prevent the wrong, and/or we might think that it is appropriate to blame the person behaving wrongly. To say that an agent is doing something morally wrong is either to give an assessment of the external circumstances the agent is bringing about or to assess the internal relations of the agent’s intentions, desires, and beliefs, or a combination of the two. The external has to do with the question of permissibility, while the internal concerns questions of blame. It is important to note that blameworthiness and permissibility should be kept apart, and that a moral wrong could concern either of the two, or both.

Granted this taxonomy, a justification is easily placed on either of the two sides of the morally wrong. The reasonable belief view of justification accentuates exemption from blame, while the true belief view of justification accentuates permissibility of prevention. I want to propose a third way of using the term ‘justification’ based on who might be prevented in situations of conflict. While I shall give up the requirement of true beliefs, I retain the view that if an agent is justified in doing so and so, it should have implications concerning who ultimately might be prevented from acting in such and such a way when party to a conflict. Yet the relation between an act being justified and the question of who might be prevented is more complicated than has normally been assumed. My proposal is that when there is a question of imposing equal harm on either of two parties, the one who is justified ought to be given priority and not be prevented. Accordingly, when a person is justified in using defensive force, it is impermissible to impose any amount of harm on the defendant in order to save the (apparent) aggressor or threat from an equal amount of harm.

It may pay to mention that all instances in which adherents of the true belief view on justification would say that an agent is justified in doing so and so, the agent would be justified according to my proposal as well. However, my proposal also covers some, though not all, instances in which adherents of the reasonable belief view would say that an agent is justified in doing so and so.

1. Apparent Aggressors

Let me begin by introducing an example in which it is quite reasonable that a person uses defensive force, but nevertheless permissible to prevent it. Imagine that one person attacks another person, firing a loaded gun. Unbeknown to either of them, the remaining bullets in the chamber are flawed, and will not fire. The aggressor in this example is a culpable apparent aggressor, which means that we are not speaking of self-defence among innocents. The aggressor has an unjustified intention to kill his victim, and takes what he thinks are the necessary steps towards doing so. By chance, those steps are inadequate.

Clearly, such an attack must pass the test of reasonable fear, even if chance events mean that the aggressor no longer poses any mortal danger.5 Many people will say that use of defensive force in this case should be deemed justified. The epistemic situation for the defendant is, after all, the same with regard to this apparent aggressor as it would have been with regard to a real one.

If we assume that a justified act may be supported but not prevented, we cannot grant the victim of the apparent aggressor a justification. Since the last bullets in the aggressor’s gun were flawed, it could be permissible to prevent the defendant from killing him, for example, by snapping the gun out of his hands. She is not under any real threat, and any use of defensive force from her side would be force used in vain. If we therefore want to say that a person may be justified in using defensive force against an apparent aggressor, it looks as if we must give up the idea that there is any interesting relation between an act being justified and whether or not it might be prevented. Given the fact—if it is a fact—that it would be permissible to prevent the person under attack from defending herself, it appears to be of little interest to say that use of defensive force would be justified; nothing seems to follow from it. If all we want is to exempt from blame, an appropriate moral excuse would do.

We might observe, however, that under no circumstances can the upshot be that it is the culpable aggressor, although only apparently representing a threat, who should be selected when there is a choice of saving the

5 The example is from Arthur Ripstein in Equality, Responsibility, and the Law, p. 193. aggressor or the victim. Although the victim might be prevented from killing the aggressor, there are limits to the means that may be used in doing so. To inflict considerable harm to the prospective victim in order to save the culpable apparent aggressor is not permissible.

Innocent Apparent Aggressors

There are situations in which people have good reasons to regard a person as an aggressor and to take appropriate steps against him or her, but where the person mistaken for an aggressor has no intention to harm. A person who merely looks like an aggressor is an innocent apparent aggressor, and should be distinguished from a culpable version who intends to harm but who, unbeknown to his or her victim, does not possess the necessary means to do so.

I shall now discuss the plausibility of saying that some victims of innocent apparent aggressors may be justified in using defensive force, and what the criterion for it could be. As I said, I will give up the requirement of true beliefs, but maintain the view that if an agent is justified in doing so and so, it should have implications for the question about who ultimately might be prevented when there is a conflict between two parties. I will therefore use this insight when evaluating the different proposals. Criterion 1: Reasonable beliefs on part of the defendant. It has been argued that the use of defensive force based on mistakes of fact, where the defendant has reasonable beliefs concerning those facts, is justified. And clearly, one reason for claiming that a reasonable mistake is justified is to differentiate it from a simple excuse based on incapacity on the part of the defendant.

When a mistake is reasonable, a person acts intentionally on reasons she takes herself to have, and we can see that those reasons really were reasonable in her situation. She has demonstrated due care, and acted upon the reasons she has been able to gather in a way we want people to do. There is nothing more we should expect from her. Therefore, we could be tempted to say that her act is justified.

My proposal is that if reasonable beliefs justify, and we have to choose between imposing equal harm on either of two parties in a heated situation, we would have to give preference to the one acting upon reasonable beliefs. In talking of an escalating situation, I imagine a person perceiving herself to be under attack and therefore setting out to defend herself. Subsequently, the apparent aggressor defends himself with available means against the defendant (assuming that he carries a gun with him, for example). In such an escalated scenario, we have two people fighting each other with lethal means.

At a minimum, we can now see that talk of what is reasonable is deficient. There might be situations in which a person reasonably believes herself to be under attack, but where it would be impermissible to favour the defendant if we had to choose between imposing equal harm on either of the two. Let us assume that Mary believes she must defend herself against Bill with lethal force. She believes so because a third party has told her untrue things about him. She has been told that Bill is planning to kill her, and that he is going to do so in such and such a way. But what she has been told is false; Bill is not intending to kill anyone. Let us furthermore assume that what Bill does would not in ordinary circumstances have been enough to make Mary believe that he was an aggressor. The reason why she believes so in this case is what she has been told about him, and that he (accidentally) acts in a way that fits well with that description. Nor has she cause to suspect that what she has been told is not true. The latter is what makes her beliefs reasonable, provided, of course, that Mary has demonstrated due care, and that there is nothing more we could expect from people in such a situation.6 I shall call this example ‘OriginalMary’.

In OriginalMary, Mary acts on the basis of reasonable beliefs when she begins to use defensive force against Bill who is acting in a manner consistent with the warnings given to Mary. Yet if it comes to a choice between letting Mary kill Bill, or letting Bill kill Mary, it seems wrong not to let Bill kill Mary. This creates a problem for the reasonable belief view of justification. It seems odd to claim that Mary is justified when we would judge it wrong to save her at the cost of the life of the one she is defending herself against. We should not, I believe, on the one hand, say that Mary is justified in using defensive force against Bill, while we, on the other hand, perceive that if it came to a choice between them we would have to assist Bill. This observation indicates that acting on reasonable beliefs is not enough to grant a person a justification, nor to guarantee that a person doing so will be favoured if there is a choice between the two parties.

If one holds the view that Mary as Bill equally merit being saved, one could maintain that both are justified in using defensive force: Mary would be justified in using defensive force against Bill’s apparent attack, and Bill would be justified in using force to defend himself against Mary. This proposal is not very illuminating because what more would we in this case say

6 It is not enough that we could not expect more from Mary in particular; she must live up to the requirement of a reasonable person. that neither of them should be blamed for using defensive force. It would therefore be better to say that neither is justified in using defensive force, and merely grant each an excuse. Yet, I should think that many people would regard it permissible to save Bill at the cost of Mary, in which case we had better say that he is justified in using defensive force, while she can only be excused.

Although it might be reasonable to regard a person as an aggressor if a third party has provided untrue information about him or her, you would not be the one deserving of leniency in a situation in which we have a choice between imposing equal harm on you or the innocent apparent aggressor in order to save one person. It is after all, you that triggered the escalating process.

Criterion 2: Blame on part of the apparent aggressor. An innocent apparent aggressor may or may not be to blame for appearing aggressive. It would depend on what he has done preceding the situation in which someone believes herself to be under attack from him. Perhaps, then, when deciding whether or not a person is justified in using defensive force, what is essential is whether or not the apparent aggressor has done something he should have known could cause other people to regard him as an aggressor?

This proposal means that we add something to the reasonable belief view of justification. In order to be justified one must act on reasonable beliefs and it must be the case that the person against whom defensive force is used is to be blamed for appearing like an aggressor. If the apparent aggressor ought to have known that his behaviour would reasonably be perceived as so threatening by other people that they would think use of defensive force were necessary, and he could have avoided such behaviour, he is to be blamed for creating that impression. Accordingly, I shall call such a person ‘a blameful apparent aggressor’. Be aware that he is not blameworthy because he is a culpable aggressor with an intention to harm: the blame is only concerned with why he appears like an aggressor. The proposal at this point is that if response to a blameful behaviour by the other party can be justified on the basis of reasonable belief, and we have to choose between harming either of the two parties equally in an escalated situation, we would have to choose not to harm the one responding to the blameful behaviour. Thus, a person may be justified in using defensive force only if the apparent aggressor is to be blamed for creating an aggressive impression; he must have done something he ought to have known was menacing. Put otherwise, he must at least be partly responsible for the situation, and more so than the other party. And if none of the parties are to be blamed for the unfortunate escalated situation, none of them will be justified in using defensive force either.

Now, it may seem correct to give the defendant moral precedence if the apparent aggressor is to be blamed for appearing like an aggressor. This becomes obvious if we imagine Bill is simply having fun with Mary. If Bill deliberately creates a situation in which he appears like an aggressor, it would be unfair to let someone else carry the cost of the situation created. Moreover, if Bill behaves recklessly or negligently he ought to carry the cost of doing so.

That said, this criterion and the one on reasonable beliefs on the part of the defendant share a significant problem. They cannot help us to resolve all instances of escalated force. The above example illustrates the problem. In OriginalMary, Mary believes she must kill Bill because a third party has told her untrue things about him, not because of something he has done. Bill is therefore not responsible for appearing to Mary to be an aggressor. But Mary, on her part, has every reason to believe what the third person has told her about Bill, and she is therefore not to be blamed for believing that she must defend herself against him. Thus, in OriginalMary it is not only the case that Mary acts on reasonable beliefs, but it is also the case that neither Bill nor Mary are to be blamed for the situation.7

Notwithstanding the fact that neither is to blame, it seems wrong not to choose to save Bill if the situation escalates and we have to choose between saving him or Mary. This indicates that Bill is justified in using defensive force against Mary, although Mary is not to blame for escalating the apparent conflict. We can therefore not say that the idea of being blamed for causing an impression of impending aggression is sufficient to narrow down the reasonable belief view of justification and thereby determine when we should say that a person is justified in using defensive force. Often, however, the person who is morally responsible for appearing like an aggressor must carry the cost if the situation escalates. But there is another criterion, which in these cases coexists as an explanation.

Criterion 3: Initiating behaviour. In the example of Bill and Mary, both may be acting upon reasonable beliefs and neither of them may be to blame for appearing like an aggressor. It seems nevertheless that there is a reason

7 Circumstances in which this may happen may readily occur in wartime. Soldiers may have been led to believe (falsely) that their attack is justified.

In OriginalMary, the only person to be blamed is the informer who told Mary untrue things about Bill. But even this person could have been telling untrue things about Bill in good faith. for giving Bill priority if there is a choice of inflicting equal harm on one of them to save the other. Why?

The simple answer, I believe, is that Mary initiates the escalating process. She starts to do something, which triggers Bill’s response. Because although Bill appears like an aggressor to Mary, Bill is not doing anything that would have been regarded as threatening if it were not for the particular information given to Mary from a third person.8

The idea to be investigated at this juncture is that we should always give preference to the one who does not initiate the escalating process. If it makes sense to say that one person’s action is a response to something the other person does, the first person, who begins the process, will have to be sacrificed if it comes to a question of saving one of them. The one justified in using defensive force is the one responding to an apparently threatening behaviour by the other party.

One problem, of course, is that a person’s response might be triggered by a variety of irrelevant factors. For example, if I should regard your tapping on the desk as an attempt to blow up the building, my reaction would not be deemed justified if I attacked you with all my force. And this is true although I respond to something you do, namely your tapping on the table.9

It seems as if a requirement of reasonableness must be called upon again. Any interpretation will not do. The defendant’s interpretation of the behaviour of the apparent aggressor as threatening has to be reasonable; if not, the response cannot be justified. The interpretation of the epistemic situation must be reasonable because we need to know that it was the apparent aggressor who initiated the escalating situation and not a particular belief on part of the defendant. Thus, whether or not Mary is justified in killing Bill, or Bill is justified in killing Mary, depends (1) on which of them initiated the escalating process, and (2) whether interpreting such initiating behaviour as threatening is reasonable.10

8 While one might be justified in blaming the informer for causing an escalation of the conflict, i.e., identify her as the initiating party, it would nevertheless be correct to say that, relative to Bill and the apparent aggressor (Mary), Mary was the initiating party. In situations under discussion, it is never the case that the defendant may survive by using force against third parties. If that were the case, it could make sense to discuss the permissibility of using defensive force against such third parties. I shall not do so here.

9 What if it turns out that your tapping on the desk would have blown up the building? In that case the interpretation of your tapping on the desk as threatening would be unreasonable, but true.

10 The concept of the reasonable is not always transparent. Let me first say that my use is not normative in the sense ‘the reasonable’ is used in political theory. However, it is not

At this point I propose the principle of initiating behaviour, which reads as follows:

In order for one person to be justified in using defensive force the other party must initiate the apparently threatening behaviour, but the defendant’s interpretation of that behaviour, as being threatening, would have to be reasonable.

The principle of initiating behaviour explains the asymmetry between innocent aggressors and their (innocent) victims, and the principle separates justified and unjustified defensive force when both parties are innocent.

The principle of initiating behaviour narrows down the reasonable belief view of justification further than the criterion of blame. On many occasions, however, they will be coextensive. By accepting the principle of initiating behaviour as pertinent to identify justified agents, we may maintain the view that if an agent is justified in doing so and so, it will have implications for determining who ultimately might be prevented when two parties are locked in conflict. The implication is that when there is a question of imposing equal harm on either of two parties, the one who is justified ought to be given priority.

2. On Initiating Behaviour

The person to be sacrificed does not need to do something he or she ought to have known involved a risk.11 The only thing that is required is that it makes sense to say that what initially triggered the escalation can be traced to one of the two parties. Whenever that is possible, it determines which party should be deemed justified in using defensive force among innocent people. purely descriptive either. An interpretation of some conduct as threatening will depend upon the norms of the particular social culture, and the real level of risk one is exposed to in that culture. As has been indicated by colleagues here in Oslo, what may reasonably be seen as threatening on the streets of New York may not always reasonably be interpreted as threatening on the streets of Oslo. It is beyond the scope of this paper, however, to attempt to make the epistemic concept of the reasonable transparent.

11 Thus, I disagree with Ripstein, who says that in doing something that might reasonably be perceived as an attack, ‘I take a risk that I will be killed or wounded by the reasonable defensive measures of another’ (Ripstein, Equality, Responsibility, and the Law, p. 195).

Tracking the Initiating Behaviour

I said that interpreting an epistemic situation as threatening would have to be reasonable because we need to know whether it was the apparent aggressor who initiated the escalating situation and not particular beliefs on the part of the defendant. However, retracing such initiating behaviour is not always a straightforward process.

To begin with we could say that the requirement of a reasonable interpretation implies that most people who knew nothing about the apparent aggressor would regard him as an aggressor by simply observing the movements of his body and hearing what he says. Yet this will not always do because we may know something about the person that we would like to take into account when interpreting his behaviour. We therefore need to know whether taking such knowledge into account helps us trace the fount of the initiating behaviour.

Should, for example, things done in the past count as initiating behaviour? Imagine a person with a record of a contract killer, but who recently converted to a more peaceable occupation. You meet him, and he acts in such a way to warrant a reasonable expectation on your part of danger, especially as you believe he still pursued a career as a contract killer. You know about his record, but not his recent transformation, and reasonably believe that he is on a contract to kill you. Under such circumstances, if the conflict should escalate, and we have to choose between saving you or the assumed contract killer, it would be wrong according to the principle of initiating behaviour to save the latter. We would have to act to save you because your beliefs about your supposed opponent rest on his former proven conduct and can therefore be traced back to him. The important thing is not the exact timing, but whether an impression of danger flows from the person due to his past or present behaviour.

What about initiating behaviour that will make someone believe something about an initiator that is not true? Imagine that a person A has behaved in the past in such a manner as to persuade others that A’s present behaviour is dangerous in some way or another. In this case, too, priority should be given to the one who responds to A’s present behaviour, provided that interpreting A’s behaviour as threatening is reasonable in light of his (false) record of aggressive behaviour. Mary may, for example, be justified in using defensive force against Bill if it were reasonable for her to interpret Bill’s behaviour as threatening in light of (false) information she possesses on his putative criminal record, provided that something about Bill’s past behaviour made it reasonable to interpret him as having such a record. The difference between this version of the example and OriginalMary is that the reason for the mistake flows from Bill and not from a third party who tells Mary untrue things about Bill. It is Bill himself whose past behaviour warrants an interpretation of his present behaviour as threatening. In general, we can say that when tracing initiating behaviour it is important that the information, in light of which an interpretation of a particular behaviour as threatening can be said to be reasonable or not, in some way or another flows from the agent in question.

Particularly hard cases seem to come about concerning mistakes of identity. In such cases we might hesitate before determining the initiator. One problem is that it would be difficult to decide when it is common knowledge that looking a particular way may be dangerous. This has to do with the criterion of reasonableness. Imagine, for example, a person who grows a beard, which makes him resemble a well-known dangerous person.12 Given the fact that he now resembles such a person, it might be reasonable to interpret his behaviour as threatening. But should the growing of a beard count as the initially threatening behaviour? To me it seems as difficult to say that the beard grower is the initiator, as it is to determine who should be saved at the cost of the other. It is not clear that what makes this person appear like an aggressor actually flows from him, or whether changes in the world has made him appear like an aggressor. In certain circumstances it might prove hard to determine the initiating party, and in those cases it should come as no surprise that we do not know how to settle the question.

The principle of initiating behaviour can allow for border cases like the beard case, but will not sit well with examples in which it is clear that the party who initiates the dangerous situation is, at the same time, not the person who has to carry the costs if the situation deteriorated. The latter would be a counterexample to the principle. No such example has yet been identified. Border cases are different; they are cases the theory is not apt to settle. Yet, since we have no clear intuition about these cases, they do not indicate a failure with the theory.

Questioning the Action Requirement

So far, I have taken for granted that in order to be justified in using defensive force, the defendant must respond to something the other party does,

12 Douglas N. Husak presented this tricky example to me in conversation. or has done in the past. In spite of this, it is not clear that in order to be justified in defending oneself against an innocent person, the threat has to be constituted by a person doing a particular action.

I have already given up the requirement that a person ought to have known better when he or she acted as s/he did. In OriginalMary it was not the case that Mary was to blame for using defensive force against Bill, yet she would be the one who had to carry the costs if the situation escalated. And having accepted that the person who eventually has to bear the costs does not need to be morally responsible for causing the conflict to escalate, why should we insist that the initiating behaviour must be an action at all? If we accept that a person might be held responsible (in the sense that he or she eventually will have to carry the costs) for things he or she is not to be blamed for doing, and perhaps may not even know that he or she is doing, then there should be no reason to resist the assumption that a person might be held responsible in the same way for things he or she does not do either, as long as we can say that it is what happens with him or her that initiated the apparently dangerous situation. There seems to be no reason to require agency if we have given up the requirement of intentional action.

Let me introduce an example with an innocent threat. Although it has been contested, I shall assume the following to be an instance of justified selfdefence.13 Imagine Lisa walking in a park, and a sudden gust of wind blows her down a well. It would be permissible for a person at the bottom of it to disintegrate her falling body by using his ray gun,14 provided, of course, that he needed to do so to save himself from being crushed by Lisa’s falling body. We assume that Lisa will survive if not disintegrated, because the person already in the well acts as a protective buffer from the lethal impact.

Now, if we accept that the person in the well is justified to use defensive force, what may be the reason for it? Remember that both are morally innocent with regard to the unfortunate situation they suddenly find themselves in. It makes therefore no sense to say that defensive force is justified because Lisa ought to have known that she might unwittingly

13 A version of this example was first presented in Robert Nozick, Anarchy, State, and Utopia (New York: Basic Books, 1974). For an argument against its permissibility, see Michael Otsuka, ‘Killing the Innocent in Self-Defence’, Philosophy and Public Affairs 23 (1994), pp. 74–94. A defence of its permissibility is to be found in J. J. Thomson, ‘SelfDefence’, Philosophy and Public Affairs 20 (1991), pp. 283–310.

14 If you do not like the idea of a ray gun, imagine that the person in the well can deflect the falling person further down the shaft. threaten the life of another. A reason at hand is to say that the person in the well is justified in using defensive force because it is the falling person who initiated the dangerous situation. Lisa is a threat to the life of the person down the well, not the other way around. Although Lisa is not doing something that constitutes threatening behaviour, it seems plausible to say that she is the initiator of the situation.15

By saying that Lisa initiates the dangerous situation, we also say which of the two people should carry the cost of it. It is therefore not permissible for Lisa to save herself at the expense of the other person. This observation does not necessarily imply that we would blame her if she did. There may be reasons for granting her some form of (hardship) excuse.16 However, Lisa would at least be a candidate for blame if she took steps to defend herself from the person in the well (and not, of course, for being a threat in the first place). For Lisa to prevent the person in the well from defending himself is wrong.

What I have said seems to be true as well if the falling person is only an apparent threat. Imagine that, unbeknown to either of the two persons, a net has been strung across the well that will catch Lisa before she hits the person at the bottom. Accordingly, when the person in the well believes only his ray gun will save him, he is mistaken. He does not need to do anything to survive. To create the sort of dilemma we are discussing, the situation has to escalate and we must imagine Lisa being able to draw her gun while falling to defend herself against the ray-gun-firing person below. In such an escalated situation it would be impermissible to save Lisa at the cost of the person in the well because it is her bodily movements that set the situation on its escalatory course. And this is true, although it is not Lisa who moves her body, but wind and gravity.

15 One might question if this is a correct use of the word ‘initiate’. Normally it means something like ‘to begin’ or ‘set going’, which are actions, and not something that happens to a person. However, the point I am stressing is that to initiate is to cause something, for example, an instance of violence. If two kids are fighting, we might ask who started the fight. And although the incriminated child only happened to trip and fall towards the other, we might agree that he nevertheless initiated the fight, but that he is not to be blamed for it. One might reply that it is nevertheless wrong of the other child to respond by hitting back. Well, that is true if he hits back after the first event is over, and merely to get revenge. The point, however, is that it is not wrong for the child to act to prevent being hit by the falling body.

16 A hardship excuse is an excuse we may grant people who do something wrong, but for which we find it difficult to blame them because they are in a pressed situation. Other instances where hardship excuses may be relevant are where a person pushes another person off a plank in order to avoid drowning themselves, or when a person succumbs to violence under duress.

3. The Moral Significance of Initiating Behaviour

How do we explain the moral significance of being an initiator, when all it takes to become one is that you are taken by the wind? One reason why Lisa’s initiating behaviour makes a moral difference could be presented by appealing to the notion of having one’s rights violated. The assumption in this case would be that one could only be justified in using defensive force against another person if this other person had lost his or her right not to be harmed. We would then have to show that the initiator (Lisa) violated some rights of the defendant, which made the use of force permissible. A problem with this account is the difficulty it introduces in making sense of the idea that being blown down the well involves the violation of some of the rights of the person at the bottom. It might seem odd to say that a person violates another person’s rights merely because something happens to her; moreover, it is not clear what sort of rights may be violated by simply appearing to be an aggressor.

Those questioning the assumption that it is permissible to use defensive force against innocent aggressors and threats could ground their argument on the claim that any person has a right not to be (unjustly) killed, and that one cannot lose this right by accidental events in the world. In order to lose one’s right not to be killed, one has to be morally responsible for a particular behaviour, which causes another person to be under threat. According to David Rodin, for example, ‘to take the language of rights seriously, is to be committed to the idea that a person’s right may be infringed or forfeited only on the basis of something that the person is or does as a moral subject’.17

I am not sure if we should take the language of rights that seriously. Clearly, rights may be altered by the right-holder’s own deliberate actions, but they may be altered by what happens in the world as well.18 When you kick my leg in order to fulfil the whims of a villain who is threatening to kill you if you don’t, it is not the case that I have violated any of your rights upfront. Nevertheless, it seems that you would be justified in kicking my leg. Even though I, under normal circumstances, have a right not to be kicked in the leg by anyone, I may lose this right if people around me broach danger unless I get kicked in the leg. Perhaps you would say that this is true for trivial rights, but not about the right of not being killed, for

17 David Rodin, War and Self-Defense (Oxford: Clarendon Press, 2002), p. 88.

18 Or we might simply say that rights come with a set of almost unlimited qualifications. example. But, then, what should we say about the person on a track or a road towards whom we steer a trolley or car in order to save five other people standing in our way? Most people would think such acts are permissible, but the sacrificed person has clearly not violated anyone’s rights. Perhaps we could say that people have a right to use defensive force against anything that comes their way, be it a stone, trolley or a person. Such threats may or may not have violated the victim’s rights. Any person, who initiates threatening action but who defends him or herself against a victim, violates the victim’s right to use defensive force.19 That is why threats like that posed by Lisa are liable for moral blame if they act to prevent others from defending themselves against them. They are not to be blamed for what has happened to them, causing them to initiate threatening conduct, but they might be if they try to extricate themselves from their problems by shoving them onto others.

Unfortunately, this response will not take us very far. Saying that people have a right to use defensive force against anything that comes their way would delegate the same kind of right to the person initiating the threat. The initiator would have a right to use defensive force against her victim when he uses defensive force against her. There must be some account of the asymmetry between the two parties. A plausible theory must be able to explain why one party is justified in using defensive force, while the other is not. The reason for the asymmetry provided here is that it is only the person who defends him or herself against an initiating (apparent) threat or aggressor that is justified in using defensive force. But what we are looking for at this juncture is a reason that can explain the moral significance of being the initiating threat in the first place.

A Contractual Perspective

According to Jeff McMahan, who is looking for a justification for killing innocent aggressors but sceptical about the prospects of finding any, it might have considerable social utility to permit self-defence against any threatening person other than just aggressors because one can seldom be certain that an aggressor is indeed fully innocent. And since cases

19 Having a right to use defensive force against initiating threats should be understood as having a claim against being prevented when doing so, and no duty not to do it. See W.N. Hohfeld, Fundamental Legal Conceptions, ed. W.W. Cook (New Haven: Yale University Press, 1966), pp. 35–50. See also J.J. Thomson, The Realm of Rights (Cambridge, MA: Harvard University Press, 1990), particularly the first four chapters, pp. 37–122. involving innocent aggressors are rare, a case in which there is uncertainty is more likely to involve a culpable aggressor than an innocent one.20 Having said this, McMahan expresses a worry about morality being merely conventionalist. But the point I am about to explore is that people have good reasons to accept the principle of initiating behaviour, and not the mere fact that they have converged towards it; there are contractual grounds for accepting initiating behaviour as a demarcation principle.

Any rule that allows the killing of innocent people must be given a rationale that people in general would regard as a good reason for accepting the rule. Giving such a rationale is the essence of giving a contractual account of the rule. The point is neither that people have made a particular contract, nor that we may be confident that they would if they had an opportunity to do so. The point is that people have good reasons to accept the rules. We must focus on what we know about them in virtue of which we have reason to believe that they would have consented to the rule.21 And I should think that we know at least two things about them: (1) They think that it is a bad thing that people are killed, and (2) they think that it is worse that innocent people are killed than culpable.

Knowing this, there seem to be good reasons for accepting a rule that gives priority to a party that responds to an initiating behaviour. First, it may work as an incentive to prevent the causes of such unfortunate situations. Secondly, being an initiator will often be a proxy for culpability, and by giving priority to the responding party more innocents will be saved in the long run. Thirdly, giving priority to the responding party seems to be the best way to avoid situations in which a culpable person survives at the cost of an innocent person.

The Incentive Argument

Saying that it is always the initiating person who eventually has to carry the cost of an escalating conflict could be the best way to minimize the likelihood that such situations arise in the first place. People will try to avoid becoming initiators when they know that, whether or not they are morally responsible for so doing, they will have to carry the cost of it.

20 Jeff McMahan, ‘Self-Defence and the Problem of the Innocent Attacker’, Ethics 104 (1994), pp. 252–90, at p. 288.

21 See Thomas Nagel, ‘Rawls on Justice’, in Norman Daniels (ed.), Reading Rawls

(Stanford: Stanford University Press, 1989), p. 5, and Thomson, The Realm of Rights, p. 187.

It is not always clear whether or not someone could have avoided a certain situation; innocence comes in degrees. This is true of many kinds of innocent aggressors, for example, people who lack the mental capacity required to be held morally responsible for their actions. It is far from clear at which point they should be considered completely outside the realm of moral responsibility, and although we might think it is inappropriate to hold them responsible for their actions in the sense that they are punishable, we might think that it is better if they carry the costs rather than their victims in situations of self-defence.

One problem with the incentive explanation is the point that people cannot take steps to avoid certain situations when they do not know how to avoid them. After all, the crux of the example of Lisa falling down the well was exactly that she could not have known that she would be thrown off balance by a gust of wind. This observation appears to obstruct the prospect of explaining on contractual grounds why we should give priority to a person who responds to an initiating behaviour by another party, when what caused the initiating behaviour were merely something that happend to the person.

Such objections can be met, however. First, we could ask how the person down in the well possibly could know whether Lisa had demonstrated appropriate care. On what criterion could we decide whether he is morally justified to defend himself against the falling woman? Whether or not Lisa is to be blamed for falling down the well is difficult to decide both for prospective victims and bystanders alike. We therefore need an account that helps us converge on the party who ought to be given priority; both the defendant and intervening bystanders need external features which can enable us to pass judgement. Using the initiating behaviour as a demarcation principle provides such an external feature.

Moreover, although the point of the example with Lisa was that she could not have foreseen that she would be bowled over by the wind, deciding whether or not she ought to have known is not easy. Although Lisa is not to blame for choosing to walk in a particular direction at a particular time of day, it seems easier for her to avoid becoming a threat than for the person down in the well to foresee that she will come hurtling down towards him. The decisions that made up the dangerous situation are closer to her. It seems therefore rational to put the burdens on the person who creates the costs, and not on the one whom the costs fall. This observation helps explain why people’s responsibility is to avoid becoming a threat to another person, rather than to avoid the use of force against initiating threats with no culpability.

The Proxy Argument

Being an initiator will often be a proxy for culpability. People who are morally responsible for endangering other people will normally also be the initiating party. And since being an initiator is an external feature that it is possible for other people to observe, sacrificing the initiator is likely to save more innocent people in the long run while putting paid the culpable. That seems to be a good reason for accepting initiating behaviour as a demarcation principle.

