Challenges for Humanitarian Intervention: Ethical Demand and Political Reality 019881285X, 9780198812852

Citation preview

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

Challenges for Humanitarian Intervention

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

Challenges for Humanitarian Intervention Ethical Demand and Political Reality

EDITED BY

C. A. J. Coady, Ned Dobos, and Sagar Sanyal

1

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

3

Great Clarendon Street, Oxford, OX2 6DP, United Kingdom Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries © the several contributors 2018 The moral rights of the authors have been asserted First Edition published in 2018 Impression: 1 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by licence or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this work in any other form and you must impose this same condition on any acquirer Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY 10016, United States of America British Library Cataloguing in Publication Data Data available Library of Congress Control Number: 2017962087 ISBN 978–0–19–881285–2 Printed and bound by CPI Group (UK) Ltd, Croydon, CR0 4YY Links to third party websites are provided by Oxford in good faith and for information only. Oxford disclaims any responsibility for the materials contained in any third party website referenced in this work.

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

Contents List of Contributors Morality, Reality, and Humanitarian Intervention: An Introduction to the Debate C. A. J. Coady

vii

1

1. Complicating the Moral Case of Responsibility to Protect: Kosovo and Libya Stephen Zunes

15

2. Why Sovereignty Matters Despite Injustice: The Ethics of Intervention Richard W. Miller

38

3. Women and Humanitarian Intervention Janna Thompson

59

4. Humanitarian Intervention and Non-Ideal Theory Ramon Das

75

5. The Leeriness Objection to the Responsibility to Protect Marco Meyer

100

6. On the Uses and ‘Abuses’ of Responsibility to Protect Ned Dobos

123

7. Scrutinizing Intentions Chrisantha Hermanson

139

8. ‘Words Lying on the Table’? Norm Contestation and the Diminution of the Responsibility to Protect Aidan Hehir

165

9. Responsibility to Protect, Polarity, and Society: R2P’s Political Realities in the International Order Robert W. Murray and Tom Keating

183

10. Closing the R2P Chapter: Opening a Dissident Current within Philosophy of War Sagar Sanyal

200

Name Index General Index

219 221

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

List of Contributors C. A. J. C OADY , University of Melbourne R AMON D AS , Victoria University of Wellington N ED D OBOS , University of New South Wales, Canberra A IDAN H EHIR , University of Westminster C HRISANTHA H ERMANSON , Blue Ridge School, Virginia T OM K EATING , University of Alberta M ARCO M EYER , University of Cambridge and University of Groningen R ICHARD W. M ILLER , Cornell University R OBERT W. M URRAY , Frontier Centre for Public Policy and the University of Alberta S AGAR S ANYAL , University of Melbourne J ANNA T HOMPSON , LaTrobe University S TEPHEN Z UNES , University of San Francisco

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

Morality, Reality, and Humanitarian Intervention An Introduction to the Debate C. A. J. Coady

There are two profound points of departure for discussions of the moral evaluation of humanitarian intervention and its partial echo in international law, the Responsibility to Protect. The first is the distressingly massive damage sometimes inflicted on people by their own governments (or other politically powerful and unhindered agents), and the second concerns the appalling disasters and ravages of war. The first cries out to outsiders for action to prevent or discontinue the horror (which may itself involve forms of warfare, such as civil war), but the second cautions against those forms of intervention or protection that themselves threaten to replace the horror with something as or even more damaging. The tension between these two instincts has been a significant issue in many violent conflicts in very different parts of the world in the last part of the twentieth century and into the present day, and, of course, has earlier historical precedents. This tension continues in comment and discussion of the Responsibility to Protect doctrine firmly proclaimed by the United Nations in 2005, and, of course, the document itself stands as a sincere morally motivated response not only to various outrages and massacres that shocked the conscience of so many, but also to inaction and even perceived indifference by outsiders to those episodes of extensive inflicted suffering on innocent people. That doctrine, to be sure, is by no means exclusively concerned with military interventions, indeed its primary stress is upon early preventative measures and diplomatic and other non-violent measures within the scope of the responsibilities it proclaims. There are some admirable and some contentious features of these other responsibilities, but the R2P doctrine does include the military intervention option (though for a

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

restricted number of offences—genocide, crimes against humanity, war crimes, and ethnic cleansing). It is this military option, or some version of it, that the term ‘humanitarian intervention’ is usually taken to mean, and it raises more, and often different, problems than the non-violent responsibilities do, and that is the primary reason why the chapters in this book concentrate upon assessment of the military intervention option. We are also concerned principally with the underlying moral issues in the humanitarian intervention debate and not the legal regime that the moral concerns have inspired, important as those laws are.

Political Realism and its Flaws The genuine revulsion against major violations of human rights in foreign parts is first of all a moral impulse, and as such an unimpeachable one. Its translation into political action however is another matter. There are those who would decry any role to morality in the politics of foreign affairs, but I am not one of them. Nor, I think, are any of the contributors to this volume. The deniers are sometimes mere cynics, but more respectably they can claim the title of ‘political realists’ in the complex tradition boasting, amongst many others, such theorists as Max Weber, George Kennan, E. H. Carr, Reinhold Niebuhr, Hans Morgenthau, Dean Acheson, and Henry Kissinger, and often claiming antecedents in Nicolo Machiavelli and Thomas Hobbes. I have charted the differences and similarities amongst a number of these theorists elsewhere, and I have there also argued that the attempt to replace morality in the sphere of international relations with some overriding concept of ‘national interest’ is doomed by the fact that this concept itself can most plausibly be introduced and defended by appeal to some essentially moral considerations. Moreover, once morality is back in play the arguments supporting an exclusive role for national interest can be seen as indefensibly narrow. Such arguments focus on the undoubted responsibilities that a national political leader owes to her citizenry and often adduce as well the epistemic superiority of access to the reality of the national interest against the supposed uncertainties of moral judgement. But political leaders’ responsibilities for the welfare of their citizens are not the only responsibilities they have, nor are their citizens’ interests and moral concerns so narrow as to be limited to what promotes restricted local goods. Moreover, on closer examination what is in the national interest is often far more opaque than such theorists characteristically suppose, and on many occasions contrasts with the certainty that morality can provide. So the sort of collective egoism generated by the restriction of moral concern to an idea of national interest is unviable.

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

MORALITY , REALITY , AND INTERVENTION



A further point that can be made about many realist writings is that the picture of morality (or idealism) that realists often work with and criticize can be a very distorted one. They have a tendency to think of the morality of ordinary folk, at any rate, as something lofty, clear, rigid, and inflexible, most at home in private settings, such as the family or religion (a private affair, they suppose). As such, they say, it is not capable of handling the complex realities of politics and international affairs. This picture of a sharp contrast between private and public morality is much too simple: for one thing, the flexibility of thought and imagination required to handle justice and other moral imperatives within a family and other ‘private’ realms is very considerable: consider, for instance, the often difficult choices regarding education of children, or the work options of parents, or in wider contexts the compromises and dilemmas of workplace relations. The issues of acute and disturbing compromise are often urgent and publicized in political arenas, but they pervade all spheres of life. On the other hand, politics is not always a matter of ‘dirty hands’ decisions and often involves issues of administrative efficiency and routine justice, even if there is a tendency to bungling or corruption in politics as in other areas. On the positive side, it can be argued that many of the realists’ criticisms of morality in foreign affairs can be interpreted more generously as targeting, albeit in an unsatisfactory fashion, the dangerous role of what I have called elsewhere ‘moralism’.¹ Moralism and moralizing are distorted versions of genuine moral thinking and constitute a sort of vice that is all the more dangerous for its mimicking of morality. Moreover, like vices, moralism not only indicates a defect of character in the moralizer, but often has profoundly bad effects on others due to the actions and policies supported by the moralizer. As I have argued, moralism takes many different forms, but in some of them it involves a lack of self-awareness in the making of moral judgements on others and a sort of selfrighteousness that makes it easy to overlook facts relevant to moral action. Witness these flaws at work in the following quotation from American President William McKinley facing the question of annexing the Philippines in the early twentieth century. McKinley, a devout Methodist, had apparently spent much of the night in prayer about what to do with the Philippines after the US victory in the Spanish-American war: And one night late it came to me this way—I don’t know how it was, but it came: (1) That we could not give them back to Spain—that would be cowardly and dishonourable; (2) that we could not turn them over to France and Germany—our commercial rivals in the Orient—that would be bad business and discreditable; (3) that we could not leave them to themselves—they were unfit for self-government—and they would soon have

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

anarchy and misrule over there worse than Spain’s was; and (4) that there was nothing left for us to do but to take them all, and to educate the Filipinos, and uplift and civilize and Christianize them, and by God’s grace do the very best we could by them, as our fellowmen for whom Christ also died.²

Behind the high moral tones of ‘discreditable’, ‘cowardly’, and ‘dishonourable’, not to mention ‘unfit for government’, and ‘anarchy and misrule’, there is the clear note of commercial opportunism, though no doubt the moral sentiments are sincerely meant. In addition, there is the obfuscating denial of plain facts involved, since the Filipinos had been Christian for four hundred years. Of course, they hadn’t been McKinley’s type of Christian, but that merely shows the underlying, no doubt unconscious, partiality of his outlook. Many Filipinos were not only Christian, but civilized enough to resent Spanish rule and mount an active movement for political independence long before McKinley came on the scene. They had their martyrs too, including their famous novelist Jose Rizal. As for commercial opportunism, I cannot forebear mention of a personal anecdote that also illustrates commercial national interest in an unexpected (certainly to me) direction. In the 1980s I took part in a workshop on nuclear deterrence in Washington, DC, involving philosophers, political scientists, and strategic analysts. In one exploration of the purposes of ‘extended deterrence’ one of the strategists stated that a significant purpose of US nuclear deterrence was the protection and advancement of American business interests abroad! The term ‘extended deterrence’ was usually understood, at least for public consumption, as deterring military attacks by US enemies on countries allied to the USA or whose survival was politically significant for the security of those allies. I was astonished at the casual and perfectly overt expression of the idea that a serious threat of the nuclear devastation of hordes of innocent civilians on behalf of American business interests was part of the purposes of United States foreign policy. My surprise was compounded by the fact that the speaker seemed a decent and congenial human being. I should add that in the book eventually published as a result of the workshop there is no reference to such a broad interpretation of ‘extended’.³ McKinley’s remarks, and those of my workshop colleague, also exhibit the often hidden, and even sometimes unconscious, role of what now would be called ‘geopolitical policy’ that must always be considered as part of the context for contemporary humanitarian interventions. It will form a significant focus for a number of investigations in this book. McKinley’s comments in justification and many public statements like them subsequently often exhibit a combination of exaggerated self-confidence in the likely success of the means and power to be used in pursuit of both altruistic and self-interested objectives and a certain opacity of understanding about the precise

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

MORALITY , REALITY , AND INTERVENTION



nature of those objectives. The American neo-conservative objective of bringing democracy to the Middle East that was partly responsible for the Iraq invasion naively exaggerated the power of military means for beginning that process, and compounded this mistake by employing a simplistic picture of democracy as a secular paradise of ‘free elections’. This picture is not only an impoverishment of what a genuine, or at any rate healthy, democracy stands for but it is also absurdly adrift of the cultural and historical factors that must be in place for the movement from dictatorships and other non-democratic forms of governance to reach feasible democratic processes and relevant internalized attitudes.

‘Something Must Be Done’ So, although the strictures on morality by the political realists are confusingly presented, the gist of some of them can be given an interpretation that should encourage caution about enthusiasm for military adventures aimed at rescue and regime change even on behalf of victims of severe persecution. Such caution is likely to be met by the cry ‘We must do something!’—a cry that is understandable in the face of widespread suffering, especially when inflicted by malicious, powerful, and largely unaccountable agents. This anguished appeal, however, masks several dubious assumptions. The first is that there really is something that we can not only do, but also do effectively. Confronted with massacres, forced expulsions, deliberate starvation, and the like, the harsh reality is that sometimes there is nothing of the relevant sort that we can do. We can of course sympathize, protest, complain, and pray, but in terms of preventive action there is sometimes nothing available that is likely to be directly effective. The second dubious thought behind the cry is, often enough, the conviction that military intervention is the proper ‘something’. When the extreme violations have run their course or are coming to a head, accusations are made that ‘We (the West, the European Union, the United States or whoever) stood by and did nothing’ and the usual implication is that military intervention was the action that was neglected in favour of other ineffectual efforts like diplomacy, sanctions, or widespread condemnation. Often enough, these were indeed ‘things done’ but failed to prevent or ameliorate the disaster. What is frequently forgotten about such alternatives to violent invasion is that military efforts need have no striking superiority over these non-violent efforts when it comes to the likelihood of success. The record of military humanitarian interventions by foreigners (a number of them discussed in chapters of this book) is hardly such as to inspire confidence in their beneficial outcomes. Indeed, in most wars there is little realistic certainty of outcome, though the protagonists initially usually think otherwise, and even the victors often pay

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

dearly for their efforts. Certainly, there are a few humanitarian wars that seem on balance to have achieved a degree of humanitarian rescue or mitigation of grave harms, but very many have had either ambiguous or negative outcomes. An understandable response to such an argument is that hindsight is a wonderful thing but it is not available to political decision-makers at the time of a humanitarian emergency. But there are two replies to this. The first is that enough experience of military interventions for humanitarian rescue is now available to provide more informed foresight in the face of a new challenge. The second is that many past interventions proceeded in the face of criticism that pointed out at the time some of the pitfalls that lay in the path of the military option. Consider the humanitarian aspects of the 2003 Iraq invasion where domestic atrocities by Saddam Hussein were much mooted in the rhetoric surrounding the intervention and became increasingly prominent as the other ostensible reasons for the invasion fell apart, notably the weapons of mass destruction scenario. The earlier attack upon Afghanistan began as retaliation for 9/11 against Osama bin Laden’s forces in the mountains, extended to the project of ousting the Taliban government (allegedly complicit in 9/11), then proceeded largely on the basis of compassionate claims about defending the human rights of Afghan citizens from the brutalities of the Taliban and other forces. But at the time of initiation of both interventions, there were many voices raised against them (especially Iraq, less so Afghanistan) and some of the concerns expressed, for instance by politicians in the UK, Europe, and even in the American Congress, and by some intelligence sources, were related to the sort focused on earlier in this introduction.

Intervention and the Just War Framework To this point my discussion has stressed distorting effects of a moralistic outlook on the world and the negative implications of this for military humanitarian intervention. Crucial to this is a certain blindness to relevant empirical, cultural, and moral facts about the world and about oneself. My argument and many of the chapters in this book may be seen as seeking to warn against the dangers of this distortion and blindness in the context of armed humanitarian intervention. One perspective from which to view this distortion and blindness is that of just war theory. In particular, there is often amongst advocates of intervention a certain failure to heed, or a tendency to underplay, what I have elsewhere called the ‘prudential conditions’ for the jus ad bellum.⁴ In particular, in the context of humanitarian war, there is a tendency to think either that the existence of a just cause alone is sufficient to validate military intervention, or that the just cause

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

MORALITY , REALITY , AND INTERVENTION



carries such weight that the prudential considerations can be treated lightly. The jus ad bellum requires satisfaction not only of just cause, but also the conditions of legitimate authority, right intention, last resort, probability of success, and proportionality of response. Of these, the last three are clearly what I have called ‘prudential conditions’ (legitimate authority and right intention may also be prudential but have other moral elements) and, although they are secondary to just cause in that they come into play only when a just cause has been established, they function to put a brake upon imprudent resort to military violence no matter how righteous or decent its putative objective. (I should note that I am here treating prudence and imprudence as moral notions and that is contrary to the practice of many philosophers who treat prudence as somehow equivalent to selfinterest which, in turn, they treat as external, if not antithetical to morality. My reasons for this I have treated more fully elsewhere, but suffice to say here that prudence should be seen, as the ancients tended to view it, as a virtue concerned with the best implementation of moral values and concerns, whether those involve the agent’s own wellbeing or those of others. Certainly, we can exercise prudential judgement about the best way to ensure that something valuable we have found in the street can be returned to its owner. Likewise, a politician’s prudent concern to explore other ways than violence to solve a political problem needn’t involve an interest only in his own political survival or even that of his own nation.) There are of course many complex philosophical issues raised by these just war conditions, such as questions about how to interpret the stringency or scope of last resort criterion or the meaning of ‘success’, the degree of probability involved in calculating the prospect of success, and the meaning to be given to proportionality, but we cannot delve into those here. Suffice to say that all these conditions call attention to the practical difficulties in terms of implementation and outcomes that can afflict otherwise morally good intentions. So it is not enough to perceive and emotionally respond to the unjustified, gross, and massive violations of human rights in some distant country and to want to save the afflicted and prevent further outrages, because attention must be given to the circumstances and the likely consequences of any interventional action, especially where this means the employment of extreme lethal violence. This is to say that a war with a just cause is not eo ipso a just war, a point worth making because it is often ignored, especially in public debate. There is much talk these days about ‘unnecessary wars’ and that terminology echoes a widely accepted necessity condition on the conduct of war that is often cashed in in terms of ‘military necessity’. Military necessity has some connection with legitimate restrictions on tactics available in waging wars, the area normally

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

covered by the jus in bello. But, as often understood, it is much narrower than the proportionality and other requirements of the jus in bello, just because ‘military objectives’ themselves can be, and often are, understood without reference to moral considerations. That echo of necessity has gained currency in recent years, as in the common criticism, for instance, that the 2003 invasion of Iraq was ‘an unnecessary war’. In this usage, necessity has a more moral flavour and is more akin to the prudential conditions of the jus ad bellum discussed above. When the Australian historian, Henry Reynolds recently wrote a blistering critique of Australia’s involvement in the Boer War, and spelled out the deep significance of that involvement for future Australian attitudes to participation in distant wars and its developing ‘love affair with the warrior’, he called his book ‘Unnecessary Wars’.⁵ I have indicated the ways that precautionary attention to practical realities can be accommodated by, indeed fit neatly within, the framework of just war thinking, but other approaches to such realities may also be possible. Indeed, one of the chapters in this book, that of Sagar Sanyal, explicitly rejects the just war framework as too abstract to fully deal with the role of states in the international order, and he advocates a critical approach to humanitarian intervention that makes use of Marxist analysis. My own view is that the substance of many of his criticisms is compatible with a suitably nuanced version of the jus ad bellum, but readers can make up their own mind about his challenging argument. Other problems with humanitarian intervention arise less from moralistic blindness to the awkward circumstances and likely bad outcomes of humanitarian intervention than from a deliberate abuse of the just war (jus ad bellum) conditions. This could be seen as the flouting of the right intention condition or even of the legitimate authority condition (for instance, a deliberate failure to gain UN Security Council approval). The prospect of such abuse is indeed a difficulty that has been urged against the resort to humanitarian intervention, and the nature and complexity of such abuse and its relevance to an overall assessment of the case for intervention are also addressed in some of the chapters that follow, especially in the chapters by Dobos and by Hermanson. A final point that is worth noting is the way that some recent comments on humanitarian intervention have pointed to another ethical consideration that can be problematic for such intervention, and that is the issue of ‘moral hazard’. Alan Kuperman has been prominent in urging this difficulty and discussing possible solutions. Kuperman’s general point is that apparent guarantees of external support for persecuted groups, for which there seem to be excellent moral reasons, can embolden such groups to embark upon actions such as armed revolution or secession that are particularly hazardous for them (and others) because of their optimistic and excessive reliance for success upon the

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

MORALITY , REALITY , AND INTERVENTION



guarantees.⁶ The persecuted groups can find that their outside help comes too little or too late or in some cases not at all. In just war terms, the persecuted may have just cause but will be prone to miscalculating their prospects of success because of such proclamations as R2P. Confidence in the availability of such outside support, especially in its military form, may also lead the persecuted into treating too lightly the possible utility of non-military options, hence violating the last resort condition. The prospect of outside intervention and the reliance upon it by the aggrieved groups may also provide incentives to the incumbent government for increased violence and propaganda to support it by rallying their own supporters against supposedly malicious foreign meddling.