Clearly, it will be crucial to explore if there could be upshots from sacrificing the initiator in the form of costs to the innocent rather than the culpable. This would concern conflicts involving an intention to harm by the responding party, and the innocence of the initiator of the threatening situation. Although we might with some ingenuity imagine such situations, they will be rare. To illustrate we may adjust the example of Lisa falling down the well. Imagine that Lisa’s fall would flatten somebody intent on killing her anyway. Evil sits deep down in an empty well waiting for Lisa to come looking for water. Evil has the intention of shooting her as soon as he sees Lisa peer over the rim. But because his concentration fails him, he fails to notice Lisa who, in the darkness, is suddenly plummeting towards him at full speed. Lisa slipped on the wet ground, missed her balance and fell into the well. Evil thinks (correctly) that the only way he can survive is to use his ray gun and disintegrate Lisa before she hits him.22

People I have acquainted with this case say that it would be permissible to save Lisa at the cost of Evil defending himself. This they continue to believe, although they maintain that in situations in which both parties are innocent, priority should be given to the party in the well who is responding to the threatening behaviour of the falling person. Thus, although initiating behaviour normally sways the balance, it seems as if culpability prevails whenever it comes into play. But this should not surprise us. We comply with the principle of initiating behaviour in order to reduce the likelihood of saving culpable people at the cost of innocent; we should therefore not comply with the principle when we know that one party is culpable. If we do not know which of the two is culpable, the principle of initiating behaviour will dictate us to save Evil because intentions

22 There might or might not be a net across the well that will catch falling Lisa before she hits Evil. are not external while being the initiating party is. That is unfortunate, but, as I said, such instances will be rare.

Fairness

An important observation is that people in general have reasons not to accept rules which would save culpable people at the cost of innocent. We know that it may be difficult for a person under attack to decide whether or not an aggressor or threat is innocent. A rule which required the defendant to make sure that the aggressor/threat was culpable before defensive force could be deemed permissible, would lead people to give up their life on wrong occasions. Wrong occasions would be instances in which the defendant may mistakenly believe a culpable aggressor to be innocent, and therefore infer that he ought not to defend himself. In such cases a culpable person would survive at the cost of an innocent person. This may look unfair.

To avoid such unfortunate results, one could say that whenever the defendant knows him or herself to be innocent—a question easer to settle for the defendant than the question of the aggressor’s innocence—the defendant will not be blamed if he or she uses defensive force against apparent threats or aggressors. People in general have good reason to allow a rule permitting selfdefence against initiators because compliance to the rule does not (normally) lead to the deaths of the innocent while saving the culpable. On the contrary, such action will normally save the innocent at the cost of the culpable. Giving priority to the person who responds to an initiating behaviour will enable people to avoid giving up their life on wrong occasions.

In OriginalMary, where Mary has been told untrue things about Bill, Mary initiates the escalating situation, but she does not know that by doing so she inadvertently makes it permissible to kill her if need be. Thus, by allowing a person who responds to an initiating party to use defensive force, innocent people will on occasion be killed. But that would have been the upshot whatever rule were chosen, unless we followed a rule prohibiting the use of any defensive force, or permitted it only when it was known that it was used against a culpable aggressor.

On some occasions, a rule prohibiting any use of defensive force would save people from killing an apparent aggressor. This would be a pacifist rule. It is questionable, however, whether such a rule would cause fewer deaths in the long run. A more modest version would say that it is permissible to use defensive force if and only if the aggressor is a culpable real aggressor. This would be the proposal presented by Michael Otsuka, and recently supported by David Rodin; namely that it is only permissible to use defensive force against a person who is morally responsible for being an aggressor or threat.23 An attractive feature about this proposal is that it points to a property about which it is easy to agree has moral significance, namely moral responsibility. However, it fails with regard to a variety of examples, such as those discussed in this paper. Neither would it, I believe, be the preferred solution on contractual grounds. This is because it will lead to plenty of situations in which a culpable person is saved at the cost of an innocent person.

Concluding Remarks

Someone comes at you, and to all intents and purposes he appears to be intent on killing you. Perhaps he is screaming blue murder, perhaps you are in a war zone and he is wearing a uniform like those who killed your friends. What should you do? Well, first of all you should consider whether he represents a real threat, and not simply someone appearing like one. The standard of evidence you have to satisfy in order to determine whether the situation actually is dangerous is the standard of reasonableness. Secondly, you have a duty to consider necessity and proportionality. Any use of defensive force should be the least available, and be proportional to the threat. If these requirements are satisfied, that is, you reasonably perceive that the threat is real and assume you cannot escape unless you kill the aggressor, you may do so. If it is reasonable to interpret the behaviour of the other party as threatening, you are justified in using defensive force even though it should turn out that the threat is only an apparent one. But even then, we might prevent you from wounding the aggressor if we know what you don’t, i.e. that the threat is only apparent, and that we can defuse the situation at moderate costs to you. The fact that you are justified only dictates that we ought to give priority to you if there is a choice of imposing equal harm on either of the two in order to halt the escalating process.

23 See Otsuka, ‘Killing the Innocent in Self-Defence’, pp. 74–94, and Rodin, War and SelfDefense, pp. 70–89.

So far there may be little disagreement. The real controversy is about culpability. I claim that although you should be able to figure out that the aggressor is not morally responsible for his or her actions, the implications of that information would be limited. Imagine that you receive a phone call telling you that the person attacking you is insane, sleepwalking, a soldier believing falsely that he is fighting a just war, or something else removing his moral responsibility. It would nevertheless not be the case that you ought to stop defending yourself, and allow the aggressor to have his or her way. If it comes to a choice between your life and the aggressor’s, you are, according to the principle of initiating behaviour, permitted to choose your own, and it would be wrong for everybody else to suggest that the deranged but innocent aggressor should be saved.24

It could be objected that the innocent aggressor and his or her victim are both innocent victims thrown together by circumstances for which neither is morally responsible, and that even if the innocent aggressor loses the immunity to attack, he or she retains a right to self-defence.25 A common objection to any reasonable belief view on justification is that it is likely to allow two parties to be justified in killing each other. But this objection does not apply to the idea put forward here. Two parties cannot both respond to each other’s initially threatening behaviour. It is part of the idea that one party takes the initial step, while the other responds to it.26 According to the principle of initiating behaviour, the innocent aggressor and his or her victim have not been thrown together. One party is thrown against the other and it is wrong of the former to thrust his or her bad luck upon the victim.

In this paper I have indicated that in order for an act of self-defence to be justified against an apparent aggressor or threat, the other person must initially have behaved in such a way—some physical movement, some verbal exclamation—that it is reasonable to interpret it as threatening. Given this account of justification, we can maintain the view that there is a significant relation between an act being justified and which party ought

24 I said that the implications of being informed that the aggressor was innocence would be limited, not that it would be completely negligible. If you know that the aggressor is innocent you may have to endure more harm in order to avoid killing the aggressor than if he or she is culpable.

25 See McMahan, ‘Self-Defence and the Problem of the Innocent Attacker’, p. 283.

26 If they were pitted against each other, neither would be justified in killing the other. And it would be permissible to save either of the two; a procedure using lottery could be preferable. eventually to be given priority in a conflict situation. If we know that an act is justified, we know that we have to give preference to its agent when there is a choice between imposing equal harm on him or her and his or her victim.27 We should always give preference to the one who does not initiate the escalating process.

27 Another view on the matter could be to say that as long as the situation has not escalated, use of force is only excused. If the conflict escalates, the ultimately justified party is the one that did not initiate the situation.

Killing in War and Moral Equality*

Stephen R. Shalom

1. Introduction

Do innocent civilians who will be killed in a justified attack on a nearby military target have a right to defend themselves by shooting down the bomber pilot? I will argue that they do not, and that Jeff McMahan’s view that they do have such a right—that there is a moral equivalence between pilot and civilian—is flawed in much the same way that Michael Walzer’s moral equivalence of combatants—a position that McMahan has so persuasively refuted—is flawed.

Michael Walzer has been the most eloquent advocate of the concept of the “moral equality of soldiers,” which holds that soldiers, so long as they adhere to jus in bello norms, are entitled to kill each other, regardless of whether or not they are fighting for a just cause.1 So, for example, if country A has unjustly attacked country B, and country B is engaging in justified self-defense against this attack, a soldier from country A has just as much right under this view to kill a soldier from country B, as the soldier from country B has to kill the soldier from country A. Soldiers do wrong and can be held accountable, says Walzer, if they violate jus in bello—if they target civilians, say, or mistreat prisoners of war—but only the leaders of states do wrong and can be held accountable for violating jus ad bellum, for waging a war of aggression, for example.

There of course might be epistemic reasons for excusing individual soldiers for violating jus ad bellum. It is often hard for a young private to have access to the information necessary to carry out a jus ad bellum analysis of a particular war his or her country is involved in.2 But that we excuse people in cases where they could not reasonably have been expected to know

I Am Grateful to Jeff Mcmahan for Extremely Generous and Valuable Comments. Thanks as Well to an Anonymous Reviewer for Helpful Suggestions.

1 Michael Walzer, Just and Unjust Wars, 4th ed. (New York: Basic Books, 1977, 2006), pp. 34–37.

2 In his critique of Walzer’s view, Jeff McMahan (Killing in War [New York: Clarendon Press, 2009], p. 3) phrases the standard just war theory view as “no one does wrong, or acts impermissibly, merely by fighting in a war that turns out to be unjust.” But the phrase that they were doing wrong, does not mean that the wrongful act in general becomes right. Nor does it mean that all who commit the act are excused independent of whether they personally could not have been expected to know the act was wrong. Rather, we still consider the act to be wrong in general, while excusing the particular actor given the particular mitigating circumstances.

Consider Alice, who kills a man who comes to her door with a mask on and a bloody axe in his hand. She believed she was acting in selfdefense; as a recent immigrant, Alice was unaware of the American holiday of Halloween, and she was unaware too that her victim had mistakenly thought Halloween was in September. So we might excuse Alice for the killing of her harmless neighbor, but we certainly wouldn’t conclude that killing harmless neighbors is generally right or that someone else, Barb, who also killed a neighbor, should be excused even though she was well aware that her neighbor was harmless.

It is true that many soldiers do not have the information to judge whether the war in which they fight is a just one.3 But it is also true that many soldiers often do not have the information to judge whether they meet jus in bello standards. For example, jus in bello criteria include proportionality—does the military value of the target outweigh the collateral harm that will be inflicted on innocent non-combatants? But does the individual soldier have the information to assess the military value of particular targets? Maybe that large enemy fort, given its location far from the area of combat, is of negligible military value, while a smaller fort closer to the scene of the fighting is much more significant. Or when a factory produces ball-bearings: are these ball-bearings—rather than ones of a different size—militarily significant? We certainly might find the epistemic difficulties here serious enough to mitigate or excuse jus in bello

“turns out to be” suggests that the unjustness of the war is only discovered later, after the fact. In this sense, however, the standard view is unassailable. If I participate in a war that at the time seemed just, and only afterwards was evidence made available that showed it to have been unjust (perhaps through the release of something like the Pentagon Papers), then my participation should rightly be excused. McMahan’s full discussion of the epistemic excuse indicates that the wording on p. 3 is just a careless formulation and does not represent his considered view.

3 I am assuming an objective account of permissibility here. It is easy to construct an example of a war where both sides fight justly from a subjective point of view: each may believe that the other side attacked first and their own side is therefore engaged in justified self-defense. But objectively speaking, only one side can have attacked first and therefore only one side is objectively engaged in self-defense. Aside from my discussion of epistemic excuses in the fourth paragraph of this essay, I am considering only objective justification. responsibility. We wouldn’t, however, thereby conclude that soldiers as a general rule are not responsible for their violations of jus in bello rules. Likewise, we might frequently excuse a soldier for jus ad bellum violations because of epistemic difficulties, but this is no reason to provide blanket absolution from any jus ad bellum responsibility or to automatically excuse soldiers for such violations even in those cases where they had good reason to know that they were fighting in an unjust cause.

Note that Walzer’s argument is not simply that the higher up one is in the chain of command, the greater one’s responsibility for any wrongdoing carried out. This principle is a truism, accepted as well by those who disagree with Walzer regarding the moral equality of soldiers. And the principle applies just as much to jus in bello crimes as to jus ad bellum crimes. The commandant of Auschwitz was more responsible than individual guards who served at the death camp, at least because the commandant presumably had more freedom of action and more access to information than did his subordinates. But the guards are not blameless, certainly not as an automatic matter. And the same logic applies to jus ad bellum concerns. The generals bear more responsibility for launching an unjust war than do the privates, and the political leaders who ordered the generals into action bear greater responsibility than the generals. But there is no reason this should lead us to conclude that the soldiers are ipso facto blameless.

Jeff McMahan, among others, has criticized Walzer’s notion of the moral equality of soldiers.4 Killing, he argues, is generally wrong. Certain narrow exceptions can make it permissible: among them, defending yourself from a person who attacks you unjustly. But you don’t have the right to kill someone who is attacking you justly. Walzer gives a domestic analogy here, but dismisses it; McMahan argues that the analogy is entirely apposite. A bank robber goes into a bank with gun drawn. The guard tries to draw her own gun, but the robber shoots her first, in “self-defense.” But we wouldn’t call this permissible self-defense because the guard was herself acting justly (putting aside the question of whether the bank, through its predatory lending practices, was itself guilty of robbery). You may kill someone in self-defense if your attacker is trying to unjustly kill you. The right of self-defense, however, does not extend to the killing of those who are acting justly. In the same way, a Polish soldier in September 1939 had the moral right to kill invading German soldiers. This was self-defense

4 McMahan, Killing in War. against unjust attackers. The German soldiers, however, had no reciprocal right to kill Polish soldiers, even those Polish soldiers trying to kill them, for these Polish soldiers were acting justly and one is not entitled to kill in self-defense against someone acting justly.

McMahan’s argument is powerful and intuitively compelling. That is, differentiating between the bank robber and the guard or between the German invader and the Polish defender corresponds with our intuitive moral judgments.

Imagine that we observed two soldiers in September 1939—one a member of the invading German army and the other a member of the defending Polish army. The two soldiers are within artillery range of one another. What would we say ought to happen? If we were pacifists, we might call for both soldiers to abandon their artillery. Otherwise, we would want the German soldier to switch sides and join the Polish defenders in trying to repel the Nazi onslaught. More simply, we would say that the Polish soldier, fighting in a just cause, has the right to fire at the German soldier, but the German soldier, as part of an unjust cause, has no reciprocal right. And surely we would not declare that the determination of which soldier should survive and which should get blown up ought to be left up to which one is the better or quicker shot. Such an outcome would hardly be moral, given that military prowess has no moral basis.

Note that I am not arguing that we can never be indifferent between two choices. If option A would be favored in some circumstances and option B in other circumstances, then, by the intermediate value theorem, there exists some set of hypothetical circumstances that would lead to indifference.5 The point here is that I don’t think anyone believes that we should be indifferent in the situation at hand. Surely no one with Walzerian values would be indifferent as to which soldier—the Pole or the German—prevailed.

Now it might be claimed that leaving the result up to who is the better shot is just a rough way of leaving the result up to chance. There are numerous moral dilemmas where chance is used to resolve situations where there is no moral reason to prefer one outcome to another. (Who gets to go first? We flip a coin. Who should be drafted when only a fraction of those eligible are needed as soldiers? We use a draft lottery.) But in situations where we are not morally indifferent as to the outcome, we would not want to leave the result to chance. We don’t decide the outcome of judicial

5 I assume that hypothetical conditions represent a continuous function. proceedings by flipping a coin, nor give out food stamps by lottery. Indeed, it would be highly immoral to depend on randomness in such cases. And it would be immoral too to depend on people’s relative skill with a sword or a gun to resolve a moral conflict where the moral standing of the parties is not equal.

Yet resolving moral conflicts in this way is precisely what Walzer’s moral equality of soldiers leads to. For him, the two soldiers are morally indistinguishable, even though their causes are polar opposites: naked conquest versus self-defense. The fact is, however, that no one would find the parties to this conflict morally indistinguishable.

2. McMahan’s Moral Equivalence

Interestingly, this same objection can also be applied to one aspect of McMahan’s analysis. McMahan considers a wartime case where he says two people act permissibly when they try to kill each other.6 One country, Aggressia, has unjustly attacked another country, Defendland. Defendland is engaged in justified self-defense against this unjust assault. A pilot from Defendland is dispatched to bomb a military base in Aggressia. There are a few civilians who live right next to the base, and bombing the base will almost certainly harm these civilians. Nevertheless, the attack on the base would be morally permissible under the doctrine of double effect if (a) there were no intention to harm the civilians, either as an end in itself or as a means to an end, and (b) the proportionality criterion were met: that is, the military value of the target was sufficient to outweigh7 the unintentional but foreseeable harm to the civilians.8 In addition, it would have to

6 The example is referred to in general terms in Killing in War and spelled out in Jeff McMahan, “The Basis of Moral Liability to Defensive Killing,” Philosophical Issues, vol. 15, no. 1 (Oct. 2005), p. 388. McMahan places the innocent civilians just across the border in a neutral country. For the time being, I ignore this point. The country names are mine.

7 How much greater does the military benefit have to be than the harm? This is a complex question that I will not discuss here, but see section 4 below for some brief comments on what I take to be the misuse of the proportionality criterion.

8 One could argue that the civilians bear some responsibility here for not moving away from an obvious military target. It might be thought that we can avoid this problem by assuming that the target is a secret military facility, but, as will be seen, we want the civilians to be aware that their potential deaths are the result of a morally permissible attack. So, instead, let us assume that the civilians and the military target are in a small valley, and roads leading out are impassable due to weather conditions.

Walzer persuasively argues (Just and Unjust Wars, pp. 152–59) that the traditional doctrine of double effect is an inadequate moral principle and that we ought to insist not just be the case that there was no means of obtaining the same military advantage without endangering the civilians at least as much.

The question McMahan poses is whether the civilians in this case have the right to try to shoot down the plane in order to save their own lives. He argues that they do. And then does the pilot have the right to intentionally shoot at these armed civilians who are shooting at him (or who will soon be shooting at him)? Again he answers yes. So both have the moral right to shoot at each other and the outcome is left up to who is militarily more proficient. But this means that we are indifferent as to who kills whom, even though the bombing raid was a just mission in pursuit of a just cause. Shooting the pilot down will make the success of the just cause more difficult and will make the possibility that the unjust side will prevail in the war more likely. Should we really be indifferent here?

Some might argue that, while shooting down the pilot might not be justified from an impartial point of view, the civilians do have an agentrelative permission to shoot down the pilot. Many moral philosophers believe that we are permitted to assign greater weight to our own interests than to those of others, and that therefore we may favor our own interests compared to what would be required by strict impartiality.9 Some have extended this general position to the case of self-defense, arguing that we are not obligated to flip a coin before favoring our own survival over that of another, who, like us, is morally blameless.10 But obviously agentrelative permission can only go so far without collapsing into egotism. It is one thing to favor one’s own life when faced with two morally equivalent options—as might be the case if it were just a question of the life of the morally innocent civilian versus that of the morally innocent pilot. It is something else again, however, when the two options are morally quite unequal: on one side, there’s an innocent civilian; on the other, there’s an innocent pilot plus the additional innocent lives that—by hypothesis—will be saved as a result of the successful attack on the that pilots not intend the civilian deaths, but that they take reasonable measures to avoid the civilian deaths, even at some risk to themselves. So let us assume that the pilot is taking those precautions and thus that the raid meets not just the traditional double effect criteria, but Walzer’s more demanding standard as well. There are other philosophical challenges to the doctrine of double effect, but I will put aside those arguments here.

9 See Samuel Scheffler, The Rejection of Consequentialism, revised ed. (New York: Oxford University Press, 1982, 1994).

10 See, e.g., Nancy Davis, “Abortion and Self-Defense,” Philosophy and Public Affairs, vol. 13, no. 3 (Summer, 1984): 175–207; Jonathan Quong, “Killing in Self-Defense,” Ethics, 119 (April 2009): 507–537. military target.11 And if there are multiple enemy civilians whose lives would be directly saved by shooting down the pilot, double effect proportionality would mean that even more innocent lives would be saved by the success of the mission.

In support of the moral equality of the pilot and the civilians in this situation, McMahan can appeal to the important right not to be killed, which the civilians in question have neither waived nor forfeited. But consider a slightly different example. Say that the military target, instead of being located near some civilians’ homes, is located next to a small prisoner of war camp.12 Defendland knows of the POW camp and that some prisoners will die—unintentionally but foreseeably—in an attack on the military target. But, as in the previous example, Defendland’s policy makers assess (correctly) that by the doctrine of double effect the attack is morally permissible in pursuit of a just cause. Now what if one of the prisoners has managed to smuggle a high-powered rifle into the prison camp? Knowing that it is quite likely that she and the other prisoners will be unintentionally killed in the attack, does the prisoner have the right to shoot at her own country’s pilot? That is, does she have the right to thwart a mission which on balance—weighing all the morally relevant factors: the deaths of the prisoners and the value of the military target—furthers a just cause?

It seems clear to me that she does not. Does she just have to accept her death? No, she can hide under her bed, or try to escape using her rifle. But it can’t be right that she could be permitted to shoot at the pilot. Justice precludes her saving herself by causing more harm to others—and by hypothesis, given the double effect proportionality calculation that was done, shooting the pilot causes more harm. This seems no different from someone who discovers a live hand grenade at her feet: she may not save

11 Thus, even if one accepts that in McMahan’s example (Killing in War, pp. 27–28) of gladiatorial combat, where two warriors are forced to fight to the death or else they both will be killed, each gladiator is morally entitled to try to kill the other, this is not an analogy to the pilot case. There may be no moral reason to prefer one gladiator to the other, but there is a reason to prefer the pilot over the civilian: namely, that the pilot’s death leads to the failure of the morally justified mission, which mission, by hypothesis, will save lives.

12 Just as a country violates international humanitarian law by placing a military base near civilians, so too it violates the IV Geneva Convention by “set[ting] up places of internment in areas particularly exposed to the dangers of war.” (Convention [IV] relative to the Protection of Civilian Persons in Time of War. Geneva, August 12, 1949, art. 83, in The Laws of War: A Comprehensive Collection of Primary Documents on International Laws Governing Armed Conflict, ed. W. Michael Reisman and Chris T. Antoniou [New York: Vintage Books, 1994], p. 255.) herself by throwing it into a crowd of school children. If she is in a closed room with a single window, outside of which school children are playing, then she may not throw it.

It might be argued that what distinguishes the innocent civilian case from the prisoner of war case is that the latter owes an obligation to her own country and to the pilot who is her compatriot that is not owed by the civilian. On this view, that would explain why the civilian is justified in shooting down the pilot, but the prisoner of war is not. This argument, however, is unavailable to McMahan, for he believes (correctly, in my view) that it is not only combatants from the just side who have an obligation to the just cause. For McMahan, a combatant whose country is involved in an unjust war has the obligation of “refusing to continue to fight,” or of “sabotaging his own side’s military efforts, or even … defecting to the just side.”13 But if unjust combatants ought to take up arms against their own compatriots, then what is crucial is not which country is one’s own, but which cause is just, in which case, a civilian from Aggressia would seem to have as strong an obligation to the just cause as the Defendland POW does. And if this obligation prevents the POW from shooting down the pilot, it should likewise prevent the Aggressian civilian from doing so. Might the difference between the civilian case and the POW case be that the latter is a soldier and as such has certain obligations to her fellow soldiers that enemy civilians do not have? Well, in that case, let us replace the POWs with a group of Defendland civilians who were in Aggressia when war broke out and have been kept against their will as “guests” of Aggressia, confined to quarters near the military base—and thus essentially involuntary human shields.14 If these civilians somehow obtained an anti-aircraft weapon, would they be justified in shooting down their coun- try’s plane? That seems implausible.

13 McMahan, Killing in War, p. 51.

14 An example of this might be the Westerners held by Saddam Hussein in the weeks following his invasion of Kuwait in 1990. He ultimately let them go before the Gulf War began, but had he not, Washington and its allies would have had to decide whether to proceed with their bombing despite the presence of their non-combatant citizens.

I don’t mean to imply by this example that the U.S. decision to go to war was just or that the targets of U.S. bombing were always permissible. I believe the opposite to be the case. See Jeff McMahan and Robert McKim, “The Just War and the Gulf War,” The Canadian Journal of Philosophy, vol. 23, no. 4 (Dec. 1993): 501–541; Barton Gellman, “Allied Air War Struck Broadly in Iraq; Officials Acknowledge Strategy Went Beyond Purely Military Targets,” Washington Post, June 23, 1991; and Thomas J. Nagy, “The Secret Behind the Sanctions: How the U.S. Intentionally Destroyed Iraq’s Water Supply,” The Progressive, Sept. 2001.

3. Infringing Rights

McMahan says that although the pilot is justified in killing the Aggressia civilians, the pilot is nevertheless infringing the civilians’ rights.15 He distinguishes this case from that of the bank robber who is about to be shot by the bank guard. There too someone is about to be justifiably killed, but the bank robber’s rights are not being infringed. By his act of robbing the bank he has forfeited his right not to be killed, while the Aggressia civilians have done nothing to forfeit their right not be killed.

But what has to be kept in mind in the bombing case is that the mission is a just one because it serves a just cause and satisfies the criteria of the doctrine of double effect: proportionality (saving more lives than the number of civilians that will be killed), that civilians are not being intentionally targeted, and that there is no way to destroy the military target with less harm to civilians. This must make the shooting down of the pilot by the Aggressia civilians impermissible. To see that this is so, consider the following set of examples:

1. A villainous aggressor is driving at you, trying to kill you. The only way to save yourself is to use your anti-tank weapon to blow up his vehicle. Is that permissible? Most everyone will agree that it is.16

2. You are locked in a room into which a villain throws a live hand grenade. The only window leads to a yard where a group of deaf children are playing. May you throw the grenade out the window? Most everyone will agree that this would not be just, even though you are innocent and the action is necessary to save your life.

3. Now combine the two examples. A villainous aggressor is driving a bus at you, trying to kill you. In the back of the bus a bunch of innocent children are tied up. May you fire your anti-tank weapon at the bus to save your life even though you know this will kill all the children? If we accept the conclusions in (1) and (2), then we must agree here that blowing up the bus is impermissible.17

15 McMahan, Killing in War, p. 47.

16 This example, and the terms “villainous aggressor,” “innocent aggressor,” and “innocent threat” below are taken from Judith Jarvis Thomson’s seminal article, “Self-Defense,” Philosophy and Public Affairs, vol. 20, no. 4 (Autumn 1991): 283–310.

17 Quong (“Killing in Self-Defense,” p. 533) gives the example of an Innocent Shield “trapped inside a runaway trolley that is speeding toward you, and the only way to stop the trolley is by firing a small missile at the trolley which will kill the Shield (if you do not fire the missile they [i.e. the Innocent Shield] will survive).” He considers it permissible to fire the missile and states that “many people believe such killings are permissible.” I’m not sure

4. But if it is impermissible to blow up the bus in case (3), then it would be at least as impermissible to blow up the bus if it were being driven by an innocent aggressor (say, a driver who had been hypnotized or drugged by a villain into trying to run you down).

5. And if it’s impermissible to blow up the bus in case (4), then surely it is impermissible to blow up the bus when we change the situation to one of an innocent threat. This time a villain has tied up a bunch of children in a bus, put on their seat belts so they won’t be harmed in a crash, and then rolled the bus down a hill toward you. Alas, the only way you can stop the bus from killing you is to destroy the bus and all its passengers with your anti-tank weapon.18

But there is no more justification for shooting down the plane bombing the military base than there is for blowing up the bus in case 5. In both situations, an innocent person faces death and the only way for the potential victim to prevent that death is to take an action that will remove the immediate threat, but also kill many additional innocent people. If it is wrong to do this in the bus case, where the additional innocents are sitting in the back of the bus, then it must be wrong as well to do it when the innocents (aside from the pilot) are those whose lives the bombing attack is designed to save—which, to repeat, follows from the attack being proportionate.

It might be objected that there is a crucial difference between the bus case and the pilot case: in the former, the potential victim directly kills the innocent children, while in the latter, shooting down the pilot doesn’t directly kill other innocents (apart from the pilot), but merely prevents them from being saved (as a result of the military benefit of the mission, had it been successful). To some, preventing someone from being saved is morally more akin to letting die than to killing, and both letting die and preventing from being saved are not as morally objectionable as killing.19 So consider case 6. what fraction of people would agree that this is permissible, but I doubt many would agree when there were several Innocent Shields involved. This would then no longer simply be a case of one innocent life versus another, where agent-relative permissibility would arguably allow you to favor yourself over the shield. Rather it would be a case of saving one life or more than one.

18 If it is the existence of the villain in this case that makes one think that blowing up the bus is justifiable, assume the kids, all disabled but seat-belted in, were waiting in the bus for their driver when a gust of wind propelled the bus down the hill toward you.

19 See Matthew Hanser, “Killing, Letting Die and Preventing People from Being Saved,”

Utilitas, vol. 11, no. 3 (Nov. 1999): 277–295.

6. Rather than children in the back of the bus, there are a dozen frozen hearts being rushed to a children’s hospital where a dozen recipients with just minutes to live await heart transplants.

Here too it seems to me wrong to blow up the bus, though admittedly our intuitions will not be nearly as strong as in case 5. Of course, one of the reasons for our intuitive ambivalence is that in real life we’d wonder whether some other transplant organs might not become available at the last minute, so that destroying the bus would not lead to the deaths of any innocents.20 But the point of the proportionality calculation in the pilot case is that there isn’t another method of obtaining the same military benefit with less harm to civilians; if there were, then by definition the air mission would not be proportional.

Consider one further example, an extension of the famous trolley problems.21 Say a trolley is hurtling down a track towards a car that has stalled on the track with its doors jammed shut and with five people trapped inside. You can flip a switch, which will divert the trolley onto a different track where another car is similarly stalled, but this one has only a single person trapped inside. Most people would agree that it is permissible to flip the switch, and many would argue that it is required. But now imagine that the single trapped person has a gun that cannot stop the trolley (which has bulletproof glass), but can shoot you in the head, preventing you from flipping the switch. Is it permissible to shoot you? The single passenger is innocent, is being threatened with death by someone who is (presumably) acting justly, but shooting you will not just kill you, but ensure the deaths of the five innocent people trapped in the other car. For anyone who agrees that you are acting permissibly in flipping the switch, it is hard to see how shooting the switch flipper can be justified.22

20 In real life we are never as sure of the situation as we are presumed to be in philosophical examples. The condition “and there is no other way to save these lives” is far less likely to be known with certainty in reality than in hypotheticals, and the more distant in time the presumed deaths are from the present, the greater our uncertainty. This is one powerful reason why it would be wrong to (foreseeably but unintentionally) kill 100 innocent civilians today in order to save the lives of 101 people in the future. Many things can intervene before those 101 deaths occur, rendering the 100 deaths unnecessary. This uncertainty is a further argument against too permissive an understanding of double effect proportionality, as discussed in section 4 below.

21 I have altered this problem, originally posed by Philippa Foot and analyzed by Judith Jarvis Thomson, to remove all villains.

22 Moreover, if the single passenger has the right to shoot the switch-flipper, then the five passengers surely have the right to shoot the single passenger to try to prevent her from taking an action (shooting the switch-flipper) that will lead to their deaths. If the single passenger manages to get off her lethal shot before the five kill her, then all seven people

4. The Problem of Proportionality

McMahan offers a caveat to his position that may be thought to get him out of these difficulties. He writes:

Unless what is at stake is so important that the civilians are morally required to sacrifice themselves for the sake of the just combatants’ mission, they are entitled to protect their rights against infringement.23

But this exception is already covered in the notion of proportionality. If there is a military mission that it is known will have no bearing on the outcome of the war and will not substantially reduce the harm to just combatants (or just non-combatants), and yet will cause foreseeable, though unintentional, deaths to civilians in the country that is pursuing an unjust cause, then that mission fails the test of proportionality. It doesn’t matter whether 500 tanks will be destroyed in the process and only a single civilian killed. If the destruction of these tanks doesn’t change the war’s outcome or costs, then there can be no justification for killing even one innocent civilian (or, more precisely, for destroying those tanks with the knowledge that there will be this foreseeable civilian death). Now, of course, in real life, nothing is known with certainty, and one can only make probabilistic guesses as to the military benefit of any particular attack. But I am not convinced that McMahan is right in suggesting that there are cases where the proportionality principle is met but yet where the stake is not so important that civilians are morally required to sacrifice themselves for the sake of the mission.24

A supporter of McMahan’s position might offer the following argument. In a war, the just side might carry out thousands of bombing raids, many of which will foreseeably kill some civilians, but the raids are justified nonetheless by double effect. If all or a large number of the raids were thwarted, this might make a difference in the outcome of the war or in the cost of the war in terms of additional lives lost. But the failure of a single raid is hardly likely to make a significant difference in the outcome or will end up dead, and yet—if we assume the single passenger has the right to shoot at the switch-flipper—everyone has acted justly.

23 McMahan, Killing in War, p. 47.

24 Note that when we say “sacrificing themselves” we do not mean committing suicide. Rather, we mean “refraining from killing innocents in an attempt to save ourselves.” So while there is no moral obligation to run over to a crowd of children into which a hand grenade has been tossed and throw oneself on top of the grenade, there is a moral obligation to refrain from flinging the grenade that has landed at your feet into a crowd of children even if this is the only way to save yourself. costs. The outcome would be at stake only if the military prospects of the two sides were virtually identical so that this one raid would be the deciding factor. But the probability of this being the case is extremely low, and indeed approaches zero. Therefore, although this raid (and all the others) meets the proportionality test, the odds of it making a difference that this one particular pilot is shot down are close to zero.25 Accordingly, it is permissible for the civilian to shoot down the pilot.

This argument fails, however, because it violates the basic requirement of moral principles that they be generalizable. That is, a moral principle cannot be something that applies to a single actor or a single instance. If it is right for this one civilian to shoot down this one just pilot from Defendland, then it must be right for every threatened civilian to shoot down every one of Defendland’s pilots that threatened them. And while the shoot-down of a single pilot would be unlikely to change the outcome or costs of the war, the same is not true of a generalized policy of shoot-downs.

One reason for believing that there is a non-empty set of instances where double effect proportionality allows an attack but yet where the attack is not important is that the standard understanding of double effect proportionality is much too permissive.

Consider, for example, this explanation of proportionality by the Israeli High Court of Justice:

Take the usual case of a combatant, or of a terrorist sniper shooting at soldiers or civilians from his porch. Shooting at him is proportionate even if as a result, an innocent civilian neighbor or passerby is harmed. That is not the case if the building is bombed from the air and scores of its residents and passersby are harmed.26

That the latter example is disproportionate and impermissible is evident, but the first example may or may not be permissible; the example does not provide us with adequate information to judge the situation. If the sniper is randomly picking off civilians who are trapped in an open area, then it might be permissible to shoot at the sniper, even when there is substantial

25 Of course, if the raids are all sequential, it may well be known before the last raid that the outcome of the war or its costs will not be affected by subsequent raids. But if so, then those later raids do not meet the proportionality test. So let us assume that the raids are nearly all simultaneous.

26 Israeli Supreme Court, HCJ 769/02, The Public Committee Against Torture in Israel v. The Government of Israel, December 2006, art. 46, http://elyon1.court.gov.il/files risk to a bystander. But if the sniper is shooting at soldiers who are behind fortifications and there is no pressing need to advance, and if the sniper does not pose a future threat significantly greater than that of any other enemy soldier, then it would be wrong to return fire, even if the odds of harming the bystander were slight. Certainly we would be outraged if police officers fired into a house from which a gunman was shooting and killed a bystander they knew was there, even though the gunman posed no danger to the well-protected police. Given that we ought to take as much care in avoiding harm to enemy civilians as to our own,27 proportionality does not mean that every time you can take out two enemy soldiers when killing only one enemy civilian, it is thereby proportional.

Consider again the case of the prisoner-of-war or the involuntary human shield. If two, or even 20, enemy soldiers held a P.O.W. or civilian hostage, it would be outrageous to drop a bomb on the group of them, despite the favorable ratio of two to one or even 20-to-one of enemy soldiers to hostages, unless these enemy soldiers were an immediate and likely threat to kill others. Domestic hostage negotiators are not praised for killing the bad guys when the hostage gets killed too: indeed, their primary objective is the preservation of lives, and apprehension of the perpetrator only secondary.28

McMahan argues against the view that shooting down the pilot would be unjustified by stating that it is incorrect to think that the fact that the mission will cause more good than harm (the proportionality criterion) necessarily means that thwarting the mission is unjustified. Thinking this, he says, involves a subtle mistake, which he illustrates with an example: you are entitled to kill a malicious and culpable aggressor who is trying to kill you, even though this villain is a virtuoso surgeon who would otherwise soon perform five life-saving operations that no one else could perform.29 McMahan emphasizes that he does not think this is an analogy to the bomber case, but only an illustration of the point that a simple proportionality cost benefit analysis is an inadequate basis for determining whether one has the right to act in self-defense; one needs also to consider

27 Avishai Margalit and Michael Walzer, “Israel: Civilians & Combatants,” New York Review of Books, vol. 56, no. 8, May 14, 2009, http://www.nybooks.com/articles/22664 (“Conduct your war in the presence of noncombatants on the other side with the same care as if your citizens were the noncombatants.”). This assumes, of course, that the enemy civilians have not done anything to compromise their non-combatant status.

28 See, for example, Truro [Mass.] Police Department, “Hostage Situations,” Policy Number: OPS 6.21 Effective Date: Dec. 20, 2000, http://www.truropolice.org/On%20 Line%20Manuals/Hostages.pdf.

29 McMahan, Killing in War, pp. 41–42. issues of causation, agency, rights, and so on. But why did McMahan use an example that is so non-analogous to the bomber case (the bomber acts justly while the villain does not)? It may be that good analogies supporting McMahan’s point are hard to come by.

Here’s what a satisfactory analogy must do. It must provide an instance of an act that is morally permissible from an impartial viewpoint that would harm an innocent person who is then morally entitled to use lethal force to prevent the harm.

Say someone wanted to kill you in order to harvest your organs to save the lives of several good people. You are indeed permitted to thwart this “humanitarian” killer with lethal force, but this example does not meet the condition of there being an act that is morally permissible from an impartial viewpoint. That is, almost everyone will agree that it is impermissible to kill someone for their organs.30 On the other hand, consider another case. Say your blood type is extremely rare and that by your donating a pint of it you could save the lives of a dozen people who have no other hope of surviving. You refuse to donate the blood, insisting on your right to bodily autonomy. Many might conclude that society is permitted to overrule your objection and infringe your right, compelling you to give the blood. Are you then permitted to use force to prevent your blood from being taken against your will? It seems clear that in deciding that society has the right to compel you, we are necessarily concluding that you have no right to thwart society’s effort. To put it another way, if there’s anyone who believes that you have the right to shoot society’s agents as they come to collect your blood, such a person cannot also believe that society is acting justly in trying to compel your donation.31

30 I conclude that few living moral philosophers believe that it is required that people offer up their healthy bodies to provide organs to others; any who do believe this are presumably no longer with us.

31 McMahan suggests that it is possible to imagine a war in which an attack would be “morally justified” but yet would “be unjust”—because an act can be morally justified if there is a lesser-evil justification, but still unjust because it lacks just cause under traditional just war criteria. (McMahan, Killing in War, pp. 27–28.) His example involves a country, A, threatened by B and needing to use C’s territory to adequately defend itself from

B. C, however, refuses to allow its territory to be so used, and, says McMahan, has a right to do so. In such a case, it would seem that if A sent its troops into C’s territory, and C resisted by force of arms A’s incursion, both sides would be morally justified. This might then offer an example of where a moral act is justly opposed by lethal force. But I am not at all persuaded by McMahan’s example—which in any event McMahan says is “so unlikely in practice” (ibid., p. 28). For anyone who judged A’s incursion as morally acceptable, C’s armed resistance would be deemed impermissible. And those who approve C’s taking up arms against A’s incursion would not judge the incursion as morally acceptable.

Now McMahan’s version of the bomber case actually refers to potential civilian victims not in the territory of the unjust state, but just over the border in a neutral country. Does this make a difference? I suppose one might argue that while the government of Aggressia has an obligation to “endeavor to remove the civilian population, individual civilians and civilian objects under their control from the vicinity of military objectives,” the neutral nation, Neutralia, has no such obligation, since Protocol I of the Geneva Convention imposes an obligation to move civilians out of harm’s way upon “the parties to the conflict.”32 But this limitation in Protocol I is only because it is so unusual for the civilians of one country to be directly endangered by attacks on another country. Surely it would be wrong for the neutral nation, Neutralia, to intentionally keep its citizens in the vicinity of Aggressia’s military targets, thereby making attacks on those military targets more difficult, and thus lending aid to the unjust side in this war. Such tacit complicity on the side of the aggressor does not justify attacks by Defendland that are directed at Neutralia’s civilians or even that cause harm to Neutralia’s civilians beyond that permitted by double effect proportionality. But the complicity does seem to weaken the justification for Neutralia’s civilians to shoot down Defendland’s pilot, especially if they are willing instruments of their government.

Neutralia might argue that it should not have to relocate any of its citizens because of the war between its neighbors, and it might demand that Defendland refrain from any air strikes in the border region. Of course in real life Defendland would want to bend over backwards not to antagonize Neutralia and risk having it join the war on the other side. But if we ask in the abstract what should Neutralia do, it seems to me that Neutralia is not behaving very neutrally (not from the point of view of international law— since international law is silent on the matter, given that the situation of a country’s civilians potentially shielding the military target of another country is so artificial—but from a moral point of view). In the face of their own government’s immoral behavior indirectly supporting the actions of an aggressive state, citizens of Neutralia have an obligation to oppose their government and certainly to avoid actions that will help the unjust side prevail in this war.

32 Protocol I Additional to the Geneva Convention of 1949, adopted 1977, Article 58, excerpted in The Laws of War (see note 13 above), p. 93.

5. Defense of Others

McMahan poses the following question:33 If the civilians have the right to shoot down the just pilot, do unjust combatants have the right to come to the aid of the civilians and shoot at the pilot as well? McMahan says no, arguing that some things that are permitted depend on who the agent is who is carrying out the action. The unjust combatants are themselves part of the reason the civilians are in danger in the first place; they can best help those civilians by working for the early defeat of their side, says McMahan.

But what happens if we change the situation a little bit? Civilians generally don’t have access to anti-aircraft weapons. But they might spot the bomber approaching in the distance. Do they have the right to protect themselves by calling the nearby military base and warning them of the impending attack? Each possible answer to this question seems to present difficulties for McMahan’s view.

It would be strange if the civilians were allowed to call on the services of the unjust combatants who make up their country’s military, but those unjust combatants were then prohibited from coming to their aid. On the other hand, if the unjust combatants were permitted to come to their aid, then we have the anomalous situation where unjust combatants don’t have the right to come to the aid of the innocent civilians if the soldiers detect the incoming plane themselves, but they do have the right to do so if it’s the civilians who detect the plane and notify them. Finally, if the civilians are not allowed to call on the services of their country’s soldiers, then they are allowed to protect themselves by using a means almost never within the capability of civilians—for how many civilians have their own anti-aircraft weapons?—but are not allowed to protect themselves by the only means likely to be effective, namely, calling on the anti-aircraft weapons of their country.

The most plausible way out of this mess is to conclude that the civilians had no right to shoot down the just pilot in the first place—whether by themselves or by calling on their unjust combatant compatriots. Thus, this is another reason why we must reject McMahan’s claim that civilians are permitted to try to shoot down a pilot who justifiably threatens them.

McMahan thinks there are plausible arguments on either side of the question as to whether unjust combatants may come to the aid of the

33 McMahan, Killing in War, pp. 48–51. innocent civilians of Aggressia, but what he finds decisive is that the unjust combatants will save more civilians by surrendering.34 But what about a “defense of others” that doesn’t involve action by guilty unjust combatants?

Imagine that a civilian resident of Aggressia sees the plane flying overhead on its way to bomb the factory. (She knows that there is nothing else in that direction, so she can be certain that the factory is the target.) And she also knows that there are innocent civilians living in the vicinity of the factory who face a high risk of death in the attack. May she fire at the plane to save her innocent compatriots? That she cannot, can be seen by asking whether a civilian living in Defendland may fire on the plane (as it heads toward the Aggressia border) in order to save innocent Aggressian civilians? Obviously she cannot, because the mission is a just one: it meets the double effect and proportionality standards and helps the just side prevail in the war. And what if the plane flew near the border of a neutral country: may an innocent civilian in the neutral country fire on the plane to save innocent civilians in Aggressia? Again, she cannot. She has no right to interfere with a mission that is just and that will save lives.

6. Conclusion

McMahan is persuasive is his argument that there is no moral equality between soldiers pursuing a just cause and those pursuing an unjust cause. The latter, no matter how careful they are to avoid harm to civilians and mistreatment of prisoners, cannot fight justly, for they are fighting on behalf of an unjust cause. We cannot be indifferent in a contest between the just and the unjust combatant. In the same way, however, we cannot be indifferent between the success and failure of a mission that is morally justified and will further a just cause. So the innocent civilian has no right to shoot down the plane carrying out such a mission.

34 McMahan, Killing in War, p. 50.

Self-defence and the Principle of Non-combatant Immunity

Helen Frowe

I. The Problem of Scope

Accounts of self-defence are often divided into two camps: extended accounts (sometimes called broad accounts) and restricted accounts (sometimes called narrow accounts). The biggest difference between these two camps is their stance on the permissibility of killing innocent people in self-defence. Extended accounts permit the killing of at least some innocent people in self-defence.1 For example, they permit the killing of Falling Person, who is hurtling helplessly towards Victim as Victim stands trapped at the bottom of a well. If the only way that Victim can stop Falling Person from crushing him to death is by vapourising her with his ray gun, ESD accounts permit him to do so.2 Restricted accounts permit the killing of fewer people, usually because they require something like moral responsibility for a threat to render a person a legitimate target.3

Zohar argues that their inclusion of innocent threats as legitimate targets has made Extended Self-Defence Doctrines (ESDs) look like good candidates for the rules of war because they can deal with the putative problem of a mutual permission to kill between combatants.4 Historically,

1 See Judith Jarvis Thomson (1991), ‘Self-Defense’, Philosophy and Public Affairs 20, pp. 283–310, F.M. Kamm (1992), Creation and Abortion (New York: Oxford University Press), Jonathan Quong (2009), ‘Killing in Self-Defence’, Ethics 119 No. 3, pp. 507–537, Helen Frowe (2008), ‘Equating Innocent Threats and Bystanders’, Journal of Applied Philosophy 25 No. 3, pp. 277–290 and Helen Frowe, Defensive Killing: An Essay on War and Self-Defence, (manuscript in progress).

2 As I discuss in Section III below, my account of self-defence holds that Falling Person is similarly permitted defend herself against Victim.

3 See, e.g. Jeff McMahan (2005), ‘The Basis of Moral Liability to Defensive Killing’, Philosophical Issues 15: Normativity, pp. 386–405, Michael Otsuka (1994), ‘Killing the Innocent in Self-Defence’, Philosophy and Public Affairs 23, No. 1, pp. 74–94.

4 Noam Zohar (2004), ‘Innocence and Complex Threats: Upholding the War Ethic and the Condemnation of Terrorism’, Ethics 114, pp. 734–751, p. 745. Zohar claims that Thomson, “explicitly adduces ESD to justify the killing of enemy soldiers” (p. 741). There is no reference for this claim and I have to say that I’m unclear about how Thomson’s account would generate the moral equality between combatants that Zohar envisages. After all, Thomson all combatants obeying jus in bello have been deemed morally innocent, and yet the rules of war permit the killing of these apparently innocent people. ESDs seem able to explain this, because these accounts don’t require culpability for liability to defensive killing. The combatants might be morally innocent, but they nonetheless pose a threat and are thus liable to be killed.

Despite this apparent advantage of ESDs, Zohar argues that such accounts cannot currently support a second key tenet of the war ethic, namely a permission to kill even non-attacking members of the armed forces. Zohar claims that in order to cover all the permissible war-time targets, an account of self-defence would need to be ‘amplified’:

If the ESD is to justify the war ethic, the concept of ‘threat’ cannot be restricted to those actually trying to shoot at me; rather, it must extend to soldiers generally. The enemy soldiers as a class, other than the actual shooters, can be said to be threats only in the sense that they contribute to the threats posed directly by the shooters. Some non-shooting soldiers contribute to specific threats and, in terms borrowed from the criminal law, can be described as ‘accomplices’ of the attackers… [But many] of those who are encamped far back from the front, whether they fulfil administrative roles or are even trained combat soldiers, are neither attackers nor accomplices—in an analogous, non-military setting they would not be described as threats at all.5

Zohar argues that these ‘non-attacking’ soldiers are best understood as obstructors.6 For example, a group of soldiers might not be attacking anyone, but still be positioned in such a way as to rule out an escape route or a manoeuvre that would otherwise be open to the enemy.

In a domestic context, obstructors are usually regarded as a special sort of bystander. People like Judith Thomson, who endorse an ESD, explicitly deny a permission to kill bystanders, including obstructors, to save one’s own life.7 But Zohar argues that it is only by extending the range of legitimate defensive targets to include obstructors that an ESD can hope to capture non-attacking soldiers in the scope of the right of defence.8 is explicit that an innocent threat has no right of defence against an innocent victim (Falling Person may not defend herself against Victim on Thomson’s view). It is therefore unclear how we would get the mutual permissions to kill that Zohar is after, even if he is correct that unjust combatants are morally innocent. However, we can set this aside for the purposes of this paper.

5 Zohar, ‘Innocence and Complex Threats’, p. 742.

6 Zohar, ‘Innocence and Complex Threats’, pp. 744–745.

7 See e.g. Thomson ‘Self-Defense’, pp. 289–291.

8 Of course, we could also deny that such people are legitimate targets. However, since I hold a strong inequality position, I don’t want to do this.

Moreover, he claims that not only do proponents of the ESD need to extend their range of targets in this way, but that they have no way to resist such an extension, since there is no principled difference between an obstructor and an innocent threat.

He illustrates this claim by thinking about variations on the well case that I just described. Imagine that you are trapped between a person’s body and a metal plate. In scenario 1, the plate is stationary, and the person is moving towards you to where she will crush you against the plate. In scenario 2, the person is stationary, and it is the plate that is moving towards you. In each case, says Zohar, the moving object is dangerous only because of the presence of the stationary object. The threat to your life consists in the conjunction of the two objects in their relative motion.9 Yet the stationary person is traditionally labelled a bystander whom it would be impermissible to kill, whereas the moving person is identified as an innocent threat whom ESDs grant a permission to kill. Zohar claims that there is no plausible way to distinguish between the stationary person and the moving person. They are both what Zohar calls ‘passive threats’.

Using the model of the Israeli / Palestinian conflict, Zohar suggests that both combatants and non-combatants often pose these passive threats. Non-combatants form settlements that enable terrorists to hide in their midst. Children will grow up to be soldiers. Non-combatants form part of an expanding population in an area that doesn’t have room for both the Israelis and the Palestinians. They provide a reason for the combatants to fight, perhaps by encouraging the soldiers, or simply by being something that the soldiers wish to protect. Zohar argues that:

[I]n terms of individual contribution, it seems hard to assert any essential difference between such civilian contributions and those of non-shooting soldiers. Under ESD, once killing in ‘self-defence’ has been permitted against the innocent and extended to enemy soldiers in general, a plausible case can be made for extending it to civilians as well… We require a moral argument [for the wrongness of terrorism], but if ESD is correct, applying to innocent obstructors and hence to non-attacking soldiers, it seems to follow that siege and terrorism can in principle be justified by the same token. Since ESD allows killing innocent people insofar as they passively contribute to the threat against us, it can often plausibly apply to noncombatants as well— perhaps to the entire enemy population. If, then, we are to justify the war ethic in such a way that precludes terrorism and preserves noncombatant immunity, it cannot be by means of endorsing ESD.10

9 Zohar, ‘Innocence and Complex Threats’, p. 744.

10 Zohar, ‘Innocence and Complex Threats’, p. 748.

I think that Zohar is right that the moral status of obstructors poses a problem for many accounts of self-defence, including Thomson’s ESD.11 Briefly, I suggest that thinking about obstructors forces us to widen our conception of what it is to pose a threat to a person. It is not only those people who are going to kill me, like Falling Person, who can be said to pose a threat to me. A person can also threaten by obstructing an escape route or by conspiring with a person who is going to kill me. A useful test for identifying whether a person is a bystander or a threat is to think about whether it is possible that she play her role in a given scenario culpably. Someone who deliberately blocks my only escape route as I flee some threat to my life is liable to be killed by me, and I think that any plausible account of self-defence will want to endorse such a permission. If, however, we also want to retain the claim that bystanders are not legitimate targets, we had better not call such a person a bystander. And, since being a bystander is about causal facts, and these facts remain constant irrespective of culpability, we shouldn’t regard innocent obstructors as bystanders either.12

The problem is that more conventional ESDs of the sort Zohar has in mind cannot endorse a permission to kill culpable obstructors, since they insist that obstructors aren’t appropriately causally involved in a threat, and are thus bystanders.13 If the causal role of an obstructor is such that she cannot be described as threatening, it shouldn’t matter if she plays that role culpably. But I think it clear that this does matter, and it strongly suggests that obstructors are not bystanders, but threats.

Now, the point that Zohar will press (and it is a very good point) is as follows. If one can kill innocent threats, and a person who blocks an escape route is a threat, surely this person is a legitimate target even if she is innocent. In other words, Zohar seems to be right that an ESD cannot help but endorse a permission to kill innocent obstructors in the course of saving one’s own life. I think Zohar is right that some ESD accounts will indeed be committed to this. For example, Thomson’s account of permissible defence holds that I may use defensive force only if the target is going to violate

11 See Helen Frowe, ‘Equating Innocent Threats and Bystanders’ and Helen Frowe (2008) ‘Threats, Bystanders and Obstructors’, The Proceedings of the Aristotelian Society 108 No. 3, pp. 365–372 for discussion of this argument and its implications for permissible defence.

12 Some writers have tried to argue for a class of culpable bystanders whom, unlike innocent bystanders, it is permissible to kill. I develop several arguments against this idea in Defensive Killing, but will not pursue them here.

13 See Thomson, ‘Self-Defense’, pp. 298–299. some important right of mine. If I can kill a culpable obstructor, then, it must be because he is going to violate some right of mine. But Thomson famously argues that one can violate a right even if one is not acting (hence the permission to kill innocent Falling Person). Thus, a person who helplessly obstructs my escape can also violate my rights, and be a legitimate target.14 But it seems counter-intuitive that I could, for example, lethally trample over an innocent person as I flee something else, and Thomson herself explicitly rejects such a permission.15 And, as Zohar points out, if we think self-defence underpins the war ethic, widening our range of targets in this way will raise difficulties on the national level for non-combatant immunity. We don’t want combatants trampling over civilians simply because they are in the way.

Zohar concludes that ESDs fail as both a ground of the war ethic and as accounts of self-defence. He suggests that the correct account of selfdefence is a restricted or narrow view that does not permit the killing of the innocent. Since Zohar takes a permission to kill the innocent to be integral to the war ethic, he argues that the rules of war cannot be reduced to the rules of self-defence.

II. Defending the Reductive Strategy

Zohar’s argument presents two challenges for those who wish to defend a reductive strategy based upon an ESD account. The first is to show that what Zohar calls the ‘amplification’ of the range of defensive targets for the purposes of war is in fact mirrored on the domestic stage. In other words, I want to show that what Zohar takes to be an extension of the scope of self-defence is not an extension at all. The domestic equivalents

14 Gerald Lang has suggested to me that Thomson could respond by saying that the right the culpable obstructor violates is something like ‘a right that people not block my escape when they could get out of the way’. We might think that this has something to do with people being where they shouldn’t: the obstructor is entitled to be on the bridge when she can’t help it, but not when she can help it. I think this reply fails for two reasons. The first is Thomson’s explicit defence of the irrelevance of intention to permissibility. So, if it’s impermissible for the obstructor to be there when she intends that Victim be trapped, it’s impermissible even when she doesn’t intend it. The second is that once Thomson lets intention into the content of a right, she will have trouble explaining why one can kill innocent threats. If she allowed that what makes it permissible to kill culpable obstructors is that they intend to occupy the space in a way that is detrimental to Victim, but not innocent obstructors who have no such intention, then she needs a new argument for why it’s permissible to kill innocent threats who don’t intend to occupy the space in a way detrimental to you.

15 Thomson, ‘Self-Defense’, p. 290. of those he identifies as targets in war are legitimate targets of selfdefence. The second challenge is showing that ESDs can resist the apparent progression from non-attacking soldiers to terrorism.

I suggest that whilst Zohar is correct that obstructors are threats, he is mistaken to assume that we cannot distinguish between different sorts of threat. I will argue that by invoking a distinction between direct and indirect threats, my account can meet both the challenges that Zohar’s argument poses. However, whilst my view will not endorse terrorism, it will not afford non-combatants blanket protection. In this respect Zohar is right that a plausible account of self-defence cannot support any wholesale Principle of Non-Combatant Immunity (PNI). However, this gives us grounds to be suspicious of such a principle, rather than to reject the reductive strategy.

(i) Direct and Indirect Threats

I am not going to attempt to devise any sophisticated metaphysical account of the difference between a direct and an indirect threat, or between a cause and a causal condition. Rather, I will define a direct lethal threat as a person who is going to kill you, using a pretty commonsensical notion of what it is to kill someone. Direct threats are what we might call the ‘agent’ of harm, where the term ‘agent’ does not imply moral agency, but rather refers to ‘being the thing that harms you’. So, disease or starvation can be the agent of a person’s death. This will include as direct threats people like Falling Person, who is going to crush you to death, and also those who act through tools, like a person firing a gun, injecting a poison or driving a vehicle.16

An indirect lethal threat is a person who is not going to kill you, but whose movement or presence lessens your present chances of survival in some other way. A person who blocks your escape route counts as an indirect threat, as does a person driving the car for a drive-by shooting, and the mafia boss who hires an assassin to kill you.

On my account of self-defence, you can use proportionate force against even an innocent person if they will otherwise directly inflict upon you a harm to which you are not liable. A person is not liable to bear a harm if she has not consented to bear it, and is not party to some relevant wrongdoing. So, my account of self-defence is an extended account of the sort

16 There might be some cases in which it is unclear whether a person counts as killing you, but I will not explore those here. that Zohar objects to, because I permit an innocent person to kill an innocent direct threat like Falling Person to save her own life. However, I do not want to permit the killing of innocent indirect threats. I think that indirect threats can be killed only when they are morally responsible for posing an unjust threat.

I want to go at least some way to supporting a morally significant distinction between direct and indirect threats by suggesting that the fact that someone is on course to kill you, or to inflict serious harm upon you, is morally significant even in the absence of their having moral responsibility for that fact. It seems to me that an essential part of what it means to say that one is not liable to a harm is that one may use proportionate force against a thing that will otherwise inflict it, including those things that are themselves innocent. Since Victim is no more liable to a harm simply because it will be innocently, rather than intentionally, inflicted, he may defend himself against its infliction. The fact that Falling Person is going to kill Victim creates asymmetrical permissions between these two equally innocent characters. Victim may do to Falling Person things that Falling Person may not do to Victim. Thinking about this asymmetry undermines the idea, favoured by people like Michael Otsuka, that there is no moral difference between Falling Person and Victim, and that both have the moral status of bystanders.17

The following two cases illustrate the asymmetry I have in mind:

Tractor Beam 1: Falling Person is heading towards Victim. She will lethally crush him, but his body will cushion her fall. Victim can use his tractor gun to alter Falling Person’s trajectory. He can move her into the wall, which will paralyse her from the waist down. Given her altered trajectory, Falling Person will no longer hit Victim, but hitting the wall will break her fall so that she will not suffer any additional injury when she hits the ground.

Tractor Beam 2: Falling Person is heading towards the bottom of the well. She will be killed if she hits the well floor. She can use her tractor gun to pull Victim underneath her, where he will cushion her landing, but he will be paralysed from the waist down.

17 Michael Otsuka, ‘Killing the Innocent in Self-Defence’, p. 84–85, 87.

It seems to me clearly permissible that Victim proceed in the first case, using his tractor beam to move Falling Person so that she doesn’t kill him, even when doing so will cause her serious injury. But it seems equally clear that Falling Person may not pull Victim underneath her to cushion her fall when doing so will cause Victim serious injury. To do so would be for Falling Person to employ Victim—a bystander—as a human shield between herself and the ground. But if it is true both that (a) Falling Person and Victim are equally innocent, and (b) under such circumstances, the fact that one is about to kill the other is morally irrelevant, it is hard to explain the different permissions in the two cases. What could explain why Victim may harmfully move Falling Person to save his life, when Falling Person cannot harmfully move Victim to save her life?

Well, as I have just noted, Victim is a bystander. And it is wrong to intentionally inflict serious harm on bystanders, even in the course of saving one’s own life. But of course, to say that Victim is a bystander really amounts to nothing more than the observation that Victim does not pose a threat to Falling Person. If this is the difference that makes the difference, we seem to be saying that it matters that Falling Person will inflict harm upon Victim even when she is not responsible for this infliction.

We might say, instead, that Falling Person acts wrongly in Tractor Beam 2 because she is forcing Victim to save her, but there is no duty to save at very serious cost to oneself. But again, this says nothing more than that whilst Falling Person will be forcing Victim to save her, Victim will be forcing Falling Person not to kill him. Again, if we think this makes the difference, we are committed to saying that it matters that Falling Person is going to kill Victim. Moreover, the lack of any duty on Victim’s part to save Falling Person in TB2 can be a relevant difference only if there is a duty on Falling Person’s part not to kill Victim in TB1, such that Victim’s harmful action is justified because he is forcing Falling Person to comply with this duty. But those who deny a permission to kill innocent threats will not want to endorse the idea that Falling Person fails in her duty if she kills Victim, and so this reply doesn’t seem open to them.