Chapter Summaries I turn now to summarize briefly the chapters that follow. In Chapter 1, Stephen Zunes examines the military interventions in Kosovo and Libya (often advanced as successful humanitarian interventions), and argues that the interventions did more harm than good. They escalated the level of killings (by regime and rebels) in both cases, fanned nationalism in the first case and sectarian militias in the second. The general explanations behind both cases are that the intervening powers are rarely neutral or impartial, and that military intervention changes the strategies of the target regime and of the rebelling parties. However, these explanations are not principles or assumptions that drive the argument. Rather, the bulk of the chapter details the political history and aftermath of the two cases, and the form of argument is to base judgements on detailed specifics, not all of which will be summarized here. Two themes however stand out. One is an argument for the efficacy of strategic non-violent action internally, and preventive diplomacy externally, as ways to contest or overthrow repressive regimes. Zunes notes the successes of the non-violent movements in both case studies, and critical moments at which they could have been supported by preventive diplomacy but were not. A second theme is sensitivity to broader ramifications of military intervention for international affairs. The Kosovo intervention persuaded many Russians that NATO is an aggressive alliance bent on eastward expansion, and this fanned Russian nationalistic sentiment, prompting military build-up. NATO’s (and their allies’) arming of various rebel groups in Libya led to a subsequent flow of arms and conflict into neighbouring countries. Richard Miller’s chapter advances a position of great reluctance to launch humanitarian military intervention, but does not appeal to the inherent value of sovereignty or of autonomous political community. Rather he appeals to the

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

likely consequences of such intervention—both within the target country and for diplomacy and international relations. He considers four types of candidate cases for intervention: stable tyrannies, unstable tyrannies, popular secession, and ongoing large-scale killing and displacement. In the last of these, we ought to be disposed to support interventions (absent contrary reasons) since the likely consequences that plague the other three types are less of a challenge. Stable tyrannies are usually maintained because the regime has engineered a wide base of support among elites, perhaps harnessing religious, ethnic, and other divisions to maintain support and divide opposition. External overthrow of such a regime risks violent conflict between divided groups in society, but can also foster patriotic sentiment among militia to repel what is perceived as foreign invasion. In unstable tyrannies there is a rebel group seriously contesting power. The internally driven overthrow of the tyrannical regime is preferable, but any case for external intervention should rest on likely consequences (rather than the regime’s forfeiture of its sovereignty). In popular secession, not only is it difficult to judge the prospects for autonomy short of secession, but external intervention can stoke Great Power worries about spheres of influence and stoke military build-up. Janna Thompson examines the prospects for the R2P framework being used to combat women’s oppression, with specific attention to attacks on women in Bosnia in the 1990s and in Afghanistan under the Taliban regime. She argues that while mass rape in Bosnia would constitute just cause for humanitarian intervention, other desiderata, like prospects of success or avoiding disproportionate harm, present greater difficulty. Thompson notes two assumptions commonly present in the humanitarian intervention debate. One, humanitarian intervention is supposed to target repressive states or murderous militias, not the cultural traditions of the population. Two, the repressing regime is ethnically, racially, religiously, or ideologically distinct from the repressed. She argues that these assumptions did not hold in Afghanistan, since the Taliban’s repressive laws regarding women had some prior cultural basis in parts of the society. Hence, women can hardly be put in separate enclaves, given territory of their own or effectively protected by a peacekeeping operation. She is more optimistic about a focus on women’s oppression in aspects of R2P other than military intervention: namely in prevention and rebuilding. She considers ways diplomacy might have worked in the early years of Taliban rule, and possible roles for women in post-intervention reconstruction and rebuilding, especially in working with community groups and NGOs. Ramon Das argues that the philosophical discussion of humanitarian intervention (HI) and R2P could be improved if it were less ideal-theoretic. He identifies two ideal-theoretic assumptions. One, in target states where

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

MORALITY , REALITY , AND INTERVENTION



humanitarian intervention is being considered, there are two distinct and easily identified groups: ‘bad guys’ committing serious human rights abuses, and innocent civilians against whom the abuses are being committed. Two, external to the target state in question there are suitably qualified ‘good guys’—prospective interveners who possess both the military power and the moral integrity. If the assumptions hold, the prospects for successful humanitarian intervention are much greater. As a contrast, some possible non-ideal assumptions are that (i) there are many bad guys in a civil war, and (ii) the good guy intervener is itself supporting some of the bad guys. If these non-ideal assumptions hold, prospects for successful HI are small. Normative conclusions about an intervention— particularly insofar as they depend on the accuracy of the two ideal assumptions—are to a large extent ‘driven’ by relevant empirical considerations. To demonstrate the fruitfulness of his theoretical framework, Das examines the Rwanda and Kosovo wars and argues against the ‘consensus position’ on both. Marco Meyer argues that even non-abusive interventions (those that are motivated purely by altruistic concern, have a just cause, are a last resort, etc.) are morally problematic due to their effects on the international order. The trouble, according to Meyer, is that ‘bystander states’—those that are neither prosecuting the intervention nor targeted by it—usually do not have sufficient direct evidence that the intervention is just and properly motivated, nor can they accept the testimony of the intervening state. Thus, for all that bystander states know, any and every instance of R2P enforcement is abusive: an act of unjust international aggression masquerading as something else. This, in turn, weakens the willingness of these bystander states to comply with the non-aggression norm themselves, since states are ‘conditional co-operators’ according to Meyer—they abide by norms only insofar as they are reasonably assured that other states in the international arena are abiding. Critics of the R2P doctrine routinely warn of its abuse potential, but often leave under-described what this abuse consists of, and how it differs from the proper, legitimate use of R2P (if there is such a thing). While Ned Dobos thinks that even ‘abusive’ interventions can sometimes be justified, his chapter seeks to remedy this descriptive neglect. He distinguishes three kinds of foreign intervention that might, for different reasons, be considered abuses or misappropriations of the R2P norm. The first is where the language of R2P is used to publicly justify an intervention that in fact has little to do with protecting vulnerable populations. These are cases in which humanitarian rhetoric is used as window-dressing for economic or political self-aggrandizement. The second kind of R2P abuse involves its unilateral implementation, or enforcement by agents other than the UN Security Council. Third, the R2P norm is arguably abused when it is

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

over-extended, as where it is invoked to justify the forcible democratization of undemocratic states, or the liberalization of illiberal ones, rather than being limited to the prevention of grave and systematic human rights violations (massacre, ethnic cleansing, and the like). Chrisantha Hermanson’s contribution is also about abuses of R2P, but she concentrates upon those that arise from the failure to meet the right intention clause of just intervention. She teases apart our intuitions on when the presence of a secondary motivation counts against an intervening state, and when it does not. The mere presence of a secondary motivation is distinguished from actions pursuing that motive beyond actions commensurate with the just cause. These additional actions themselves may be just or unjust, and might or might not require the use of force. Or they might or might not have gained the prior consent of the target population. While consent to the primary aim of rescue from mass killings can be assumed in emergency cases, consent to the secondary aims must typically be sought and not assumed. Each of these points is relevant to whether the intervention would constitute an abuse of the framework. Concerning post-atrocity reconstruction, there are again possibilities of abuse. Here, difficulties around ‘right intention’ arise in offers of aid or loans for reconstruction that are conditional on imposed criteria. This picks up the discussion thread about consent and coercion (from the lack of acceptable alternatives). While it is often unjust to impose conditions on aid, Hermanson identifies an interesting set of cases where the intervening state can justly impose conditions, for example, where they are necessary to prevent moral wrongs that would likely occur in the event of unconditional aid. In such circumstances, there may even be an obligation to impose appropriate conditions. Aidan Hehir laments that the Responsibility to Protect has become nothing more than ‘a largely ineffective empty signifier’: it may have found its way into state discourse, but it has not meaningfully influenced state behaviour. At least fifty Security Council resolutions to date have mentioned R2P, but by Ban Ki-moon’s own admission ‘the frequency and scale of atrocity crimes have increased’. This, according to Hehir (echoing to some extent the discussion of abuses), is the result of R2P having been ‘co-opted’ over time. Co-option occurs when parties hostile to a norm publicly endorse it, but work to redirect and manipulate its evolution so that the norm comes to serve their interests and values. Such is the case with R2P and the emerging BRIC powers says Hehir, especially Russia. Most importantly, the BRICs have succeeded in almost entirely expunging Pillar III—the part of the R2P doctrine that calls on the international community to intervene and protect human rights where sovereign states cannot or will not. Without Pillar III, however, we are left with a rendering of R2P that

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

MORALITY , REALITY , AND INTERVENTION



makes no provisions for its actual enforcement. All we have is another highsounding rhetorical device. In Chapter 9 Robert Keating and Tom Murray start by showing that our unipolar international system is particularly conducive to the propagation and enforcement of universal human rights norms. To this they add that a strong rhetorical consensus has built up around R2P over the last decade. And yet, despite all of this, the implementation of R2P has been a failure. Murray and Keating seek to explain why that is so. The behaviour of the US and its NATO allies—in Kosovo and later in Libya—is no small part of the explanation offered. They ignored the UN Security Council over Kosovo. Other world powers, such as the BRICs, took this as an affront and a challenge. Normative defiance to the liberal world order was the reaction: Russia in particular became less willing to support humanitarian intervention than it had been throughout the 1990s. Similarly, in Libya, NATO refused to conform to the limitations on the intervention imposed by UNSC Resolution 1973. This weakened confidence in the Security Council’s ability to manage interventions once set in train, further undermining support for such operations generally. Thus the manner in which Western powers have sought to implement R2P—in particular their failure to adhere to the procedural norms of international society—has alienated the emerging powers on whose support successful R2P implementation depends. Sadly, Keating and Murray suggest that the prospects for R2P enforcement will only grow weaker as the international system becomes more multipolar. In the final chapter Sagar Sanyal implores us to look beyond the analytical just war tradition in our thinking about R2P. The standards and assumptions built into this approach are its trappings: they necessarily limit our appreciation of what is at stake, morally and politically. What is more, if we persist in this approach Sanyal fears that philosophers of war and military ethicists will become ‘decreasingly relevant to political reality and increasingly seen as ivory tower ideologues’. To avoid this fate, Sanyal argues, Marxist concepts and precepts must be taken more seriously in our appraisal of war and conflict. The point is not so much to produce new answers to old questions, but rather to raise questions that we otherwise would never have asked. If we start with a Marxist understanding of states and the wars they wage—as rooted in economic competition and driven by imperialist ambitions—rather than the liberal conceptions taken for granted by most just war thinkers, we get a radically different problem-set. And the problems in this set are more deserving of our intellectual energy, Sanyal thinks, because they more closely approximate the problems of the real world.

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

The editors would like to thank the Australian Research Council for its support of this project with a Discovery Grant, and the University of Melbourne also for its hosting of the project.

Notes 1. C. A. J. Coady, Messy Morality: The Challenge of Politics (Oxford: Oxford University Press, 2008). See especially chapters 1 and 2. See also the essays in C. A. J. Coady (ed.), What’s Wrong with Moralism? (Oxford: Blackwell, 2006), including Coady ‘The Moral Reality in Realism’. 2. General James Rusling, ‘Interview with President William McKinley,’ The Christian Advocate, 22 Jan. 1903, 17. Reprinted in Daniel Schirmer and Stephen Rosskamm Shalom (eds), The Philippines Reader (Boston: South End Press, 1987), 22–3. Curiously, there are several slightly different versions of this statement attributed to McKinley, and, to be fair, there is some controversy about the authenticity of the Christianize claim in (4). 3. The book in question is Henry Shue (ed.), Nuclear Deterrence and Moral Restraint; Critical Choices for American Strategy (Cambridge: Cambridge University Press: 1989). 4. The jus ad bellum specifies the justifying criteria for resort to war in contrast to the jus in bello that specifies the criteria for waging the war in a just fashion. There is much debate about the relation between these two categories, and about the best way to understand the just war tradition, but the theory has a lot of intellectual currency and political influence so it is worth taking seriously. 5. Henry Reynolds, Unnecessary Wars (Sydney: NewSouth Publishing, 2016). For the reference to Australia’s ‘love affair with the warrior’, see p. xiii. 6. For Kuperman’s accounts, with illustrations from recent conflicts, see: Alan Kuperman, ‘The Moral Hazard of Humanitarian Intervention: Lessons from the Balkans’; International Studies Quarterly, 52 (2008): 49–80; Alan Kuperman, ‘A Model Humanitarian Intervention? Reassessing NATO’s Libya Campaign’, International Security, 38/1 (2013): 105–36.

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi

1 Complicating the Moral Case of Responsibility to Protect Kosovo and Libya Stephen Zunes

Even if one is to assume that there indeed exists a moral and legal basis for foreign military intervention in the name of the Responsibility to Protect in theory, recent research has raised serious questions regarding such foreign military intervention on a utilitarian level. Indeed, there is an increasing amount of evidence which suggests that, on average, foreign military intervention in the name of R2P has actually done more harm than good. For example, Kathman and Wood, using a statistical analysis of such interventions to stop mass killings over a fifty-year period between 1955 and 2005, conclude that international military interventions in cases of severe repression actually exacerbate violence in the short term and can only reduce violence in the longer term if the intervention is impartial or neutral, which it rarely is. Similarly, Wood, Kathman, and Gent, noting how changes in the balance of power in a civil war influence the strategies of combatants as they relate to the level of violence, conclude that, with foreign intervention, the state engages in increasingly violent tactics towards civilians due to the increased difficulties in resource extraction and threats to their power.¹ Indeed, not only would the regime recognize that it has little to lose once foreign forces have intervened, the armed insurgency—now with foreign forces fighting on their behalf—would have a lessened incentive to negotiate or compromise so to end the bloodshed. These quantitative studies do not assume that foreign interventions in the case of large-scale killings always make things worse or that R2P should never be invoked under any circumstances. It does, however, raise certain questions regarding R2P’s efficacy. This is particularly important in regard to the tendency

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

for governments which have engaged in foreign combat operations in the name of ‘humanitarian intervention’ to exaggerate their successes and downplay the failures. This chapter looks at two of the major Western military interventions of recent decades taken in the name of the Responsibility to Protect—the bombing of Serbia in 1999 and the bombing of Libya in 2011. Both were widely seen as justifiable, yet both appear to vindicate these and other empirical surveys that raise questions regarding the utility, and thus the morality, of the interventions. In both cases, it appears that non-violent alternatives were not fully supported, violence dramatically escalated as a result of the intervention, and hardline elements among the resistance came to dominate, resulting in increasing instability and ongoing conflict.

Kosovo Kosovo had been an autonomous region of Serbia, the largest of the six republics in the Federal Republic of Yugoslavia. Ethnic Albanians consisted of 80 per cent of the population, ethnic Serbs around 10 per cent, with the remaining 10 per cent composed of other minorities. Starting in 1974, under Communist rule, the Yugoslav constitution effectively provided Kosovo de facto self-government. By 1989, however, with growing nationalism among the ethnic Albanian population in Kosovo and the rise of Serbian nationalist leader Slobodan Milosevic in the federal government, Kosovo’s autonomy was revoked and Serbian was imposed as the sole language of government and education. In the summer of 1990, local governance was dissolved and over 100,000 ethnic Albanian government employees were fired, prompting a general strike. Ethnic Albanian leaders declared independence from Serbia and from the Federal Republic of Yugoslavia. In subsequent years, as the Yugoslav Federation began its bloody disintegration, the Kosovar Albanian resistance remained largely non-violent. A parallel Albanian-language education system emerged along with a parallel government, which held democratic elections resulting in the election of pacifist academic Ibrahim Rugova as president. Younger Albanians, concerned that such parallel institutions while still under Serbian rule were not enough, began engaging in public protests in urban areas, often met by severe repression. Meanwhile, a shadowy armed group calling itself the Kosovo Liberation Army (KLA) emerged in the countryside and—flushed with arms suddenly made available from a temporary collapse of government authority in the neighbouring Republic of Albania—launched an armed struggle, resulting in even harsher countermeasures from federal troops and Serbian militia.

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi

COMPLICATING THE MORAL CASE OF R  P



By the end of 1998, with increasing attacks by the KLA and an increasingly brutal counter-insurgency campaign by Serbian forces just three years following the end of the carnage in Bosnia, NATO made a series of demands on the Serbian regime to stop the oppression, restore Kosovo’s autonomy, and withdraw their forces. Milosevic rejected the proposal, resulting in a seventy-eight-day NATO bombing campaign during the spring of 1999, leading to a Serbian withdrawal from Kosovo.

Opposition to Intervention The NATO war on Yugoslavia in the name of humanitarian intervention resulted in unusual and unprecedented political divisions in Western capitals. In the United States, it created what one Congresswoman referred to as an ‘aviary conundrum’, where traditional hawks became doves and doves became hawks. In Germany, Vice-Chancellor and Foreign Minister Joschka Fischer, unofficial leader of the Green Party and a former pacifist, split his party by supporting Germany’s participation in the war. With many liberals and moderate leftists supporting the war, much of the arguments against intervention came from an unlikely coalition of conservative and far left elements. Though there were some valid reasons to question the NATO intervention, which will be outlined below, many of the objections raised by opponents were highly questionable. For example, there was the fatalistic argument that the people of the Balkans had been ‘killing each other for years’ and that efforts to end the bloodshed were therefore hopeless. While there certainly have been conflicts between various ethnic and national identities in the Balkans going back for centuries, the peoples of the region have been at peace more often than not, including nearly a half century from the end of the Second World War until the breakup of Yugoslavia in the early 1990s. Ethnic conflicts in Yugoslavia, as elsewhere, are largely elite constructs; Balkan peoples do not have some genetic or even cultural predisposition to hate each other. They have been used by various overlapping empires to advance their aims in the region, as outside powers played one group against the other. Indeed, a major purpose behind the founding of Yugoslavia (‘land of south Slavs’) following the First World War was to prevent such manipulation, which makes its breakup such a tragedy. While some allege a Western imperialist plot in the rebalkanization of the Balkans, the roots of modern ethno-nationalism by the Serbs, Croats, and others seemed to come more from a desperate attempt by various Yugoslav elites to find a unifying ideology to replace the discredited Marxist–Leninist system. And it should be noted that Serbian leader Slobodan Milosevic was not the only one to

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

do so; Western-backed Croatian leader Franjo Tudjman also manipulated ethnic hatred in his rise to power. The use of the ethnic card also has tended to emerge, as such chauvinistic campaigns generally do, during times of great economic crisis. Yugoslavia, under heavy debt to Western banks, was under a draconian structural adjustment programme imposed by the International Monetary Fund. More than two million people found themselves unemployed, largely as a result of the forced de-nationalization of state industries, most of which were part of the socialist country’s innovative system of worker self-management. In this way, the West does share some of the blame for Yugoslavia’s tragedy. Indeed, the premature recognition by Western countries of moves for independence seemed to come from the mistaken belief that it was part of the process of democratization rather than a potentially reactionary trend which would threaten the rights of minorities within each of the republics. By contrast, it is doubtful the West would have been as willing to support Belgium, for example, splitting into separate countries for its constituent ethnic groups. Another highly problematic objection to the war was the assertion that the Serbs were not the aggressors in Kosovo. Some on the far left, in the name of antiimperialism, even made a strange alliance with right-wing ultra-nationalist Serbs repeating highly dubious claims that the Serbs of Kosovo were actually victims of Albanian terror and oppression and that the reports of massacres and ethnic cleansing by Serb forces were simply NATO propaganda. As a result of the history of US officials fabricating and exaggerating alleged atrocities by Sandinista Nicaraguans, Vietnamese communists, and other governments and movements the United States have opposed, there was a natural tendency by those sceptical of US foreign policy to question all such stories. Like the little boy who cried wolf, however, in this particular case—despite understandable scepticism— the alarm was all too real. Though there certainly were some exaggerations of Serb atrocities to manipulate public support for the war, most of the transgressions by Serb forces in Kosovo cited by Western governments were welldocumented by such reputable human rights monitoring groups as Amnesty International, Human Rights Watch, Doctors Without Borders, and others. And the repression predates the bombing and the earlier rise of the Kosovo Liberation Army; indeed, it had been going on for almost a decade, following the revocation of Kosovo’s autonomy, with the 10 per cent Serb minority essentially imposing an apartheid-style system on the country’s ethnic Albanian majority. Most ethnic Albanians in the public sector and Serbian-owned enterprises were fired from their jobs, forbidden to use their language in schools or government, severely limited in their right to non-violent dissent, and were subjected to widespread arrest, torture, and extra-judicial killings.

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi

COMPLICATING THE MORAL CASE OF R  P



Despite such dubious reasons for opposing the war, however, there were still reasons to question the legitimacy of NATO intervention. Indeed, the NATO war against Yugoslavia continued for more than ten weeks despite the many ways it could have been avoided or ended sooner, and despite the growing opposition and increasing uneasiness even among its supporters.

Alternatives to War There are reasons to believe that had Western powers taken a more proactive approach earlier, the war could have been prevented. For most of the decade prior to the NATO intervention, the Kosovar Albanians waged their struggle non-violently, using strikes, boycotts, peaceful demonstrations, and alternative institutions. The democratically elected parallel government provided schooling and social services and pressed their cause to the outside world. Taken together, it was one of the most widespread, comprehensive, and sustained non-violent campaigns since Gandhi’s struggle for Indian independence earlier that century. This was the time for Western powers to have engaged in preventive diplomacy. However, the world chose to ignore the Kosovars’ non-violent movement and resisted the consistent pleas by the moderate Kosovar Albanian leadership to take action. It was only after the emergence of the KLA in 1998 that the Clinton Administration and other Western governments finally took notice. By waiting for the emergence of guerrilla warfare before seeking a solution, the West gave Milosevic the opportunity to crack down with an even greater level of savagery than before. The delay allowed the Kosovar independence movement to be taken over by armed ultra-nationalists who quickly proved to be no more ready to compromise or respect the rights of the other ethnic group than their Serbian counterparts. Indeed, the KLA murdered Serb officials and ethnic Albanian moderates, destroyed Serbian villages, and attacked other minority communities. It is a tragedy that the West squandered a full eight years when preventive diplomacy could have worked. The US rejected calls to expand missions set up by the United Nations and the Organization for Security and Cooperation in Europe in neighbouring Macedonia into Kosovo or to bring Kosovo constituencies together for negotiations. Waiting for a full-scale armed insurrection to break out before acting also gave oppressed people around the world a very bad message: in order to get the West to pay attention to your plight, you need to take up arms. Though many Kosovars and others expected that the 1995 Dayton Accords would include provisions which would lead to an end to the Serbian occupation and oppression of Kosovo, the United States and other parties decided it did not merit

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

attention. Indeed, in contrast to the subsequent demonization of Milosevic, the United States boosted the Serbian leader’s standing throughout the Bosnia peace process; chief US negotiator Richard Holbrooke’s memoirs repeatedly praised the Serbian strongman. Had the peace process instead included other forces in Serbian society than the government—such as the Serbian Orthodox Church, the democratic opposition, and the Serbian minority in Kosovo—it is doubtful Milosevic would have found himself in such a powerful position. When Western powers finally took decisive action towards the longsimmering crisis in autumn 1998, a ceasefire was arranged where the Organization for Security and Cooperation (OSCE) in Europe sent in unarmed monitors. However, they were given little support. The OSCE monitors were largely untrained, they were too few in number, and NATO refused to supply them with helicopters, night vision binoculars, or other basic equipment which could have made them more effective. As Serb violations of the ceasefire—including a number of atrocities— increased, Western diplomatic efforts accelerated, producing the Rambouillet proposal which called for the restoration of Kosovo’s autonomous status within a greater Serbia. While such a political settlement was quite reasonable, and the Serbs appeared willing to seriously consider such an agreement, it was sabotaged by NATO’s insistence that they be allowed to send a large armed force into Kosovo along with rights to move freely without permission throughout the entire Federal Republic of Yugoslavia. Also problematic was that it was presented essentially as a final document without much room for negotiations. One of the fundamental principles of international conflict resolution is that all interested parties are part of the peace process. Some outside pressure may be necessary— particularly against the stronger party—to secure an agreement, but it cannot be presented as a fait accompli. This ‘sign this or we’ll bomb you’ attitude also doomed the diplomatic initiative to failure. Few national leaders would sign an agreement under such terms, which amount to a treaty of surrender: allowing foreign forces free reign of your territory and issuing such a proposal as an ultimatum. Smarter and earlier diplomacy could have probably prevented the war.

The Military Campaign The bombing campaign initially made things worse for the Kosovar Albanians. Not only were hundreds of ethnic Albanians accidently killed by NATO bombing raids, but the Serbs—unable to respond to NATO air attacks—turned their wrath against the most vulnerable segments of the population: the very Kosovar Albanians NATO claimed it would be defending. While the Serbs may have

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi

COMPLICATING THE MORAL CASE OF R  P



indeed been planning some sort of large-scale forced removal of the population in areas of KLA infiltration, similar to what the United States did in South Vietnam in the 1960s and early 1970s, both the scale and savagery of the Serbian repression which came were undoubtedly a direct consequence of NATO actions. By forcing the evacuation of the OSCE monitors, which—despite their limitations—were playing something of a deterrent role against the worst Serbian atrocities, NATO gave the Serbs an opportunity to unleash unprecedented savagery. By bombing Yugoslavia, NATO gave the Serbs nothing to lose. While there had been severe repression, including war crimes, committed by Yugoslav forces and Serbian paramilitaries in the course of the counter-insurgency operations for many months, the wholesale ethnic cleansing of hundreds of thousands of ethnic Albanians commenced only after the OSCE monitors were evacuated and the bombing appeared imminent. The war against Yugoslavia was also illegal. Any such use of force is illegal under international law unless in self-defence or unless authorized by the United Nations as an act of collective security. Kosovo was internationally recognized as part of Serbia; it was, legally speaking, an internal conflict. In addition, the democratically elected president of the self-proclaimed (if unrecognized) Kosovar Albanian Republic, Ibrahim Rugova, did not request such intervention; indeed, he opposed it. Many of those who have traditionally opposed US military intervention elsewhere recognized the strong humanitarian imperative in supporting the Kosovar Albanians and challenging Serbian ethno-fascism and thereby initially supported the war. Indeed, had a few days of bombing Yugoslav military positions really led to a Serbian withdrawal, one could make the moral case that such limited violence prevented far greater violence. This, however, as many of those knowledgeable about Yugoslavia predicted, was not the case. More than ten weeks of bombing, resulting in the widespread destruction of Yugoslavia’s civilian infrastructure, the killing of many hundreds of civilians, and severe environmental damage (through the bombing of chemical factories, the use of depleted uranium ammunition, and more) did not lead to substantially greater compromises than the Serbs originally entertained prior to the war. The final settlement ending the bombing was at least as close to the Serbian parliament’s counter-offer as it was to the original Rambouillet proposal, raising questions as to whether a similar compromise could have been reached without war. A number of human rights groups which challenged the morality and legality of Serbian actions in Kosovo also condemned NATO attacks which endangered the health and safety of millions of people by disrupting water supplies, sewage treatment, and hospitals. Indeed, the number of Yugoslav civilians who died from

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

NATO bombing was not significantly less than—and may have even surpassed— the number of Kosovar Albanian civilians killed by Serb forces prior to the onset of the bombing. Rather than curb the violence, the Serbian troop presence in Kosovo increased as the bombing continued and the repression of the Kosovar Albanians dramatically escalated. Those doing the killing in Kosovo were small paramilitary groups, death squads, and police units which could not be effectively challenged by high-altitude bombing and were not affected by bridges or factories hundreds of miles to the north being destroyed. As a result, as the bombing of Serbia and Montenegro extended into its third month, the repression in Kosovo continued unabated. Milosevic’s power had been based on the myth of Serbian victimhood and the belief that the whole world is against them, resulting in many Serbians developing pathological fear and resentment over non-Serbs. From the centuries-long conquest and occupation by Ottoman Turks, to the massacres of 300,000 Serbs by Croatian fascists in the 1940s, to the ethnic cleansing of more than 100,000 Serbs by US-backed Croat forces in the Krajina region of Croatia in 1995, Serbians had internalized within their culture a strong distrust and resentment of outside forces wishing to conquer and control them. Their songs and epic poems imply that righteous martyrdom is better than avoiding conflict. There is little evidence to suggest strategic bombing alone works in accomplishing political goals in most circumstances, but it would seem particularly difficult in the case of the Serbs. Having much of their society destroyed by a multinational force did not make them think more rationally. Such destruction was no more likely to bring quick compromise than Palestinian terrorism has encouraged compromise by the Israelis. Though NATO had hoped bombing would encourage defections in the military, few such defections took place, particularly since Western nations refused to grant temporary asylum for draft resisters and deserters. As a result, the war lasted much longer than many anticipated and, as a result, casualties—including those of civilians—escalated dramatically. There is a fair amount of evidence to suggest that NATO falsely assumed that simply the threat of bombing would lead to a last-minute capitulation by Milosevic but, having made the threat, felt obligated to follow through. Once bombing commenced, there was an equally unrealistic assumption that a few days of bombing would lead to a Yugoslav submission. Many used the analogy of the short bombing campaign against rebel Serb positions in Bosnia in August 1995 which preceded the signing of the Dayton Accords. However, there were several important differences: in Bosnia, there was a simultaneous Croatian ground offensive; ethnic Serbs in Bosnia were to be granted their own republic with substantial autonomy, unlike the ethnic Serbs in Kosovo; Bosnia-Herzegovina was an

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi

COMPLICATING THE MORAL CASE OF R  P



internationally recognized independent nation-state, whereas Kosovo was considered part of Serbia itself; and Kosovo was far more significant, culturally and religiously, to the Serbs than were the Serbian-populated parts of Bosnia. Despite this, NATO was apparently confident enough about their use of force that there was no Plan B when it quickly became apparent that the bombing was not working. Similarly, there was no preparation for the massive outflow of refugees resulting from the Serbian offensive in Kosovo against the civilian population. The only response NATO had was to bomb some more and to address the refugee crisis in an ad hoc fashion. In short, if the Serbs carry some cultural baggage around victimhood, some policy makers in the West have cultural baggage around this idea that the utilization of fancy, expensive, destructive firepower will automatically lead to the achievement of political goals. The attitude during the air campaign appeared to be that, if bombing doesn’t work, bomb some more and maybe it will work eventually. After the first couple of weeks of futile air strikes, pro-war guests on television talk shows began to talk less and less about saving Kosovar Albanians than about saving face. Like the periodic US/British bombing of Iraq going on at that time, Yugoslav policy became a case of foreign policy by catharsis, an assumption that, even if it does not really accomplish the stated political goals, at least it appears we are ‘doing something’. It is here where the much overused Vietnam analogy may have been appropriate: recently released documents reveal that, as far back as 1967, US leaders recognized that the US could not win the war, yet the war continued to be prosecuted. A ground invasion would have more likely stopped the repression sooner, but it would have come at a cost seen as unacceptable by the leaders of the NATO countries. Unlike Operation Desert Storm in Iraq eight years earlier, it would not be implemented on flat open terrain under cloudless skies. Kosovo and the surrounding areas are mountainous and wooded, and often overcast. Instead of confronting a conventional army hunkered down in one place, they would be facing a people with nearly 700 years of experience in guerrilla warfare, who tied down more German regiments per capita than anyone else during the Second World War. As retired Joint Chiefs chairman, Colin Powell, put it in an interview some years earlier, ‘We don’t do mountains.’ There are strong indications that the Yugoslavs had largely accepted the idea of a Serbian withdrawal from Kosovo, Kosovar autonomy within a greater Serbia, a return of refugees, a disarmed KLA, and an international monitoring force to guarantee the agreement. Where NATO could have been more flexible at the outset (indeed, what a number of NATO members had suggested but the Clinton Administration refused to consider) was the deployment of a force under the

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

control of the United Nations or the OSCE. There is little question that neither NATO nor the United States were the only forces capable of leading such a deployment and that compromise in this area could have enforced the political agreement while allowing the Yugoslavs to save face by not having it appear to be an occupation by a military alliance which had attacked their country. Yet the Clinton Administration seemed more intent on dominating the post-war order politically and militarily than reaching an agreement earlier which could prevent further bloodshed and allow the refugees to return sooner.

Ramifications The war had serious consequences besides death and destruction in Serbia and Kosovo. One of the original justifications was to prevent a broader war, yet the bombing campaign destabilized the region to a greater degree than Milosevic’s campaign of repression against the Kosovar Albanians. It emboldened ethnic Albanian chauvinists, not just in Kosovo where they had come to dominate, but the restive ethnic Albanian minority in the neighbouring Republic of Macedonia, which twice launched armed rebellions against the Slavic majority in the years immediately following the 1999 war. The bombing campaign also initially strengthened ultra-nationalist Serbian hardliners at the expense of the burgeoning pro-democracy movement. A large and active non-violent movement challenged Serbian leader Slobodan Milosevic and his alliance with right-wing ultra-nationalists on several occasions during the 1990s. This movement, led by young people whose lives were shattered by the Serbian regime’s endless wars, supported a more pluralistic and democratic Yugoslavia, and an end to human rights abuses against both Serbs and nonSerbs. In the winter of 1996–7, for example, a mass non-violent movement threatened the Milosevic regime, but it received no help or encouragement from Western governments. Indeed, Richard Holbrooke, the Clinton Administration’s point man for the Balkans and architect for the Dayton Accords, was among those who pressured Clinton to back Milosevic as a stabilizing influence in the region. (Ironically, Holbrooke became one of the most virulent supporters of the war two years later.) The Serbian government succeeded in crushing the pro-democracy movement. In 1999, a re-energized student-led pro-democracy movement, coalescing around a group called Otpor! (‘Resistance’), had emerged. Despite efforts by Milosevic to depict the opposition as Western agents, the vast majority of students involved were actually left-of-centre nationalists, motivated by opposition to their government’s increasing corruption and authoritarianism. Once the United States launched airstrikes against their country, however, they suspended

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi

COMPLICATING THE MORAL CASE OF R  P



their anti-government activities and joined their compatriots in opposing the NATO bombing. The US-led war gave the Milosevic regime an excuse to jail, drive underground, or force into exile many leading pro-democracy activists, shutting down their independent media and seriously curtailing their public activities. Journalists were not exempt: Milosevic’s secret police murdered Slavko Curuvija, publisher of the independent Dnevni Telegraf, soon after the bombing campaign began. Most of the population, meanwhile, rallied in support of the regime, as is the tendency when a country finds itself under attack from foreign forces. Ironically, NATO bombs targeted urban areas that were mostly anti-Milosevic. Air raids struck parts of northern Serbia in the autonomous region of Vojvodina, including areas where ethnic Serbs were a minority. NATO planes also struck the Republic of Montenegro, the junior partner in the Yugoslav federation, setting back its efforts at becoming closer to the West and more independent from Serbia. Fortunately, a year and a half later, pro-democracy forces led by Otpor! were able to regroup and—when Milosovic tried to steal the election in October 2000—a massive wave of non-violent action succeeded in driving him from power. The people of Serbia were able to do non-violently what over ten weeks of NATO bombs could not. As with the democratic revolutions that swept Eastern Europe in 1989, it wasn’t the military prowess of the Western alliance bringing freedom to an Eastern European country, but the power of non-violent action by the subjugated peoples themselves. Unfortunately, through both appeasement and war, the United States allowed Milosevic to remain in power far longer than he would have otherwise. As Milosevic’s nationalist successor Vojislav Kostunica put it, ‘The Americans assisted Milosevic not only when they supported him, but also when they attacked him. In a way, Milosevic is an American creation.’² Anger at the NATO bombing helped maintain Serbian ultra-nationalists as a powerful force in the coming years, at several points threatening to defeat more liberal elements. Anti-American riots broke out soon after Kosovo declared its independence and the Kosovo issue remains the major obstacle to Serbia joining the European Union. The most negative ramifications of the bombing campaign, however, remain in Kosovo itself. While the Democratic League of Kosovo (LDK), which grew out of the non-violent resistance movement of the 1990s, dominated Kosovo politics in the initial post-war period, the NATO intervention gave unprecedented legitimacy to the ultra-nationalist KLA. Two parties allied with that extremist group— the Democratic Party of Kosovo (PDK) led by former KLA leader Hashim Thaçi and the Alliance for the Future of Kosovo (AAK) led by former KLA commander Ramush Haradinaj—have subsequently dominated Kosovo politics. Haradinaj

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

was serving as prime minister in 2005 when he was indicted to face the UN war crimes tribunal in The Hague. Thaçi went on to serve as president and prime minister despite a 2010 Council of Europe report alleging that, under his leadership, KLA rebels engaged in organ trafficking and other crimes subsequent to the end of the 1999 war. Even more seriously, hardline nationalists have effectively forced two-thirds of the non-Albanian population from Kosovo since coming to power, with the ethnic Serb population having been reduced by 80 per cent. In the months following the NATO intervention, the International Crisis Group reported that ‘Systematic attacks upon the Serb population, and to a lesser degree upon other minority groups, suggest that at least some elements of the ethnic Albanian majority are determined to rid the province of all non-Albanians.’ The report went on to note how ‘Serb owned properties have suffered grenade attacks or been set alight, and individuals and groups of Serbs have been routinely kidnapped or murdered.’³ One of the few areas in the country where ethnic Serbs remain is in a strip of northern Kosovo where they govern themselves separate from Kosovo government. Armed clashes are not uncommon, with particularly serious outbreaks of violence taking place in 2004, 2009, and 2011. Kosovo remains the second poorest country in Europe and the legacy of the NATO intervention has been a divided country. Kosovo unilaterally declared independence in 2008, bringing an end to nine years of United Nations administration. Unlike its previous declaration of independence in 1990, the country was immediately recognized by the United States and scores of other Western countries, despite a ruling by the International Court of Justice declaring it illegal and despite the refusal of most of those countries to recognize the declarations of independence by Western Sahara or by Palestine, both of which have a much stronger legal basis. The ramifications of the ‘humanitarian intervention’ went well beyond the Balkans. At the NATO summit in April 1999, the member states approved a structure for ‘non-Article 5 crisis response’, essentially a euphemism for war. (Article 5 of the NATO charter provides for collective self-defence; non-Article 5 refers to an offensive military action like Yugoslavia.) According to the document, such an action could take place anywhere on the broad periphery of NATO’s realm, such as North Africa, Eastern Europe, the Middle East, and Central Asia, essentially paving the way for NATO’s ongoing war in Afghanistan. This expanded role for NATO wasn’t approved by any of the respective countries’ legislatures, raising serious questions about democratic civilian control over military alliances.

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi

COMPLICATING THE MORAL CASE OF R  P



The occupation by NATO troops of Serbia’s autonomous Kosovo region, and the subsequent recognition of Kosovar independence by the United States and a number of Western European powers helped provide Russia with an excuse to maintain its large military presence in Georgia’s autonomous South Ossetia and Abkhazia regions, and to recognize their unilateral declarations of independence. This, in turn, led to the 2006 war between Russia and Georgia. Indeed, much of the post-Soviet tension between the United States and Russia can be traced to the 1999 war on Yugoslavia. Russia was quite critical of Serbian actions in Kosovo and supported the non-military aspects of the Rambouillet proposals, yet was deeply disturbed by this first military action waged by NATO that followed. Indeed, the war resulted in unprecedented Russian anger towards the United States, less out of some vague sense of pan-Slavic or pan-Orthodox solidarity, but more because it was seen as an act of aggression against a sovereign nation. The Russians had assumed NATO would dissolve at the end of the Cold War. Instead, not only did NATO expand, it went to war over an internal dispute in a Slavic Eastern European country. This stoked the paranoid fear of many Russian nationalists that NATO would later find an excuse to intervene in Russia itself. While in reality this is extremely unlikely, the history of invasions from the West no doubt strengthened the hold of Vladimir Putin and other semiautocratic nationalists, setting back reform efforts, political liberalization, and disarmament. The war also had political repercussions in the United States. It provided a precedent of Democratic Party lawmakers supporting an illegal war and allowing for extraordinary executive power to wage war, of which the Bush Administration was able to fully take advantage in leading the country into its debacle in Iraq. Indeed, the illegality of the US-led NATO war on Yugoslavia helped undermine the United Nations Charter and thereby paved the way for the US invasion of Iraq, perhaps the most flagrant violation of the international legal order by a major power since the Second World War. It also served as an important precedent for NATO intervention in Libya.