The sole difference between TB1 and TB2 seems to be that in harming Falling Person, Victim is averting a greater harm to himself that Falling Person will inflict. This fact must have not insignificant moral force if it is indeed permissible for Victim to inflict very serious harm on Falling Person. Falling Person cannot invoke a similar justification for inflicting great harm upon Victim. It is not enough that she, like Victim, is aiming to prevent herself from being killed; it matters, morally, that Victim is not going to kill her. Hers is a paradigm ‘use of a bystander’ case: she aims to exploit Victim, making herself better off for his presence in a way that should strike us as impermissible. But if Falling Person, lacking responsibility for what threatens Victim, were really morally equivalent to a bystander, it would be similarly impermissible for Victim to inflict great harm upon her to save himself. I do not think that it is impermissible, and I think that this gives us pretty good grounds for thinking that being the agent of harm can be morally relevant even without moral responsibility. Thinking about this asymmetry undermines the idea that there is no moral difference between Falling Person and Victim, and that both have the moral status of bystanders.18

So, the fact that Falling Person is going to kill Victim grounds Victim’s permission to kill Falling Person. But I do not think that the asymmetry between them rules out Falling Person’s defending herself. Consider Defensive Ray Gun 1:

Defensive Ray Gun 1: Falling Person is falling towards Victim. She will kill him if she hits him. Victim powers up his ray gun in order to lethally defend himself. Falling Person has a revolver that she can use to stop Victim from killing her.

Falling Person is morally innocent, and thus on my account she is not liable to bear the harm that Victim will inflict. As I argue elsewhere, that Victim may inflict harm on Falling Person does not show that the harm itself is just—that Falling Person will not be wronged by being harmed. I therefore hold that Falling Person is permitted to defend herself against Victim.19

It might be thought that this claim undermines the idea that Falling Person may not inflict harm upon Victim in Tractor Beam 2. Is it plausible to hold that Falling Person can kill Victim when she sees him power up his ray gun to kill her, if she can’t force him to save her by using her tractor gun to move him underneath her? I think that it is, precisely because it makes a moral difference that when Victim is about to use his ray gun on Falling Person, he is about to inflict harm upon her. Now that he poses a direct threat of harm to her, and she is not liable to be killed, she may defend

18 Michael Otsuka makes the opposing claim in, ‘Killing the Innocent in Self-Defence’, p. 84–85, 87.

19 I expand and defend this view in Helen Frowe (2009) ‘The Justified Infliction of Unjust Harm’ The Proceedings of the Aristotelian Society 109 No. 3, pp. 345–351 and Helen Frowe (200), ‘A Practical Account of Self-Defence’, Law and Philosophy 29 No. 3, pp. 245–272. herself against him. I think this is also true when she realises that Victim is going to paralyse her in TB1. Again, it matters that now she will be preventing the infliction of a harm, not forcing Victim to save her (even if his body will save her if she manages to defend herself against him).

These cases suggest that inflicting harm is morally significant, even if it is innocent infliction. However, someone might object that the relevant feature here is that of posing a threat, rather than a direct threat. So, let’s take a third case, Comfy Well:

Comfy Well: Falling Person is falling towards Victim, who will be killed if she lands on him. Since this well has a nice soft floor, Falling Person would survive even without Victim’s body to cushion her. Victim could get out of the well, were it not for the fact that Obstructor is lying unconscious in the doorway. Victim powers up his ray gun, intending to kill Falling Person. Falling Person has her own ray gun. She can either

(a) kill Victim before Victim kills her, or (b) kill Obstructor, enabling Victim to move out of the way.20

In this case, both Victim and Obstructor threaten Falling Person. Victim poses a direct threat, since he is going to kill her. Obstructor poses an indirect threat, since in Obstructor’s absence, Victim would have no need to lethally defend himself against Falling Person. She is thus worse off for Obstructor’s presence. But it seems very implausible to me to think that Falling Person may kill Obstructor rather than Victim. My view explains why this is. If Falling Person is not liable to bear the harm that Victim will inflict, it follows on my account that she may defend herself against its infliction. This means that she may take option (a), killing Victim before he kills her, but she may not take option (b), killing Obstructor to enable Victim to move. Such a killing would not be defensive, but preservative.

My position holds, then, that one can kill even an innocent direct threat, because inflicting unjust harm is morally significant even without agency. But when a person threatens only indirectly, Victim needs additional justification for killing her. The primary justification for this will, I suggest, be

20 The cushioned floor matters because this means that Obstructor indirectly threatens both Falling Person and Victim. If the well had a hard floor, Falling Person would be better off for Obstructor’s presence, because it is only Obstructor’s presence that keeps Victim in the well where he will cushion her landing. She is worse off for his presence when the well is cushioned, because now Victim cannot move, and thus she threatens Victim’s life. moral responsibility on the part of the indirect threat for unjustly endangering Victim’s life.21

(ii) Responsible Indirect Threats and Reasonable Opportunities

I take a person to be morally responsible for endangering you if she intentionally fails to avail herself of a reasonable opportunity to avoid posing an unjust threat to you. I used to think that an opportunity was reasonable provided that it didn’t require the agent to bear a harm comparable to that facing Victim.22 If, for example, a person who blocks your escape route from some lethal threat could get out of your way, but refuses to do so because this will mean getting her feet wet, she is a responsible indirect threat. She had a reasonable opportunity to avoid endangering you, and she intentionally failed to take it. But if she can avoid blocking your escape route only by throwing herself off a bridge into the ravine below, that would not count as a reasonable opportunity, and she would not count as a responsible threat if she decided not to take it.

But this seems to conflict with our intuitions about the role of duress in excusing wrongdoing. Take the Lynch case of 1975.23 Lynch was forced under pain of death to drive some IRA members to the shooting of a policeman. Lynch was originally tried and convicted of murder along with the IRA members, because duress is not a legal defence to homicide. I think it plausible that Lynch ought not to have been punished (although we should be reluctant to say that he acted rightly).24 But the question that I am interested in is not whether Lynch should be punished or blamed, but whether the policeman whom the IRA went to kill would have been permitted to shoot Lynch if doing so was the only way in which he could have saved himself. Imagine that whilst he could have had no hope of defending himself against all three IRA shooters, he could have sent the whole car off the road if he had shot Lynch, thus saving his life. Would this have been permissible? I think it would. If I am right, things are more complicated than we might have thought. It looks like an indirect threat can render herself liable to be killed even if the alternative involves a comparable harm. Perhaps, then, an indirectly threatening person is not liable to be killed only if refusing to endanger someone else would require her to

21 The other obvious candidate is numbers.

22 See Helen Frowe, ‘Equating Innocent Threats and Bystanders’.

23 DPP for Northern Ireland vs. Lynch, 1975 AC 653.

24 In a similar case, Abbot vs. R (1977, AC 775), the defendant was the principal killer, and received a life sentence for murder. bear a greater harm than that to which she will expose them. Had the IRA been going to just beat the policeman up, for example, it would not have been permissible for the policeman to kill Lynch, given the comparatively large cost that Lynch would have faced had he refused to endanger the policeman.

We might go some way to drawing a distinction between Lynch and the person who must throw herself off the bridge into the ravine if we think that there is a morally relevant difference between requiring that a person refuse to endanger Victim even if someone else will then harm her, and requiring that a person harm herself in order to avoid endangering Victim. It seems to me psychologically harder, and therefore more demanding, to require that, for example, Lynch drive the car off a cliff himself than to say that he ought to refuse to drive the car even if the IRA will shoot him (although of course, this too is very demanding).25 However, I’m not sure that this will greatly affect what I want to say about indirect threats in war, so we can perhaps set it aside for now.

There is a further question of whether or not Lynch acted impermissibly. I return to this in the discussion of civilian liability below.

III. Indirect Threats and Self-Defence

So how does this help me to resist Zohar’s objections to a reductive strategy based on an ESD? Well, let’s go back to the two problems that Zohar’s challenge raises. The first is whether the range of legitimate targets in war, including those who threaten more abstractly, like administrators and other non-attacking soldiers, can be plausibly mimicked on the domestic stage in a way that supports the reductive strategy. Is Zohar right that such people could not count as threats in a domestic context, much less as legitimate targets?

Imagine that a particularly well-organised branch of the mafia have set up offices, and employed various people to assist them in killing off people

25 Kamm makes a related point in her discussion of killing innocent threats in Creation and Abortion, arguing that whereas Falling Person ought to redirect herself if doing so will cause her harm (a broken leg), she would not be required to fatally redirect herself: “it is not appropriate to require an innocent threat to do this. Because she is morally innocent of threatening her victim, she does not owe it to him to give up her life to save him” (p. 48). But, Kamm also suggests that others would be permitted to lethally redirect Falling Person to save Victim from great harm: “Suppose that the innocent threat would paralyze her victim and could be stopped only by being killed. Although the threat would not need to press her button and fatally redirect herself… I believe that a third party may permissibly kill” (p. 54). who stand in the way of their illegal and unjust activities. They track the movements of potential targets, establishing their routines, translating their phone calls and monitoring police radios to enable the shooters to make their getaway. All of these people seem like legitimate targets of defence to me. If, somehow, shooting one of them would enable the intended target of a hit to save his life, and there is no less harmful way in which he can do this, then the fact that a person works in an office rather than driving the getaway car does not grant her immunity. There is no principled difference between self-defence and national defence here.26 There are contingent differences about the likelihood of being in this sort of situation. But thinking about this analogous, non-military setting undermines Zohar’s claim that in such a scenario, these people would not be seen as threats at all.

The same seems true of those ‘passive threats’ whom Zohar describes as threatening ‘simply by being there’, perhaps by setting up camp in a way that forces the opposing army to travel across more dangerous territory. A person who intentionally lessens the survival chances of another person in this way would be a legitimate target of self-defence, and this explains why such ‘non-attacking’ soldiers on the unjust side of the war are legitimate targets of national defence. They are indirect threats whom, given their responsibility for the unjust threat that they pose, it is permissible to kill to save one’s own life.27 So, I think that my version of the ESD applies across both war and self-defence in a way that is compatible with, and supports, the reductive strategy.28

26 Remember, that I have in mind administrative staff working on the unjust side of a war, and thus we cannot appeal to the legality of the non-attacking soldier’s actions compared to the illegality of the mafia office staff. Unjust wars of aggression are no more legal than mafia hits.

27 I’m assuming here that all those who sign up for the armed forces are morally responsible for doing so.

28 Of course, all of this takes place against the background assumption that the threatened person or group is not liable to be killed. Given this, my account will not support what Zohar identifies as a crucial tenet of the war ethic: the mutual permission to kill between opposing combatants. A person may kill an innocent person if she will otherwise inflict a harm to which he is not liable. But he may not use force to avoid a harm if he is liable to bear it: as I suggested, this is the most important feature of liability to harm. But I don’t think that this is problematic. A proponent of an ESD need not accept Zohar’s claim that many or all unjust combatants are morally innocent, and that a battle is thus a bunch of innocent threats killing each other. As mentioned above, it is also hard to see how Thomson’s account of self-defence could be thought to support the moral equivalence of combatants. She argues that one may kill an innocent person in self-defence if, and only if, they will otherwise violate your right not to be killed, or some other comparable right. If combatant A may kill combatant B, it must be because B will violate his rights. But if this is true, B has no right not to be killed by A that he can lethally defend.

IV. Culpability in War

As Zohar points out, however, getting the permission to kill in war to cover non-attacking soldiers is supposed to be only half the battle. The other half is making sure that it doesn’t cover non-combatants as well. But if, as Zohar plausibly argues, lots of non-combatants make comparable contributions to the war effort, how can my account restrict the permission of defensive killing to only enemy soldiers?

The short, and probably unwelcome answer, is that it often won’t. Zohar is right that once we recognise that many people indirectly threaten, the range of permissible targets becomes much wider. Lots of civilians will count as indirect threats: those manufacturing weapons, those working in code-breaking or as translators. Imagine that the political leader waging the war won his election campaign on the promise of waging this war. Aren’t the people who voted for him, and gave him political power, responsible in no small part for the war? It seems to me that they are, and that therefore it matters enormously whether or not the war is just. If it is a just war of defence against aggression that these people support, they do not act wrongly and as innocent indirect threats they are not legitimate targets. Nor are their frontline combatants legitimate targets. Endangering a person or group that is liable to be killed does not render one a permissible defensive target.

But if a country is engaged in an unjust war of aggression, contributing to the war effort might well render a civilian a legitimate target if her support lessens the survival chances of innocent combatants and civilians in another country. A person’s contribution may be indirect, but much like the person who gathers information for the mafia, this fact cannot grant her immunity when she is morally responsible for the threat that she poses. It is now that it starts to matter a lot what we think counts as being morally responsible for the fact that one poses a threat. Is a person who, for example, faces imprisonment if she does not join her county’s war effort liable to be killed if she succumbs to the duress and contributes (assuming that the harm that she will help inflict is greater than that she will suffer)?

I think that the answer to this question has to be yes, with two important caveats. Before I outline the caveats, let me explain two possible ways in which the ‘yes’ could be interpreted. We could take what I call the Hard Line, which is to say that one acts impermissibly when one intentionally exposes others to risk unless one would have to bear a greater harm oneself to avoid so exposing them. Since one has acted wrongly, one is liable to be killed. Or we could take what I call the Soft Line, which is to say that it is permissible for people to decide not to bear significant costs to avoid endangering someone else. However, when they decide to expose the person to this risk, they become liable to be killed if this is necessary for the other person to save their life.

I have in mind here something like what Jeff McMahan says about driving.29 If a careful, conscientious driver decides to drive down the street, he knows that he exposes pedestrians to a small risk of great harm. His doing so is permissible: it is not always wrong to expose others to even lethal risks. But since he is responsible for creating the risk, it is fair that he bear the costs if the risk eventuates in harm. Should he lose control of the car and career towards a pedestrian, she may zap the car with her ray gun to save herself. We might say that in choosing to drive, the driver consents to bear any resulting harm if it becomes necessary that someone bear it.

I am sympathetic towards the Soft Line when the costs of refusing to indirectly threaten are sufficiently high. It seems pretty demanding to require that, for example, a person in Nazi Germany not join the Nazi party even when this is the only way that he can get a job and feed his family. But it also seems very demanding to insist that just combatants cannot kill such a person if it turns out that doing so is necessary to defend their lives, and the lives of others. We might want to say something like this about the Lynch case that I discussed earlier. Perhaps Lynch does not act impermissibly in acquiescing to the IRA’s demands when the cost to him of refusing is so high. This might explain our reluctance to blame or punish him. But, as in McMahan’s case, even permissible activity can render one liable to be killed. Again, it seems very demanding to insist that the policeman cannot shoot Lynch to save his own life.

The first caveat that attaches to finding such civilians liable, then, is to remember that again, what we’re interested in here is liability to defensive killing, and not blame or punishment. The second caveat is to remember that killings of this sort, like all defensive killings, will be permissible only if they are necessary. Part of a harm’s being necessary is that it is the least harmful means available of averting the threat to oneself. There will probably be comparatively few instances in which killing an indirectly threatening civilian, at least one posing a relatively minor indirect threat, will be the least harmful way of defending oneself. A related point is that part

29 Jeff McMahan, ‘The Basis of Moral Liability to Defensive Killing’, Philosophical Issues

15, Normativity, (2005), p. 393–394. of the justification for killing in war is that one gains a military advantage as a result of inflicting damage or harm.30 This is particularly important when we think about civilian contributions. There might be many people who, on my account, have rendered themselves liable to be killed, but whom it would afford one no advantage to kill, perhaps because they will be easily replaced, or because their contribution is very small. Again, I suspect that actually, there will be comparatively few cases in which killing a person who poses a non-military, indirect threat will meet this condition.31

But when they do, such killings will not be terrorism. Zohar describes terrorism as the “purposeful slaying of the innocent”.32 But on my account, only those indirectly threatening people who are morally responsible for the fact that they pose a threat will be legitimate targets. As Michael Walzer points out, terrorism is about attacking people because of who they are, not what they have done.33 What I am proposing is that a civilian can, through her actions, render herself liable to be killed. I am not suggesting that a civilian might be a legitimate target simply because, as Zohar claims, of ‘who and where she is’. Killing a person because she is morally responsible for contributing to an unjust threat is not an act of terrorism. Moreover, many of the sorts of threat that Zohar claims noncombatants might pose—being a reason why the army fights, or being a person who might grow up to be a soldier, being born in contested territory—are just not the sorts of threat that one can pose responsibly. So Zohar’s claim that using an extended account of self-defence to underpin the war ethic will end up labelling the ‘entire enemy population’ fair game seems unfounded. If we can recognise a distinction between direct and indirect threats, an extended account of self-defence will have the resources to cover the range of permissible wartime targets without

30 Suzanne Uniacke has recently argued that the ad bellum requirement that one have a reasonable prospect of success is disanalogous to self-defence, where there is no such requirement ((2009), ‘Self-Defence, Necessary Force and a Reasonable Prospect of Success’, Unpublished Manuscript). We might think that the requirement of gaining military advantage is the in bello cousin of the reasonable prospect of success requirement. I think that there are interesting questions about how this informs our understanding of the relationship between war and self-defence, but I cannot address those here.

31 Note, then, that I reject McMahan’s view that liability has an internal necessity condition, such that one can be liable to a harm only if that harm will be instrumental in averting a threat for which one bears moral responsibility (see, for example, Jeff McMahan (2009) Killing in War, (Oxford, OUP) p. 225). On my account, whether or not a person is liable to a harm depends on what they have done, not on the usefulness of killing them.

32 Zohar, ‘Innocence and Complex Threats’, p. 735.

33 Walzer (1997), Just and Unjust Wars, (New York: Basic Books), p. 200. endorsing terrorism. But, our understanding of permissible wartime targets might well have to change in light of our observations about permissible self-defence.

Conclusion

I have argued that accounts of self-defence that permit the killing of innocent threats can be used to underpin the rules of war. Zohar is correct that existing accounts of self-defence are mistaken to conceive of obstructors as bystanders. Obstructors are threats. However, he is mistaken to think that we cannot distinguish between different sorts of threat. I have argued that the fact that a person is going to kill me is morally significant even intheabsenceof theirhavingmoralresponsibilityforthatfact. Recognising this gives us a way to morally differentiate direct and indirect threats. I then suggested that indirect threats become liable to be killed if they are morally responsible for the fact that they pose a threat, and that this applies both on an individual level and during war. Many non-combatants indirectly threaten, and some are morally responsible for doing so. My account does indeed undermine the idea of a universal Principle of NonCombatant Immunity. But it does not undo our grounds for condemning terrorism, because the defensive killing of a person who is morally responsible for an unjust indirect threat is not terrorism.

Partiality and Weighing Harm to Non-combatants

David Lefkowitz

Suppose that under certain conditions, combatants waging a just war are morally permitted to engage in acts that cause collateral damage; that is, harm done to non-combatants as a side effect of an attack on a morally permissible target.1 Does it matter morally whether those non-combatants who will be harmed are citizens of the same state as the combatants who carry out the attack, or are instead citizens of the state against whom these combatants are waging war? Frances Myrna Kamm and Thomas Hurka each claim that it does.2 They argue that combatants ought to be partial to their compatriot non-combatants, assigning greater weight to harm done to them than they assign to harm done to enemy non-combatants. This entails that in some cases whether a collateral damage causing act of war satisfies the jus in bello criterion of proportionality depends on the citizenship of the non-combatants harmed by it.3 Moreover, Kamm suggests that combatants ought to assign even greater weight to harm done to neutral non-combatants than they assign to their own compatriot noncombatants. In some cases, and holding all else equal, a combatant ought to carry out an attack that will cause a greater number of compatriot

1 For purposes of this article, I will assume that no non-combatant is morally liable to attack as part of the conduct of a just war; combatants are not morally permitted to deliberately or negligently harm them. Both Kamm and Hurka note (rightly, in my view) that degree of moral culpability for an unjust war may provide a superior criterion for distinguishing those who are morally liable to attack from those who are not than does the combatant/non-combatant distinction. But since they carry out their discussion in terms of the latter distinction, I do so here as well.

For discussion of various attempts to provide a moral justification of acts of war that cause collateral damage, see D. Lefkowitz, ‘Collateral Damage’, in D. Lefkowitz, ‘Collateral Damage’, in L. May (ed.), War: Essays in Political Philosophy (Cambridge: Cambridge University Press, 2008), pp. 145–64.

2 F.M. Kamm, ‘Terror and Collateral Damage: Are They Permissible?’ Journal of Ethics 9 (2005): 381–401; F.M. Kamm, ‘Failures of Just War Theory: Terror, Harm, and Justice’, Ethics 114 (July 2004): 650–92; Thomas Hurka, ‘Proportionality in the Morality of War’, Philosophy and Public Affairs 33.1 (2005): 34–66.

3 Throughout this article, I write as if harm to non-combatants is the sole factor to be taken into account when assessing whether a given act of war meets the jus in bello criterion of proportionality. In fact, other factors, such as the destruction of infrastructure or harm to the natural environment, may also be relevant. But whether there are other relevant factors, and whether their relevance is simply a matter of the effect that harm to them will have on people’s well-being, makes no difference to the issues under examination here. non-combatant deaths as collateral damage, rather than one that will collaterally kill a lesser number of neutral non-combatants. As Kamm puts it, ‘non-combatants in one country have a higher degree of inviolability than those in another country’.4 In what follows, however, I will demonstrate that neither Kamm nor Hurka provide a plausible justification for their conclusions. Absent further argument, combatants ought to treat all non-combatants impartially, weighing harm to them equally regardless of their citizenship.

I begin by describing several ways in which partiality to compatriot non-combatants affects the moral permissibility of acts that cause collateral damage. I then refute Kamm’s attempt to defend partiality to compatriot non-combatants by appeal to an alleged asymmetry between the sacrifice one must make to save others, and the sacrifice one may impose on a third party in order to save those same people. Next I focus on both Kamm’s and Hurka’s efforts to defend partiality to compatriot noncombatants by appeal to an analogy with parents’ justifiable partiality toward their children. I contend that Kamm and Hurka employ the wrong analogies, and that the proper ones generate less or no intuitive support for the conclusion they seek to defend. Finally, I argue that Kamm fails to provide a compelling reason to believe that neutral non-combatants enjoy a higher degree of inviolability than do compatriot or enemy noncombatants. Nevertheless, I suggest that given a certain understanding of the nature and moral value of state sovereignty, it is possible to provide a non-instrumental justification for Kamm’s claim that given a choice between causing a greater number of compatriot non-combatant deaths and a lesser number of neutral deaths, a combatant ought to choose the former over the latter.

1. Partiality and Jus in Bello Proportionality

Practically all contributors to the just war tradition argue that only those acts of war that meet the criteria of discrimination and proportionality are morally permissible. Discrimination requires those who carry out military operations to distinguish between combatants, whom they are morally permitted to target, and non-combatants, whom they may not target, and whom they should make reasonable efforts to avoid harming. Discrimination does not absolutely prohibit engaging in acts of war that

4 Kamm, ‘Failures of Just War Theory’, p. 673. cause harm to non-combatants, however. Many just war theorists contend that attacks on legitimate targets that also involve or result in unintended harm to non-combatants may still be morally permissible if the harm done is proportional to the good achieved by that act.5

Though proportionality has applications in just war theory other than to acts that inflict collateral damage, it is with these cases that I am concerned here, for it is the moral permissibility of such acts that is affected if partiality to compatriot non-combatants is morally justifiable. Hurka argues that such partiality warrants engaging in acts of war (and so, ultimately, an entire war) even if it is foreseen that the number of enemy noncombatants killed as collateral damage will exceed the number of compatriot non-combatants whose deaths the war prevents. Thus with respect to military attacks undertaken by the United States and Israel to prevent terrorist attacks against them (or their citizens), Hurka contends that US and Israeli combatants ought to assign less weight to the harm done to Pakistani, Afghan, or Palestinian non-combatants killed as a side effect than we should assign to the harm prevented to Israeli or American non-combatants.6 If Hurka is right about this, then the claim that Israel’s 2006 attacks on Hizbullah forces in Lebanon caused a disproportionate amount of collateral damage may not be as obvious as numerous commentators around the world have thought.7

Kamm claims that in a case where a combatant waging a just war can achieve the same good by conducting one of two aerial bombardments,

5 Two additional criteria must also be met. First, the act the agent does intentionally, and which causes the unintended harm, must be one he is morally permitted to do, and second, the unintentional harm must be no more than is necessary to achieve the good end. Together with the requirements that the harm to non-combatants be unintended and proportional, these criteria constitute the doctrine of double effect. Only the proportionality criterion will be at issue here; I will assume that all of the examples under discussion satisfy the other conditions for the doctrine of double effect’s applicability—or whatever alternative principle justifies collateral damage.

With respect to justifying collateral damage, Kamm rejects the doctrine of double effect in favor of what she labels the principle of permissible harm. See F. M. Kamm, Morality, Mortality Volume II (New York: Oxford University Press, 1996), pp. 143–204. Exactly what justifies acts of war that cause collateral damage is unimportant for the issue under discussion in this article, since Kamm, Hurka, and I all argue under the assumption that some principle does. See Kamm, ‘Failures of Just War Theory’, p. 673; Hurka, ‘Proportionality’, p. 61.

6 Hurka, ‘Proportionality’, pp. 59–60.

7 See, for example, ‘Too High a Price’, The Nation, 14 July 2006; Larry Derfner, ‘Flirting with State Terrorism’, The Jerusalem Post, 17 August 2006. Hurka does not believe Israel’s 2006 invasion of Lebanon met the jus in bello proportionality condition (personal communication). both of which qualify as permissible, but where bombardment A will result in the death of 100 compatriot non-combatants, while bombardment B will result in the death of 200 enemy non-combatants, the combatant ought to choose B over A.8 Though Kamm’s example may seem artificial, it is not difficult to conceive of an actual scenario with the same essential features. For example, members of the Free French Air Force during World War II might have faced a choice between bombing a German munitions factory located in occupied France, or one in Germany proper, with the expectation that attacking the latter would result in twice as many non-combatant casualties as attacking the former. According to Kamm, the Free French would not only have been permitted to bomb the factory located in Germany, they would have acted incorrectly had they bombed the French plant instead, even though doing so likely would have produced fewer non-combatant casualties.

Partiality to compatriot non-combatants also permits warriors to bear less risk of harm to themselves at the expense of greater risk of harm to non-combatants if those non-combatants are enemy ones rather than compatriots. As Michael Walzer writes, if saving [non-combatant’s] lives means risking soldier’s lives, the risk must be accepted. But there is a limit to the risks that we require. These are, after all, unintended deaths and legitimate military operations, and the absolute rule against attacking [harming?] civilians does not apply. War necessarily places civilians in danger; that is another aspect of its hellishness. We can only ask soldiers to minimize the dangers they impose.9

The idea that soldiers ought to seek to minimize the risk of harm to noncombatants, even at the cost of increasing their own chances of suffering harm, suggests that combatants’ lives and/or well-being have value proportionate to that of non-combatants’ lives and/or well-being.10 If combatants ought to assign greater importance to the lives of compatriot

8 Kamm, ‘Failures of Just War Theory’, p. 672; ‘Terror and Collateral Damage’, p. 397. It would beg the question against Kamm to assert that, given the possibility of carrying out bombardment A, bombardment B violates the no more harm than necessary condition on the applicability of the doctrine of double effect.

9 Michael Walzer, Just and Unjust Wars (New York: Basic Books, 3rd edn, 2000 [1977]), p. 156. Walzer claims that ‘the degree of risk that is permissible is going to vary with the nature of the target, the urgency of the moment, the available technology, and so on’ (ibid., 156). As will become clear below, however, Walzer does not believe that the nationality or citizenship of the at-risk non-combatants affects the degree of risk combatants may permissibly impose upon them. See n. 27.

10 Seealso Kamm,‘Failuresof JustWarTheory’ and Kamm,‘Terrorand Collateral Damage’. non-combatants than they do to enemy ones, it follows that the number of enemy non-combatant lives ‘equal’ to the life of a combatant will be less than the number of compatriot non-combatant lives ‘equal’ to the life of a combatant. Assuming that risk to the combatant varies inversely with the likelihood of harm to non-combatants, it follows that combatants may bear less risk of harm to themselves when carrying out an attack that will cause collateral damage to enemy non-combatants than when carrying out one that will collaterally harm compatriot non-combatants. Thus were the French pilots to attack the munitions factory in Germany, they would be morally permitted to fly higher, and so likely inflict more harm on noncombatants, than they would be permitted to fly were they to attack the plant in occupied France.11

As these examples indicate, if partiality to compatriot non-combatants were morally justifiable, it would significantly shape the form that the morally permissible conduct of war may, or must, take. As I will demonstrate, however, neither Kamm nor Hurka provides a plausible defense of such partiality.12

1. Self/Other Asymmetry and Partiality to Compatriot Non-Combatants

One argument Kamm offers to buttress her claim that combatants ought to be partial to compatriot non-combatants rests on the alleged permissibility of imposing on a third party the cost of saving others where one is not morally required to bear the same cost in order to achieve the same end. Call this permissible self-other asymmetry. Kamm contends that an agent who confronts a scenario in which she can prevent a trolley from killing five people only by directing it down either a second track, on which she stands, or a third track, on which one (or two) strangers stand, may choose to direct it down the third track, rather than the second one (or flipping a coin). Individuals enjoy ‘a prerogative not to make big sacrifices in order to aid others’.13 Kamm then draws an analogy between this

11 As Pierre Mendes-France, a Free French pilot, writes: ‘It was … this persistent question of bombing France itself which led us to specialize more and more in precision bombing—that is, flying at a very low altitude. It was more risky, but it also permitted greater precision …’ (quoted in Walzer, Just and Unjust Wars, p. 157).

12 Note that I will not be arguing that partiality to compatriots is never morally justifiable, but only that neither Kamm nor Hurka demonstrate that combatants ought to be partial to their compatriot non-combatants when choosing between various collateral damage causing acts of war.