Libya The modern nation-state of Libya is made up of what were three distinct regions—Cyrenaica, Tripolitania, and Fezzan—pulled together by the United Nations following the end of the Second World War and the defeat of fascist Italy, the colonial power. The pro-Western monarch installed by Western powers, King Idris, did little to develop the country despite its small population and enormous oil wealth. He was overthrown in 1969 by Muammar Gaddafi, a

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

young colonel, who nationalized the country’s oil, closed a major US air base, and embraced a radical nationalism that put him at odds with both Western powers and most of his North African neighbours. An initially non-violent pro-democracy struggle centred in Benghazi and other western cities began in February 2011, but spread to other cities, including the capital of Tripoli. Most urban areas in the eastern part of the country fell under the control of the largely non-violent revolutionaries and popular councils took over governance. Defections from the government and the armed forces increased in response to orders to attack unarmed protesters. Some defecting soldiers brought their weapons with them and, within a week, an armed uprising commenced. More adept at fighting armed militias than unarmed crowds, regime forces recaptured many of the liberated cities. Fearing widespread massacres if Gaddafi’s forces recaptured Benghazi, NATO—with the support of some Arab monarchies—began active support for the rebels, primarily with air strikes. The regime was overthrown by the end of August, followed by Gaddafi’s summary execution and the last pockets of pro-regime resistance being eliminated by October.

Opposition to Intervention As with the 1999 NATO intervention in Serbia, there were some reasonable and some less reasonable criticism of Western involvement in Libya’s civil war. Some of the critiques of the NATO intervention in Libya from the anti-imperialist left had little merit: for example, it was not simply a war for oil. Gaddafi had long ago opened his oil fields to the West, with Occidental, BP, and ENI among the biggest beneficiaries. Relations between Big Oil and the Libyan regime were doing just fine and the NATO-backed war was highly disruptive to their interests. Since 2003, following the extradition of Libyan intelligence operatives indicted for the bombing of a Pan Am airliner over Scotland in 1988 and paying reparations to families of this and another airline bombing in which the regime was implicated, as well as giving up non-traditional weapons development, relations with Western nations and their corporate interests had been relatively good. Similarly, Libya under Gaddafi was hardly a good progressive alternative to the right-wing Arab rulers favoured by the West as some in the Western left had sometimes portrayed it. Despite some socialist initiatives early in his reign which led Libya to impressive gains in health care, education, housing, and other needs, his final two decades witnessed increased corruption, regional and tribal favouritism, capricious investment policies, an increasingly predatory bureaucracy, and a degree of poverty and inadequate infrastructure inexcusable for a country of such vast potential wealth.

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi

COMPLICATING THE MORAL CASE OF R  P



Libya’s human rights record under Gaddafi had always been decidedly poor, but dramatically worsened in response to the popular uprising, including the killing of hundreds of non-violent protesters. Unlike the recently deposed dictator Zine el-Abidine Ben Ali in neighbouring Tunisia, whose forces had refused orders to massacre protesters, and Hosni Mubarak in neighbouring Egypt, whose forces eventually deposed him, Gaddafi’s forces were tribal loyalists and foreign mercenaries quite willing to kill unarmed demonstrators in large numbers. Another difference was that the Tunisian and Egyptian militaries were dependent on Western democracies (primarily France and the United States, respectively) for security assistance and other strategic cooperation. Public opinion in Europe and North America would have demanded cutting such ties in the event of largescale massacres, and many in the military leadership of those two countries were unwilling to lose that relationship. Similarly, influential sectors of the government, intelligentsia, and business class did not want to risk international isolation. By contrast, the Libyan elites had been largely isolated in the international community for most of Gaddafi’s reign, so they had little to lose. Similarly, Tunisian and Egyptian civil society was much stronger, better organized, and thereby more resilient in the face of oppression. These factors were what primarily contributed to the staying power of the Libyan regime, not a reflection of the popularity of the government or lack of support for the opposition. It also means that Gaddafi’s pledge to show ‘no mercy’ to those who rose up against him, particularly in Benghazi, where the opposition was strongest, should not have been dismissed as summarily as some war critics have done. Though comparisons to Rwanda and other genocides were grossly exaggerated, there were some genuine humanitarian concern of the fate of thousands of Libyans had government forces been able to take the city. As with the case of Yugoslavia twelve years earlier, the spurious nature of some arguments against intervention did not mean that it was a good decision. Indeed, without the NATO intervention, the Libyans might have been able to defeat Gaddafi on their own without many of the tragic consequences which have followed.

Alternatives to War NATO played a critical role during the six-month civil war in disrupting the heavy weapons capability of the repressive Libyan regime and blocking its fuel and ammunition supplies through massive air strikes and providing armaments and logistical support for the rebels. However, both the militaristic triumphalism of the pro-intervention hawks and the more cynical conspiracy-mongering of some on the left ignore that this was indeed a popular revolution which may have

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

been able to succeed without NATO, particularly if the opposition had not focused primarily on the military strategy. Engaging in an armed struggle against the heavily armed despot essentially took on Gaddafi where he was strongest rather than taking greater advantage of where he was weakest—his lack of popular support. There has been little attention paid to the fact that the reason the anti-Gaddafi rebels were able to unexpectedly march into Tripoli on 22 August with so little resistance appears to have been a result of a massive, and largely unarmed, civil insurrection which had erupted in neighbourhoods throughout the city. Indeed, much of the capital had already been liberated by the time the rebel columns entered and began mopping up the remaining pockets of pro-regime forces. As Juan Cole noted in an interview on Democracy Now, ‘the city had already overthrown the regime’ by the time the rebels arrived. The University of Michigan professor observed how, ‘Beginning Saturday night, working-class districts rose up, in the hundreds of thousands, and just threw off the regime.’⁴ Similarly, Khaled Darwish’s 24 August article in the New York Times describes how unarmed Tripolitanians rushed into the streets prior to the rebels entering the capital, blocked suspected snipers from apartment rooftops, and sang and chanted over loudspeakers to mobilize the population against Gaddafi’s regime.⁵ Though NATO helped direct the final pincer movement of the rebels as they approached the Libyan capital and continued to bomb government targets, Gaddafi’s ouster from power appeared to be more the result of a civil insurrection than a foreign military intervention. It should also be noted that the initial uprising against Gaddafi that February was overwhelmingly non-violent. In less than a week, this unarmed insurrection had resulted in pro-democracy forces taking over most of the cities in the eastern part of the country, a number of key cities in the west, and even some neighbourhoods in Tripoli. It was also during this period when most of the resignations of cabinet members and other important aides of Gaddafi, Libyan ambassadors in foreign capitals, and top military officers took place. Thousands of soldiers defected or refused to fire on crowds, despite threats of execution. It was only when the rebellion took a more violent turn, however, that the revolution’s progress was dramatically reversed and Gaddafi gave his infamous 22 February speech threatening massacres in rebel strongholds, which in turn led to the United States and its NATO allies to enter the war. Indeed, it was only a week or so before Gaddafi’s collapse that the armed rebels succeeded in recapturing most of the territory that had originally been liberated by their unarmed counterparts six months earlier.

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi

COMPLICATING THE MORAL CASE OF R  P



It can certainly be argued that, once the revolutionaries shifted to armed struggle, NATO air support proved critical in severely weakening Gaddafi’s ability to counter-attack and that Western arms and advisers were important in enabling rebel forces to make crucial gains in the northwestern part of the country prior to the final assault on Tripoli. At the same time, there is little question that foreign intervention in a country with a history of brutal foreign conquest, domination, and subversion was successfully manipulated by Gaddafi to rally far more support to his side in his final months than would have been the case had he been faced with a largely non-violent indigenous civil insurrection. It isn’t certain that the destruction of his military capabilities by the NATO strikes was more significant than the ways in which such Western intervention in the civil war enabled the besieged dictator to shore up what had been rapidly deteriorating support in Tripoli and other areas under government control. Military force can eventually wear down an autocratic militarized regime, but—as the ouster of oppressive regimes in Egypt, Tunisia, the Philippines, Poland, Chile, Serbia, and scores of other countries through mass non-violent action in recent years has indicated—there are ways of undermining a regime’s pillars of support to the extent that it collapses under its own weight. Ultimately, a despot’s power comes not from the armed forces under his command, but the willingness of a people to recognize his authority and obey his orders. This is not to say that the largely non-violent struggle launched in February would have achieved a quick and easy victory had it not turned to armed struggle with foreign support. The weakness of Libyan civil society, combined with the movement’s questionable tactical decision to engage primarily in demonstrations rather than diversifying their methods of civil resistance, made them particularly vulnerable to the brutality of Gaddafi’s foreign mercenaries and other forces. In addition, unlike the well-coordinated non-violent anti-Mubarak campaign in Egypt, the Libyan opposition’s campaign was largely spontaneous. However, insisting that the Libyan opposition ‘tried non-violence and it didn’t work’, because peaceful protesters were killed and it did not succeed in toppling the regime after a few days of public demonstrations makes little sense, particularly since the armed struggle took more than six months. And it does not mean there were no other alternatives but to launch a civil war.

The Military Campaign As with the Kosovo intervention twelve years earlier, there are serious reasons to question whether NATO intervention was necessary and whether it did more harm than good.

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

President Obama justified the NATO intervention on the grounds that it was necessary in ‘preventing genocide’, declaring: ‘If we waited one more day, Benghazi . . . could suffer a massacre that would have reverberated across the region and stained the conscience of the world.’⁶ This is far less than certain, however, given what happened following the reconquest by government forces of Misurata, Zawiya, and Ajdabiya—Libyan cities which collectively have a larger population in Benghazi. While there was serious collateral damage in the counter-insurgency operations which likely violated the law of war, there was no evidence of deliberate massacres of civilians by Gaddafi’s troops. Immediately following the initial NATO intervention, rebels were emboldened to escalate their attacks, recapturing coastal cities. With supply lines too thin, however, they soon retreated and the cities fell yet again to Gaddafi’s forces until yet another offensive thanks to escalating NATO support several weeks later. Each time these cities became battlefields, civilian lives were lost due to operations by both sides. Meanwhile, the escalation led Gaddafi to crack down on dissent even harder in areas still under his control. Gaddafi’s leadership style had always been repressive, impulsive, unpredictable, and seemingly arbitrary. Yet his nationalism, anti-imperialism, and professed socialism led many educated Libyans who formed the backbone of the government to stay loyal despite their misgivings, in large part in reaction to what was seen as the punitive and hypocritical sanctions imposed by Western nations and the constant threat of renewed US air and missile strikes against the country. It was only when the sanctions and threats of war subsided that there began to be a dramatic increase in resignations and self-imposed exile by prominent Libyans who had been members and supporters of the government. In short, the US-led efforts to isolate, punish, and threaten the regime likely contributed to Gaddafi’s longevity as dictator. Once relations were normalized and the isolation and threats subsided, Gaddafi was seen less as the strong leader defending his nation against Western imperialism and more as the mercurial and brutal tyrant that he was. As with the case of Serbia, however, NATO intervention initially appeared to strengthen Gaddafi. The crimes committed over the years by his regime, while frequently exaggerated and not always unique, were very real. Yet Libya’s most serious offenses, in the eyes of Western policy makers, were not in the areas of human rights, terrorism, nuclear ambitions, subversion, or conquest, but in daring to challenge Western hegemony in the Middle East. Serving as an impediment to American ambitions gave such regimes credibility and legitimacy they would not otherwise receive from large numbers of Middle Eastern peoples resentful of such foreign domination, thereby strengthening these regimes’ rule

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi

COMPLICATING THE MORAL CASE OF R  P



at home, as well as their influence throughout the Middle East and beyond. As a result, the intervention encouraged nationalist reaction that could play into Gaddafi’s hands. It is no accident that Gaddafi chose as the backdrop to his bizarre and frighteningly belligerent speech on 22 February a building in Tripoli destroyed in the 1986 US bombing. (And this history of US intervention in Libya predates the Gaddafi era—by 165 years, under President Thomas Jefferson, who sent Lt. Stephen Decatur to attack the base of the Barbary Pirates, later referenced in the opening line of the Marine Hymn: ‘to the shores of Tripoli’.) For a full four months following the NATO intervention, there was a bloody stalemate. The Health Ministry of the National Transitional Council which succeeded the Gaddafi regime initially estimated as many as 30,000 were killed, though some estimates are closer to half that figures. In any case, the death toll was far greater than those killed in the largely non-violent uprisings against Arab dictatorships during this period, including the first nine months of the Syrian uprising. In addition, there was the widespread destruction of key segments of the country’s infrastructure and the resulting economic and environmental damage. A report from a Human Rights Watch (2012) investigation of just eight of the hundreds of NATO air strikes—ostensibly designed to protect civilians— reported seventy-two civilian deaths, including twenty women and twenty-four children.⁷ There is reason, therefore, to believe that the ‘humanitarian intervention’ by NATO likely resulted in a far greater death toll than had there been no such foreign military involvement. Furthermore, while it is certainly possible that Gaddafi would have continued to refuse to step down in any case, the NATO intervention emboldened the rebels to refuse offers by the regime for a provisional ceasefire and direct negotiations, thereby eliminating even the possibility of ending the bloodshed months earlier.

Ramifications There were serious legal questions regarding the NATO intervention in Libya’s civil war. Unlike the 1999 war on Yugoslavia, there was authorization from the United Nations Security Council regarding the use of force to protect civilians. However, NATO and allied Arab forces went beyond the mandate provided by the United Nations Security Council to simply protect the civilian population through the establishment of a no-fly zone. Instead, NATO became an active participant in a civil war, providing arms, intelligence, advisers, and conducting over 7,500 air and missile strikes against military and government facilities. Armed units from the Saudi and Qatari dictatorships contributed to the fighting on the ground. This overreach led Russia and China, which did not oppose the

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

UN Security Council resolution authorizing the use of force to protect civilian lives, to become alarmed at how NATO essentially became the air force for the rebels. As a result, not wanting a limited UN resolution to be used as an excuse for a full-scale Western intervention, Russia and China have subsequently vetoed very moderate resolutions addressing the war in Syria that are critical of the Assad regime’s repression, thereby crippling the world body in addressing the ongoing crisis. Indeed, such abuse of the UN system will create even more scepticism regarding the implementation of the Responsibility to Protect should there really be an incipient genocide somewhere where foreign intervention may indeed be the only realistic option. Another problem with an armed overthrow of a dictator, as opposed to a largely non-violent overthrow of a dictator, is that you have many armed individuals who are now convinced that power comes from guns. The martial values and the strict military hierarchy inherent in armed struggle can become accepted as the norm, particularly if the military leaders of the rebellion become the political leaders of the nation, as is usually the case. Indeed, history has shown that countries in which dictatorships are overthrown by force of arms are far more likely to suffer from instability and/or slide into another dictatorship. By contrast, dictatorships overthrown in largely non-violent insurrections more often than not evolve into democracies within a few years. Unlike the peaceful and relatively orderly transition to democracy going on in neighbouring Tunisia, where largely non-violent actions toppled the hated Ben Ali dictatorship just prior to the Libyan uprising, Libya is still struggling with armed militias fighting each other for the spoils. Though the 2012 Libyan elections resulted in victories by relative moderates, the new government was unable to exert its power in the face of more than 200,000 militiamen under arms in a country with a population of barely six million. Scores of the militias have proclaimed themselves as ‘guardians of the revolution’ and, in many cities, were soon asserting more power than any government officials. Some of these armed groups have engaged in massacres and mass incarceration of alleged supporters of the old regime. In addition to Gaddafi himself, hundreds of suspected supporters of the old regime have been summarily executed. Black Libyans and other black Africans living in the country have been targeted in particular, with hundreds killed and thousands driven from their homes. A radical Islamist group attacked the US Consulate in Benghazi in September 2012, killing the US ambassador and three other Americans. Salafi extremists have attacked and destroyed Sufi schools, mosques, and holy sites, with Human Rights Watch reporting that the Libyan government had ‘failed both to protect

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi

COMPLICATING THE MORAL CASE OF R  P



sites sacred to Libyan Muslims who follow Sufi practices and to arrest those who have destroyed the tombs’.⁸ By mid-2014, the Islamist extremist Ansar al-Sharia militia had seized all or parts of several Libyan cities, declaring the establishment of an ‘emirate’. In August 2014, Tripoli’s international airport fell to Islamist forces following a ten-day battle. The deteriorating security situation led to air strikes by Libya and the United Arab Emirates. Indeed, since 2011, Libya has gone from an authoritarian but relatively stable country to that of a failed state. The largest and most well-equipped groups are made up of radical Islamists. In 2013, the Libyan Ministry of the Interior reported that the murder rate was five times what it had been prior to the uprising and robberies had increased by almost as much. Cabinet members, other government officials—including the prime minister—and family members have been kidnapped. By 2015, the country was divided into a series of fiefdoms consisting of the internationally recognized coalition government in the east, the Islamist Libya Dawn coalition which controls most of the northwest, various ethnic militia in the southwest, the Islamic State, Ansar al-Sharia, and other Islamist extremists which control all or part of a number of northern coastal cities. Hundreds of thousands of Libyans have been displaced by the fighting. The destabilization resulting from the NATO intervention went well beyond Libya. The African Union—while highly critical of Gaddafi’s repression— condemned the NATO intervention, fearing that the ensuing chaos would result in Libya’s vast storehouse of arms fuelling local and regional conflicts elsewhere in Africa and destabilizing the region. This is exactly what happened. Whereas the successful non-violent revolution against the neighbouring Tunisian dictatorship resulted in a positive contagion of unarmed pro-democracy civil insurrections, the successful armed revolution in Libya—made possible by Western intervention—resulted in a negative contagion of armed rebellions. Disparate armed groups—some allied with the regime and some allied with the rebels—ended up liberating major stores of armaments. Among those carrying off these vast caches of weapons were Tuareg tribesmen who passed them on to fellow Tuaregs in Mali who, now having the means to effectively challenge the Malian government militarily, dramatically escalated their long-dormant rebellion under the leadership of the National Movement for the Liberation of Azawad (MNLA). Advances by the rebels prompted a coup by a US-trained army officer against Mali’s democratically elected government. With supporters of the ousted democratic government protesting in the capital and the army divided by the coup, Tuareg rebels took advantage of the chaos in the south and quickly consolidated their hold on the northern part of the country, declaring an independent state. Then, with the Malian army routed and Tuareg forces stretched

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

thin, radical Islamist groups—also flush with new arms resulting from the Libya war—seized most of the towns and cities in the north, which in turn prompted the French military intervention in January 2013. This renewed Western military intervention resulted in a terrorist attack at a natural gas installation in Algeria, killing thirty-eight foreign hostages. How much additional blowback will come from the 2011 Libyan intervention remains to be seen.

Conclusion Given that—despite the negative consequences of foreign intervention in the cases described and despite the possibilities of non-military alternatives to challenge the repression in both situations—Kosovo and Libya are still cited as among the most ‘successful’ applications of R2P, it raises serious questions as to whether the Responsibility to Protect is indeed a valid doctrine. On the other hand, supporters of R2P could acknowledge the validity of the arguments in this chapter regarding these two cases and still defend the doctrine on the grounds that these two interventions did not in fact meet the criteria under the R2P doctrine. The R2P doctrine includes a number of ‘operational principles’, including that there must be ‘rules of engagement which fit the operational concept; are precise; reflect the principle of proportionality; and involve total adherence to international humanitarian law’ and that there must be an ‘acceptance that force projection cannot become the principal objective’. This chapter has argued that non-military options were not reasonably explored in these cases and the military operations went well beyond the original concept, were not precise or proportionate, gave active assistance to allied forces, and violated international humanitarian law. As a result, it could therefore be argued that these two cases were actually misuses of the R2P doctrine since they neither met the ‘last resort’ criteria nor these key operational principles and should therefore not be used to invalidate R2P. However, economic interdependency and growing recognition of the power of large-scale strategic non-violent action has made it increasingly difficult to argue that all non-military means have been exhausted. Similarly, the lethality of modern warfare, the inherent inaccuracy of air power, and the tendency for ‘mission creep’ and other unforeseen escalations raise questions regarding proportionality and other criteria regarding the operational principles. In short, while the Responsibility to Protect can be morally defensible in theory, there are serious questions as to whether it can actually be applied.

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi

COMPLICATING THE MORAL CASE OF R  P



In any case, in the face of ongoing serious war crimes, as were occurring in Kosovo in 1999 and in Libya in 2011, Western nations should not be forced into the false choice of going to war and doing nothing. These tragic conflicts and their problematic aftermaths should be reminders that, moral and legal arguments aside, military force is a very blunt and not very effective instrument to promote human rights, and that bloated military budgets and archaic military alliances are not the best way to bring peace and security. As long as such ‘conflict resolution’ efforts are placed exclusively in the hands of governments, there will be a propensity towards war. Only when global civil society seizes the initiative and recognizes the power of strategic non-violent action, and the necessity of preventative diplomacy, can there be hope that such conflicts can be resolved peacefully.

Notes 1. Jacob D. Kathman and Reed M. Wood, ‘Managing Threat, Cost, and Incentive to Kill: The Short- and Long-Term Effects of Intervention in Mass Killings’, Journal of Conflict Resolution, 55 (Oct. 2011): 735–60. Reed M. Wood, Jacob D. Kathman, and Stephen E. Gent, ‘Armed Intervention and Civilian Victimization in Intrastate Conflicts’, Journal of Peace Research, 49/5 (Sept. 2012): 647–60. 2. Nenad Lj. Stefanovic, ‘Empty Pocket and State: Interview: Dr Vojislav Kostunica, Presidential Candidate’, Vreme, Belgrade, 12 Aug. 2000. 3. International Crisis Group, Violence in Kosovo: Who’s Killing Whom, CG Balkans Report, 78, Prishtinë/Pristina, London, and Washington, 2 Nov. 1999. 4. Juan Cole, ‘Libya has Reignited the Flame of Liberty in the Arab World’, Democracy Now!, 22 Aug. 2011. 5. Khaled Darwish, ‘When Libya Grew Wings’, New York Times, 24 Aug. 2011. 6. Barack Obama, ‘Remarks by the President in Address to the Nation on Libya’, White House Press Office, Washington, DC, 28 Mar. 2011. 7. Human Rights Watch, ‘Unacknowledged Deaths: Civilian Casualties in NATO’s Air Campaign in Libya’, 14 May 2012. 8. Human Rights Watch, ‘Libya: Stop Attacks on Sufi Sites: Protect Religious Sites, Investigate Crimes’, 31 Aug. 2012.

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi

2 Why Sovereignty Matters Despite Injustice The Ethics of Intervention Richard W. Miller

This chapter derives from a lecture at Peking University. Discussions of common ground between just war theory and Confucian masters, recent diplomatic stances of China as well as the United States, and dangers of conflict between the United States and China have been retained in light of the importance of these two powers in shaping prospects for peace and conflict and the importance of finding common standards of moral judgement in containing dangers of conflict between them.

Nearly all of the world’s land and a fair amount of ocean are contained by borders within which a single organization claims a right to issue directives and back them up by force and denies that outsiders have this authority. Within these territories, there are sometimes sub-territories, such as American states, in which the central authority recognizes an independent right of command of a local organization, but the scope and nature of the right are meant to eliminate serious conflicts of commands. In one meaning of the slippery term, the central territorial organizations of command are sovereign governments. I will generally call them ‘governments’ for short. Many of them are seriously unjust in their treatment of people in their territories. This gives rise to the question, ‘What is it right for a government to do in response to another government’s injustice toward those it governs, in order to reduce or remove the injustice?’ From case to case, governments and individuals answer this question differently. For example, the United States claimed that it was right to intervene militarily against the Serbian government in Kosovo, to end brutal injustice toward the local ethnically Albanian majority, while China and Russia disagreed. In the Syrian civil

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi

WHY SOVEREIGNTY MATTERS DESPITE INJUSTICE



war, when the United States has proposed resolutions in the Security Council that might provide a basis for military intervention to end brutal injustice of the Assad regime, China and Russia have disagreed. Within the United States, even limited bombing to create disincentives to further use of poison gas by the Assad regime was not supported by most citizens, but a large minority supported this response. It would be naive to deny that the conflicting positions of different governments reflect national strategies in foreign policy. When US allies have inflicted injustice more severe than Milosevic’s in Kosovo, the US response has often been to arm the perpetrators, not to intervene. The US Ambassador to the UN at the time of Indonesia’s subjugation of East Timor had this description of his instructions to respond to the initial reports of widespread massacres, ‘The Department of State desired that the United Nations prove utterly ineffective in whatever measures it undertook. This task was given to me, and I carried it forward with no inconsiderable success.’¹ US-supplied counterinsurgency aircraft used in forcing Timorese out of their villages were crucial to the carnage, in which about 200,000, mostly civilians, died, carnage inflicted by armed forces deriving 90 per cent of their weapons from the United States, as the flow of US arms more than tripled in the three years after the invasion.² In 1994, at the height of a conflict in Turkey over Kurdish autonomy, characterized by widespread government brutality and suppression of minimal expressions of Kurdish identity, the Turkish minister for human rights (who was soon replaced) described the government’s campaign of burning villages as ‘state terrorism’,³ condemnation preceded by a UN committee’s assertion of widespread and systematic torture.⁴ In 1994, the US provided Turkey with $600 million in US military aid,⁵ including 80 per cent of the armaments used by Turkish forces in their offensives against Kurdish villages.⁶ Nonetheless, moral judgements have an impact on foreign policy. Widespread moral disapproval at home or abroad of a government’s foreign policy reduces its ability to get its way, and moral support increases it. By affecting these public attitudes, moral advocacy changes the calculus of power, providing strategic reasons to limit harm in order to avoid weakening allegiance. So it is neither naive nor pointless to seek guidelines for the moral judgement of responses to foreign injustice. With this guidance, people can better engage in public discussions with real impact on the many lives affected by injustice and foreign responses to it. War is the most severe violation of the exclusive authority that sovereign governments claim, and disagreements over humanitarian war-making are, quite properly, the fiercest disagreements over how to respond to foreign injustice. The disagreements in which the government of the United States stands on one side, the government of China on another, are sometimes said to reveal distinctive moral perspectives characteristic of two different cultures. I will argue that this is wrong.

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

A sensible and humane application of the so-called ‘theory of just war’, the enduring framework for moral judgement of wars in the West, with ancient roots in St Augustine’s writings, supports the current characteristic reluctance of China to endorse humanitarian military intervention and justifies the lines that China draws between appropriate and inappropriate grounds for such intervention. On the other hand, someone who is even more respectful of sovereignty than the just-war-theory approach requires will face severe and insightful criticism from the legacy of the Confucian masters, not just from the legacy of Catholic saints. I will argue for great reluctance to launch humanitarian wars. In a sense, this will be an argument for respect for sovereignty in conditions of injustice: the conclusion will be that launching force across sovereign borders, violating the sovereignty of a government in response to its serious, systematic injustice, is almost always morally wrong. But the premises of the argument will not treat either sovereign government under conditions of serious systematic justice or the autonomy of political struggles under such a government as valuable as such. The borders of a seriously, systematically unjust sovereign government should almost always be respected, but the harms caused by intrusion, not the value of what is intruded on, are the reasons why. In the assessment of humanitarian war, this basis for respecting unjust sovereignty will help to explain the exceptions to the rule that war should not be launched to reduce injustice within a foreign country, as well as the normal validity of the rule. Turning from war to the many non-violent ways in which governments can respond to foreign injustice, I will show how the same perspective supports only a very limited range of responses, even when mere condemnation is concerned. Finally, I will briefly turn away from governments and look at criticisms of foreign injustice by individuals and non-governmental organizations. I will argue that transnational moral criticism from these sources is, in contrast, vitally important and that its importance is sometimes strengthened by the concerns that give moral criticism very little place in interactions among governments.

Just War and Common Sense What do moral philosophers really support when they say that they are supporters of humanitarian military intervention? Unless they are crazy, they have not supported invasion of the US by the UK to correct injustices of the US health care system, and would not have even if the relations in military power were reversed. They would not have favoured a grave risk of nuclear war by invading the Soviet Union to end Stalin’s injustices. Instead, they are apt to use the language of just war theory to express their qualified position: ending grave

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi

WHY SOVEREIGNTY MATTERS DESPITE INJUSTICE



and systematic injustice is a just cause for war, an aim that a morally permissible war could have, but war in this cause is only morally permissible if other criteria are met as well; these criteria include likelihood of success in advancing the just cause, avoidance of harm that is morally disproportionate to achievement of the just cause, and choice of war only as a last resort. It might seem that they have found a secure basis for a distinctive position on humanitarian military intervention, more interventionist than other current views. But they have not, if just war theory is properly understood. The word ‘theory’ might suggest that there is a reasonably precise, independent, systematic way of establishing whether each of the standard necessary conditions is met. No thoughtful just war theorist believes this. People in China rightly made defensive war against the Japanese invasion without establishing that it was probable that they would win. ‘Likelihood of success’ means sufficient likelihood given the value of the war aim and the expectations of harm and further benefit to people worthy of concern. In the two millennia of ‘just war theory’, no remotely adequate rule has been developed for moving from this value and these expectations to that conclusion of sufficiency without independent moral judgement of the case at hand. Similarly, when thoughtful theorists say that war must be a last resort, they do not mean that no possibility of peaceful resolution must exist. If the genocidal regime in Rwanda said that they were interested in negotiations as they continued to kill hundreds of thousands of Tutsis and moderate Hutus, this would not have made it wrong to invade to stop the mass killing. ‘Last resort’, like ‘likelihood of success’, involves a moral assessment of likelihood of harm and benefit: peace must be pursued unless the risks of not going to war are more morally serious than the risks of launching it. With the same recognition of interdependence among moral dimensions and the pervasive need for moral judgement, proportionality should be viewed as a super-criterion, with ‘likelihood of success’ and ‘last resort’ as relevant aspects: the expected harms of launching a war, guided by the cause in question, must not be so great that they are reasons not to launch one, when likely benefits of going to war and harms and benefits of giving peace a chance are taken into account. As sensible just war theorists always admit, no general rule can take the place of reflective moral judgement in applying this standard. What general rule could the people of China have used in deciding whether war against the Japanese invasion was worth its expected horrible costs? Certainly not, ‘any chance of ending unjust territorial takeover is more weighty than any expected costs’. In the 1962 border dispute between China and India, each side thought the other was engaged in an unjust territorial takeover, and both sides would have been wrong to inflict the terrible costs of launching a war meant to completely vindicate its claims whatever the costs.

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

Even though this additional step is not essential to the arguments that follow, the claim of interdependence among criteria for morally permissible war-making could be pressed further, making the assessment of an aim as a just cause dependent on the other criteria. A war aim, it might plausibly be said, is a just cause in virtue of the fact that in the current enduring background circumstances of political order and military force there is sometimes adequate reason to take its successful pursuit not to be disproportionate, in the super-criterial sense. In premodern circumstances, among groups relying only on small arms, it may sometimes have been proportionate for one group to make war on another to gain access to farmland, grazing territory, or water-holes that many members needed to overcome material deprivation. In modern circumstances, war for such economic goals is always disproportionate, so the cause is no longer just. Killing to expand the glory of a ruling family or the power of a nation is always disproportionate, and so these aims have never been just causes. There is an alternative view of just cause, with parallels within the Confucian tradition, that ties it more directly to a requirement that those attacked be engaged in a specific wrongful form of conduct making attack permissible, so that attack is, in effect, a form of punishment. But this cannot be a requirement in a complete account of when wars are morally permissible. Russia’s attack on Finland, to make Leningrad defensible, may have been morally justifiable as needed to prevent Nazi victory, but not as a response to wrongful conduct by Finns. The whole strict and literal truth about the moral permissibility of launching a war can be stated in this Great Test: it is wrong to launch a war with a given aim if the moral reasons to do so are not as strong as the moral reasons to continue on the most promising path of peace. The word ‘theory’ in ‘just war theory’ is an impressive title for what is really a useful checklist in deciding whether the Great Test is met. What is the war aim, the goal whose achievement would constitute victory? How likely is it to be reached, and how likely are further benefits from reaching it? How likely are harms to those morally worthy of concern if this goal is rationally pursued? What costs are risked in giving peace a chance? These are always relevant questions, even though we have no general rule that moves from answers to them to the verdict that the Great Test is passed or failed, without the need for further moral judgement. If we cannot develop a general rule, is there any way we can develop guidance in advance for judging particular proposals to go to war? If we cannot, this is a depressing fact about the moral judgement of war. It would mean that people cannot prepare for morally responsible participation in one of the most important discussions that citizens can have. Fortunately, there is an alternative to the search for general rules, namely, examination of constellations of factors that

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi

WHY SOVEREIGNTY MATTERS DESPITE INJUSTICE



occur again and again in modern circumstances, to see whether they always, nearly always, or typically lead to war’s passing or failing the Great Test, and why. A survey of the major constellations will provide a basis for strong opposition to war to end foreign injustice, opposition as strong as actually found among people who say they highly value national sovereignty. But the opposition does not depend on any distinctive view of the inherent value of unjust sovereign government.

The Spectrum of Constellations: The Restrictive End We can start with a clear negative finding, which considerably narrows the search for constellations that pass the Great Test. Because of the terrible dangers of organized, large-scale transnational violence using modern weapons, war to end injustice will not be worth its dangers unless its aim is to end or drastically reduce grave systematic injustice that will otherwise endure. Putting to one side these cases of insufficient injustice for humanitarian war, it is helpful to consider a constellation that is an appallingly common form of deep injustice yet (I shall argue) should not be attacked from outside except in rare and special circumstances. This is stable tyranny, the sort of appalling regime exemplified by Saddam Hussein’s in 2003: reasonably stable, owing to repression that includes death and torture to deter political opposition, with enough arbitrariness to create a pervasive fear of the regime’s terrible power but without ongoing widespread indiscriminate killing. Ending the tyranny and replacing it with a much less unjust government would, as such, be a great improvement morally speaking. Certainly, harms to the tyrant and his henchmen as they fight to preserve their unjust repressive regime are not important reasons not to improve the world in this way. In recurrent modern circumstances, are there reasons not to do so, nonetheless, that produce failure to pass the Great Test when the question is posed of whether to launch a war to end the tyranny? There almost always are. The destructive violence that wrecked or ended hundreds of thousands of lives as a result of the overthrow of Saddam Hussein illustrates one very serious reason: dangers to innocents when an otherwise stable tyranny is violently deposed from outside. Stable tyrannies have a substantial base of support in elites and social groups in their sovereign country. They skilfully exploit religious, ethnic, or other divisions to weaken opposition, and control the loyalties of most of those who enforce their authority, authority embodied in ordinary law and order as well as unjust repression. So their violent overthrow from abroad is apt to trigger violent disorder. Violent crime flourishes. Militias battle. Some of the fighting is invigorated by appeals to patriotic duties to repel foreign invaders. Violent suppression of these insurgent forces by the invaders is apt to create a

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

deadly spiral in which repression recruits insurgents through increased hatred of the outsiders or fury at civilian casualties, and the increase stimulates more repression, killing some but recruiting more. Those recruited into insurgent violence will often be people who would have done nothing to make them appropriate targets for deadly force if there were no invasion. Those who invade without taking account of this danger of making currently innocent people legitimate targets show inadequate regard for the lives of those they will kill as combatants, as well as for the lives of those who will remain innocent civilians. In modern circumstances, could these various dangers be worth the expectation of overthrowing the tyranny and replacing it with a much less unjust regime? The crucial question for answering this question—in many ways, the crucial question for the whole issue of humanitarian war—is, ‘Whose judgement is morally decisive?’ If the goal is liberation of people in a foreign country from vicious oppression, then they should be the judges of whether this endeavour is worth the risks it inflicts on them. After all, the war is launched for the sake of their free pursuit of legitimate goals, but violent disorder also gravely interferes with this freedom. War to cure tyranny is a potentially deadly operation, and it is wrong to perform a hazardous operation on a patient without sufficient assurance of her informed consent. In general, those who invade to reduce injustice must have warranted confidence that the majority of those whose lives will be endangered and who have moral standing to object to the invasion want to take on the deadly risks. Of course, tyrants do not invite in pollsters to resolve this question. If a large and growing proportion of the population are in rebellion, this is a very positive indication. But then, we are not dealing with a stable tyranny. Given the grave risks to innocent insiders of invasion, warranted confidence of adequate informed consent to invasion to overthrow a stable tyranny is normally unavailable. The infliction of this danger to help the oppressed foreigners is, then, unjustified. Of course, insiders contemplating rebellion against a stable tyranny must also take account of costs to unwilling victims. But adequate consideration of these dangers can permit rebellion when humanitarian invasion would be wrong. The rebels willingly exercise a prerogative of self-defence against oppression. If their movement stays small, the violence it provokes is apt to be limited and highly concentrated on them, unlike the widespread violent disorder unleashed by a humanitarian invasion of a stable tyranny. If the rebellion does not stay small but grows into a widespread revolution, this will be evidence of widespread willingness to risk death to overthrow the tyranny. But then we are no longer dealing with a stable tyranny. For these reasons, insiders’ decisions to launch a violent

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi

WHY SOVEREIGNTY MATTERS DESPITE INJUSTICE



attack on a stable unjust tyranny more readily meet the test of adequate confidence of informed consent than outsiders’ decisions. This distinction between insiders and outsiders has nothing to do with any inherent value of gravely unjust sovereignty. So far, only dangers to people oppressed by the gravely unjust tyranny have been considered. But there are severe dangers to others. Of course, one should consider the death and maiming of the soldiers who will be ordered to invade. Further dangers loom, if one adopts an appropriately realistic view of the international relations affected by a decision to launch a war to overthrow a foreign tyrant. If a great power or a regional power invades a stable unjust tyranny, this threatens international order. Stable unjust tyrannies typically cultivate client relationships with outside powers. The outside power will be alarmed by the threat to its own power, strengthen its military preparations and reduce its trust in other powers, heightening long-run or intermediate-run risks of an outbreak of proxy wars, if not direct military conflict. (Think of what would have happened in China if India, much less the United States, had responded to injustice in Myanmar with a humanitarian invasion.) Invasion by a weak and unthreatening government may have much less chance of disturbing international power relations, but it typically has much less chance of success. A humanitarian invasion encourages further humanitarian wars, appealing to judgements of proportionality that are bound to be imperfectly informed and bound to be shaped by national interests that blunt awareness of harmful consequences for a tyrant’s current victims. Agreement among all great powers, sustaining Security Council approval, could reduce these dangers to international order, but this agreement is very unlikely. The great powers’ foreign policy interests are apt to differ in the case at hand. In any case, they will not want to prepare the way for invasion to cure similar injustice where those interests are at stake. Moreover, great powers are separated by different views of what constitutes grave injustice. In explaining why Saddam was so unjust, George W. Bush included offences against free enterprise along with violations of freedom of religion. It is much better for international order if debates based on these moral differences are not a central feature of intergovernmental deliberations over war and peace. (I will say more about this need for caution later, in connection with human rights.) These dangers within the unjustly governed country and in the world at large are moral reasons for reluctance to support war to remove a stable tyranny that are at least as strong as the reluctance to lie of morally conscientious people. There are exceptions, in both cases. But an understanding of the reasons for general reluctance explains the occasional appropriateness of departure from it.

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

Sometimes, for example, the tyranny is in the generally respected sphere of influence of a great power which has shaped governance in the afflicted territory; as part of its long experience of overlordship, the great power has strong reasons to believe that the tyranny has little popular support and can be overthrown without grave costs if the overlord invades to back a more just domestic coalition. The United States overthrew the Cédras regime in Haiti and France has sometimes overthrown unjust regimes in its former African colonies, without waiting for pervasive instability and consequent widespread violence to envelop the country. In such cases, there is often sufficient assurance that harms are proportionate to gains, as assessed by most of those worthy of concern, and the special subordinate relationship prevents disturbance of international order. But no one wants the exception to be embodied in an explicit norm, applied in public deliberations among governments, about what great powers may do in their spheres of influence. Diplomatic discourse is more effective and supportive of peace if it observes the courtesy of not elevating differences in power among sovereign governments into explicit norms used in justifications by representatives of governments addressed to governments in the world at large.