13 Kamm, ‘Failures of Just War Theory’, p. 674. case, and the permissible self-other asymmetry she claims applies in it, and the example of a state (or state official, such as a bomber pilot) faced with a choice between carrying out bombardment A, which will kill 100 compatriot non-combatants, and bombardment B, which will kill 200 enemy non-combatants. She writes, one may conceive of the situation as one in which country A must decide whether to harm itself or harm someone else, both options involving (assumed) permissible ways to harm non-combatants. It may simply be supererogatory … to direct a threat to oneself rather than send it to someone else to whom it is also permissible to send it.14

Unfortunately for Kamm, several disanalogies between the two cases she considers undermine her attempt to extend the application of permissible self/other asymmetry from the above variation on the trolley problem to the bombing example.15

First, Kamm’s argument ignores the difference between the threat facing the agent on the second track, and the threat facing the state if it opts to carry out bombardment A. Choosing to direct the trolley down the second track will result in an enormous loss to the agent who stands on it, namely her death; it is the ‘ultimate’ nature of the sacrifice that likely leads many people to form the intuition that the agent is permitted to direct the trolley down track three rather than track two. The potential loss to the state in the bombardment example, however, is of nowhere near the same magnitude. In practically every scenario, the state will continue to exist even if the bomber pilot chooses bombardment A over bombardment

B. Nor is it clear that harm to some of the state’s subjects will necessarily harm the state. It is conceivable, at least, that the individuals the bombing will kill are engaged in acts that interfere with the proper functioning of the state, such that their deaths actually benefit the state, rather than harming it.16

Second, Kamm describes the trolley problem as an instance of deciding whom to save from harm. But that is not the question the bomber faces;

14 Ibid.

15 For the sake of argument, I accept permissible self/other asymmetry in the trolley problem, though in fact I am skeptical of it even in this case.

16 The reader might take this objection to Kamm’s argument, together with some of the criticisms set out later in this article, as reasons to reject the use of domestic analogies as a method for defending particular moral principles for the just conduct of war. I do not believe the wholesale rejection of this method is necessary, though as the text indicates, careful attention must be paid to specific disanalogies between domestic and war scenarios that may have morally important implications. his choice is a matter of deciding whom to harm, not whom to save from an independently originating harm. Even if we assume that either bombardment is morally permissible, it could be that partiality to compatriots is morally justifiable in the case of preventing harm, but never in the case of causing it. Given that many moral theorists think there is an important moral distinction between causing harm and giving aid, this discrepancy between the two cases Kamm considers should raise a red flag regarding her extension of a moral principle from one to the other.17

A third disanalogy between Kamm’s two examples involves the moral conclusions she draws in them. In the trolley problem, Kamm defends permissible self/other asymmetry; agents are permitted, but not required, to forgo making big sacrifices in order to aid others. In the bombing example, however, Kamm claims that it would be incorrect to carry out bombardment A rather than bombardment B, and also speaks of a ‘duty not to harm one’s own non-combatants rather than non-citizens, if someone must be harmed in a permissible manner’.18 Permissible self/other asymmetry cannot give rise to a duty to be partial, nor can it entail that an agent who elects not to be partial acts incorrectly. Thus even if the two cases Kamm considers are sufficiently analogous to warrant the extension of the principle of permissible self/other asymmetry to the bombing example, that principle does not justify the conclusion Kamm draws.

1. Parental Partiality and Partiality to Compatriot Non-Combatants

Kamm’s second argument in defense of partiality to compatriot noncombatants also takes the form of an argument by analogy. She begins with the assumption that it is generally permissible for an agent to direct a runaway trolley away from a track with five people on it and toward a track occupied by only one person. However, Kamm contends that if the one person is the agent’s child, then the agent has a special duty to the child not to direct the trolley down the track he occupies. Similarly in the bombing example, both bombardment A and bombardment B are of a type assumed to be morally permissible. But just as the parent owes a

17 It should be noted, however, that Kamm is unlikely to find in this disanalogy a compelling objection to her argument, since she does not think the distinction between causing harm and giving aid a morally relevant one. See her principle of permissible harm, which she defends as the true governing principle for cases like the trolley problem, the munitions factory, and so on (and on). Kamm, Morality, Mortality Volume II.

18 Kamm, ‘Failures of Just War Theory’, p. 674, italics mine. special duty to his child, so too the bomber pilot owes a special duty to his compatriot non-combatants, one that requires him to give their lives and well-being greater weight than he assigns to the lives of enemy noncombatants. Thus it is the existence of a special duty to compatriots that requires the pilot to choose bombardment B, and the resulting death of 200 enemy non-combatants as collateral damage, over bombardment A, and the 100 compatriot non-combatant collateral deaths it will cause.

Hurka also rests his case for partiality to compatriot non-combatants on an analogy to special duties parents owe their children. He describes a scenario in which the only way for an agent to save an innocent victim from a fatal attack by an aggressor involves the use of a grenade that will kill the aggressor, but also kill an innocent bystander. Hurka concedes that it is arguable whether the agent may throw the grenade in this scenario, sacrificing one innocent person’s life (albeit unintentionally) for another’s. He draws a different conclusion, however, if the innocent victim threatened by the aggressor is the child of the agent in position to carry out the rescue.

It seems to me that he [the father] may throw the grenade, and may do so even if this will kill some number of bystanders greater than one. If he is not aiming at the bystanders but killing them collaterally, he may show some preference for his daughter.19

Hurka suggests that a combatant may display an analogous partiality by unintentionally killing a greater number of innocent enemy noncombatants in order to prevent the killing of a lesser number of his compatriot non-combatants.20

Both Kamm’s and Hurka’s arguments depend on the extension of an intuition from a non-war scenario to a war one.21 I suggest, however, that they employ the wrong non-war scenarios. Take Kamm’s argument: if the question concerns the extent of the partiality compatriots may justifiably

19 Hurka, ‘Proportionality’, p. 61.

20 Hurka’s use of the term ‘may’ implies a permission to give greater weight to the lives of compatriot non-combatants over enemy non-combatants, but in an earlier article where he develops a general account of partiality to compatriots, he talks of duties of partiality (Thomas Hurka, ‘The Justification of National Partiality’, in Robert McKim and Jeff McMahan (eds), The Morality of Nationalism [New York: Oxford University Press, 1997], p. 152).

21 They also depend, of course, on the transfer of an intuition regarding parents’ obligations to their children to compatriots’ obligations to one another. This transfer may also be challenged, though I will not do so here. See Christopher Heath Wellman, ‘Associative Allegiances and Political Obligations’, Social Theory and Practice 23 (1997): 181–204. display to one another, then a more closely analogous non-war scenario involves the following variation on the trolley problem. Suppose that a runaway trolley will run over and kill either one person who is a compatriot, or five people who are foreigners, but who are alike in all other morally relevant respects. Are you permitted, or even required, to direct the trolley toward the five people, rather than the one? The intuition that you may, or must, do so would provide the kind of intuition in a non-war scenario that would best support partiality in a case like the bombing example, since it more closely mirrors the choice the bomber pilot confronts than does Kamm’s parent/child variation. For what it is worth, my intuition is that the compatriot relationship does not justify directing the trolley so that it kills the five foreigners rather than the one compatriot. Interestingly, David Miller, one of the most prominent defenders of partiality to compatriots, agrees. He writes:

If we think about cases modeled on the trolley problem made famous by Judith Thomson, I don’t think it would be justifiable to switch the trolley from a track on which it was hurtling towards a compatriot on to a track on which it would hurtle towards a foreigner. Nor do I think, if one takes the view that when the difference between the numbers on the two tracks becomes large enough, one ought to switch the trolley, that there should be any additional weighting in favor of compatriots. If one should switch the trolley to kill one in order to save ten, then the identity of the ten and the one is irrelevant. At this level, morality appears to me to require strict impartiality at least as far as nationality is concerned.22

At a minimum, then, employing the more closely analogous variation on the trolley problem weakens the argument for partiality to compatriot noncombatants, and it may well undermine it altogether.23

As for Hurka’s non-war scenario, a variation on it more closely analogous to choosing between harm to compatriot or to enemy non-combatants would involve throwing a grenade that will kill one or even several foreigners in order to save one compatriot. I contend that such an act is not

22 David Miller, ‘Reasonable Partiality Towards Compatriots’, Ethical Theory and Moral Practice 8 (2005): 75.

23 Exactly when, and with what detail, to pose this variation on the trolley problem raises some difficult questions regarding the weight we ought to assign to people’s intuitions, and what counts as a genuinely considered judgment. For example, asking current citizens of the United States whether it would be permissible or required to direct the trolley at five Iraqis rather than one American might elicit a different response than if ‘Irishmen’ or ‘Canadians’ were substituted for ‘Iraqis’. It may also make a difference if the question is posed while a state is at war, since feelings of nationalism (compatriotism) tend to be highest at such times. morally justifiable. The moral duty not to kill an innocent person is commonly thought to be stronger than the moral duty to save an innocent person from being killed. It follows, then, that holding all else equal, the duty a combatant has to her compatriots not to kill them is stronger than the duty she has to save them from being killed by a third party. Supposing that to be the case, if it is not justifiable to kill a greater number of enemy non-combatants in order to avoid killing a lesser number of compatriot non-combatants, as I maintain against Kamm, then surely it is not justifiable to kill a greater number of enemy non-combatants in order to save a lesser number of compatriot non-combatants from being killed by a third party.

Note, too, that Hurka’s original grenade throwing example seems closely akin to (if not an example of) cases of duress that practically all theorists agree provide an excuse, but not a justification, for an agent’s conduct. I am slightly more sympathetic to the claim that agents may point to the fact of common membership in a political community as a partial excuse, rather than justification, for giving greater weight to the lives of compatriot non-combatants in comparison to enemy ones. I suspect, however, that whatever it is about being a parent, or the parent–child relationship, that provides the basis for an excuse will be realized to a significantly weaker degree in the case of compatriots, and so the degree of exculpation it provides will be far less.

It may be that the compatriot/foreigner variation on the trolley problem, while better than the one Kamm employs, still contains an important disanalogy to the kind of wartime scenarios that are of concern in this article. The actors on whom we focus in the latter scenarios are state officials. Therefore, in order to maintain the analogy, we ought to consider whether a state official is permitted, or even required, to turn the trolley toward five foreigners rather than one subject of her state, or whether in the above variation on Hurka’s example, the addition of the fact that the agent in position to affect a rescue is a state official makes any difference to the intuition we form. I suggest that the answer is no in both cases. Surely we do not think a New York City policeman would be justified in killing three innocent German tourists, even unintentionally, in order to save one innocent American. Suppose, however, that I am wrong about this, and that some of those who consider this last variation on Hurka’s example act reasonably when they conclude that the official should save the American, though in doing so he will unintentionally kill the Germans. Their intuition may follow not from the belief that partiality to compatriots is morally justifiable in this case, but rather from certain views about what follows from the agent’s status as a state official. It is at least conceivable that an agent might occupy a state office without himself being a citizen of that state; indeed, in countries where military service is a path to citizenship, a significant number of combatants may not bear the relationship of compatriot to non-combatant citizens of that state.24 The question would then be why occupying the role of state official permits or requires that, when he engages in a permissible act that will cause collateral damage, the agent should give greater weight to avoiding harm to citizens of the state he serves than he gives to avoiding harm to foreigners. One reason to think that being a state official makes no difference in this case is Thomas Pogge’s observation that it would be bizarre if agents could circumvent moral prohibitions on certain types of acts simply by having officials do those acts for them.25 If I am right to think that an individual may not appeal to the compatriot relationship to justify throwing a grenade that kills three innocent foreigners as a side effect of saving one compatriot, then it should not be permissible to circumvent this prohibition by having a state official carry out the same act.

Both Kamm and Hurka recognize that all non-combatants have a (possibly defeasible) claim not to be targeted by combatants as part of the latter’s conduct of a war. Given that none of the non-combatants have done anything to make themselves liable to attack, they enjoy an equal claim against all combatants that they not be intentionally harmed, as well as a claim that all combatants exercise due care to avoid harming them unintentionally.26 However, Kamm and Hurka both go on to claim that combatants have an additional reason to avoid harming their compatriot non-combatants, one that they do not have vis-à-vis enemy noncombatants. This reason is the compatriot relationship, which in the paradigm case consists of being full members, or citizens, of the same political community. In virtue of this reason, combatants may, or must, assign a greater weight to harm done to a compatriot non-combatant than they assign to harm done to an enemy non-combatant. As I have tried to

24 As this example illustrates, the relationship between legal citizenship and the relationship of the compatriot is somewhat unclear. Suppose that in Kamm’s bombing example, the 100 people who would be killed by bombardment A lived in the territory of the combatant’s state, but were not citizens of it. Should the combatant still be partial to them? Does it matter whether they arrived in the territory legally or illegally, or whether they are guest workers or resident aliens?

25 Thomas Pogge, World Poverty and Human Rights (Oxford: Blackwell, 2002), p. 126. Note that Pogge’s claim is compatible with officials enjoying moral entitlements to do certain things that private individuals are forbidden from doing.

26 But see n. 1. show, however, neither Kamm nor Hurka provides a plausible case for the claim that merely because they are members of the same political community combatants ought to be partial to their compatriot noncombatants. If my arguments succeed, then it follows that Kamm and Hurka should accept that combatants ought to be impartial with respect to the morally permissible collateral damage they cause. This is so because the prohibition on targeting non-combatants follows from a property had by all non-combatants, namely their not having done anything to make themselves liable to attack.27 Therefore, holding all else equal, when deciding between two munitions factories a pilot ought to choose to bomb whichever factory’s destruction will result in less collateral harm to noncombatants. And when engaging in a war justified at least in part as necessary to prevent future attacks on their state and its citizens, as in the case of Israel’s 2006 war against Hizbullah, combatants must at the very least refrain from causing more harm to enemy (and neutral) noncombatants than they prevent their enemy from inflicting on their compatriot non-combatants.

All else may not be equal, however. Suppose, for instance, that the enemy combatants are responsible for placing the enemy non-combatants at risk. That is, had the enemy combatants not chosen to take cover in the middle of a village, it would have been possible for our combatant to attack them without posing a risk of harm to either compatriot or enemy non-combatants. In this case, the enemy combatants’ responsibility for placing their compatriot non-combatants at risk, when, let us assume, it was not morally permissible to do so, may provide a consideration sufficient to justify our combatant choosing the act that will collaterally harm a greater number of enemy non-combatants, rather than a lesser number of compatriot non-combatants.28 I am not sure that the enemy combatants’ (share of?) responsibility for putting the enemy non-combatants at risk does provide such a consideration.29 All I wish to claim here, though, is that if it does, then the rejection of partiality to compatriot

27 Walzer agrees, though with his typical lack of clarity: ‘the structure of rights stands independently of political allegiance; it establishes obligations that are owed, so to speak, to humanity itself and to particular human beings and not merely to one’s fellow citizens. The right of German civilians [during WWII]—who did no fighting and were not engaged in supplying the armed forces with the means of fighting—were no different from those of their French counterparts …’ (Walzer, Just and Unjust Wars, p. 158).

28 As this example illustrates, combatants’ duty to reduce the risk of harm to noncombatants, even at the cost of increasing the risk to themselves, affects not only the permissibility of various modes of attack, but also various modes of defense.

29 For discussion of this point, see Hurka, ‘Proportionality’, pp. 46–50, and Walzer, Just and Unjust Wars, pp. 158–96. non-combatants does not entail that combatants are never morally permitted to collaterally kill a greater number of enemy non-combatants, rather than a lesser number of compatriot non-combatants.

Another possibility is that all (adult? voting?) members of an aggressor state share collective responsibility for their state’s aggression, and that therefore non-combatant members of that state are either (a) liable to attack by just combatants (though perhaps less liable than unjust combatants, who may bear a greater degree of responsibility for the unjust war), or (b) not liable to attack, but left with a weaker claim not to be collaterally harmed than is had by those who do not share in the collective responsibility for launching or perpetuating an unjust war. The former implication of collective responsibility for an unjust war threatens to undermine the distinction between legitimate and illegitimate target of war that is essential to the very notion of collateral damage, and so I set it aside here. If it can be successfully defended, the latter implication of collective responsibility for an unjust war may warrant a just combatant’s collaterally killing a greater number of enemy non-combatants, rather than a lesser number of compatriot non-combatants. The justification in this case, however, does not involve appeal to the compatriot relationship as a reason to give greater weight to the harm that will likely be suffered by compatriot noncombatants than to the harm that will likely be suffered by enemy noncombatants. Instead, it involves an appeal to the enemy non-combatants’ collective responsibility for an unjust war as a reason to give less weight to the harm that will likely be done to them than they would merit had they not acted wrongly. As with the foregoing discussion of responsibility for placing non-combatants at risk of being collaterally killed, I take no stand here on the correctness of the argument from (alleged) collective responsibility for an unjust war to a lesser claim against being collaterally harmed.30 Once again, I aim only to make clear that rejecting Kamm’s and Hurka’s defense of partiality to compatriot non-combatants does not necessarily entail that combatants are never morally justified in collaterally killing a greater number of enemy non-combatants, rather than a lesser number of compatriot non-combatants.31

30 For a defense of non-combatant liability in virtue of collective responsibility for an unjust war, see Erin Kelly, ‘The Burdens of Collective Liability’, in Deen K. Chatterjee and Don E. Scheid (eds.), Ethics and Foreign Intervention (New York: Cambridge University Press, 2003), pp. 177–92. For a critique of this argument, see Larry May, Aggression and Crimes Against Peace (New York: Cambridge University Press, 2008), chapter 13.

31 Yet another possible justification for collaterally killing more enemy non-combatants, instead of fewer compatriot non-combatants, is if each of the former bear a degree or type

In the remainder of this article, I consider another reason (in addition to the two just mentioned) that may morally permit, or even require, a combatant to carry out an attack that will result in collateral harm to a greater number of non-combatants than would have been harmed by an alternative, equally effective, attack. I do so by way of a critical discussion of Kamm’s claim that given a choice between two courses of action, equally desirable from a military standpoint, a combatant ought to choose the one that will inflict collateral harm on a greater number of compatriot non-combatants, rather than the one that will collaterally harm a lesser number of neutral non-combatants.

1. Weighing Harm to Neutrals

While Kamm claims that compatriot non-combatants have a greater inviolability than do enemy non-combatants, she also suggests that their inviolability is less than that enjoyed by neutral non-combatants.32 Kamm briefly notes two arguments that might be given in support of this claim. The first is an argument by analogy:

War by its nature involves engagement between designated parties who are expected to absorb all costs. The appropriate analogy for war is a prize fight; people in the audience are not liable at all to being punched.33

As stated, there is one obvious disanalogy between a war and a prize fight, namely that audience members at a prize fight are said not to be liable at all to being harmed by the fighters, while Kamm claims only that neutrals enjoy a high degree of, but not absolute, inviolability. This disanalogy can be addressed easily enough, however, by weakening the claim regarding the inviolability of the audience members at a prize fight. If a prize fighter of individual culpability for an unjust war that warrants giving less weight to harm done to them than would be the case had they not acted wrongly (or wrongly enough). How often such cases would arise depends on what type or degree of individual responsibility for an unjust war renders a non-combatant liable to attack. For discussion of this issue, see Jeff McMahan, ‘The Ethics of Killing in War’, Ethics 114.4 (2004): 693–733.

32 It may be that the claim should be stronger—neutral combatants may enjoy a greater degree of inviolability than do compatriot non-combatants.

33 Kamm, ‘Failures of Just War Theory’, p. 675. The first sentence in this quote suggests an analytic argument for the claim that neutrals enjoy a greater degree of inviolability (indeed, total inviolability) than do members of the states or groups at war with one another. Such an argument seems implausible, since many paradigm examples of war have been fought by parties who had no expectation that they would absorb all of the costs. Perhaps the parties to a war should expect to absorb all of the costs, but this is a moral claim, not an analytic one regarding what it means (or is) to wage war. can save himself from death at the hands of his opponent only by causing a relatively minor harm to one of the audience members, doing so seems permissible. If so, then audience members enjoy a (very) high degree of inviolability vis-à-vis the prize fighters, but not total inviolability.

A more significant difficulty with Kamm’s suggested argument by analogy is that many non-combatants in the warring states are also like audience members at a prize fight, in that they are not involved in the fight itself, but simply bystanders caught up in it.34 Moreover, unlike the typical audience member at a prize fight, many non-combatants in warring states (as well as most neutrals) have not voluntarily chosen to ‘attend’ the fight; indeed, some of them may have actively opposed it (and continue to do so). Non-combatant members of a warring state may be designated as participants in the war (as the first sentence in the above quote suggests), but that they have the legal status of citizen in a state that happens to be at war does not ipso facto entail that they are participants in the fight itself. Were it to do so, then it would be impossible to distinguish between combatants and non-combatants, a distinction that Kamm wishes to retain.

The defense of the greater inviolability of neutrals by appeal to an analogy with audience members at a prize fight may seem more compelling if we focus on states rather than individuals. But even states may not be voluntary participants in a war. As Walzer points out, ‘the wrong the aggressor [state] commits is to force men and women to risk their lives for the sake of their rights’.35 States (or their members) may choose to sacrifice the enjoyment of their rights or to risk their lives to defend them, but in either case their choice is not a free one. If what accounts for the fact that the prize fighters are liable to be harmed, while audience members are not so liable, is that the former but not the latter have voluntarily chosen to participate in the fight, then members of a state that has not voluntarily chosen to wage war should enjoy the same level of inviolability as neutrals. Such a conclusion conflicts, however, with Kamm’s claim that compatriot non-combatants enjoy a lesser degree of inviolability than do neutrals, at least on the assumption that the former are citizens of a state that declares and wages war justly.

Moreover, as I noted earlier in this article, harm to some members of the state is not the same thing as harm to the state itself. One can imagine, for instance, that the neutrals who will be killed are members of a

34 Or at least this is so if the collective responsibility argument mentioned earlier proves to be false.

35 Walzer, Just and Unjust Wars, p. 51. minority political or ethnic group that is complicating the governance of the neutral state, so that their deaths actually strengthen the state, rather than harming it. This is not to say that these neutrals ought to be killed, of course, but only to point out that harm done to some of the state’s subjects need not equate to harm done to the state itself. If so, then the argument that neutral states ought not to be harmed by the participants in the war does not necessarily rule out an attack that collaterally kills some members of the neutral state.

Kamm briefly notes a second, instrumental, argument for the claim that neutrals enjoy a higher degree of inviolability than do compatriot non-combatants. According to this line of argument, the weightier prohibition on causing harm to neutrals is justified because it reduces the likelihood that the neutral state will be drawn into the conflict, with the likely consequence that even greater harm and/or wrong will occur. It seems that such an argument will often provide a sufficient moral justification for not carrying out an attack that will harm neutrals, even when the only other permissible option is an attack that will harm a greater number of compatriot non-combatants. However, Kamm does not think this argument identifies a necessary condition for the assignment of greater inviolability to neutrals in comparison to compatriot non-combatants. She writes that ‘even if C [the neutral state] were weak and posed no threat to A [the state whose combatants carry out the attack], it seems to have greater inviolability’.36 For Kamm, then, the greater inviolability enjoyed by neutrals is not simply a matter of the consequences that may follow from an attack that causes harm to them.

Once again, the description of states as if they were individuals misleads. Even if the neutral state enjoys greater inviolability than the warring states, it does not follow that the latter states’ non-combatants enjoy a lesser degree of inviolability than do the members of the neutral state. It might mean only that the combatants of the neutral state enjoy greater inviolability than that enjoyed by combatants of the warring states (or at least those of the warring state without a just cause for war). Alternatively, or in addition, it might mean (only) that the neutral state enjoys a greater degree of inviolability with respect to the use of its territory by the warring states than does either of those states vis-à-vis their opponent. For example, it could mean that state A’s seizure of some of its enemy’s

36 Kamm, ‘Failures of Just War Theory’, p. 675. public assets—e.g. bridges and highways—would not violate the moral constraints on war, but A’s seizure of a neutral state’s public assets would. Whatever the details, the general point should be clear enough: there is no obvious inference from the claim that a neutral state enjoys a greater degree of inviolability than do warring states to the claim that the noncombatant members of a neutral state enjoy a greater degree of inviolability than do non-combatant members of warring states.

Is it the case, then, that when faced with a choice between permissible collateral damage causing acts of war, a combatant ought to select whichever one will result in the fewest number of non-combatant deaths, regardless of the citizenship of those non-combatants? In particular, is Kamm mistaken when she claims that a combatant ought to collaterally kill a greater number of compatriot non-combatants rather than collaterally kill a lesser number of neutrals, at least up to a point? Perhaps not. Given a certain understanding of the value and nature of state sovereignty, Kamm’s claim may still be correct, though not on the basis of the reasoning she employs. Space does not permit a thorough defense of this understanding of state sovereignty, nor am I at all certain that a satisfactory defense of it can be offered. Therefore my argument is conditional: if a certain understanding of the value and nature of state sovereignty can be defended, then in some cases combatants ought to choose a course of action that will result in the death of a larger number of compatriot non-combatants, rather than one that will result in the death of a lesser number of neutrals, and the violation of the neutral state’s sovereignty.

Suppose that the neutral state enjoys a morally justified claim to sovereignty, understood to consist in a particular status I will characterize vaguely as a right to self-government, or self-determination. A selfgoverning state is the final arbiter of its affairs, in principle and (usually) in fact, and it exercises this normative and de facto authority within a given territory.37 Perhaps this notion of state sovereignty should be understood reductively, in terms of the individual rights of the state’s members, or perhaps it must be understood non-reductively, in terms of moral entitlements held collectively by the state’s members or by the group as a whole. However it is understood, though, any act that violates the neutral state’s sovereignty wrongs it because it fails to respect the states’ status as

37 This characterization of sovereignty is consistent with both the voluntary transfer of authority, so that the final arbiter over some of a state’s affairs may be a body created by an international treaty with other states, and with the scope of a state’s moral claim to sovereignty being limited by, and conditional upon respect for, principles of justice. self-governing. In wronging the neutral state, the warring state either wrongs the individual members of the neutral state, or wrongs the members of that state as a group. But just as in the case of a violation of individual self-determination, it is not essential to the wrong at issue that the state whose sovereignty is violated be harmed by that violation. The complaint is not ‘we are worse off because of what you did’, but rather ‘you are not entitled to do what you did; you have encroached upon our moral dominion’. Thus, in the case of an attack that impacts the territory and/or citizens of a neutral state, even if the ability of the neutral state to govern itself is not reduced at all, and even if no one in the neutral state is harmed, the warring state still wrongs it.

Suppose that something like this account of state sovereignty as a moral status can be fleshed out and defended. If it can, then we have a possible explanation for why state A ought to choose a course of action that collaterally kills more of its own subjects rather than one that will result in collateral harm to a lesser number of neutrals. Every non-combatant (and perhaps neutral combatant) enjoys the same level of inviolability, since none have done anything to make themselves liable to being targeted as part of the conduct of war. However, an attack that causes harm to neutrals also violates the neutral state’s sovereignty; the latter is a particular kind of wrong that does not occur if the combatant elects the course of action that involves collaterally killing some of his compatriot noncombatants. It may be, then, that the moral prohibition on violating a state’s sovereignty provides a justification for Kamm’s intuition that the combatant has a non-instrumental moral reason to choose the act that will result in a greater number of compatriot non-combatant deaths, instead of the one that will result in a lesser number of neutral deaths and the violation of the neutral state’s sovereignty. Of course, the neutral state surely does not enjoy an absolute right not to have its sovereignty infringed; for example, a combatant with the choice between killing a far greater number of compatriot non-combatants or a very few neutrals, but violating the neutral state’s sovereignty, may still be morally permitted (or even required) to choose the latter course.

What about a choice between killing a larger group of enemy noncombatants or a lesser number of neutrals? The correct answer to this question depends on whether the enemy state has a just cause for war. If it does (e.g. it is fighting in self-defense), then regardless of which act the combatant chooses, he violates the sovereignty of the state whose citizens he harms. More importantly, it is extremely unlikely that the combatant will have a just cause for war if the enemy state has one (indeed, this may even be conceptually impossible), in which case I believe the combatant is not morally permitted to kill anyone, collaterally or not.38 Suppose, however, that the just conduct of war does not depend on the justice of one’s cause (or on having met all six of the jus ad bellum criteria). Then holding all else equal, the deciding factor is the number of people harmed who are not morally liable to being harmed by an act of war.39 However, if the enemy state does not have a just cause for war, then it is unlikely that the harm done to enemy non-combatants will also constitute the violation of the enemy state’s sovereignty.40 In that case, the same reasoning that applies in the case of a choice between killing a larger number of compatriot non-combatants or a lesser number of neutrals holds in the case of a choice between killing a larger number of enemy non-combatants or a lesser number of neutrals.

In sum, Kamm is wrong to think that combatants ought to be partial to their compatriot non-combatants, assigning greater weight to harm done to them than they assign to harm done to enemy non-combatants. All non-combatants enjoy the same degree of inviolability, and so holding all else equal, a combatant ought to choose whichever course of action will result in the fewest number of non-combatant deaths. Kamm may well be correct, however, when she asserts that a combatant ought to kill a greater number of compatriot non-combatants rather than a lesser number of neutrals. But the reason for this is not that the lives and well-being of compatriot non-combatants count for less than those of neutrals. Rather, it is because an attack that causes harm to neutrals involves a distinct moral wrong absent from an attack that harms only compatriot noncombatants, namely a violation of the neutral state’s sovereignty. The disvalue of this wrong is great enough that, at least up to a point, it requires choosing an act that will result in more non-combatant deaths than would result from committing the alternative act under consideration.41

38 Foa defense of the claim that in order for an agent to wage war in conformity with the criteria of jus in bello, he or she must have a just cause for going to war (or be part of a political organization that has a just cause for going to war), see McMahan, ‘The Ethics of Killing in War’.

39 All else may not be equal, of course. If the violation of the neutral state’s sovereignty would bring it into the war, and thereby increase the amount of harm and/or number of wrongs done, then this would count in favor of an attack that harmed a greater number of enemy non-combatants over an attack that harmed a lesser number of neutrals (at least up to a point).

40 Without a fully fleshed-out account of the moral justification for state sovereignty, it is not possible to be certain of this. But, for example, if a state enjoys a moral claim to sovereignty only if it does not wage unjust wars, then harm done to its non-combatants will not count as a violation of the state’s sovereignty.

41 I wish to thank Craig Derksen, Tom Hurka, and Scott James for their comments on an earlier version of this paper.

Part Four: Just War Theory and Terrorism

Defining Terrorism for Public Policy Purposes: the Group-target Definition

Eric Reitan

Introduction

Any public policy response to terrorism achieves coherence only to the extent that there is a clear understanding of what is being targeted. Put another way, in order to fashion and assess public policy responses to the kind of threat exemplified by paradigms of ‘terrorism’, we need to agree on what ‘terrorism’ will mean for public policy purposes. Stated simply, we need a clear public definition.

Unfortunately, ordinary usage does not provide the needed clarity. At the same time, a public definition must strive to avoid being so divorced from ordinary usage that it amounts to a technical definition—one that frames a discourse of specialists disconnected from the pervasive public discourse and largely inaccessible to the general public. In a democratic society, active public engagement in the discourse culminating in policy decisions should be encouraged and facilitated. A public definition ought therefore to be true to ordinary usage in some sense, even if such usage is too messy to immediately provide us with an adequate public definition.

In this essay, I do four things: first, I discuss the prospects for a public definition of terrorism that will minimally comport with ordinary usage while still providing the kind of precision that ordinary usage does not offer; second, I argue that such a definition will encompass dominant paradigms of terrorism while having essentially pragmatic goals in view, and then construct criteria for an adequate definition with these ideas in mind; third, I develop what I call the ‘group-target’ definition of terrorism, which identifies a feature of terrorism that may be especially helpful for public policy purposes; and finally, I argue that this definition fits the criteria for a good public definition especially well. In doing the last, I hope also to demonstrate the useful role that a public definition can play in evaluating public policy.