The Abysmal Constellation At the opposite pole from invasion to reduce the injustice of an ordinary stable tyranny is invasion to stop ongoing large-scale killing and brutal displacement of people because of who they are, not what they do. Close to this opposite pole is ongoing widespread killing and maiming of non-combatants in the course of military operations that show contempt for their lives. When people are directly, gravely threatened by these horrors, it is not guesswork that they would give their informed consent to efforts to rescue them despite the risks. Rather than constituting tainted law and order, such conduct is vile violent disorder. A great power whose trust should be invited in the interest of world order would be willing to discourage such large-scale massacre. No controversial moral judgement is required for severe condemnation. So intervention to rescue from such horrors does not typically endanger world order to anything like the same extent as intervention to end ordinary gravely unjust tyranny. These features make it much easier to justify military intervention to stop ‘the four international crimes of “genocide, war crimes, ethnic cleansing and crimes against humanity”’ perpetuated on a very wide scale. The list of four is taken from a 2009 statement of Liu Zhenmin, China’s UN Ambassador.⁷ As always, there are exceptions. Intervention to stop Saddam Hussein’s ethnic cleansing of Kurdish border regions at the end of the Iraq–Iran war would have revived the

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi

WHY SOVEREIGNTY MATTERS DESPITE INJUSTICE



fires of a decade-long war that had consumed too many lives already. Intervention against Croatia’s ethnic cleansing of Serbian regions had too great a risk of prolonging deadly conflict in the former Yugoslavia and intensifying the viciousness in Croatia; better to pursue a negotiated Balkan settlement. More urgently today, even if the Assad regime is guilty of very widespread war crimes, intervention might strengthen the violence of vicious forces on the other side and postpone the achievement of stable division into fairly peaceful but mutually distrustful cantons that is Syrians’ best hope for basic safety. Still, when a regime perpetuates widespread atrocity, the disposition to intervene should be very different from the case of ordinary stable tyranny. One ought to be disposed to support rescue from such injustice, in the absence of special contrary reasons. However, those contrary reasons are not rare and exotic. So the phrase, ‘Responsibility to Protect’, with its implication that non-intervention is negligent, can do much harm.

Unstable Tyranny and Popular Secession In between the poles are two constellations of factors posing special difficulties. One is an unstable, severely challenged gravely unjust tyranny. If the opposition is rebelling against injustice and aspires to a more just order, justice will be served by their victory. But their progress in the face of severe repression typically makes it better for outside governments to stand aside: the rebels, acting on their own, creating new ties of allegiance and political capacities, are more likely to establish a reasonably stable juster political order than if their victory depends on outside intervention; their unaided victory is less likely to inspire fear and distrust among non-intervening powers. Sometimes, however, the prospect, in the absence of intervention, is the interminable violence of life under a failed state. If a defender of the inherent value of sovereignty were to say, ‘There is no effective sovereignty here, because the asserted monopoly of permission to use force is not effective’, she would not be much of a defender. Most proposals to intervene to reduce injustice are aimed at governments challenged by rebel groups who make their authority ineffective in substantial regions of their territory. Better to say that stopping the interminable violence of a failed state is all right, if the success in intervening is to be followed by measures to construct a state fitting local circumstances and aspirations and if the intervention is organized in ways that do not threaten long-term international order by offence to threatened powers. The other difficult topic is secession. Sometimes, mounting popular support for a secessionist movement produces strong evidence that people in a sub-territory

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

would give their consent to outside help in resisting the militarily dominant central government, despite the consequent dangers. After all, this is the sort of cause for which people willingly risk their lives, and here, military dominance can be a bad measure of what those most affected want. For example, it was reasonable to suppose that the majority of Kosovars welcomed NATO intervention, despite its risks. The relatively frequent warrant for belief in informed consent makes it tempting to locate popular secessionist movements at the permissive end of the spectrum of constellations, putting widespread pleas for support for independence close to pleas for rescue from massacre. This temptation should be resisted. The prospects of eventual accommodation of legitimate aspirations through forms of autonomy short of secession are very hard to judge by outsiders, and can be severely underestimated in the heat of struggle by insiders; in this respect, their consent to violence may be inadequately informed. Moreover, successful intervention in support of secession, even if it takes the indirect form of military supplies to rebels, can pose important dangers elsewhere. Loss of secure control of a territory can threaten the central government’s vital strategic interests. This threat can lead to militarization, distrust, and heightened international conflict. Even if the central government has little international power and no powerful allies (a very infrequent circumstance), support for secession poses the risk of further disorder elsewhere. For it encourages hope of outside support among the great many militants throughout the world who, rightly or wrongly, regard their ethnic group or nationality as unjustly bound to their sovereign regime. Dogged violent pursuit of independence, in the hope of outside support, is encouraged, as opposed to willingness to compromise on autonomy that achieves central aims. If the four crimes listed by Ambassador Liu are not being perpetrated on a large scale, it is wrong for an outsider to intervene militarily in support of a popular movement for secession and internationalize a domestic conflict. When a foreign government intervenes in support of secession, the question can arise for another foreign government of whether to oppose this initiative, through condemnation or support for the central government. Dangers of stoking conflict within the old borders or internationally, which count against intervention in support of secession, can also provide reasons to avoid or mute such international governmental opposition. For example, they justify China’s neutral response to Russia’s takeover of Crimea and support for secessionist movements in the eastern Ukraine. On the one hand, endorsement of America’s protests would have supported meddling that endangers global stability. In the leaked recording of a telephone conversation between Assistant Secretary of State Victoria Nuland and US Ambassador to the Ukraine Geoffrey Pyatt, the

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi

WHY SOVEREIGNTY MATTERS DESPITE INJUSTICE



American diplomats discuss the next moves for Ukrainian opposition leaders in the overthrow of the Yanukovych regime as overbearing bosses might plan the work of employees. (‘I don’t think Klitsch [Vladimir Klitschko] should go into the government. . . . I don’t think it’s a good idea.’ ‘Yeah . . . Just let him stay out and do his political homework and stuff.’ ‘I think Yats [Arseniy Yatseniuk] is the guy who’s got . . . the governing experience . . . what he needs is Klitsch and Tyahnybok on the outside. He needs to be talking to them four times a week.’⁸) The divisions in language, ethnicity, family ties, and economic interests between the east and west in the Ukraine made US meddling dangerous for Ukrainians. Combined with US aid and encouragement to the ‘Orange Revolution’ of 2004, the US responses to the overthrow of Yanukovych advanced a strategic goal of moving pro-US alliances, ultimately including NATO, eastward in the old Soviet sphere, as deeply as possible into the old Soviet Union. Increasingly strong and military Russian resistance is the predictable response. It would have been wrong of China to encourage America’s destabilizing eastward march. On the other hand, support for Russia’s intervention would have encouraged disorder in its own way. While the border of the Ukraine had been the outcome of a decision made by Khrushchev without reflection on local needs or the possibility of foresight of serious future consequences, widespread intervention to revise arbitrary borders would produce widespread disorder in our often arbitrarily segmented world.

Agreement in Practice, Disagreement in Foundations If one looks at what the government of China supports and opposes, it fits the patterns that I have described. If one looks at what the US government says in defending particular interventions, it observes these limits in the terms of justification, whatever the factual validity of its claims. The intervention in Kosovo was supposed to respond to widespread massacre (although it probably did not, but stimulated widespread massacre, instead).⁹ The overthrow of Saddam Hussein was primarily based on false and unwarranted claims concerning weapons of mass destruction, not appeals to his injustice. The US emphasizes allegations of widespread atrocity in Syria, not Assad’s tyranny as such, and, so far, has been responsive to grave dangers that would be triggered by invasion or massive additional military aid to Syrian rebels. So my arguments are supportive of the international diplomatic practice of justification, case by case, even if the US and China favour somewhat different formulas in their occasional summaries of their general views. The various general UN declarations on sovereignty, human rights, and intervention include passages which, in isolation, favour positions that are either

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

more pro-interventionist or anti-interventionist than my survey of recurrent constellations allows. That is all to the good. The anti-interventionist passages are a disincentive to governments whose interests draw them to excessive intervention. If they intervene too frequently, their international standing is threatened by frequent criticism making use of these formulas about respect for sovereignty. The interventionist passages are a disincentive to tyrants who might do more damage if they did not face a heightened risk of intervention. Real law, administered by courts and enforced by an effective executive authority, has to seek more coherence and precision. But this part of international law is not real law. However, even though my arguments largely fit current practice, they undermine a frequent view among those who fear US foreign conduct, namely, that the restriction of humanitarian intervention honours the proper value of sovereignty as such. For the value of sovereignty as such, as opposed to the typical dangers of its violation, has played no role in my arguments. In a nutshell: the reason why it would have been wrong to invade Iraq to end the unjust tyranny of Saddam Hussein is not that the exclusive authority of a central political organization is valuable even when it inflicts systematic grave injustice, but that all hell could be expected to break loose in case of invasion, inflicting grave harms without warranted confidence of consent of those who deserved protection. Such emphasis on likely consequences of intervention, rather than its inherent intrusiveness, contrasts with the views of Michael Walzer, the Euro-American theorist who has done the most to argue for strict moral limits to intervention. He regards intervention to end injustice within a political community which stops short of widespread massacre as wrong. He says that this prohibition can be based on the inherent value of political community even in circumstances of injustice. In his view, we should be guided by the precept, ‘Always act so as to recognize and uphold communal autonomy’, where the relevant community is ‘a single political community’, whose members have ‘a right . . . to suffer only at one another’s hands’.¹⁰ Whether this precept has the anti-interventionist implications that Walzer derives depends on the appropriate interpretation of ‘political community’. There is no interpretation on which the proper valuing of political community is a sound basis for his prohibition. In one natural usage, genuine political community requires mutual concern and respect among those living under a government. Of course, such a political process should not be vulnerable to intervention. But when tyrannies inflict systematic grave injustice, they are not in genuine political community with their victims, in this sense. In a much less demanding usage, mere governance

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi

WHY SOVEREIGNTY MATTERS DESPITE INJUSTICE



by a sovereign power constitutes political community. But the great virtues of this unity are its consequences in peace, order, and efficacy, which would be taken into account by standard considerations of proportionality. In between the morally demanding conception and the mere ascription of sovereign power, political community might be seen as resting on shared cultural ties and a shared view of their political importance. This is, clearly, Walzer’s conception, the basis for his assertion that outside help to a popular secessionist movement honours the value of political community. Suppose that a group of people regard their shared territory as their common homeland and seek to establish the terms on which they live together through sufficiently shared interpretation of norms that are part of their common culture. If the vast majority in the territory seek this reconciliation through mutual engagement, proper valuing of their project might seem a powerful reason for condemning interference by outsiders that interferes with governance by insiders. Again, whether a proper valuing of the internal process precludes intervention depends on the further specification of the process. If the vast majority of the people of a territory, including the vast majority of those with cause for complaint of injustice, willingly, self-respectfully take part in the establishment of the terms of their common life in their territory through mutual persuasion appealing to shared traditions and convictions, this joint project should not be violated by outside intervention. But a tyranny inflicting grave and systematic injustice interferes with this project as surely as a foreign invader. Such a government, even if it wholly consists of locals and appeals to an interpretation of local norms, derives its power from fear of its repression, fear of the disorder that would be unleashed by its overthrow, and manipulative indoctrination and misinformation, not from mutual persuasion appealing to shared norms and convictions. The assumption that typical victims of its grave injustices would reject foreign rescue as too external, consequent dangers to one side, conflicts with their frequent willingness to take on wrenching burdens of emigration, with the celebrations in the immediate wake of successful intervention, and with the unashamed requests of rebels for outside help. The value of the internal process in which a gravely unjust tyranny stably dominates will not make an intervention disproportionate when its benefits would otherwise be proportionate to its harms. This is not, remotely, to deny that humanitarian military intervention can interfere with movement toward the valuable form of political community. The foreignness of the intervening country, the shaping of its conduct as occupier by distinctive geopolitical interests, the reduction of incentives among insiders to seek great breadth in a coalition opposed to the tyranny, and the discouragement

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

of independent leadership by potentially unifying figures with popular bases of their own may well seriously retard the emergence of mutual persuasion appealing to shared norms as the basis for a shared political life. This consideration will be accompanied by other dangers of disorder, already surveyed. It is hard to imagine its rendering disproportionate an intervention which would otherwise be proportionate, despite those dangers. In any case, moral anxiety about the postponement of a future common political life based on mutual persuasion does not honour the inherent value of subordination to one’s own tyrant. I have argued that only sufficiently just sovereignty has inherent value. This assessment is powerfully expressed by the Confucian masters. Reflecting on the overthrow and killing of the tyrannical Emperor Zhou, King Xuan asks Mengzi (Mencius) whether it is permissible for subjects to kill their rulers and gets this reply. ‘One who violates benevolence should be called a “thief.” One who violates righteousness is called a “mutilator.” A mutilator and thief is called a mere “fellow.” I have heard of the execution of a mere fellow “Zhou,” but I have not heard of the killing of one’s ruler.’¹¹ Far from possessing inherent value, gravely unjust sovereignty, in this view, merits contempt. The wars that Mengzi approves are punitive expeditions, military responses to grave injustice. For example, he endorses a chronicle’s praise of the Emperor Tang for launching a war against a callous ruler who has attacked a group of old and young people bringing food relief, killing a boy: ‘this is . . . because he wanted to avenge the common men and common women’.¹² The horrors of disorder of the Warring States period led Xunzi to emphasize the importance of social structures that sustain order, yet he says that the highest standard of conduct is the rule, ‘Follow the dictates of the Way rather than those of one’s lord . . . follow the requirements of morality rather than the wishes of one’s father.’¹³

Beyond War Very fortunately, war is rarely treated as a live option for responding to foreign injustice. Equipped with a better understanding of this extreme, rare proposal for curing injustice, we can assess the morality of less drastic measures. Substantial military support for insurgents is, essentially, an act of war. I have already alluded to it. Its morality is the same as the morality of humanitarian invasion. But at least one source of potentially severe pressure meant to force change toward justice in sovereign countries is significantly different from war, namely, economic sanctions. These measures impose costs on innocent people in the damaged economy. But the harms are, typically, much less than the harms of violent disorder

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi

WHY SOVEREIGNTY MATTERS DESPITE INJUSTICE



inflicted by modern warfare. Informed consent of those with moral standing is more apt to be given, evidence of such consent is easier to obtain, and the threshold for adequate evidence of informed consent is lower, because of the less dire risks of the operation. The general risks to international order are less severe, as well. These are reasons why the international, government-supported boycott of chromium from Ian Smith’s Rhodesia and the banning of new trade and investment in South Africa by the US Anti-Apartheid Act of 1986 merited the support that invasion should not have received. If sovereignty were inherently valuable above the threshold of widespread atrocity, this would be a reason to oppose those initiatives, since they were directed at weakening and ultimately ending the authority of sovereign governments. So, the rejection of that value makes a difference for this response to foreign injustice. On the other hand, the harm to innocents is real; occasionally, it is devastating, as in the sanctions imposed on Iraq, through US initiative and guidance, after the first Gulf War.¹⁴ In addition, potential targets for economic sanctions are frequently allied with major powers, giving rise to global risks to peace. Finally, tactics this severe prevent trusting and respected participation of the target in deliberations that can be productive on topics removed from the issue of justice. So this mode of coercive pressure ought to be limited to serious violations of dignity that most victims can be expected to count as more serious than losses due to the economic sanctions, and such measures should only be supported by a government on the basis of broad international support of the goal of the economic sanction, if not the tactic itself. What remains is a vast moral terrain: the transnational condemnations of injustice in which the language of human rights is spoken. Here, most of the activity of governments consists of endorsement of general declarations that certain human interests deserve protection everywhere. Even if these documents do not specify a way of responding to violations of the rights that are declared, the implication is that some official response—at the very least, an occasional diplomatic expression of concern—may be appropriate. The violations of the rights affirmed in international declarations rarely rise to the level of large-scale atrocity. The several UN resolutions and UN-sponsored conventions on torture condemn even occasional torture by a government. Saudi Arabia clearly, pervasively violates the 1979 Convention on the Elimination of All Forms of Discrimination Against Women and the prior General Assembly resolution on this topic, but it would degrade an important category to say that this injustice has the same seriousness as widespread massacre. Human rights declarations do some harm. Their sparse and uneven implementation and the blatant hypocrisy of many signers creates cynicism

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

concerning moral condemnation, condemnation that would, in any case, be expressed by advocates and NGOs. The discourse of human rights that the declarations support makes it easier for the United States and its European allies to intervene militarily when this is not justified and to pick and choose among targets for intervention and sanctions in ways that strengthen domineering influence that the US has abused. There is no strong evidence that signing a human rights declaration makes a government more apt to respect human rights.¹⁵ The good that justifies human rights declarations, nonetheless, is help in overcoming loneliness, isolation, and despair among victims of injustice. Awareness that what the apartheid regime said was right and normal was a gross violation of justice in the eyes of the rest of the world must have lent encouragement and comfort to victims of apartheid during its many decades. Today, women in Saudi Arabia must be helped in similar ways by international insistence that what they seek is liberation and respect, not, as their oppressors say, self-degradation. Of course, this is also the message of NGOs. But there are vast numbers of NGOs on every side of nearly every issue. Because it is exceptional, painstakingly negotiated, and achieved at some cost to alliances with obvious violators, the endorsement of a human rights declaration by many governments plays a special role in overcoming isolation. Now, we have come to a final and extremely sensitive question about the appropriate response of governments to foreign injustice. General declarations of human rights are one thing. A government’s condemning another government for violating human rights is quite another. Some say that this condemnation should be avoided because it conflicts with the sort of autonomy that Walzer seems to value as such, struggle over the terms of a common life confined to those who share it. But if people can expect to continue to suffer serious deprivation of human rights if the current process of governance endures, the absence of outside voices offering support for their grievances does not seem valuable as such. Is there, then, no morally serious reason for general reluctance to accuse another government of violations? The reason lies in the normal humane benefits of diplomacy. Official conversations among governments, bilateral, multilateral, and through global organizations, are the way in which governments resolve differences peacefully and productively, creating bonds of trust that are an important supplement to commerce and culture. If, in a forum for this cooperation, governments are subjected to official condemnation, they have reason to devalue the forum; they also have reason not to rely on the good will of those who condemn them when they might be partners in negotiations leading to binding compromises that require trust.

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi

WHY SOVEREIGNTY MATTERS DESPITE INJUSTICE



Suppose that the failure of the United States to relieve burdens of poverty whose moral significance is stated in human rights declarations were affirmed in a proposed UN resolution. This resolution could be accurate. China, the world leader in liberation from poverty, would not be hypocritical to support it. But China wisely would not. The grave cost of encouraging the United States to devalue a useful forum for negotiations would not be worth the gains. Indeed, as often happens, the gains would be nil, since the most important opponents of US neglect of the poor, namely, Americans who support doing more, would now have to cope with patriotic suspicions that they are unpatriotic supporters of foreign powers. Of course, the responses of a superpower have special impact. But many governments have the same combination of moral flaws that could be condemned as violating general declarations of human rights together with a capacity to engage in humanely beneficial cooperation and reliance on patriotic allegiance that resists yielding to concerted outside pressure. The general reluctance of governments to engage in strong, explicit public condemnation of other governments for violating human rights is not just strategically but morally justifiable. It typically sustains humanely beneficial deal-making and avoids intrusion that would heighten, rather than relieving, the isolation of victims of injustice.

Beyond Governments The question of how to respond to perceived foreign injustice is faced not only by governments, but by individuals, as well. The question is especially pressing for relations between people in the United States and China because many Americans believe that the government of China is gravely, systematically unjust. My case for governmental reluctance to condemn other governments provides them with reasons why they should not want their government to do more at present than occasionally mildly express concern at isolated problems, as a minor aspect of diplomatic engagement. But it would be cowardly to restrict the question of condemnation to governments. Each of us can take part in discussions and nongovernmental activities that address moral and political issues. How productive are non-governmental criticisms of foreign injustice meant to reach other countries including the one we take to be unjust? If we are speaking as individuals or through non-governmental organizations, not as official representatives of a government, the arguments about the humane benefits of diplomacy undisturbed by extensive condemnation of foreign governments do not apply. Sometimes, non-governmental condemnation of foreign injustice plays an important role in reducing the isolation of its victims and creating incentives

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

for world powers to reduce their support for the perpetrators. The widespread non-governmental opposition to injustices of the government of Israel against Palestinians is, I believe, a case in point. How beneficial is such criticism, directed at the government of China, on the part of citizens of the United States? Of course, an American’s answer will, in part, depend on his or her view of governance in China. On one view, the basic processes of governance in each country are morally defensible even though governance in China, like governance in the United States, leads to injustices that are not simply the inevitable failings of individual officials. But many Americans regard one-party rule in China as morally indefensible on account of unjustifiable violation of values of liberty and democracy and inevitable grave harms. Still, they should recognize the limited value and adverse side-effects of addressing their criticisms to people in China. Because argument and criticism by people in China are vigorous, in any case, critical Americans do little to reduce isolation; indeed, they often add to it by making internal critics seem agents of the sole superpower. Nonetheless, criticism of China’s governance from the United States can do much good if it is part of a genuine conversation about justice and views of justice among people in both countries. This is not because Americans have wonderful insights to offer in their criticisms when they intervene in Chinese political discourse. They are usually less informed and insightful than millions of people in China. Rather, the conversation advances the cause of peace. Critics of China in the United States need to learn from people in China who are basically supportive of Chinese governance as well as people deeply opposed to it and need to gain a sense of the broad range of opinions and arguments in China including these two poles. People in China who engage with criticisms of their country from abroad, rejecting what they take to be wrong in the criticisms while acknowledging the real problems, play a central role in this learning process. Especially effective are those who display the mixture of criticism and acceptance, outrage and pride characteristic of thoughtful Americans surveying their own country. Rather than turning away and minding their own business or substituting pity or scorn for listening, individuals in both countries can advance mutual engagement through discussion and debate about injustice in both countries. The major gain from such openness is not new knowledge, but help in avoiding a disaster that threatens humanity. The challenge to US pre-eminence of the rise of China may eventually lead to struggles over regional influence in which militarists and ultra-nationalists on both sides call for resort to violence via proxy wars and sponsored rebellions and civil wars. Revival of the very hot

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi

WHY SOVEREIGNTY MATTERS DESPITE INJUSTICE



calamities of the Cold War will be much more likely if many people on one side or both see people on the other as a mix of oppressors, oppressed, and dupes, people whose outlooks are shaped by interests in power, patriotic fervour, fear, ignorance, and deference to their superiors. Who would not fear the international power of the government of such people? How could it be contained except by force? (That Americans and their British cultural-historical cousins did not see one another in this way insured against disaster when the one country overtook the other.) Polite avoidance of delicate topics will not prevent this estrangement. The vaccine is mutual engagement in which moral judgement is an invitation, not an insult. The prospect of killing people with whom one is engaged in serious discussion, in which their current form of government is taken seriously as a possible path to justice, is disgusting. For this reason, the search for common ground in responding to shared questions of justice is not just intellectually revealing. It may be life-saving. If keeping struggle over justice and injustice in a country confined to compatriots were inherently valuable then the right Chinese response to my critical compatriots would be ‘Mind your own business’. But, in fact, this moral insulation is not inherently valuable, and the barriers it creates to engagement in criticism and response would be a moral loss.

Notes 1. Daniel Patrick Moynihan, A Dangerous Place (Boston: Little, Brown, 1978), 247. 2. See William Hartung, ‘U.S. Arms Transfers to Indonesia 1975–1997’ (New York: World Policy Institute, n.d.), , 5–6. 3. See Jonathan Randal, After Such Knowledge, What Forgiveness? (New York: Farrar, Straus & Giroux, 1997), 259. 4. Amnesty International, ‘Turkey: No Security without Human Rights,’ [1996], 3. 5. United States Statistical Abstract 1995 (Washington, DC: US Census Bureau, 1995), table 1305 and, on the basis of further government figures, Tamar Gabelnick, William Hartung, and Jennifer Washburn, Arming Repression: U.S. Arms Sales to Turkey during the Clinton Administration (Washington, DC: World Policy Institute and Federation of American Scientists, 1999), , 10–12. 6. See Randal, After Such Knowledge, 269; see also Gabelnick et al., Arming Repression, passim. 7. UN Mission of the People’s Republic of China, ‘Statement by Ambassador Liu Zhenmin at the Plenary Session of the General Assembly on the Question of the “Responsibility to Protect” ’, 24 July 2009, .

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

8. BBC News, ‘Ukraine Crisis: Transcript of Leaked Nuland-Pyatt Call’, 7 Feb. 2014, . Ukrainian politics developed along these lines. Yatseniuk became prime minister. Klitschko was elected mayor of Kiev and headed the winning parliamentary bloc in the May 2014 elections. Tyahnybok, a far-right politician known for appalling anti-Semitic diatribes, led a party which was part of Yatseniuk’s first coalition government and which provided a militia fighting in the contested eastern regions. 9. Anti-Kosovar repression in the year before the NATO bombing campaign began seems to have resulted in a death toll of about 2,000, including Kosovo Liberation Army combatants, with declining civilian casualties in the previous six months. During the seventy-eight days of the bombings about 10,000 Kosovars were killed. See Independent International Commission on Kosovo (a UN-supported committee of inquiry, largely financed by the Swedish government), Kosovo Report (Oxford: Oxford University Press, 2000), 83 and 97. 10. See Michael Walzer, Just and Unjust Wars (New York: Basic Books, 1977), 90, 87, and 86. 11. Philip Ivanhoe and Bryan Van Norden (eds), Readings in Classical Chinese Philosophy (Indianapolis: Hackett, 2005), 125; from book I, part B, 8 in the traditional compilation. 12. Mencius, tr. D. C. Lau (London: Penguin, 2003), 68; book III, part B, 5. 13. Xunzi, vol. iii, tr. John Knoblock (Stanford, CA: Stanford University Press, 1994), 251; book 29, 1. 14. Assessments by UN agencies reported ‘an explosive rise in the incidence of . . . cholera and typhoid’, ‘severe hunger and malnutrition’, and ‘massive human suffering’. See World Health Organization, ‘The Health Conditions of the Population of Iraq since the Gulf Crisis’, 1996, , 14; Food and Agricultural Organization/ World Food Programme, ‘Crop and Food Supply Assessment Mission to Iraq’, 1993, , 1. 15. See e.g. Emilie Hafner-Burton and Kiyoteri Tsutsui, ‘Justice Lost! The Failure of International Human Rights Law to Matter Where Needed Most’, Journal of Peace Research, 44 (2007): 407–25.

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

3 Women and Humanitarian Intervention Janna Thompson

Michael Walzer says that humanitarian intervention in the affairs of another state can be justified when it is a response to events that shock the conscience of humankind.¹ Two attacks on women in recent times have had this disturbing effect. The first was the mass rape by Serb military forces of Muslim women in Bosnia. These women, many of them still children, were imprisoned, abused, repeatedly raped, and deliberately impregnated. Those who survived this ordeal were often rejected by their families and communities. The second shocking affair was the treatment of women by the Taliban when it came to power in Afghanistan. Women were denied education and opportunities for employment. They were severely punished for any infraction of strict rules of dress or deportment. They were not allowed to go anywhere without a male family member. They were not allowed to contact each other. Some women died of hunger or ill health because of their inability to leave their homes or lack of access to health care. These are not the only recent attacks on women that can be described as shocking. In the course of the genocide in Rwanda, Hutu men systematically raped, mutilated, and killed Tutsi women. In its advance through the Middle East ISIS is enslaving women of minority groups and forcing restrictions on others that are similar to those imposed by the Taliban. However, in this discussion, I will concentrate on the attacks on women in Bosnia and Afghanistan. Guided by Walzer’s test, should we agree that humanitarian intervention, in some form or another, was, or would have been, justified to rescue women from mass rape in Bosnia or suppression in Afghanistan? The rapes committed by Serb militias were widely condemned and the International Criminal Tribunal for the Former Yugoslavia ruled that rape is a crime against humanity. First Lady Laura Bush drew attention to the plight of Afghan

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

women in a speech that described the war on terrorism as a fight for the rights and dignity of women.² Nevertheless, it is difficult to suppose that these attacks on women had much to do with motivating or justifying interventions in former Yugoslavia or Afghanistan. Only one Serbian leader appearing before criminal tribunals was convicted of organizing rape. Other deeds loomed much larger. The pledge to liberate women from the Taliban was, perhaps, more than a public relations stunt to justify the subsequent invasion of Afghanistan but the suppression had been going on for many years and it is unlikely that the intention to rescue women was very high on the agenda of the invaders.³ There is thus a mismatch between the shocking nature of the events and the way they were treated by international agents and organizations. The reason, I think, is not merely that national and international affairs are largely run by men who have other priorities. Attacks on women do not fit very comfortably into the justifications for humanitarian intervention that are generally accepted by international agents or philosophers. Nor do they connect well with values and assumptions that underlie debates about humanitarian intervention. I will first of all explain why this is so and will then consider whether the recent emphasis on ‘responsibility to protect’ could make a difference to how international actors regard and respond to oppression of women and violence against them.

Humanitarian Intervention and Rape of Bosniak Women The international bodies that condemned the imprisonment and mass rapes of Bosniak women understood these acts not merely as crimes against women but also as an attack on Bosnian Muslim communities. It was clear that this was the intent of the perpetrators. Women are central to family and community life and to its perpetuation through the generations. So by attacking women the perpetrators were attacking and terrorizing their communities. They hoped that the fear that they caused would drive people away from the territories they wanted to claim for a Greater Serbia. Moreover, in traditional societies where female purity is valued, where men require their wives to be faithful and families want their bloodline to be continued, rape of their women is shameful and degrading—not only to fathers and husbands but to women themselves. The militias demonstrated their hatred and contempt for Bosnian Muslims through organizing mass rapes of Bosniak women. And by impregnating these women so they would bear their children, they intended to undermine the intergenerational integrity of Bosniak communities.

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

WOMEN AND HUMANITARIAN INTERVENTION



The United Nations Convention on the Prevention and Punishment of Genocide includes as genocidal acts not merely mass murder but other crimes calculated to bring about the destruction or partial destruction of a group, including prevention of births in the group or the transfer of children to another group. Though not specifically mentioned, systematic rape committed with the aim of terrorizing and undermining a group fits the intent of the Convention. The International Tribunal for Rwanda recognized this by ruling that rape done with the intent to destroy or undermine a group is a genocidal act. Genocide is one of the worst crimes against humanity. As Karen Engle says, ‘If one wants to see military intervention for reasons other than self-defense, genocide provides the firmest ground. Particularly in situations that might otherwise be characterized as civil war, genocide is the trump card that permits, if not requires, intervention.’⁴ Nevertheless, identifying the mass rape of Bosniak women as genocidal is problematic. First of all, it invites a comparison between this crime and other acts of genocide that the Serbs or others committed. Compared to acts of mass slaughter, such as the massacre by the Serb military of Bosniak men and boys at Srebrenica, the mass rape of women, many of whom survived, is likely to be regarded as a lesser crime. So perhaps it is not surprising that it has rarely figured in the trials of the perpetrators of genocide. Moreover, despite the Genocide Convention, most people assume that genocide only occurs when perpetrators engage in systematic, large-scale acts of murder.⁵ The crime against Bosniak women does not fit this conception of genocide and thus it is not so obvious that it would have counted, in most people’s opinion, as a sufficient justification for humanitarian intervention. If the Serbs had not engaged in mass murder, if they had confined themselves to rounding up Bosnian women and raping them, would this deed have been regarded as bad enough by leaders of Western powers and citizens to count as a just cause for humanitarian intervention?⁶ A further problem with identifying organized mass rape as genocide is that it does not do justice to what most of us regarded as shocking about the treatment of the Bosniak women. Genocide is an attack on a group, and those who condemn mass rape as genocide are focusing on the harm that it does to the religious, racial, or ethnic group from which the women come. Rape as a crime against women themselves disappears from this account of the wrongdoing.⁷ But what was so horrible about the actions of the Serb militias was the way that they treated the women, who were not only physically assaulted and imprisoned, but were also subject to acts that were intended to humiliate and degrade them. The men who attacked them were not merely engaged in an action against their communities. Their acts demonstrated contempt and hatred for the women themselves. To add to the harm, the crime condemned the survivors to shoulder

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

the burden of degradation for the rest of their lives—and was calculated to do so. The Serbian militias seriously violated these women’s human rights. To condemn them solely for attacking Bosniak communities draws attention away from the crime against the women. Rape is a violation of human rights. But is it a legitimate reason for intervention even if committed on a massive scale? Walzer asks us to make a distinction between acts that are truly shocking and ordinary cases of tyranny. The former in his view can provide a legitimate reason for military intervention. The second do not. How that distinction should be made is a matter on which judgements differ. But there are many communities in the world where women are forced into marriage and required to have sexual intercourse with their husbands and to bear their children whether they want to or not. There are many countries where violence against women and rape are not regarded as serious crimes and are seldom punished. Catherine MacKinnon points out that societies that condemn acts of terrorists tolerate violence that terrorizes women.⁸ One of her points is that context, and not the badness of the crime, is often the explanation for why we are shocked in one case and not in others. She thinks that societies ought to take violence against women much more seriously than they do. But it is possible to turn her argument around to reach the conclusion that the treatment meted out by the Serbs to Bosniak women was ordinary tyranny against women— grabbing our attention only because it occurred in such a concentrated way. This view is supported by the way rape in war is regarded and treated. Rapes are common in war and have sometimes been committed on a massive scale. The Geneva Convention prohibits rape in war but there have been few prosecutions of military leaders for failing to prevent it and it is easy to reach the conclusion that this prohibition is not taken as seriously as prohibitions against other war crimes. However, MacKinnon’s point—that crimes against women are not regarded with sufficient seriousness—can be applied to the case of the Bosniak women. Rape is not merely a physical assault but a violation of the integrity of a person—a violation to her (or his) sense of dignity and worth. In the circumstances in which the Serbs committed their crime this violation was magnified by the way the women were treated and used. Moreover, it is not possible to regard as separate injuries the harm done to the women and the harm done to their communities. The women were members of their community and suffered as the result of the harm done to it—particularly because they were used as instruments for undermining it. A proper appreciation of the wrong done provides a case for saying that the mass, systematic rape of Bosniak women would have been a just cause for

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

WOMEN AND HUMANITARIAN INTERVENTION



humanitarian intervention. But justification of humanitarian intervention also requires the satisfaction of further criteria. Walzer says that it must have a reasonable expectation of success and interveners must also have reason to believe that their actions will not cause disproportionate harm in achieving their aims. Whether an intervention to save Bosniak women from imprisonment and mass rape would have satisfied these conditions is open to doubt. Interveners would have had to consider whether their intervention would actually save the women. In the face of a threatened attack from military forces, the Serbs might have decided to the kill their victims rather than to let them go. In any case, a military action, however successful in freeing the women from enslavement, would not have prevented their rejection by their communities. If ‘success’ requires preventing further harm then military intervention would not have been entirely successful. Moreover, rescuing women from enslavement as well as prevention of other crimes against humanity in former Yugoslavia would have required a direct engagement with militias. When NATO decided to intervene it judged that a ground war would be a disproportionate risk to its own forces. Given that leaders are morally required to minimize harm to their own soldiers, this decision might be regarded reasonable. But the form of intervention they judged to be proportionate—aerial bombardment of strategic targets—did nothing to prevent genocidal acts and some argue that NATO intervention encouraged these acts.⁹ So even if it is accepted that rescuing women from imprisonment, violence, and mass rape is a just cause, it is doubtful that intervention to save Bosniak women could have been justified, given the way restrictions on intervention are generally understood and given the means interveners have available for preventing further harm.

Humanitarian Intervention and the Suppression of Afghani Women The Taliban put women under a harsh regime that prevented them from participating in public and political activities, denied them employment, basic opportunities, and services, and kept them under the control of the men in their families. As women they were assigned a separate and lesser status and the Taliban government enforced their subjection. The system of gender discrimination enforced by the Taliban is thus comparable to apartheid as it was practised in South Africa on racial grounds.¹⁰ Apartheid is widely regarded as a crime serious enough to be grounds for humanitarian intervention, at least in the form of sanctions against regimes that practise it.¹¹ The International Criminal Court recognizes it as a crime against

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

humanity. The United Nations General Assembly passed a resolution condemning apartheid and also a number of resolutions calling on member countries to impose sanctions on trade with and investment in South Africa and to terminate sporting and cultural links. These sanctions, adopted by the US and many other nations, played an important role in forcing the South African government to end apartheid. If intervention is justified to end racial apartheid then it seems that intervention in some form could also be justified to combat gender apartheid. Indeed, the restrictions imposed by the Taliban on women in Afghanistan were more extreme than those imposed on black South Africans under apartheid—and thus it is arguable that they could have justified a more vigorous form of intervention. South African blacks were not so restricted in their movements or in their opportunities. They were not so tightly controlled in their everyday lives or in their ability to make a living. Indeed, the system imposed by the Taliban on women had features in common with slavery, and enslavement, even more so than apartheid, provides grounds for humanitarian intervention. However, women’s position under the Taliban was different in important respects from the situation of black South Africans under apartheid. It was also different from the situation of those whose persecution or murder makes them paradigm cases for humanitarian rescue. When philosophers and others consider what counts as a just cause for humanitarian intervention they generally make two assumptions, neither of which holds in the case of Afghani women. They assume, first of all, that the agent wholly or largely responsible for the wrongdoing is a state—though this category of wrongdoers has been widened to include militias, terrorist groups, and other non-state political organizations. Intervention has the purpose of putting pressure on this agent so that it will change its behaviour or, in extreme cases, of defeating or removing it so that citizens are free to put something better in its place. Humanitarian intervention is supposed to target repressive states or murderous militias. It is not supposed to be an attack on the cultural traditions of the population. Justifications of intervention assume, secondly, that the victims of the repressive or murderous regimes or militias are either the citizens of the repressive state or members of a particular ethnic, racial, or religious group. The actions that serve as a justification for intervention are those that seek to drive out members of the group or to massacre them, discriminate against them, or enslave them. It can be assumed that the victims strongly oppose these actions and would welcome intervention. In many cases they make their desire for intervention well known. The African National Congress, representing a majority of black South Africans, supported sanctions. Intervention thus has the purpose of stopping ethnic

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

WOMEN AND HUMANITARIAN INTERVENTION



cleansing or genocide, freeing the slaves, and if necessary providing continued protection of the group that was the object of attack—perhaps by providing them with protected enclaves or by enabling them to change their government. These assumptions play an important role in giving intervention a just cause. They also figure in reasoning about the consequentialist considerations that necessarily play a role in justifying intervention. If interveners can put a stop to the wrongdoing by putting pressure on the unjust regime or defeating the repressive or murderous political organization then they have reason to believe that their intervention will be successful. If they can protect the victims of wrongdoing from further harm by the means that they, or the UN, have available, then they have a reason to believe that their intervention will result in more good than harm. If they have the support of the victims, and if victims have means of taking advantage of their liberation (as did the African National Congress), then the interveners have further reason to think that their intervention will have positive results. In the case of the Afghani women the assumptions that play such a large role in justifying interventions did not hold or did so to a much lesser degree than in South Africa. Their repression was not wholly the result of an imposition of political force. The measures imposed by the Taliban were based on interpretations of religious doctrines and traditional tribal conventions accepted by many Afghani people—particularly those who lived in tribal areas. Though these doctrines and customs were in conflict with the way of life that many people had adopted in Kabul and other urban centres, the views about the place of women that the Taliban held were also held, if not generally in such a radical form, by many people in Afghanistan, including many women. Responsibility for discrimination against women belonged not merely to the Taliban government but also to traditional tribal and familial practices and those who upheld them. The second assumption—that those who attack, enslave, or repress are ethnically, racially, religiously, or ideologically distinct from those who are attacked, enslaved, or repressed—also did not hold for Afghani women. Women were members of the community that suppressed them. They lived in the same houses, towns, and villages as their oppressors. They were the wives, daughters, and sisters of the men who were expected to play a role in enforcing restrictions on their behaviour and in many cases did so willingly. The failure of the usual assumptions to hold not only raises questions about what it means, and what it would take, to liberate Afghani women. It presents serious difficulties for those who think that their liberation counts as a just cause for humanitarian intervention. One of the difficulties is justifying intervention to liberate a group that has not requested intervention. There were indeed groups of