The Problem with Defining ‘Terrorism’

Whenever we develop public policies to fight terrorism, we are supposing that ‘terrorism’ picks out a distinctive class of actions whose members are sufficiently alike that one can devise a coherent terrorism policy that effectively addresses the actions within the class. But given what ‘terrorism’ means in ordinary usage, this is simply not true. ‘Terrorism’, in its ordinary use, simply does not pick out a class of actions sufficiently unified to guide public policy.

There are two reasons why this is so. First, there is the diversity of phenomena falling within the scope of ordinary use. As Jenny Teichman puts it, ‘if we list all the different phenomena which are at one time or another described as terrorism in ordinary conversation, or in ordinary newspapers, or by ordinary politicians, we will end up with a huge rag-bag of not very similar items’.1 Second, there is the polemical use of the term. As Virginia Held notes, ‘[u]sage characteristically applies the term to violent acts performed by those of whose positions and goals the speaker disapproves and fails to apply it to similar acts by those with whose positions and goals the speaker identifies’.2 This is true because ‘terrorism’ serves as a label of condemnation. ‘To identify someone as a terrorist’, Claudia Card observes, ‘is to render a judgment on them, not simply to make a discovery’.3

Both of these features of ordinary usage can be made sense of if terrorism is construed as an essentially contested concept in W.B. Gallie’s sense—that is, as a concept characterized by competing definitions unified by a common appraisive meaning and a shared set of paradigms.4 Gallie argues that essentially contested concepts regulate moral discourse in which the issue at stake is which phenomena deserve the appraisal we attach to the paradigms. An essentially contested concept is therefore not merely descriptive, but evaluative. To attach the concept to a phenomenon is to pass judgment on it. The paradigms provide a kind of point of

1. Jenny Teichman, ‘How to Define Terrorism’, Philosophy 64 (1989), pp. 505–517 (506).

2. Virginia Held, ‘Terrorism, Rights, and Political Goals’, in James. B. Brady and Newton Garver (eds.), Justice, Law, and Violence (Philadelphia: Temple University Press, 1991), pp. 223–40 (223).

3. Claudia Card, ‘Making War on Terrorism in Response to 9/11’, in James P. Sterba (ed.),

Terrorism and International Justice (Oxford: Oxford University Press, 2003), p. 166.

4. W.B. Gallie, ‘Essentially Contested Concepts’, Proceedings of the Aristotelian Society 56 (1956), pp. 167–98. For a development of Gallie’s understanding of essentially contested concepts and an application of the idea to negative appraisive terms, see Eric Reitan, ‘Rape as an Essentially Contested Concept’, Hypatia 16 (2001): 43–66. unification in ordinary usage, insofar as there is general agreement that each paradigm deserves to be judged or evaluated in the relevant way. But the paradigms are complex, and so there is disagreement concerning which features of these paradigms justify the distinctive appraisal that the concept carries. Competing uses therefore arise, and these are expressed in competing definitions that aim to pick out those features of the paradigms most salient for the purposes of assigning the relevant appraisal. Ordinary usage is not captured by any one of these definitions, but by the set of overlapping contenders. A proposed definition is consonant with ordinary usage so long as it extends to the paradigms, excludes what no one would classify under the concept, and preserves the concept’s distinctive appraisive meaning.

For Gallie, essentially contested concepts perform the valuable function of preventing some voices from being cut out of public debate by a kind of definitional fiat. If the ‘true’ meaning of the concept is identified with one particular competing definition—that is, by one perspectival assessment of what it is about the paradigms that justifies the relevant appraisal—the danger is that other perspectives that deserve a hearing will be silenced.

But even if this is right, essentially contested concepts are perilous when it comes to making public policy. For the public to participate democratically in the development and assessment of terrorism policies, we must be able to examine how proposed and actual policies contribute, if at all, to the prevention or reduction of the threat that ‘terrorism’ designates. And to do that, we need a precise, shared understanding of what ‘terrorism’ is taken to mean, at least for the sake of these policies.

Since ordinary usage cannot provide such an understanding, a public definition must therefore be stipulative. But a stipulative public definition may do the very thing that the recognition of essential contestability is supposed to prevent—namely, trammel public discourse by silencing significant moral perspectives by definitional fiat. Consider those who observe that domestic battery has much in common with terrorism paradigms. Their claim that domestic violence is a form of terrorism, and that comprehensive efforts to combat the distinctive evil of terrorism ought therefore to include it, may be unfairly dismissed if the reigning public definition has latched onto features of the paradigms other than those which justify the comparison with domestic abuse.

I suspect the only way to avoid this danger while acknowledging the need for a coherent public definition is to regard any such definition as provisional and situational. A public definition is introduced in a particular sociopolitical and historical context, in order to formulate policies to address a particular salient problem. Outside this context, it has no special authority. Furthermore, as I will argue in the next section, the criteria for choosing a public definition should seek to avoid any favoritism in the definitional debate that essential contestability is supposed to regulate. Insofar as that definitional debate turns on the question of which features of the paradigms justify our moral condemnation of them, it follows that a public definition should answer a different question. But what question?

Criteria for an Adequate Public Definition

A clear understanding of the kind of problem terrorism poses will require us to say both what terrorism is and why it is problematic. Ordinarily, the first task would seem to be the definitional one, and the latter a normative question that we address only once we have our definition in place. David Rodin, however, has argued that we should define terrorism in terms of what makes the paradigm cases morally problematic. As he sees it, we need ‘a clear and coherent understanding of what the morally relevant features of terrorism are’.5 To this end, we should seek a ‘moral definition’ of terrorism by beginning with the question of why the ‘core instances of terrorism’ (what I am calling the paradigms) are morally problematic, and then defining as terrorism those things that share the identified wrongmaking properties.6

It should be clear from what I have said, however, that this definitional strategy—while valuable in a number of ways—will not serve for arriving at a public definition of terrorism. What drives the essential contestability of a concept such as terrorism is precisely the fact that different constituencies disagree about which features of the paradigms justify the negative appraisal that accompanies the term’s use. That is, contested definitions arise precisely because the strategy favored by Rodin is invoked. Acknowledging essential contestability is supposed to allow for competing moral definitions so as to prevent one account of terrorism’s chief wrongmaking property from silencing, by definitional fiat, those who would locate the central moral problem elsewhere.

5. David Rodin, ‘Terrorism without Intention’, Ethics 114 (2004), pp. 752–71 (752); italics are mine.

6 Ibid., p. 753.

While offering moral definitions as Rodin does is thus constitutive of what we might call the ‘language game’ of terrorism, employing this definitional strategy in order to arrive at a public definition is dangerous. A precise, stipulative public definition may well be required for public policy purposes, but it ought not to stifle the ongoing moral discourse facilitated by the concept’s essential contestability. To privilege one of the competing moral definitions in that discourse over others by making it the definition for public policy purposes cannot help but stifle the discourse that essential contestability is supposed to protect.7

How, then, should we arrive at a public definition? My proposal is that, insofar as a public definition is required for formulating coherent public policy responses to a class of problems or threats to society, pragmatic usefulness for developing such responses may be the most salient basis for forming such a definition. What a public definition ought to do, on my view, is to identify a class of problematic phenomena which is such that

(A) the relevant paradigms belong to the class; (B) phenomena in this class can be addressed under a reasonably unified public policy rubric; and (C) phenomena in this class are not adequately addressed under existing policies targeting other established classes of phenomena.

Criterion (A) serves to ensure minimal fidelity to ordinary usage. Even if ordinary usage cannot provide us with an adequate definition, any definition that is to guide public policy must at least be consonant with ordinary usage in the minimal sense that it cannot include things within its scope that no one would call terrorism (such as nonviolent acts), nor exclude paradigms of terrorism.8 With respect to the latter, it is particularly important to keep in mind the context within which anti-terrorism policies are introduced. In post-9/11 America, there is little doubt that when ‘terrorism’ is uttered, what come to mind first are the coordinated attacks on the World Trade Center and the Pentagon (even if other paradigms, such as the bombing of the Murrah Building in Oklahoma City, remain vivid in the American consciousness). And when the Bush administration launched its ‘war on terrorism’, it presented itself to the American public as initiating a concerted public policy effort to bring to an end the kind of phenomenon exemplified in the 9/11 attacks. Thus, for the sake of

7 It should be clear that this is not intended as a criticism of Rodin, since Rodin nowhere recommends that his proposed definition serve as the public definition of terrorism.

8 Walter Sinnot-Armstrong makes fruitful use of essentially this criterion in his critique of Primoratz’s definition of terrorism. See his ‘On Primoratz’s Definition of Terrorism’, Journal of Applied Philosophy 8 (1991), pp. 115–20. guiding public policy in post-9/11 America (and for evaluating the efficacy of the Bush administration’s policies), we need a definition of terrorism that is especially suited to these attacks.

But the 9/11 attacks are not the only terrorism paradigms. And even if we cannot include within our definition everything that is called terrorism in ordinary use, a public definition would be improved to the extent that it can at least encompass all the paradigms. Achieving this aim is complicated by the fact that there are a diversity of paradigms, ranging from the lone pipe-bombing campaign of the reclusive Unabomber, to the coordinated efforts of Al Qaeda in the 9/11 attacks, to the French ‘Reign of Terror’. The aim of preserving minimal fidelity to ordinary usage may therefore come into some tension with criterion (B). A public policy initiative that may be effective in response to one paradigm may not be effective in response to another.

Hence, to define ‘terrorism’ in a way that is helpful in shaping public policies while respecting ordinary usage enough to encompass all the paradigms, we may need to specify subtypes, or kinds of terrorism. And a public policy response to one kind of terrorism should not be represented as targeting terrorism in every sense. For example, the current ‘war on terrorism’ is represented as a concerted public policy effort to reduce or eliminate the kind of thing exemplified in the 9/11 attacks. To treat public support for such a policy as carte blanche permission to target terrorism of every kind would be a mistake, as would supposing that policies suited to reducing terrorism of the given kind are equally suited to eliminating terrorism of other kinds.

Turning to (C), this criterion reminds us that if public policy is going to be shaped by the imperative to fight terrorism, then ‘terrorism’ must denote a particular kind of violent threat to public welfare distinct from other species of violence for which there already exist established and effective public policy programs and institutions. Specifically, terrorism must be distinguished from criminal violence and acts of war. ‘Terrorism’ as a public category makes sense only if there is a distinctive class of violence not adequately addressed through conventional criminal justice and military institutions.9 In the absence of such a distinctive class of violence, the use of ‘terrorism’ in public discourse becomes merely polemical. It ceases to be a useful concept for public policy purposes (which would be better

9 I believe that there is such a class (and it is partly for this reason that I view the very literal way in which the current administration is pursuing its ‘war’ on terrorism to be gravely misguided). served by reclassifying the things we call terrorism into the categories of ‘crime’ and ‘acts of war’). And so, even if a public account recognizes distinct kinds of terrorism, these kinds should ideally fall under an overarching definition that identifies how terrorism of every kind is distinct (in a pragmatically salient way) from both ordinary criminal violence and acts of war.

In what follows, I construct a public definition of terrorism with these criteria in mind. By treating terrorism as a species of violence, I avoid including within the scope of the concept things that no one, in ordinary usage, would call terrorism.10 In terms of including all of the agreed paradigms, my task is more challenging for reasons noted above. Given the current sociopolitical and historical context, a public definition should aim to be especially suited for framing public policy responses to the class of phenomena exemplified in the 9/11 attacks. Hence, I am most interested in offering a definition for the ‘kind’ of terrorism at issue in the ongoing ‘war on terrorism’. But I will locate this kind under a broader definition which, as I will argue, specifies a distinct class of threats not adequately addressed by standard criminal justice and military institutions.

The Inadequacy of Traditional Defining Characteristics

Philosophical definitions of terrorism have tended to emphasize one or more of the following characteristics:11 (1) It directly targets civilians/ noncombatants;12 (2) It seeks to achieve some goal by creating terror in a

10 Carl Wellman’s contention that terrorism may be nonviolent defies ordinary use, and is problematic for precisely this reason. His is a technical philosophical definition that might have some value as a stipulative definition in some contexts, but will only be confusing if offered up as a public definition. See C. Wellman, ‘On Terrorism Itself’, Journal of Value Inquiry 13 (1979), pp. 250–58 (251).

11 I leave out two characteristics that have sometimes appeared in official definitions of terrorism, namely that the violence is perpetrated by subnational groups and that the violence is illegal or illegitimate. I exclude these from consideration because they have both been subjected to compelling philosophical criticism in the literature.

12 Coady offers a good example of a definition of terrorism that emphasizes this characteristic. He defines terrorism as ‘A political act, ordinarily committed by an organized group, which involves the intentional killing or other severe harming of non-combatants or the threat of the same or intentional severe damage to the property of non-combatants or the threat of the same’. See his essay, ‘The Morality of Terrorism’, Philosophy 60 (1985), pp. 47–69 (52). Primoratz’s definition of terrorism emphasizes innocence, rather than noncombatant status, but it seems that by ‘innocence’ he intends roughly the same as Coady intends by ‘concombatant’ (see esp. Igor Primoratz, ‘What is Terrorism?’, Journal of Applied Philosophy 7 (1990), pp. 129–138 (p. 131). The difficulty with defining terrorism in terms of population;13 (3) It directly targets one victim in order to coerce another;14

(4) It aims to achieve some political goal—especially that of ‘destabilizing (or maintaining) an existing political or social order’;15 (5) It aims to call public attention to the perpetrator’s cause.16 Are any of these suitable candidates for constructing a public definition of terrorism?

Brief reflection should reveal that none of (1)–(5) is singly sufficient to distinguish terrorism from both criminal violence and acts of war.17 While some combination might be sufficient, the deeper problem is that none seems necessary—that is, none would extend to all the things that we would like to include within a public definition. (1) would exclude the attack on the U.S.S. Cole and the 1983 suicide bombing of the Marine barracks in Lebanon.18 More significantly, insofar as the Pentagon is arguably not a civilian target, we have the problematic result of treating the 2001 attack on the World Trade Center as an act of terrorism, but not similarly treating the tandem attack on the Pentagon. In general, two violent acts, innocence is that, as J. Angelo Corlett has recently pointed out, most terrorists perceive their targets to be wrongdoers (even if they are not combatants). See Corlett, Terrorism: A Philosophical Analysis (Dordrecht, Netherlands: Kluwer Academic, 2003), p. 117.

13 The creation of terror features prominently in both Carl Wellman’s definition and Martin Hughes’s understanding. See Wellman, ‘On Terrorism Itself’, Journal of Value Inquiry 13 (1979), pp. 250–58, and Hughes, ‘Terrorism and National Security’, Philosophy 57 (1982), pp. 5–26. Primoratz agrees that the creation of terror is an important characteristic of terrorism, but does not treat it as a defining characteristic. See Primoratz, ‘What is Terrorism?’ p. 134.

14 In line with both Wellman and Primoratz, Claudia Mills distinguishes terrorist coercion from other forms of coercion by characterizing terrorism as ‘third-party’ coercion: A attacks C in order to coerce B. See Mills, ‘The Distinctive Wrong of Terrorism’, International Journal of Applied Philosophy 10 (1995), pp. 57–60.

15 Corlett emphasizes this feature in his definition of terrorism. See Corlett, Terrorism, p. 119. He is here following Burleigh Taylor Wilkins, Terrorism and Collective Responsibility

(London: Routledge, 1992), p. 6.

16 Again, see Corlett, Terrorism, p. 119.

17 (1) is typical of criminal violence. (2) is typical of many military tactics that seek to demoralize the enemy forces and the civilian population through displays of terrifying force (consider the ‘shock and awe’ strategy that launched the US invasion of Iraq in 2003).

(3) is not uncommon in ordinary criminal violence (for example, kidnapping for ransom, or threatening a spouse or child in order to get a victim to unlock a safe). (4) is characteristic of many if not most wars (consider the goal of ‘regime change’ that motivated the Iraq war). (5) is probably not uncommon of military acts in the early stages of a revolutionary war, and can also be observed in certain acts of criminal violence. Consider a man who, outraged because his son is being denied medical coverage for a potentially life-saving experimental treatment, publicly murders an executive at his HMO while shouting, ‘This is what you are doing to my son!’

18 Held thinks the attack on the Marine barracks in Lebanon is ‘among the leading candidates for inclusion’ within the scope of the concept, and that its exclusion would be ‘arbitrary’. Held, ‘Terrorism, Rights, and Political Goals’, p. 225. one targeting civilians and the other targeting the military, may be sufficiently similar that they should be treated under the same policy guidelines. To make the targeting of civilians the chief defining characteristic of terrorism would therefore compromise the pragmatic usefulness of a public definition.

(2)–(5) all identify objectives (intermediate or final) that perpetrators of terrorist violence intend to achieve (creating fear, coercing a third party, destabilizing or maintaining some political order, or publicizing a cause). With respect to each, we can apply C.A. Coady’s objection to defining terrorism in terms of the intent to cause terror. As Coady puts it, when we do so ‘we are prejudging an empirical investigation into the specific motives’ of terrorists.19 Suppose that when Timothy McVeigh blew up the Murrah building in Oklahoma City, his intent was not to inspire fear in a population but to punish the federal government for infringing on civil liberties. Would we be willing to say, on learning these motives, that the Oklahoma City bombing was, after all, not an act of terrorism? Clearly not. Hence,

(2) is not necessary.

We can argue along similar lines with respect to (3)–(5). With respect to (3), it is helpful to consider Claudia Card’s recent discussion of the 9/11 attacks in terms of Carl Wellman’s definition of terrorism—according to which an act is terrorist if it directly harms one target in order to coerce another. Card, noting the ambiguity of the motives behind the 9/11 attacks, argues that ‘if the attacks’ intent was not coercive but was, say, punitive, or if it was simply to show the world that the United States is not invulnerable, then the deed was not basically terrorist, in Wellman’s sense’.20 But surely the proper conclusion to reach here is that Wellman’s definition is defective, not that the 9/11 attacks wouldn’t be terrorism if these proposed motives proved correct. To be minimally consonant with ordinary usage (at least ordinary usage post-9/11), the 9/11 attacks must count as terrorism regardless of what the motives of the terrorists might turn out to be. And for the purposes of shaping contemporary public policy responses to terrorism, the 9/11 attacks must be included within the scope of a public definition for the simple reason that these public policy responses to terrorism are fundamentally conceived as responses to the kind of thing exemplified in the 9/11 attacks.

19 Coady, ‘The Morality of Terrorism’, p. 53.

20 Card, ‘Making War on Terrorism’, p. 173.

But if all of this is right, then terrorism needs to be defined without reference to the specific motives of terrorists. And if that is so, then neither

(4) nor (5) will be a necessary characteristic of terrorism either.

The reason why none of (1)–(5) appear to be necessary might be because the extension of ‘terrorism’ is determined by Wittgensteinian family resemblances, such that no common feature is to be found even among paradigm cases. In fact, however, I think the paradigms of terrorism widely share a common characteristic that can also serve to distinguish terrorism from both criminal violence and acts of war. In the next section, I discuss this characteristic and, on the basis of it, propose a definition that is broadly inclusive of terrorism paradigms.

The Terrorist Principle of Discrimination

Those who define terrorism in terms of the targeting of civilians are, I think, on the right track. But for reasons already mentioned, their definition does not quite work. My alternative proposal is this: what makes an act terrorist is not the fact that civilians are targeted, but rather the fact that targets of terrorist violence are selected according to a distinctive principle of discrimination, one which bears a superficial similarity to that endorsed in just war theory but differs crucially in that it includes civilians within the class of legitimate targets.

A review of the just war principle of discrimination may be helpful here. In so-called ‘just’ wars, three principles actually operate in selecting targets. The first is that acts of war are directed against some opposing group as a whole. Thus, in order to be a legitimate target, you must be a member or agent of the group (usually a nation) against which war is waged. But in just war, mere membership in the targeted group is insufficient. The second principle for selecting targets, which is what is typically referred to as the principle of discrimination, distinguishes between legitimate targets and illegitimate targets according to combatant status. Only if you are a combatant may you be directly targeted (although it might be permissible to kill noncombatants as a side-effect of attacking military targets). The third principle, what might be called a principle of efficacy, chooses which of the legitimate targets to actually attack based on the extent to which doing so furthers the goals of the war.

There is a sense in which the targeting of enemy combatants in war is morally indiscriminate: there need not be anything special about this combatant (who is attacked) that distinguishes him or her in any morally relevant way from another combatant (who is not attacked). Combatant targets are typically selected for strategic or tactical reasons, not by virtue of any special moral desert that distinguishes them from other legitimate targets. Along these same lines, one is not in the class of legitimate targets because one is an actual threat. One is a member of that class because one has a designated role in the prosecution of the war, a role whose faithful execution would pose a threat. Thus, a soldier who consistently shoots to miss is still a legitimate target.

Terrorism, like war, is directed against an enemy group as a whole. But what distinguishes terrorism from war waged according to just war conventions is that membership in the enemy group is treated as sufficient to render one a legitimate target. In other words, legitimate targets are distinguished, not by virtue of whether they occupy a combatant role, but by virtue of their membership in an enemy group that admits of a diversity of roles, including noncombatant roles that do not pose any direct threat to the lives of terrorists or those on whose behalf the terrorists perceive themselves as acting.

The targeted group might be a nation or a group defined by such properties as ethnicity, nationality, race, gender, sexual orientation, religion, or even participation in a given economic system. What is crucial is that the criterion of group membership is not that one occupies a role whose function is to cause harm to the terrorists (or those close to them). One isn’t a member of the enemy group because one is a combatant whose role poses a direct threat to terrorists (or those on whose behalf they see themselves as acting). Rather, one is a member of the enemy group because one is Jewish, or American, or a citizen of a capitalist nation—some group marker that encompasses a variety of roles, many of which pose no direct threat to anyone.21

As in war, it is the group taken as a collective that is the main or primary target of terrorism, not the individuals who are directly attacked. This distinguishes terrorism from ordinary criminal violence. But unlike in war (at least ‘just’ war), relevant group membership is taken as sufficient to render one a legitimate target. Specific targets are chosen for attack, not by virtue of some special moral property that distinguishes them from other members of the group, but by virtue of accessibility, symbolic value, or

21 Interestingly, if I despise soldiers and seek to terrorize soldiers everywhere through selective executions, I am a terrorist even though my targets are all combatants. The reason is that, unless all soldiers everywhere have been formally mobilized to seek me out and destroy me, it is not the case that all members of the group I have targeted occupy a role whose explicit function involves directly threatening me. other strategic considerations. Membership in the targeted group, like combatant status in just war theory, is thus treated as decisive grounds for forfeiture of the ordinary moral right not to be harmed. Subsequent selection of targets is morally (but not strategically) indiscriminate.

Since the group is the ultimate target of terrorism, the effect on the immediate targets is not the primary effect sought. The real goal is some effect on the group conceived as a whole. In this respect, immediate victims are mere instruments, and the violence directed against them is instrumental. But it is instrumental only in the minimal sense that it aims to produce some effect on the broader group to which the immediate victims belong. The effect might be to coerce the group’s leaders into making decisions favored by the terrorists, or to disrupt the group’s political structure, or to perpetuate the group’s subordinated status—but it might also be something as abstract as humiliating the group or exacting vengeance. In fact, the violence might not be construed as instrumental by the terrorists, insofar as there may be a metaphysical reification of the group such that its members are viewed as parts of a whole, and an attack on one part is instrumentally directed towards harming the whole only because one cannot strike every part of the whole at once. If I take harming you to be valuable in itself, but I can only reach your little finger, when I break your little finger I will not conceive of the act as merely instrumental. Nevertheless, it is instrumental in the sense that my aim or goal is not to injure your little finger (I have nothing special against it). I injure your little finger in order to harm you.

Even when the violence is conceived of as instrumental by the terrorists, this does not mean that the immediate targets are seen as instruments for influencing some other target. I might twist back your left little finger in the hope of coercing you to stop hitting me with your right fist. In this case the violence is clearly instrumental, but I conceive of the act as attacking you in order to get you to change your behavior—even though your little finger is not striking me, and even though it is not your little finger that is responsible for choosing whether to continue hitting me. Analogously, even though a terrorist attack on civilians may aim to coerce the group’s leaders, the attack is still an attack on the group to achieve an effect on the group. The common idea that one immediate but secondary target is directly attacked in order to affect an indirect but primary target22

22 This idea is endorsed by Wellman, Primoratz, Mills and Card. A particularly concise formulation of it is offered by Primoratz, ‘What is Terrorism?’, p. 131. fails to appreciate that terrorism is directed against a group as a whole, even if the part attacked is not the same as the part in which a practical effect is sought. Metaphysical reification of the group may be what ultimately justifies the terrorist principle of discrimination, as well as explaining the motivational connection between the choice of target and the effect sought.

With these considerations in mind, we can define terrorism as follows:

‘Terrorism’ is any act or pattern of violence such that (a) the primary or ultimate target is a group conceived of as a whole; (b) the immediate targets are members of the targeted group; (c) membership in the targeted group is regarded as sufficient to render one a legitimate target; (d) the violence against targeted group members is instrumental to producing some effect on the group conceived of as a whole (which may mean influencing the group’s perceived leadership).

I will call this the ‘group-target’ definition of terrorism. Defined in this way, terrorism turns out to be closely related to the concept of collective violence as developed by John Ladd in ‘The Idea of Collective Violence’.23 In fact, I would like to suggest that terrorism can be construed as a species of collective violence in Ladd’s sense. Exploring this possibility will both shed light on the group-target definition and introduce some ideas that will be useful, in a later section, for addressing its pragmatic value.

Terrorism as Ideological Violence

Ladd defines collective violence as ‘violence that is practiced by one group on another and that pertains to individuals, as agents or as victims, only by virtue of their (perceived) association with a particular group’.24 Perpetrators of collective violence act as agents of a group, and their targets are chosen, not for any individual characteristics they possess, but by virtue of their perceived membership in the targeted group.

As Ladd sees it, perpetrators of collective violence typically rationalize their acts according to a distinctive ideological framework, which he sketches out in terms of five key ‘premises’ or ‘doctrines’: (i) the Doctrine of Bifurcation, according to which the world is divided into two groups—a ‘Chosen Group’ and an ‘Other Group’—that are ‘irretrievably separated

23 John Ladd, ‘The Idea of Collective Violence’, in Brady and Garver, Justice, Law, and Violence, pp. 19–47.

24 Ibid., p. 19. and divided’;25 (ii) the Doctrine of Moral Disqualification, according to which members of the Other Group are perceived as ‘moral outcasts’ who ‘lack the minimum attributes necessary for being members of the moral community, or even for being human’;26 (iii) the Doctrine of the Double Standard, which establishes two distinct moralities, one for interactions with those who belong to the Chosen Group and hence to the moral community (a standard that demands respect and concern), and a different one for interactions with Others, who fall outside the moral community and may thus be treated in ways that would never be permitted in relation to the Chosen;27 (iv) the Doctrine of Group Mission, ‘which assigns a plenary mission, often divinely commanded, to the members of the Chosen Group to protect the Chosen Group and its values from perceived threats to it by the Other Group’;28 and (v) the Doctrine of Zero-Sum Struggle, according to which the two groups are ‘locked in a conflict for which compromise or reconciliation are absolutely inconceivable’, and which is such that the flourishing of one group can only be achieved by the defeat or destruction of the other.29

Ladd stresses that agents of collective violence do not see themselves as private individuals acting for private reasons, but as agents of the Chosen Group acting on behalf of that group and ready to sacrifice themselves for the sake of the group mission. This feature of collective violence may at first seem to offer a point of contrast with terrorism as I’m defining it here. On the group-target definition, there is at least no explicit requirement that the perpetrators see themselves as agents of a group. Given the grouptarget definition, terrorism requires the existence of an ‘Other Group’ that is the target of terrorist violence, but it does not clearly demand a ‘Chosen Group’ on whose behalf the terrorist acts. A terrorist might be a loner such as Theodore Kaczynski.

But there may be a sense in which even the lone terrorist is acting as an agent: the terrorist is acting for a cause that is perceived to be greater than personal welfare, a cause that at least in principle could serve as a rallying point for others. While this cause might not be the welfare of some group (it might be, say, the glory of God), it nevertheless seems to be true that

25 Ibid., p. 40.

26 Ibid.

27 Ibid., pp. 40–41.

28 Ibid., p. 41.

29 Ibid. Ladd does not name this fifth premise. I do so here to preserve parallel structure. those who share the terrorist’s cause (and do not fall into the targeted group) constitute a kind of chosen group on whose behalf the terrorist acts. This idea is supported by the fact that Kaczynski, even though he was a hermit who acted entirely alone, represented himself to the world as an agent of a group called ‘F.C.’. (probably short for ‘Freedom Club’ or ‘Freedom Collective’) in his famous manifesto, ‘Industrial Society and its Future’.30

The Unabomber example raises an interesting possibility: In order to act on the group-target principle of discrimination, it may be psychologically if not theoretically necessary to absorb, in a general way, the identity of individuals into their group affiliations.31 In order to target individuals simply by virtue of their group memberships, terrorists may need to see the world in terms of groups, and see individuals as nothing more than agents of those groups—a perception that extends not only to their targets, but inevitably to themselves. Something like the ideology sketched by Ladd may offer the Weltanschauung necessary for acting on a grouptarget principle of discrimination.

But another basis for acting on a group-target principle of discrimination may also be possible—specifically, Michael Walzer’s doctrine of supreme emergency. For Walzer, when a political community faces a dire threat to its existence from an enemy group, it may—when the threat is imminent and there is no other way to overcome it—set aside the ordinary injunction against targeting noncombatants.32 Thus, Walzer offers a restricted justification for engaging in terrorism in the group-target sense. What is interesting about Walzer’s doctrine is that, while it makes no explicit commitment to premises (i)–(iii) of Ladd’s ideological framework, it endorses premises (iv) and (v). A ‘supreme emergency’ exists when the political community has a solemn duty to preserve itself from a dire and imminent threat posed by an enemy group, and the nature of the threat is so extreme and intractable that nothing short of inflicting dire harms on the enemy, without regard for traditional distinctions, has any hope of saving the community. Stated inversely, premises (iv) and (v) of Ladd’s framework essentially amount to the claim that a supreme emergency in Walzer’s sense obtains. Premises (i)–(iii) serve the preparatory role of

30 See Kirkpatrick Sale, ‘The Unabomber’s Secret Treatise: Is There Method in his Madness?’ The Nation 261 (25 September 1995), pp. 305–311.

31 Some may wonder whether Theodore Kaczynski actually followed the group-target principle of discrimination. On this, see my discussion in the next section.

32 See Michael Walzer, Just and Unjust Wars (New York: Basic Books, 2nd edn, 1992), especially pp. 251–54. rendering indiscriminate attacks permissible by virtue of the subhuman status of the enemy.

Walzer, by contrast, makes no attempt to dehumanize the enemy so as to justify violence under conditions of supreme emergency. For him, what serves this role is ‘the rule of necessity (and necessity knows no rules)’.33 It is worth noting, however, that Walzer’s chief example of a supreme emergency is the threat posed by the Nazis during the Second World War. And, according to Walzer, the Nazis were an enemy group organized in terms of and committed to ‘an ideology and a practice of domination so murderous, so degrading even to those who might survive, that the consequences of its final victory were literally beyond calculation, immeasurably awful’.34 Thus, while the enemy group may not be morally disqualified in the way Ladd envisions, the enemy group distinguishes itself in a morally relevant way from other groups—namely, by its collective commitment to evil so profound as to be incalculable.

Perhaps, then, what is most essential to the ideological framework that Ladd sketches out is that it establishes both that (a) the Other Group poses a threat to the Chosen Group that amounts to a supreme emergency and (b) traditional moral constraints against harming the innocent are waived in relation to the Other Group. As Walzer’s thinking demonstrates,

(b) might well be thought to follow from (a) without additional ideological premises. However, it may be psychologically necessary for those engaged in indiscriminate violence against a group to embrace something like premises (i)–(iii).

Even if this is not the case, what this review of Walzer suggests is that collective violence in general, and group-target terrorism in particular, requires its practitioners to embrace something substantially similar to Ladd’s premises—at least (iv) and (v) and probably some premises that do the work that (i)–(iii) do for Ladd. We might say that, for Walzer, the truth of an ideology of this sort is what gives rise to the condition of supreme emergency.35

I say that terrorism on the group-target definition can be construed as a species of collective violence, rather than a synonym for it, because

33 Ibid., p. 254.

34 Ibid., p. 253.

35 Although I doubt Walzer would be happy with labeling as ‘ideology’ a set of beliefs entailing the existence of a supreme emergency, the fact that many such belief sets are merely ideological seems clear enough. Hence, I will follow Ladd’s lead in calling the beliefs that justify the group-target principle of discrimination ideological, but with the understanding that these beliefs being true is not thereby strictly ruled out. collective violence seems to be a broader category. First, those who adhere to an ideology of collective violence, while their ultimate target will be the enemy group as a whole, may nevertheless restrict themselves to targeting only enemy combatants. Imagine a warring nation whose dominant ideology in relation to the enemy nation fits with the ideology outlined by Ladd. But imagine that its soldiers are strictly trained to target only enemy combatants—purely out of allegiance to international treaties that bar targeting civilians. In such a case, what rules out the legitimacy of civilian targets has nothing to do with the moral qualities of enemy civilians. The commitment to the principle of noncombatant immunity is, in this case, purely formal. Given their substantive beliefs about the vitiated moral status of all members of the enemy nation, their war effort is collective violence in Ladd’s sense. But it is not terrorism.

Furthermore, it is part of my definition that a part of the group is targeted in order to affect the whole. In this respect terrorism differs from a campaign of extermination in which the objective is to directly target every part in order to eliminate the whole. The Nazi ‘final solution’ was worse than terrorism. Sometimes, the only reason terrorism does not rise to the level of genocide may be because the terrorists lack the power to pursue the latter; but it might also be because the terrorists’ goals are less extreme than extermination (for example, territorial expulsion).

Respect for Ordinary Usage

My claim is that the group-target definition of terrorism is particularly suitable as a public definition for the following reasons: (a) It respects ordinary usage by encompassing the dominant contemporary paradigms of terrorism, including the Unabomber attacks, the Oklahoma City bombing, the 1998 Omagh car bombing in Northern Ireland, the 9/11 attacks, ongoing suicide bombings in Israel (such as the August 2004 suicide bombings of two buses in Beersheba), the 2002 nightclub bombings in Bali, the public transit bombings in Madrid (in March 2004) and London (July 2005), etc.; (b) It identifies a common feature of these paradigms that distinguishes them in important ways from ordinary criminal acts and acts of war, thereby facilitating the development of public policy responses uniquely suited to a class of violence not adequately addressed by existing policies aimed at criminal violence and acts of war.

I begin with a consideration of (a). To comprehensively demonstrate (a)—that is, to show that the group-target definition encompasses most or all of the paradigmatic terrorism cases—I would need to examine each such case, in part by contextualizing the individual attacks within the broader practices and ideologies of those responsible. Such an examination falls outside the scope of this essentially philosophical essay, but I suspect that (a) is sufficiently plausible, given what is generally known and suspected about these attacks and their perpetrators, that a somewhat less comprehensive review is sufficient here.

Specifically, I want to highlight the fact that the paradigms of terrorism most vivid in our public consciousness typically fit a common model characterized by two criteria: First, they are perpetrated by sub-national groups engaged in ongoing ideological conflicts with other groups (the ‘Real’ IRA, the various militant groups such as Hamas organized to oppose Israel, the radical Islamist groups such as al-Qaeda and Jemaah Islamiah, etc.). Second, these groups routinely ignore the combatant/noncombatant distinction (even though they sometimes target combatants) in the pursuit of their goals. We might call this the ‘extremist group model’ of terrorism. The group-target definition of terrorism obviously extends to exemplars of terrorism corresponding to this model, and thus extends readily to the most prominent paradigms of terrorism today, including the 9/11 attacks.

But some paradigms of terrorism deviate from the extremist group model, including such exemplars as the Oklahoma City bombing and the Unabomber attacks. Does the group-target definition encompass them? I think so, which is one reason I favor it over a definition in terms of characteristics unique to the extremist group model. While the groups targeted in the Oklahoma City bombing and the Unabomber attacks were not ones that are usually recognized as distinctive social groups, they nevertheless functioned in the attackers’ psychologies in the same way that more conventionally recognized social groups function in the psychologies of, say, Palestinian terrorists. Timothy McVeigh clearly conceived of himself as targeting the federal government, construed as encompassing not only elected officials but employees in a federal building. For McVeigh, ‘the federal government’ took on the stature of an enemy group, so fully reified that an attack on a fairly remote federal building could be conceived of as a way to attack the group as such, to punish the federal government for its perceived wrongs.36

36 There are a number of book-length examinations of Timothy McVeigh’s psychological evolution into the person who carried out the Oklahoma City bombing. See especially Mark S. Hamm, Apocalypse in Oklahoma: Waco and Ruby Ridge Revenged (Boston: Northeastern University Press, 1997); and Lou Michel and Dan Herbeck, American Terrorist:

The case of Theodore Kaczynski, the so-called ‘Unabomber’, is more complicated. In his manifesto, Kaczynski sees his mission as the revolutionary dismantling of modern technological society. At stake, for Kaczynski, is the capacity of humans to live and flourish according to their nature, which Kaczynski claims is stifled by technological society. Thus, his self-conceived mission sets him up as a kind of liberator of humanity, and his enemy is an oppressive system.37

It might therefore seem as if Kaczynski’s bombing campaign did not operate in terms of the group-target principle of discrimination. He was an agent of humanity attacking, not a group, but a system. But the fact is that even though Kaczynski styled himself a savior of humanity against a system that had taken on a life of its own, as he understood it the system is maintained and expanded by the activities of human beings. Human beings have, for Kaczinsky, been reduced to components of a vast machine. To destroy the machine, the key components—which include human beings—have to be targeted.38 What we see, then, is a sophisticated variant of the group-targeting characteristic of terrorism. Kaczynski’s ideology treats a system, constituted by human beings in connection with technological infrastructure, as a reified whole. It is this whole that he thinks must be destroyed, and the way to destroy it is to attack its parts, choosing targets based on strategic efficacy. Insofar as the targeted parts include human beings, and the whole is a system largely constituted by human beings organized in a certain way, the Unabomber case may actually highlight the value of the group-target definition as a tool for recognizing the motivations and convictions shaping a terrorist campaign.

Timothy McVeigh and the Oklahoma City Bombing (New York: HarperCollins, 2001). Both works dramatically reveal how the federal government came to be perceived, in McVeigh’s eyes, as a monolithic evil force that needed to be violently opposed.

37 Kaczynski’s manifesto is widely available over the internet, and is available in monograph from Filiquarian Publishing.

38 In paragraphs 190 and 191 of his manifesto, ‘Industrial Society and its Future’ (available online at http://www.gseis.ucla.edu/~howard/Anarchism/Unabon/manifesto.html, accessed 13 November 2009), Kaczynski represents the fundamental human conflict as existing between ‘the power-holding elite (which wields technology) and the general public (over which technology exerts its power)’ (paragraph 191). Included within the ‘powerholding elite’ are ‘politicians, scientists, upper-level business executives, government officials, etc.’ (paragraph 190). The power-holding elite, while not only beneficiaries of the system but also victims of it, are nevertheless such integral components of the ‘system’ that needs to be destroyed that Kaczynski conceived them to be legitimate targets of violent attack. But their appropriateness as targets was not based on any personal fault of theirs, but rather their status as components of the system that needed to be destroyed.

A critic may worry, however, that the group-target definition fails to adequately capture ordinary usage, because it falls prey to the very same Coady-style challenge I used to reject the adequacy of definitions that rely on characteristics (2)–(5). Earlier, I argued that numerous attempts to define terrorism are inadequate because they prejudge terrorist motives. Insofar as the 9/11 attacks are the dominant paradigm of terrorism in the US, a definition of terrorism must extend reliably to these attacks, whatever the motives of the perpetrators should turn out to be. But wouldn’t this also apply to the principle of discrimination employed by these terrorists? Whatever their principle of discrimination, or lack thereof, shouldn’t the 9/11 attacks still qualify as terrorism?

Of course, these attacks would not have been perpetrated by anyone adhering to the principle of noncombatant immunity. Thus, that this principle was ignored can be known just by looking at the acts. But on the group-target view, terrorism is characterized not merely by the failure to apply the just war principle of discrimination, but also by the presence of an alternative principle that takes group membership as sufficient to render one a legitimate target. Does this requirement fall prey to the Coadystyle challenge?

There is good reason to think that it does not. Consider the following examples. Suppose a plane is piloted into the World Trade Center by an enraged former employee of a company housed in the building, in a maddened attempt to kill his boss. Consider, alternatively, a case in which the plane is piloted by a schizophrenic driven by voices urging her to kill herself and take as many random people as possible with her, in order to feed the God of Death. The former has more in common with the case of the outraged former employee who goes on a killing spree in his former office than it does with the paradigms of terrorism. And the latter is an example of insanity culminating in a single act of inexplicable violence. In both of these cases, we are more inclined to call the acts ‘mass murder’ than we are to call them ‘terrorism’.

These examples highlight an important relationship between terrorism and the creation of terror. On my definition, the creation of fear in a larger population needn’t be an explicit part of the terrorist’s intent. But insofar as terrorists consider anyone who belongs to a certain group to be a legitimate target, everyone in the targeted group has reason to fear (so long as the terrorists remain capable of inflicting harm). Terrorism tells all the members of a targeted group that they could be next. It does so because it treats membership in the targeted group as sufficient to render one a legitimate target. It is because the two examples above don’t embody this feature that we’re not inclined to call them terrorism, but rather cases of horrific criminal violence (or criminal insanity).

When America awoke to reports and images of the attacks on September 11, 2001, it soon became clear that America was under attack, and that being an American was sufficient to render one a legitimate target.39 This fact was enough to justify calling the attacks terrorism, whatever their motives. Of course, we might have jumped to the same conclusion had it turned out that, by some strange coincidence, all in one morning two angry former World Trade Center employees had independently hijacked planes, a schizophrenic had hijacked another plane and steered it towards the Pentagon, and a passenger enraged about delays had attacked the cockpit crew of a fourth plane, causing it to crash in Pennsylvania. But in that case, learning the truth would have led us to change our conclusion. Learning these facts, we’d say it wasn’t terrorism after all.

In short, the group-target definition extends to what are arguably the most important contemporary paradigms of terrorism, and thus meets the condition of being minimally consonant with ordinary usage. What remains to be seen, then, is how pragmatically useful such a definition can be.

Reshuffling Categories: Terrorist Crimes and Wartime Terrorism

One concern regarding the pragmatic usefulness of the group-target definition is that, while it distinguishes terrorism from most criminal violence and acts of war, there is overlap in both categories—specifically, it encompasses hate crimes and such war-time military attacks as the fire-bombing of Dresden. A critic may worry that this overlap undermines the value of the definition for the purposes of framing a unified public policy response to a distinctive class of violence.

However, the fact that a stipulative public definition of terrorism requires some reshuffling with respect to other categories does not undermine the usefulness of the definition, but may actually contribute to producing more useful public definitions in the other categories. It has long been recognized that hate crimes differ importantly from more ordinary criminal violence, and cannot be satisfactorily addressed in the same

39 Although living in the middle of nowhere in Oklahoma meant that I was not a strategic target, and was probably safe. ways.40 Moving hate crimes out of the category of criminal violence committed for private reasons, and into a category that focuses on violence that targets groups, may be an improvement for the sake of formulating coherent public policy. Likewise, when armies stop attending to the principle of discrimination and begin attacking any member of the opposing nation, something fundamental has changed. Reflecting that change in our categories may be helpful: Warriors have become terrorists.

But saying that a reshuffling of our categories may be helpful is not the same as showing that it is. A pragmatic defense of the group-target definition of terrorism calls for something more: a reason to think that such a reshuffling of categories would be helpful. I think there is such a reason. Specifically, public policy responses to violence, both criminal violence and international military threats, typically have two interrelated goals: the forceful incapacitation of the perpetrators and the concomitant deterrent effect on other potential or actual perpetrators. I will argue that the group-target definition picks out a class of violence which is such that primary reliance on these goals is likely to be counterproductive.

More precisely, the likelihood that pursuing these goals will be counterproductive is sufficiently high with respect to terrorism (as defined here) that there is presumptive reason to be suspicious of policies that adopt them as primary goals. While no one can deny that some cases of terrorism can be effectively addressed through standard incapacitation/deterrence strategies, there are uniquely powerful reasons to suspect the efficacy of such strategies in response to group-target terrorism. The group-target definition of terrorism thus picks out a class of violence about which we should be uniquely cautious when it comes to public policies that rely on these strategies.

To make this case, it will be useful first to consider why someone might reject the group-target definition on pragmatic grounds. Specifically, someone might argue that the most useful categorizing principle for developing public policy responses to violence is not the means whereby targets of violence are selected, but rather what I will call the organizational character of the perpetrators—that is, how the perpetrators organize themselves in relation to one another and to external constituencies.

40 As Ladd notes, to confuse the categories of private violence and collective violence, and to treat the latter as the former, ‘is to dangerously miss the point and foolhardily lock oneself into a misdirected and inappropriate mode of response’. Ladd, ‘The Idea of Collective Violence’, p. 22.

Clearly, the organizational character of perpetrators matters for the sake of public policy. Violent perpetrators who act alone or in temporary collaboration with a few others pose a different kind of threat to society than do well-organized sub-national groups, let alone nation-states. How a society must respond to these distinctive threats is inevitably going to vary. But to treat this fact as reason to shape our most basic public categories of violence in terms of the organizational character of perpetrators betrays an implicit prejudice. It assumes that public policy responses to violence should rely primarily on the incapacitation of perpetrators through the use of force and on the deterrent effect that such forceful responses generate. When the main objective is incapacitation—capturing or killing perpetrators, disrupting their chains of command or communication, cutting off resources, etc.—knowing the organizational character of the perpetrators seems more important than knowing the principle of discrimination they use to select targets.41 But if public policy focuses at least as much on preventing violent threats by mitigating the underlying causal forces that generate them, then understanding what motivates the selection of targets, and why these targets are seen as legitimate, may well prove to be the most important thing to know.

Before developing this latter point fully, we should recall that the grouptarget definition can accommodate the pragmatic need to pursue different methods of incapacitation depending on the organizational character of terrorists, by specifying distinct species of terrorism: independent-agent terrorism (exemplified by the Unabomber attacks and the Oklahoma City Bombing), sub-national group terrorism (exemplified by Hamas and Al Qaeda and the Real IRA), and state terrorism. Within state terrorism, one can identify two further subspecies: external state terrorism, which targets groups outside the state’s established territories or sphere of influence (exemplified by civilian bombings during World War II); and internal state terrorism, which targets groups that reside within the state or its sphere of influence (exemplified by the attacks of Saddam Hussein’s regime against the Kurds and Shiite majority in Iraq, and also by the ‘Reign of Terror’ that prevailed in post-revolutionary France, which gave us the term ‘terrorism’).

In other words, the group-target definition does not preclude making distinctions in terms of organizational character, distinctions that are

41 Which is not to say that the latter is not relevant—for example, for the detectives seeking to anticipate a killer’s next move. useful for public policy purposes. The chief question, then, is whether it offers anything of particular value.

Here, it may help to note the dubiousness of assuming that the organizational character of violent perpetrators can be understood apart from the principles of discrimination they use to select targets. As noted above, what justifies the terrorist principle of discrimination will typically be either an ideology of the sort described by Ladd or a belief structure which shares many of its features (e.g., the view that a ‘supreme emergency’ in Walzer’s sense obtains). Allegiance to such an ideology will play an important role in determining the organizational character of the terrorist group—not in terms of formal structures (which are more likely to be shaped by strategic concerns), but in terms of informal relationship networks with a broader community, networks that are crucial to both terrorist operations and recruitment efforts.

For example, if a terrorist group is formally organized into independent cells, with members of each cell having limited knowledge of other cells, this will likely be explained by the effort to preserve operations in the face of opponents with a capacity to infiltrate the group or apprehend and interrogate its members. But if these terrorist cells remain well hidden in part because a regional population sympathizes with and abets their cause, this fact may be better explained by common allegiance to the ideologies that justify the group-target principle of discrimination. The terrorists are protected and concealed by members of the broader community because they are viewed as heroes—and they are viewed as heroes rather than as criminals, despite their targeting of innocent civilians, because the broader community shares the ideological commitments which imply that these civilian targets are legitimate (because they are members of the enemy group and so deserve what they get).

And if membership in the terrorist group tends to increase as their ‘enemy’ becomes more aggressive in its interdiction efforts, this may be explained by recruitment practices that depend on sympathy with the terrorists’ ideology. Terrorists may deliberately organize themselves such that aggressive efforts to incapacitate them will inevitably have ‘collateral damage’, injuring or killing members of the very group that the terrorists have identified as the chosen group caught in a zero-sum struggle with the enemy group. The anguish caused by such collateral damage may be exploited to inspire more widespread sympathy with the terrorists’ ideology, thereby expanding the numbers of those who agree that the normal prohibition against targeting innocents in the enemy group can be waived.

Even when terrorists do not deliberately court collateral damage in this way, the effect of mounting collateral damage will likely be the same. Ongoing conflicts between Palestinian terrorists and the state of Israel are instructive here.

These considerations lead directly to the most important reason why the group-target definition has pragmatic value: the policy of grouptargeting offers a presumptive reason to suspect the adequacy of responses that rely on standard incapacitation/deterrence strategies. The scope of terrorist violence depends on how many potential and actual perpetrators accept its permissive principle of discrimination. And belief systems of the sort outlined by Ladd may be the primary tools for legitimating the group-target principle. But, as Ladd argues, these ideologies do not only legitimate violence against any member of the targeted group; they inculcate in the perpetrators a sense of purpose that transcends self-interest. These ideologies pit group against group in a manner that subordinates individual identity to group identity. And this subordination not only affects how terrorists think of their targets, but how they think of themselves. It follows that such violence often involves self-sacrifice for the sake of the group’s mission. Terrorists can therefore be expected to act selflessly for the sake of the group and its goals.

And this means that standard public policies that rely on the deterrent power of forceful incapacitation are likely to backfire. Those who operate on a group-target principle of discrimination are so routinely willing to act against self-interest for the sake of their cause that they are hard to deter. Hence, policies reliant on forceful incapacitation cannot presuppose that deterrence will fill the gap of prevention where incapacitation efforts fall short. Primary reliance on such policies will therefore require an escalation in the scope of incapacitation: since the enemy group won’t be deterred, the enemy must be wiped out. But such escalation is precisely what is most likely to make the ideological picture of a zero-sum struggle between enemy groups (or belief in the existence of a supreme emergency) more plausible to the broader population with whom the terrorists identify, especially when such escalation increases the severity of collateral damage. Such escalation can thus be expected to increase the number of prospective terrorists.

In sum, insofar as adherence to the group-target principle of discrimination indicates allegiance to ideologies of collective violence, it also indicates cases of violence in which perpetrators are unlikely to be deterred by forceful responses, in which reliance on force to prevent future attacks may thus require an escalated scope of incapacitation, and in which such an escalated scope of incapacitation can reasonably be expected to facilitate the creation of new prospective perpetrators. If all of this is right, public policy responses to terrorism may need to emphasize strategies aimed at eliminating the conditions that inculcate allegiance to ideologies of collective violence, rather than strategies aimed at incapacitating and deterring terrorists. This is not to say that incapacitation and deterrence will play no role in responses to terrorism, but rather that it will not play the primary role that it plays in response to other forms of violence.

The extent to which all of this is true may depend, in any given case, on the number of terrorists and the nascent support, in the broader population, for the ideology motivating them. If one or two terrorists act on an ideology for which there is little wider sympathy—as was arguably the case with respect both to Ted Kaczynski and to Timothy McVeigh and Terry Nichols—forceful interdiction may have few of the suggested counterproductive implications. A handful of terrorists with few public supporters conspiring to protect them might be effectively tracked down through standard detective and police work. Forceful interdiction will most likely take the form of criminal apprehension and prosecution, and so will be a measured response, with little collateral damage. But had either Kaczynski or McVeigh enjoyed significantly broader public sympathy (had there been a substantially wider audience of disaffected Luddites or Federal Government conspiracy theorists), responding to the threat they posed by relying on interdiction might well have taken a more dangerous turn.

Here, it is worth remembering that McVeigh’s decision to bomb the Murrah Building was prompted by the siege and federal assault on the Branch Davidian complex in Waco, Texas (and to a lesser extent by the FBI raid on Ruby Ridge). McVeigh’s ideology—according to which he was part of a noble struggle against an oppressive government, one that viewed the citizens’ right to bear arms as the gravest threat to its totalitarian power— inspired him to identify with the Branch Davidians, who were targeted largely because of their weapons stockpile. Thus, even though the apprehension and execution of McVeigh did not apparently inspire others to take up his mission (presumably because the response was measured, had no collateral damage, and was strongly overshadowed by the horror of the bombing itself), there is a real sense in which the Federal Government’s reliance on forceful incapacitation in the Waco case created the conditions that consolidated McVeigh and his co-conspirators in their ideology.42

Of course, the Branch Davidians were not terrorists, and the Waco assault was thus not a case of force-dependent antiterrorist policies exacerbating terrorism. But insofar as the ideological affinity of the Branch Davidians to those with the potential for terrorist violence played an instrumental role in turning these potential terrorists into real ones, this example highlights the risks of blind reliance on forceful incapacitation, even in a society where we do not typically think that terrorism-prone ideologies are widespread.

In short, the effort to minimize violence that operates according to the group-target principle of discrimination requires a special awareness of the ideological affinities and sympathies that exist within society (or between and among societies), and a sensitivity to the ways that public policies affect those affinities and sympathies. Even though McVeigh was apprehended through standard police procedures, and even though his apprehension and execution did not (as far as we know) make of him an inspiring martyr or a brother to be avenged in the eyes of like-minded ideologues, the fact remains that in cases such as McVeigh’s there is a special need to be sensitive to ideological affinities and sympathies. And this is true even if such sensitivity might culminate in the judgment that, in the particular case, conventional criminal justice policies will be appropriate to defuse the threat. My point is that such a judgment needs defense in the case of violence that fits the group-target definition. It cannot be assumed (as it is so routinely assumed with respect to other kinds of violence). For this reason more than any other, making a special category for this species of violence has particular pragmatic value.

42 Mark Hamm is especially useful in tracing out the development and entrenchment of McVeigh’s terrorist ideology in response to the Bureau of Alcohol, Tobacco, and Firearm’s assault on the Branch Davidian complex. See especially chapters 7–10 of Apocalypse in Oklahoma.

Bibliography

The Journal of Moral Philosophy originally published the essays in this collection. This bibliography lists the full contents of all articles published since the JMP was launched in April 2004. The JMP acceptance rate is about 5%.

Allen, Robert Francis, ‘Robust Alternatives and Responsibility’, Journal of Moral Philosophy

1(1) (2004), pp. 21–29.

Alm, David, ‘Deontological Restrictions and the Good/Bad Asymmetry’, Journal of Moral Philosophy 6(4) (2009), pp. 464–81.

Anderson, Scott A., ‘Of Theories of Coercion, Two Axes, and the Importance of the Coercer’,

Journal of Moral Philosophy 5(3) (2008), pp. 394–422.

Andreou, Chrisoula, ‘Standards, Advice, and Practical Reason’, Journal of Moral Philosophy

3(1) (2006), pp. 57–67.

• Reprinted in Thom Brooks (ed.), Ethics and Moral Philosophy. Leiden: Brill, 2011. Bazargan, Saba, ‘The Permissibility of Aiding and Abetting Unjust Wars’, Journal of Moral

Philosophy 8(4) (2011), pp. 513–29.

• Reprinted in Thom Brooks (ed.), Just War Theory. Leiden: Brill, 2012.

Bedi, Sonu, ‘Expressive Exclusion: A Defense’, Journal of Moral Philosophy 7(4) (2010), pp. 427–40.

Bennett, Christopher, ‘Forgiveness and the Claims of Retribution’, Journal of Moral Philosophy 1(1) (2004), pp. 89–101.

Bennett, Christopher, ‘State Denunciation of Crime’, Journal of Moral Philosophy 3(3) (2006), pp. 288–304.

Besser-Jones, Lorraine, ‘Personal Integrity, Morality and Psychological Well-Being: Justifying the Demands of Morality’, Journal of Moral Philosophy 5(3) (2008), pp. 361–83. Besser-Jones, Lorraine, ‘Drawn to the Good? Brewer on Dialectical Activity’, Journal of

Moral Philosophy 8(4) (2011), pp. 621–31.

Besson, Samantha, ‘Democracy, Law, and Authority’, Journal of Moral Philosophy 2(1) (2005), pp. 89–99.

Bittner, Rüdiger, ‘A Horse in the Basement: Nietzschean Reflections on Political Philosophy’,

Journal of Moral Philosophy 7(2) (2010), pp. 321–33.

Brake, Elizabeth, ‘Rawls and Feminism: What Should Feminists Make of Liberal Neutrality?’

Journal of Moral Philosophy 1(3) (2004), pp. 293–309.

• Reprinted in Thom Brooks and Fabian Freyenhagen (eds), The Legacy of John Rawls. New York: Continuum, 2005.

Brock, Gillian, ‘The Difference Principle, Equality of Opportunity, and Cosmopolitan Justice’, Journal of Moral Philosophy 2(3) (2005), pp. 333–51.

• Reprinted in Thom Brooks (ed.), Global Justice and International Affairs. Leiden: Brill, 2012.

Broome, John, ‘Does Rationality Consist in Responding Correctly to Reasons?’ Journal of Moral Philosophy 4(3) (2007), pp. 349–74.

• Reprinted in Thom Brooks (ed.), Ethics and Moral Philosophy. Leiden: Brill, 2011. Brown, Stephen, ‘Naturalized Virtue Ethics and the Epistemological Gap’, Journal of Moral

Philosophy 1(2) (2004), pp. 197–209.

Butler, Brian E., ‘Rorty, the First Amendment and Antirealism: Is Reliance upon Truth Viewpoint-Based Speech Regulation?’ Journal of Moral Philosophy 1(1) (2004), pp. 69–88. Carens, Joseph, ‘The Integration of Immigrants’, Journal of Moral Philosophy 2(1) (2005),

pp. 29–46.

• Reprinted in Thom Brooks (ed.), Global Justice and International Affairs. Leiden: Brill, 2012.

Carter, Alan, ‘The Evolution of Rawls’s Justification of Political Compliance: Part 1 of The Problem of Political Compliance in Rawls’s Theories of Justice’, Journal of Moral Philosophy 3(1) (2006), pp. 7–21.

Carter, Alan, ‘Political Liberalism and Political Compliance: Part 2 of The Problem of Political Compliance in Rawls’s Theories of Justice’, Journal of Moral Philosophy 3(2) (2006), pp. 135–57.

Casal, Paula, ‘Global Taxes on Natural Resources’, Journal of Moral Philosophy 8(3) (2011), pp. 307–27.

Casal, Paula, ‘Rejoinder to Pogge and Steiner’, Journal of Moral Philosophy 8(3) (2011), pp. 353–65.

Clark, Michael, ‘Retribution and Organic Unities’, Journal of Moral Philosophy 3(3) (2006), pp. 351–58.

Choi, Yoon, ‘Revisiting Kant’s Ethics: Two Challenges to the Status Quo’, Journal of Moral Philosophy 5(1) (2008), pp. 137–49.

Cholbi, Michael, ‘A Kantian Defense of Prudential Suicide’, Journal of Moral Philosophy 7(4) (2010), pp. 489–515.

Christie, Tim W., ‘Natural Separateness: Why Parfit’s Reductionist Account of Persons Fails to Support Consequentialism’, Journal of Moral Philosophy 6(2) (2009), pp. 178–95.

Clayton, Matthew, ‘Equality, Justice and Legitimacy in Selection’, Journal of Moral Philosophy 9(1) (2012), pp. 8–30.

Coyle, Sean, ‘The Ideality of Law’, Journal of Moral Philosophy 6(4) (2009), pp. 521–34. Crisp, Roger, ‘Ethics Without Reasons?’ Journal of Moral Philosophy 4(1) (2007),

pp. 40–49.

• Reprinted in Thom Brooks (ed.), Ethics and Moral Philosophy. Leiden: Brill, 2011. Cruft, Rowan, ‘On the Non-instrumental Value of Basic Rights’, Journal of Moral Philosophy

7(4) (2010), pp. 441–61.

D’Agostino, Fred, ‘The Legacies of John Rawls’, Journal of Moral Philosophy 1(3) (2004), pp. 349–65.

• Reprinted in Thom Brooks and Fabian Freyenhagen (eds), The Legacy of John Rawls. New York: Continuum, 2005.

Dancy, Jonathan, ‘Defending the Right’, Journal of Moral Philosophy 4(1) (2007), pp. 85–98.

• Reprinted in Thom Brooks (ed.), Ethics and Moral Philosophy. Leiden: Brill, 2011. Debes, Remy, ‘Emotion, Value, and the Ambiguous Honor of a Handbook’, Journal of Moral

Philosophy 8(2) (2011), pp. 273–85.

DeGrazia, David, ‘Moral Vegetarianism from a Very Broad Basis’, Journal of Moral Philosophy

6(2) (2009), pp. 143–65. de Muijnk, Wim, ‘Thinking about Normativity: Ralph Wedgwood on “Ought”’, Journal of Moral Philosophy 7(1) (2009), pp. 133–44.

Deonna, Julien A., ‘The Structure of Empathy’, Journal of Moral Philosophy 4(1) (2007), pp. 99–116.

Dietsch, Peter, ‘Distributive Lessons from Division of Labour’, Journal of Moral Philosophy

5(1) (2008), pp. 96–117.

Dobos, Ned, ‘Consistency in the Armed Enforcement of Human Rights: A Moral Necessity?’

Journal of Moral Philosophy 8(1) (2011), pp. 92–109.

• Reprinted in Thom Brooks (ed.), Just War Theory. Leiden: Brill, 2012.

Dorsey, Dale, ‘The Hedonist’s Dilemma’, Journal of Moral Philosophy 8(2) (2011), pp. 173–96.

Elster, Jakob, ‘Scanlon on Permissibility and Double Effect’, Journal of Moral Philosophy 9(1) (2012), pp. 75–102.

Enoch, David and Ehud Guttel, ‘Cognitive Biases and Moral Luck’, Journal of Moral Philosophy 7(2) (2010), pp. 372–86.

Eylon, Yuval, ‘Just Threats’, Journal of Moral Philosophy 6(1) (2009), pp. 94–108.

Flikschuh, Katrin, ‘Duty, Nature, Right: Kant’s Response to Mendelssohn in Theory and Practice III’, Journal of Moral Philosophy 4(2) (2007), pp. 223–41.

Friedman, Alex, ‘Intransitive Ethics’, Journal of Moral Philosophy 6(3) (2009), pp. 277–97. Frowe, Helen, ‘Self-Defence and the Principle of Non-Combatant Immunity’, Journal of

Moral Philosophy 8(4) (2011), pp. 530–46.

• Reprinted in Thom Brooks (ed.), Just War Theory. Leiden: Brill, 2012.

Fuller, Lisa F., ‘Poverty Relief, Global Institutions, and the Problem of Compliance’, Journal of Moral Philosophy 2(3) (2005), pp. 285–97.

• Reprinted in Thom Brooks (ed.), The Global Justice Reader. Oxford: Blackwell, 2008.

• Reprinted in Thom Brooks (ed.), Global Justice and International Affairs. Leiden: Brill, 2012.

Goddu, G. C., ‘More on Blameworthiness and Alternative Possibilities’, Journal of Moral Philosophy 3(1) (2006), pp. 69–75.

Gosseries, Axel, ‘Equality and Non-discrimination in Hiring – Introduction’, Journal of Moral Philosophy 9(1) (2012), pp. 3–7.

Grau, Christopher, ‘Moral Status, Speciesism, and Liao’s Genetic Account’, Journal of Moral Philosophy 7(2) (2009), pp. 387–96.

Green, Michael, ‘Social Justice, Voluntarism, and Liberal Nationalism’, Journal of Moral Philosophy 2(3) (2005), pp. 265–83.

• Reprinted in Thom Brooks (ed.), Global Justice and International Affairs. Leiden: Brill, 2012.

Hall, Timothy, ‘Doing Harm, Allowing Harm, and Denying Resources’, Journal of Moral Philosophy 5(1) (2008), pp. 50–76.

• Reprinted in Thom Brooks (ed.), Ethics and Moral Philosophy. Leiden: Brill, 2011. Harcourt, Edward, ‘Crisp’s “Ethics Without Reasons?”: A Note on Invariance’, Journal of

Moral Philosophy 4(1) (2007), pp. 50–54.

Hassoun, Nicole, ‘Free Trade, Poverty, and Inequality’, Journal of Moral Philosophy 8(1) (2011), pp. 5–44.

Hausman, Daniel M. and Matt Sensat Waldren, ‘Egalitarianism Reconsidered’, Journal of Moral Philosophy 8(4) (2011), pp. 567–86.

Hayward, Tim, ‘Thomas Pogge’s Global Resources Dividend: A Critique and an Alternative’,

Journal of Moral Philosophy 2(3) (2005), pp. 317–32.

• Reprinted in Thom Brooks (ed.), Global Justice and International Affairs. Leiden: Brill, 2012.

Hendrix, Burke A., ‘Authenticity and Cultural Rights’, Journal of Moral Philosophy 5(2) (2008), pp. 181–203.

• Reprinted in Thom Brooks (ed.), Global Justice and International Affairs. Leiden: Brill, 2012.

Hills, Alison, ‘Practical Reason, Value and Action’, Journal of Moral Philosophy 4(3) (2007), pp. 375–92.

• Reprinted in Thom Brooks (ed.), Ethics and Moral Philosophy. Leiden: Brill, 2011. Hirose, Iwao, ‘Aggregation and Non-Utilitarian Moral Theories’, Journal of Moral Philosophy

4(2) (2007), pp. 273–84.

Hofmeyr, Benda, ‘The Power Not to Be (What We Are): The Politics and Ethics of Selfcreation in Foucault’, Journal of Moral Philosophy 3(2) (2006), pp. 215–30.

Holroyd, Jules, ‘Substantively Constrained Choice and Deference’, Journal of Moral Philosophy 7(2) (2010), pp. 180–99.

Hookway, Christopher, ‘Ethics and the Pragmatist Enlightenment’, Journal of Moral Philosophy 3(2) (2006), pp. 231–36.

Hutton, Eric, ‘Han Feizi’s Criticism of Confucianism and Its Implications for Virtue Ethics’,

Journal of Moral Philosophy 5(3) (2008), pp. 423–53.

• Reprinted in Thom Brooks (ed.), Ethics and Moral Philosophy. Leiden: Brill, 2011. Hyams, Keith, ‘When Consent Doesn’t Work: A Rights-Based Case for Limits to Consent’s

Capacity to Legitimise’, Journal of Moral Philosophy 8(1) (2011), pp. 110–38.

James, Aaron, ‘Rights and Circularity in Scanlon’s Contractualism’, Journal of Moral Philosophy 1(3) (2004), pp. 367–74.

• Reprinted in Thom Brooks and Fabian Freyenhagen (eds), The Legacy of John Rawls. New York: Continuum, 2005.

John, Stephen, ‘Security, Knowledge, and Well-being’, Journal of Moral Philosophy 8(1) (2011), pp. 68–91.

Kahn, Leonard, ‘Legal Theory, Law, and Normativity’, Journal of Moral Philosophy 9(1) (2012), pp. 115–26.

Kaufman, Alexander, ‘Rawls’s Practical Conception of Justice: Opinion, Tradition and Objectivity in Political Liberalism’, Journal of Moral Philosophy 3(1) (2006), pp. 23–43.

King, Matt and Peter Carruthers, ‘Moral Responsibility and Consciousness’, Journal of Moral Philosophy 9(2) (2012), pp. 200–28.

Kirchin, Simon, ‘Moral Particularism: An Introduction’, Journal of Moral Philosophy 4(1) (2007), pp. 8–15.

Kirchin, Simon, ‘Particularism and Default Valency’, Journal of Moral Philosophy 4(1) (2007), pp. 16–32.

Knight, Carl, ‘Egalitarian Justice and Valuational Judgment’, Journal of Moral Philosophy

6(4) (2009), pp. 482–98.

Knight, Kelvin, ‘MacIntyre’s Progress’, Journal of Moral Philosophy 6(1) (2009), pp. 115–26. Kristjánsson, Kristján, ‘A Utilitarian Justification of Desert in Distributive Justice’, Journal of

Moral Philosophy 2(2) (2005), pp. 147–70.

Laden, Anthony Simon, ‘Taking the Distinction between Persons Seriously’, Journal of Moral Philosophy 1(3) (2004), pp. 277–92.

• Reprinted in Thom Brooks and Fabian Freyenhagen (eds), The Legacy of John Rawls. New York: Continuum, 2005.

Landemore, Hélène, ‘Politics and the Economist-King: Is Rational Choice Theory the Science of Choice?’ Journal of Moral Philosophy 1(2) (2004), pp. 177–96.

Landrum, Ty, ‘Persons as Objects of Love’, Journal of Moral Philosophy 6(4) (2009), pp. 417–39.

Lang, Gerald, ‘Luck Egalitarianism, Permissible Inequalities, and Moral Hazard’, Journal of Moral Philosophy 6(3) (2009), pp. 317–38.

Lee, Win-Chiat, ‘Cosmopolitanism with Room for Nationalism’, Journal of Moral Philosophy

9(2) (2012), pp. 279–93.

Lefkowitz, David, ‘Partiality and Weighing Harm to Non-Combatants’, Journal of Moral Philosophy 6(3) (2009), pp. 298–316.

• Reprinted in Thom Brooks (ed.), Global Justice and International Affairs. Leiden: Brill, 2012.

• Reprinted in Thom Brooks (ed.), Just War Theory. Leiden: Brill, 2012.

Lenard, Patti Tamara, ‘Motivating Cosmopolitanism? A Skeptical View’, Journal of Moral Philosophy 7(2) (2010), pp. 346–71.

• Reprinted in Thom Brooks (ed.), Global Justice and International Affairs. Leiden: Brill, 2012.

Lenman, James, ‘How to Live, What to Do: A Critical Study of Allan Gibbard, Thinking How to Live’, Journal of Moral Philosophy 3(3) (2006), pp. 359–69.

Levy, Neil, ‘A Role for Consciousness After All’, Journal of Moral Philosophy 9(2) (2012), pp. 255–64.

Liao, S. Matthew, ‘Time-Relative Interests and Abortion’, Journal of Moral Philosophy 4(2) (2007), pp. 242–56.

• Reprinted in Thom Brooks (ed.), Ethics and Moral Philosophy. Leiden: Brill, 2011. Liao, S. Matthew, ‘The Basis of Human Moral Status’, Journal of Moral Philosophy 7(2)

(2010), pp. 159–79.

• Reprinted in Thom Brooks (ed.), Ethics and Moral Philosophy. Leiden: Brill, 2011. Liao, S. Matthew, ‘The Genetic Account of Moral Status: A Defense’, Journal of Moral

Philosophy 9(2) (2012), pp. 265–77.

Lippert-Rasmussen, Kasper, ‘Publicity and Egalitarian Justice’, Journal of Moral Philosophy

5(1) (2008), pp. 30–49.

Lippert-Rasmissen, Kasper, ‘Intentions and Discrimination in Hiring’, Journal of Moral Philosophy 9(1) (2012), pp. 55–74.

Lippke, Richard L., ‘Imprisonable Offenses’, Journal of Moral Philosophy 3(3) (2006), pp. 265–87.

Lippke, Richard L., ‘Punishing the Guilty, Not Punishing the Innocent’, Journal of Moral Philosophy 7(4) (2010), pp. 462–88.

Ludwig, Bernd, ‘Kant, Garve, and the Motives of Moral Action’, Journal of Moral Philosophy

4(2) (2007), pp. 183–93.

Lynch, Sterling, ‘The Fact of Diversity and Reasonable Pluralism’, Journal of Moral Philosophy 6(1) (2009), pp. 70–93.

Mahoney, Jon, ‘Public Reason and the Moral Foundation of Liberalism’, Journal of Moral Philosophy 1(3) (2004), pp. 311–31.

• Reprinted in Thom Brooks and Fabian Freyenhagen (eds), The Legacy of John Rawls. New York: Continuum, 2005.

March, Andrew F., ‘Is There a Right to Polygamy? Marriage, Equality and Subsidizing Families in Liberal Public Justification’, Journal of Moral Philosophy 8(2) (2011), pp. 246–72.

Marino, Patricia, ‘Expressivism, Deflationism and Correspondence’, Journal of Moral Philosophy 2(2) (2005), pp. 171–91.

Marino, Patricia, ‘Moral Rationalism and the Normative Status of Desiderative Coherence’,

Journal of Moral Philosophy 7(2) (2010), pp. 227–52.

Martin, Rex, ‘Twp Concepts of Rule Utilitarianism’, Journal of Moral Philosophy 5(2) (2008), pp. 227–55.

Martin, Wayne, ‘Hegel and the Philosophy of Food’, Journal of Moral Philosophy 7(2) (2009), pp. 279–90.

Matravers, Matt, ‘“Who’s Still Standing?” A Comment on Antony Duff’s Preconditions of Criminal Liability’, Journal of Moral Philosophy 3(3) (2006), pp. 320–30.

McKeever, Sean and Michael Ridge, ‘Turning on Default Reasons’, Journal of Moral Philosophy 4(1) (2007), pp. 55–76.

Merritt, Maria W., ‘Aristotelian Virtue and the Interpersonal Aspect of Ethical Character’,

Journal of Moral Philosophy 6(1) (2009), pp. 23–49.

• Reprinted in Thom Brooks (ed.), Ethics and Moral Philosophy. Leiden: Brill, 2011. Mookherjee, Monica, ‘Feminism and Multiculturalism—Putting Okin and Shachar in

Question’, Journal of Moral Philosophy 2(2) (2005), pp. 237–41.

Moyar, Dean, ‘Unstable Autonomy: Conscience and Judgment in Kant’s Moral Philosophy’,

Journal of Moral Philosophy 5(3) (2008), pp. 327–60.

Mulnix, M. J., ‘Harm, Rights, and Liberty: Towards a Non-Normative Reading of Mill’s Liberty Principle’, Journal of Moral Philosophy 6(2) (2009), pp. 196–217.

Naticchia, Chris, ‘The Law of Peoples: The Old and the New’, Journal of Moral Philosophy

2(3) (2005), pp. 353–69.

• Reprinted in Thom Brooks and Fabian Freyenhagen (eds), The Legacy of John Rawls. New York: Continuum, 2005.

Nolan, Daniel, ‘Consequentialism and Side Constraints’, Journal of Moral Philosophy 6(1) (2009), pp. 5–22.

Norman, Richard, ‘Particularism and Reasons: A Reply to Kirchin’, Journal of Moral Philosophy 4(1) (2007), pp. 33–39.

Nussbaum, Martha C., ‘Radical Evil in the Lockean State: The Neglect of the Political Emotions’, Journal of Moral Philosophy 3(2) (2006), pp. 159–78.

O’Neill, Martin, ‘The Facts of Inequality’, Journal of Moral Philosophy 7(2) (2009), pp. 397–409.

O’Neill, Onora, ‘Experts, Practitioners, and Practical Judgement’, Journal of Moral Philosophy

4(2) (2007), pp. 154–66.

O’Neill, Onora, ‘Normativity and Practical Judgement’, Journal of Moral Philosophy 4(3) (2007), pp. 393–405.

• Reprinted in Thom Brooks (ed.), Ethics and Moral Philosophy. Leiden: Brill, 2011. Øverland, Gerhard, ‘Self-defence among Innocent People’, Journal of Moral Philosophy 2(2)

(2005), pp. 127–46.

• Reprinted in Thom Brooks (ed.), Just War Theory. Leiden: Brill, 2012.

Øverland, Gerhard, ‘Poverty and the Moral Significance of Contribution’, Journal of Moral Philosophy 2(3) (2005), pp. 299–315.

• Reprinted in Thom Brooks (ed.), Global Justice and International Affairs. Leiden: Brill, 2012.

Øverland, Gerhard, ‘Conditional Threats’, Journal of Moral Philosophy 7(2) (2010), pp. 334–45.

• Reprinted in Thom Brooks (ed.), Global Justice and International Affairs. Leiden: Brill, 2012.

• Reprinted in Thom Brooks (ed.), Just War Theory. Leiden: Brill, 2012.

Øverland, Gerhard, ‘Dividing Harm’, Journal of Moral Philosophy 8(4) (2011), pp. 547–66.

• Reprinted in Thom Brooks (ed.), Just War Theory. Leiden: Brill, 2012.

Papdaki, Lina, ‘What is Objectification?’ Journal of Moral Philosophy 7(1) (2010), pp. 16–36. Peterson, Martin, ‘The Mixed Solution to the Number Problem’, Journal of Moral Philosophy

6(2) (2009), pp. 166–77.

• Reprinted in Thom Brooks (ed.), Ethics and Moral Philosophy. Leiden: Brill, 2011. Pink, Thomas, ‘Normativity and Reason’, Journal of Moral Philosophy 4(3) (2007),

pp. 406–31.

Pogge, Thomas, ‘Allowing the Poor to Share the Earth’, Journal of Moral Philosophy 8(3) (2011), pp. 335–52.

Preda, Adina, ‘Group Rights and Group Agency’, Journal of Moral Philosophy 9(2) (2012), pp. 229–54.

Primoratz, Igor, ‘Patriotism and Morality: Mapping the Terrain’, Journal of Moral Philosophy

5(2) (2008), pp. 204–26.

• Reprinted in Thom Brooks (ed.), Global Justice and International Affairs. Leiden: Brill, 2012.

Redondo, María Cristina, ‘Legal Reasons: Between Universalism and Particularism’, Journal of Moral Philosophy 2(1) (2005), pp. 47–68.

Reitan, Eric, ‘Defining Terrorism for Public Policy Purposes: The Group-Target Definition’,

Journal of Moral Philosophy 7(2) (2010), pp. 253–78.

• Reprinted in Thom Brooks (ed.), Global Justice and International Affairs. Leiden: Brill, 2012.

• Reprinted in Thom Brooks (ed.), Just War Theory. Leiden: Brill, 2012.

Richardson, Henry S., ‘Our Call: The Constitutive Importance of the People’s Judgment’,

Journal of Moral Philosophy 5(1) (2008), pp. 3–29.

Ridge, Michael, ‘Anti-Reductionism and Supervenience’, Journal of Moral Philosophy 4(3) (2007), pp. 330–48.

• Reprinted in Thom Brooks (ed.), Ethics and Moral Philosophy. Leiden: Brill, 2011. Riley, Evan, ‘Libertarian Self-Defeat’, Journal of Moral Philosophy 7(2) (2010), pp. 200–26. Riley, Patrick, ‘Kant against Hobbes in Theory and Practice’, Journal of Moral Philosophy 4(2)

(2007), pp. 194–206.

Riley, Jonathan, ‘Ethical Pluralism and Common Decency’, Journal of Moral Philosophy 1(2) (2004), pp. 211–21.

Rivera, Lisa, ‘Harmful Beneficence’, Journal of Moral Philosophy 8(2) (2011), pp. 197–222. Ronzoni, Miriam, ‘Constructivism and Practical Reason: On Intersubjectivity, Abstraction, and Judgment’, Journal of Moral Philosophy 7(1) (2010), pp. 74–104.

Rosebury, Brian, ‘Reply to Silcox on Moral Luck’, Journal of Moral Philosophy 6(1) (2009), pp. 109–13.

Sable, Andrew, ‘Virtue for Pluralists’, Journal of Moral Philosophy 2(2) (2005), pp. 207–35. Scarre, Geoffrey, ‘Corrective Justice and Reputation’, Journal of Moral Philosophy 3(3)

(2006), pp. 305–19.

Scarre, Geoffrey, ‘The “Banality of Good”?’ Journal of Moral Philosophy 6(4) (2009), pp. 499–519.

Segall, Shlomi, ‘How Devolution Upsets Distributive Justice’, Journal of Moral Philosophy

4(2) (2007), pp. 257–72.

Segall, Shlomi, ‘Should the Best Qualified Be Appointed?’ Journal of Moral Philosophy 9(1) (2012), pp. 31–54.

Seglow, Jonathan, ‘Associative Duties and Global Justice’, Journal of Moral Philosophy 7(1) (2010), pp. 54–73.

• Reprinted in Thom Brooks (ed.), Global Justice and International Affairs. Leiden: Brill, 2012.

Shafer-Landau, Russ, ‘Moral and Theological Realism: The Explanatory Argument’, Journal of Moral Philosophy 4(3) (2007), pp. 311–29.

• Reprinted in Thom Brooks (ed.), Ethics and Moral Philosophy. Leiden: Brill, 2011. Shalom, Stephen R., ‘Killing in War and Moral Equality’, Journal of Moral Philosophy 8(4)

(2011), pp. 495–512.

• Reprinted in Thom Brooks (ed.), Just War Theory. Leiden: Brill, 2012.

Sheinman, Hanoch, ‘Raz on the Social Dependence of Values’, Journal of Moral Philosophy

3(1) (2006), pp. 77–87.

Silcox, Mark, ‘Virtue Epistemology and Moral Luck’, Journal of Moral Philosophy 3(2) (2006), pp. 179–92.

Silcox, Mark, ‘Reply to Rosebury’, Journal of Moral Philosophy 6(2) (2009), pp. 245–48. Simpson, Matthew, ‘A Paradox of Sovereignty in Rousseau’s Social Contract’, Journal of

Moral Philosophy 3(1) (2006), pp. 45–56.

Sin, William, ‘Trivial Sacrifices, Great Demands’, Journal of Moral Philosophy 7(1) (2010), pp. 3–15.

• Reprinted in Thom Brooks (ed.), Ethics and Moral Philosophy. Leiden: Brill, 2011. Sinclair, Neil, ‘Expressivist Explanations’, Journal of Moral Philosophy 9(2) (2012),

pp. 147–77.

Singleton, Jane, ‘Neither Generalism nor Particularism: Ethical Correctness is Located in General Ethical Theories’, Journal of Moral Philosophy 1(2) (2004), pp. 155–75.

Slomp, Gabriella, ‘Kant against Hobbes: Reasoning and Rhetoric’, Journal of Moral Philosophy 4(2) (2007), pp. 207–22.

Smith, M. B. E., ‘Does Humanity Share a Common Moral Faculty?’ Journal of Moral Philosophy 7(1) (2010), pp. 37–53.

Smith, Stephen R., ‘Keeping Our Distance in Compassion-Based Social Relations’, Journal of Moral Philosophy 2(1) (2005), pp. 69–87.

Sobel, David, ‘The Limits of the Explanatory Power of Developmentalism’, Journal of Moral Philosophy 7(4) (2010), pp. 517–27.

Spector, Jessica, ‘The Grounds of Moral Agency: Locke’s Account of Personal Identity’,

Journal of Moral Philosophy 5(2) (2008), pp. 256–81.

Stark, Susan, ‘A Change of Heart: Moral Emotions, Transformation, and Moral Virtue’,

Journal of Moral Philosophy 1(1) (2004), pp. 31–50.

Statman, Daniel, ‘Can Wars Be Fought Justly? The Necessity Condition Put to the Test’,

Journal of Moral Philosophy 8(3) (2011), pp. 435–51.

• Reprinted in Thom Brooks (ed.), Just War Theory. Leiden: Brill, 2012.

Steiner, Hillel, ‘The Global Fund: A Reply to Casal’, Journal of Moral Philosophy 8(3) (2011), pp. 328–34.

Stern, Robert, ‘The Autonomy of Morality and the Morality of Autonomy’, Journal of Moral Philosophy 6(3) (2009), pp. 395–415.

Stohr, Karen, ‘Kantian Beneficence and the Problem of Obligatory Aid’, Journal of Moral Philosophy 8(1) (2011), pp. 45–67.

Stone, Alison, ‘Essentialism and Anti-Essentialism in Feminist Philosophy’, Journal of Moral Philosophy 1(2) (2004), pp. 135–53.

• Reprinted in Thom Brooks (ed.), Ethics and Moral Philosophy. Leiden: Brill, 2011.

Stuchlik, Joshua, ‘A Critique of Scanlon on Double Effect’, Journal of Moral Philosophy 9(2) (2012), pp. 178–99.

Talbert, Matthew, ‘Unwitting Behaviour and Responsibility’, Journal of Moral Philosophy

8(1) (2011), pp. 139–52.

Talisse, Robert B., ‘Does Value Pluralism Entail Liberalism?’ Journal of Moral Philosophy

7(2) (2010), pp. 303–20.

Taylor, Robert S., ‘Self-Realization and the Priority of Fair Equality of Opportunity’, Journal of Moral Philosophy 1(3) (2004), pp. 333–47.

• Reprinted in Thom Brooks and Fabian Freyenhagen (eds), The Legacy of John Rawls. New York: Continuum, 2005.

Telfer, Elizabeth, ‘“Animals Do It Too!”: The Franklin Defence of Meat-Eating’, Journal of Moral Philosophy 1(1) (2004), pp. 51–67.

Teroni, Fabrice and Otto Bruun, ‘Shame, Guilt, and Morality’, Journal of Moral Philosophy

8(2) (2011), pp. 223–45.

Thomas, Alan, ‘Practical Reasoning and Normative Relevance: A Reply to McKeever and Ridge’, Journal of Moral Philosophy 4(1) (2007), pp. 77–84.

Tideman, Nicolaus, ‘Secession as a Human Right’, Journal of Moral Philosophy 1(1) (2004), pp. 9–19.

• Reprinted in Thom Brooks (ed.), Global Justice and International Affairs. Leiden: Brill, 2012.

Timmermann, Jens, ‘Good but Not Required?–Assessing the Demands of Kantian Ethics’,

Journal of Moral Philosophy 2(1) (2005), pp. 9–27.

• Reprinted in Thom Brooks (ed.), Ethics and Moral Philosophy. Leiden: Brill, 2011. Timmermann, Jens, ‘Simplicity and Authority: Reflections on Theory and Practice in Kant’s

Moral Philosophy’, Journal of Moral Philosophy 4(2) (2007), pp. 167–82.

Toner, Christopher, ‘The Virtues (and a Few Vices) of Daniel Russell’s Practical Intelligence and the Virtues’, Journal of Moral Philosophy 8(3) (2011), pp. 453–68.

Tropman, Elizabeth, ‘Renewing Moral Intuitionism’, Journal of Moral Philosophy 6(4) (2009), pp. 440–63.

Tyler, Colin, ‘Brian Barry and Writings on Social Justice from the Left’, Journal of Moral Philosophy 5(2) (2008), pp. 301–12.

Upton, Candace, ‘Virtue Ethics, Character, and Normative Receptivity’, Journal of Moral Philosophy 8(1) (2008), pp. 77–95.

Valdman, Mikhail, ‘Autonomy, History, and the Origins of Our Desires’, Journal of Moral Philosophy 8(3) (2011), pp. 415–34.

Vallier, Kevin, ‘Against Public Reason Liberalism’s Accessibility Requirement’, Journal of Moral Philosophy 8(3) (2011), pp. 366–89. van Zyl, Liezl, ‘Agent-based Virtue Ethics and the Problem of Action Guidance’, Journal of Moral Philosophy 6(1) (2009), pp. 50–69.

Vargas, Manuel, ‘Taking the Highway on Skepticism, Luck, and the Value of Responsibility’,

Journal of Moral Philosophy 6(2) (2009), pp. 249–65.

Wall, Edmund, ‘The Real Direction of Dancy’s Moral Particularism’, Journal of Moral Philosophy 8(4) (2011), pp. 587–612.

Ward, Lee, ‘Locke on Punishment, Property and Moral Knowledge’, Journal of Moral Philosophy 6(2) (2009), pp. 218–44.

Warmke, Brandon, ‘Is Forgiveness the Deliberate Refusal to Punish?’ Journal of Moral Philosophy 8(4) (2011), pp. 613–20.

Webber, Jonathan, ‘Virtue, Character and Situation’, Journal of Moral Philosophy 3(2) (2006), pp. 193–213.

• Reprinted in Thom Brooks (ed.), Ethics and Moral Philosophy. Leiden: Brill, 2011. Weirich, Paul, ‘Utility Maximization Generalized’, Journal of Moral Philosophy 5(2) (2008),

pp. 282–99.

Wenar, Leif, ‘The Unity of Rawls’s Work’, Journal of Moral Philosophy 1(3) (2004), pp. 265–75.

• Reprinted in Thom Brooks and Fabian Freyenhagen (eds), The Legacy of John Rawls. New York: Continuum, 2005.

Westphal, Kenneth R., ‘From “Convention” to “Ethical Life”: Hume’s Theory of Justice in Post-Kantian Perspective’, Journal of Moral Philosophy 7(1) (2010), pp. 105–32.

White, Heath, ‘Fitting Attitudes, Wrong Kinds of Reasons, and Mind-Independent Goodness’, Journal of Moral Philosophy 6(3) (2009), pp. 339–64.

Wiland, Eric, ‘On Indirectly Self-defeating Moral Theories’, Journal of Moral Philosophy 5(3) (2008), pp. 384–93.

Wilkinson, Dominic, ‘Should We Replace Disabled Newborn Infants?’ Journal of Moral Philosophy 8(3) (2011), pp. 390–414.

Williams, Reginald, ‘Morality and Privilege’, Journal of Moral Philosophy 5(1) (2008), pp. 118–35.

Wolff, Jonathan, ‘Equality: The Recent History of an Idea’, Journal of Moral Philosophy 4(1) (2007), pp. 125–36.

Zaibert, Leo, ‘The Fitting, the Deserving, and the Beautiful’, Journal of Moral Philosophy 3(3) (2006), pp. 331–50.

Zaibert, Leo, ‘The Paradox of Forgiveness’, Journal of Moral Philosophy 6(3) (2009), pp. 365–93.

Zaibert, Leo, ‘On Forgiveness and the Deliberate Refusal to Punish: Reiterating the Differences’, Journal of Moral Philosophy 9(1) (2012), pp. 103–13.

Ziegler, Rafael, ‘Tracing Global Inequality in Eco-space: A Comment on Tim Hayward’s Proposal’, Journal of Moral Philosophy 4(1) (2007), pp. 117–24.

Thom Brooks

Just War Theory Studies in Moral Philosophy #4

Front Matter

Title Page

Publisher Details

Contents

Notes on Previous Publications

List of Contributors

Introduction

Part I: Just War and Unjust War

Part II: Just War and Self-defence

Part III: Innocent Non-combatants and Self-defence

Part IV: Just War Theory and Terrorism

Conclusion

Bibliography

Part One: Just War and Unjust War

The Permissibility of Aiding and Abetting Unjust Wars

1. Introduction

2. Preliminaries

3. Aiding and Abetting an Unjust War

4. Aiding and Abetting Narrowly Unjust Wars

5. Conclusion

References

Dividing Harm

Preliminaries

A Two-part Theory of Self-defence

The Asymmetrical Fair Share Procedure

A Fair Share

Luck Reconsidered

Contractual Considerations

Epistemic Uncertainty

Epistemic Transparency

Division

Asymmetry

Consistency in the Armed Enforcement of Human Rights: a Moral Necessity?

I

II

III

IV

V

Part Two: Just War and Self-defense

Conditional Threats

Risk

Moral Responsibility

Asymmetry

War

Can Wars Be Fought Justly? The Necessity Condition Put to the Test

for Helpful Comments on Earlier Drafts, I Am Greatly Indebted to Yitzhak Benbaji, Yuval Eylon, Michael Gross, and Saul Smilansky.

I. Necessity and Proportionality: Preliminaries

II. The Necessity Condition and Warfare

Part Three: Innocent Non-combatants and Self-defense

Self-defence Among Innocent People

Preliminaries

Killing in War and Moral Equality*

I Am Grateful to Jeff Mcmahan for Extremely Generous and Valuable Comments. Thanks as Well to an Anonymous Reviewer for Helpful Suggestions.

Self-defence and the Principle of Non-combatant Immunity

Conclusion

Partiality and Weighing Harm to Non-combatants

Part Four: Just War Theory and Terrorism

Defining Terrorism for Public Policy Purposes: the Group-target Definition

Bibliography

Just War Theory
Studies in Moral Philosophy #4