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

educated Afghani women, mostly in exile, who were strongly in favour of intervention but they could not claim to represent a majority of Afghani women in the way that the African National Congress could claim to represent a majority of black South Africans. What the majority of women wanted—what, if any, changes they would have wanted to bring about in their lives and their society—was not known. The problem was not merely that the opinions of these women were inaccessible because of restrictions on their ability to communicate with outsiders. Their lack of education meant that they were ill-equipped to make judgements about matters beyond their experience. Some feminists argue that women in this position are victims of ‘adaptive preferences’.¹² Their upbringing, daily repression, and lack of knowledge of alternatives make them unable to appreciate the advantages of liberation and to accept the life they have. But even if this is so, it is problematic to suppose that outsiders can determine what such women would want if they were fully informed and, on that basis, engage in acts on their behalf. In her discussion of the banning by the British of the practice of suttee in India, Chakravorti Spivak writes disparagingly about ‘white men rescuing brown women from brown men’.¹³ She is not saying that women should never be rescued from oppressive circumstances. Her point is that it is problematic to rush in and rescue them when they have never been given a chance to speak for themselves or to determine their own objectives. Moreover, the restrictions that many philosophers impose on what counts as a just cause for humanitarian intervention tells against intervention for the liberation of Afghani women. Walzer and many others subscribe to John Stuart Mill’s view that a people should, in general, be left to fight for its own liberation.¹⁴ By doing so, Mill contends, people learn to stand on their own feet, to appreciate properly their freedom and their sovereignty. If they are given their freedom by well-meaning interveners they will not learn these lessons and they will probably not be able to maintain it. But if people do not fight for their own liberation or do not do so with sufficient vigour then they have, in effect, chosen to live under the government they have. Walzer allows that intervention can be justified when a government is so oppressive or murderous as to render irrational and callous the demand that people liberate themselves. But, following Mill, he regards these as exceptional cases. Imagine, he says, that there is a country called Algeria where revolutionaries, though supposedly fighting for a democratic state, establish instead a theocracy that imposes patriarchal restrictions on women. But if Algerian people accept their regime as reflecting their history and values, outsiders have no right to interfere with their choice, even to help the minority who are democrats and

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

WOMEN AND HUMANITARIAN INTERVENTION



feminists. ‘For foreigners cannot judge the relative strength of such movements or allow them to substitute themselves for the people as a whole, not until they have won sufficient support to transform Algerian politics on their own.’¹⁵ The same reasoning, it seems, could be used to argue against intervention for the sake of Afghani women. The Taliban were welcomed as liberators by women and men who had suffered under corrupt governments and civil war. The restrictions they imposed reflected the history and values of many of its people. Though these restrictions were opposed by minority groups of liberals and feminists, these dissidents could not claim to represent the people of their country. Jean Cohen, who also follows in the footsteps of Mill, stresses that sovereignty is valuable even to people repressed by their state, not only because it gives them the opportunity to determine the destiny of their political society through their own efforts but also because it enables them to maintain and protect the traditions that they value. So long as a state enables them to protect these valued relations and practices it can be said to represent them. Intervention can only be justified, she says, if policies or laws treat a targeted group as ‘not one of us’, as an enemy or a people who has no right to exist.¹⁶ By doing so a government forfeits the claim to be representing the groups it oppresses. Her distinction between circumstances that justify intervention and those that do not provides no support for an intervention to liberate women from the Taliban. The Afghani women were not treated by the rest of their society as ‘not one of us’. They had rights that were defined by their traditional roles in their society. These rights were very restrictive but they nevertheless gave women a respected place in their community—a place that was reinforced by their interpretation of their religion. The positions of Walzer and Cohen thus seem to lead to the conclusion that liberating women from the Taliban was not a just cause for intervention of any kind. But those who accept this conclusion fail to appreciate the difficulties women can face in struggling for, or even envisioning, liberation. Mill, on the other hand, did understand this problem. Writing about the subjection of women in Victorian England, he not only stresses the means used to pressure and condition women to accept their lot. He also points out that women face far greater barriers to self-liberation than do other oppressed groups. Those who possess power over women, he says, have facilities in this case, greater than in any other, to prevent any uprising against it. Every one of the subjects lives under the very eye, and almost, it may be said, in the hands, of one of the masters—in closer intimacy with him than any of her fellow-subjects; with no means of combining against him, no power of even locally overmastering him . . . If ever any system of privilege and enforced subjection had its yoke tightly riveted on the necks of those who are kept down by it, this has.¹⁷

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

Mill did not suppose that women would be able to liberate themselves. He looked to well-meaning men to achieve reforms favourable to women. As it happened, women in the late nineteenth and twentieth centuries were able to conduct an organized, effective struggle for their political rights. But their ability to do so was made possible by education, and it depended on their ability to communicate, organize, and protest. In Afghanistan the Taliban systematically removed all avenues for protest, all of the means by which men, as well as women, could discover alternatives and go about questioning their traditions and changing their social and political life. If women are denied an education then they can’t learn about or make use of information critical of their traditions. If they are forbidden communication with each other or isolated in their homes, then they can’t form any kind of civil society organization, let alone one able to struggle against the regime. If men are compelled by law to uphold patriarchal restrictions on women, then they cannot make changes in their own families or communities that would give their wives and daughters more freedom. In other words, the restrictions imposed by the Taliban were designed to ensure that there could never be a popular and effective feminist movement. We can agree with Walzer and Cohen that humanitarian intervention is not justified just because rulers impose authoritarian restrictions on political activities or on the lives of citizens. It is not justified just because some people are discriminated against in ways that liberals do not approve of. It is one thing for a society to educate girls and boys in a way that encourages them to adhere to their religion and way of life. It is another to deprive girls of an education altogether. Doing the first makes it likely that women and men will accept traditional restrictions, but it does not make rejection of them impossible. Doing the second means that women will never have a chance to discover that there are alternatives. It is one thing for a government to put some restrictions on the movements of women (as is done in many Muslim countries). It is another to prevent women from communicating with each other or from leaving their homes. It is one thing for a government to enforce a gender dress code. It is another for it to assault or even kill people who infringe it in the slightest way. It is one thing for a state to have laws that favour a particular religion. It is another for it to require people to obey the precepts of a particular sect and to punish them harshly for disobedience. The first type of restriction reflects the religious and patriarchal values of the majority but it does not cut off all possibilities for change. The second type is meant to deny people the ability to criticize their traditions, reinterpret religious requirements, or change their society. Restrictions of the latter kind are worse than ‘ordinary oppression’. Their existence surely counts as a just cause for intervention. However, reaching this conclusion does not provide a sufficient

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

WOMEN AND HUMANITARIAN INTERVENTION



justification for humanitarian intervention, let alone military intervention. Consequentialist requirements must also be fulfilled and differences between the situation of Afghani women and paradigm cases of just intervention make their satisfaction doubtful. When the oppressor is a state or some other political agent then an intervention is likely to be successful if interveners have means of removing it or forcing it to reform. But the restrictions that oppressed women in Afghanistan were not the creation of the Taliban. They were more extreme versions of widely accepted practices. To liberate women interveners had to contend not only with the Taliban but also with the husbands, fathers, tribal leaders, and sometimes women themselves who supported these restrictions. This made success more difficult to define or achieve. There was also a danger that the efforts of the liberators would be counterproductive. People tend to resent foreign intervention when they fear, rightly or wrongly, that their values and traditions are under threat. An intervention that is seen, or can be presented by its opponents, as an attack on the traditions or religious requirements of a people is in danger of being resisted by large parts of the population—even by those who object to many of the actions of their government. If people react to intervention by regarding restrictions on women, and the sexual morality associated with these restrictions, to be a defining expression of their opposition to Western interference then the cause of women’s liberation is likely to be set back by intervention. This, perhaps, is the lesson that the rise of ISIS and the increasing fundamentalism of some Muslim groups is teaching us. The problem posed by an intervention that challenges cultural conventions is compounded by the fact that women are part of the communities that enforce restrictions on them. They cannot be put in separate enclaves, given territory of their own, or effectively protected by a peacekeeping operation—the means of harm prevention that interveners generally have available. What progress was made in freeing Afghani women from restrictions and giving them a political voice has depended mostly on reforms made by a government formed with Western backing. But these reforms have not done much for women in many communities, who continue to be restricted and sometimes imprisoned by their families. More girls get an education, but they are often removed from school early by parents who do not think that education is worthwhile for women. Women can participate in politics or gain employment, but they are often threatened and punished for doing so. A strict dress code is no longer enforced by the government, but gangs of men beat up women who violate what they regard as proper dress. Moreover, reforms favouring women are strongly opposed by fundamentalists and may be watered down or repealed as Westerners depart Afghanistan and as the Taliban gains more influence.

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

Responsibility to Protect I have argued that both the mass rape of Bosnian women and the suppression of women under the Taliban count as just causes for humanitarian intervention of some kind. Paradoxically, but not surprisingly, intervention to prevent or stop these violations would probably not have satisfied consequentialist requirements for a just intervention—at least as these requirements are usually understood. Although there are significant differences between the two cases, they have one thing in common: the harm done to the women was in part the result of attitudes and traditional assumptions about the role of women and their place in their community. Interveners are ill-equipped to deal with the causes of these harms and thus their interventions can make things worse, rather than better, for women. However, a failure to satisfy consequentialist conditions for humanitarian intervention may be due, at least in part, to adherence to traditional views about intervention and the responsibilities of interveners. Intervention, it is generally assumed, is an imposition of force taking either the form of sanctions or military action. When interveners use military force they assume that their job is to defeat their opponents or to make them negotiate. They have limited means to prevent further violations and are often not prepared to deal with the social consequences of their intervention, or they regard it as not their responsibility. Recent failures of interventions to achieve hoped-for results ought to motivate a reconsideration of strategies of intervention and the responsibilities it entails. One attempt to do so is the proposed international norm, Responsibility to Protect (R2P).¹⁸ Responsibility to Protect, which has been endorsed in a number of international forums, stresses that states have a responsibility to protect their population from grave harm. If they do not fulfil this responsibility, then intervention by outsiders is not only permitted, it may be required. The emphasis on protection of people can thus be understood as putting much more pressure on interveners to take risks and make sacrifices for the sake of preventing massive violation of human rights. Concentrating on defeat of military forces—as NATO did when intervening in former Yugolsavia—is not sufficient. Most of the debate about R2P has focused on its support of military intervention to prevent serious abuses of human rights. Critics fear, with some justification, that it gives a licence to powerful states to intervene for their own purposes under the cover of pursuing a humanitarian cause.¹⁹ However, R2P has aspects that deserve more discussion. According to Gareth Evans, R2P differs from humanitarian intervention, as it is usually conceived, by emphasizing the prevention of humanitarian catastrophes and reconstruction after intervention.²⁰ This change of emphasis has the virtue of opening up a discussion about how

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

WOMEN AND HUMANITARIAN INTERVENTION



populations can be protected by forms of intervention that do not necessarily involve sanctions or the use or threat of military force. It is not absurd to suppose that the Taliban could have been encouraged to relax its harsh rule by appropriate interventions at the right time. When it first came to power it was desperate to attract foreign investment and anxious to be recognized as the legitimate government of Afghanistan. This was an opportunity for negotiation that could have touched on its domestic policies. Nothing would have persuaded the Taliban to support feminism or liberal democracy but it might have been persuaded to become more like other Muslim states—especially if some of these states took responsibility for inducting it into the conventions of international society. Economic aid would have helped the process of modernization. Increasing wellbeing and new opportunities often bring with them social changes favourable to women as families recognize the value of educating their daughters as well as their sons. And resistance is less likely to occur if people themselves can see the advantages of change. These ideas about intervention might be taken in a more radical direction. Bhikhu Parekh thinks that humanitarian intervention should be understood as ‘a logical extension of the duty of mutual help which states, as representatives of their citizens, owe each other’.²¹ If we adopt this point of view we would accept other duties: for example, to divide resources more equally and to avoid agreements that are unjust. ‘We would then appreciate the moral need to share our material, political, ethical, and cultural resources with the rest of mankind and to cooperate in creating a just, peaceful and relaxed world in which many of the causes that generate the need for humanitarian intervention would disappear.’²² Parekh is perhaps over-optimistic to assume that most of the causes of severe violations of human rights would disappear with more sharing. In any case, countries that support R2P do not suppose that they have signed on to such a radical programme. But its emphasis on preventing or mitigating human rights violations by means that need not involve invasion, sanctions, or threats of force could lead to interactions that are more likely to help vulnerable people—especially women. Though Responsibility to Protect is not generally regarded as a radical programme for changing international relations, it does encourage a conversation about how military intervention, when necessary, should be conducted and what should be done to ensure that more violations do not occur. Women and children are endangered by military violence even if they are not directly attacked and some who support R2P think that intervening forces need to consider ways of protecting them from harm (and avoid doing harm themselves). Some suggestions have been made about how this can be achieved: by consulting local women about what should be done, for example, or including women in peace negotiations,

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

encouraging them to play a greater role in peacekeeping, and protecting them from rape in refugee camps. Women who have been attacked, as were the Muslim women in Bosnia, often need medical help and they need a refuge and opportunities to make a new life if they cannot go back to their communities. Interveners with a responsibility to protect, it could be argued, have an obligation to ensure that such needs are satisfied. Responsibility to Protect could also lead to more flexibility in fulfilling responsibilities for reconstruction and rebuilding communities. In many cases reconstruction that requires community involvement might be best done by supporting the work of NGOs or by encouraging the formation of civil society organizations. Ensuring that women play a central role in these organizations would be a way of making it more likely that their needs would be met. By stressing prevention and rebuilding, Responsibility to Protect could encourage a more imaginative and flexible approach to humanitarian intervention—one that focuses on preventing serious harm to populations and increasing their wellbeing rather than defeating or putting pressure on a political organisation that is already engaged in genocidal acts or is deemed a threat to security. By focusing on how military intervention should be conducted it could encourage the adoption of more effective means to protect and aid vulnerable groups. Both of these aspects of R2P, if they were widely adopted and put into practice, have the potential to provide better ways of countering crimes and oppressive practices that target women. The endorsement that R2P has received in international forums does not mean that it will be translated into the policy and practice of intervening powers. But feminists and others who care about the fate of the vulnerable have reason to support it.

Notes 1. Michael Walzer, Just and Unjust Wars: A Moral Argument with Historical Illustrations (Harmondsworth: Penguin, 1977), 107. 2. James Gerstenzang and Lisa Getter, ‘Laura Bush Addresses State of Afghan Women’, Los Angeles Times, 18 Nov. 2001, , accessed Oct. 2016. 3. According to Guglielmo Verdirame, ‘Testing the Effectiveness of International Norms: UN Humanitarian Assistance and Sexual Apartheid in Afghanistan’, Human Rights Quarterly, 23/3 (2001): 752: ‘It is quite likely that, were a conflict between different factions not occurring and were the Taliban not accused of offering sanctuary to terrorist organizations, the Security Council would have altogether abstained from intervening in Afghanistan, since the persecution of women is arguably not perceived as likely to endanger peace, let alone to constitute a threat of the peace or breach of the peace.’

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

WOMEN AND HUMANITARIAN INTERVENTION



4. Karen Engle, ‘ “Calling in the Troops”: The Uneasy Relationship among Women’s Rights, Human Rights, and Humanitarian Intervention’, Harvard Human Rights Journal, 20 (2007): 213. 5. This was demonstrated by the controversy caused by the Australian Commission of Human Rights, Bringing Them Home: Report of the National Inquiry in the Separation of Aboriginal and Torres Strait Islander Children from their Families (Canberra: Commonwealth of Australia, 1997), 234–5, when it pointed out that the removal of Aboriginal and Torres Strait Islander children from their families—a policy of Australian governments up to the middle of the twenieth century—could be described as ‘genocide’. 6. Reluctance to count any act as genocide unless it involves large-scale murder perhaps explains why only the international tribunal in Rwanda has so far recognized mass rape as a genocidal act. Hutu women who were raped were often killed, sexually mutilated, or enslaved. See Human Rights Watch, ‘Shattered Lives: Sexual Violence during the Rwandan Genocide and its Aftermath’, Sept. 1996, , accessed Oct. 2016. 7. See Sherrie L. Russell-Brown, ‘Rape as an Act of Genocide’, Berkeley Journal of International Law, 21/2 (2003): 350–74. 8. Catharine A. MacKinnon, ‘Women’s September 11th: Rethinking the International Law of Conflict’, Harvard International Law Journal, 47/1 (2006), 1–31. 9. See Richard W. Miller, ‘Respectable Oppressors, Hypocritical Liberators: Morality, Intervention and Reality’, in D. K. Chatterjee and D. E. Scheid (eds), Ethics and Foreign Intervention (Cambridge: Cambridge University Press, 2003), 215–50. 10. See Lisa M. Ayoub, ‘The Crisis in Afghanistan: When Will Gender Apartheid End?’, Tulsa Journal of Comparative and International Law, 7/2 (2000): 513–38. 11. I here understand humanitarian intervention, as do many philosophers and policy makers, as including economic and other kinds of sanctions. 12. Martha C. Nussbaum, Women and Human Development: The Capability Approach (New York: Cambridge University Press, 2000), Chapter 2, argues that adaptive preferences often explain why women accept subordination. 13. Gayatri Chakravorti Spivak, ‘Can the Subaltern Speak?’, in G. Nelson and L. Grossberg (eds), Marxism and the Interpretation of Cultures (Urbana, IL: University of Illinois Press), 280–316. 14. See John Stuart Mill, ‘A Few Words on Non-Intervention’, in Collected Works of John Stuart Mill: Essays on Equality, Law and Education, ed. John M. Robson and Stefan Collini (London: Routledge & Kegan Paul, 1984), 109–25, and the discussion of Mill’s view in Walzer, Just and Unjust Wars, 87–91. 15. Michael Walzer, ‘The Moral Standing of States: A Response to Four Critics’, Philosophy and Public Affairs, 9 (1980): 225–6. 16. Jean L. Cohen, ‘Rethinking Human Rights, Democracy and Sovereignty in the Age of Globalization’, Political Theory, 36 (2008): 587. 17. John Stuart Mill, ‘The Subjection of Women’, in Alice S. Rossi (ed.), Essays on Sex Equality (Chicago and London: University of Chicago Press, 1970), 136–7. 18. Responsibility to Protect was the product of the International Commission on Intervention and State Sovereignty sponsored by the Canadian government and cochaired by Gareth Evans. It was adopted by the World Summit in 2005.

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

19. Critics were particularly troubled by the belated humanitarian justification that the US and its allies gave for invading Iraq. See A. J. Bellamy, ‘Whither the Responsibility to Protect’, Ethics and International Affairs, 20/2 (2006): 143–69. 20. Gareth Evans, ‘From Humanitarian Intervention to Responsibility to Protect’, Wisconsin International Law Journal, 24 (2006): 703–22. 21. Bhikhu Parekh, ‘Rethinking Humanitarian Intervention’, International Political Science Review, 18 (1997), 64. 22. Parekh, ‘Rethinking Humanitarian Intervention’, 65.

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

4 Humanitarian Intervention and Non-Ideal Theory Ramon Das

The topic of humanitarian intervention has been on the philosophical radar screen for over half a century, but it is only in the last fifteen years or so that it has become a fixture in the literature on the ethics of war. This newfound and widespread philosophical interest would appear to stem from two events in particular. One, which had an immediate and lasting impact, was NATO’s 1999 armed intervention into Serbia/Kosovo—and the ‘Responsibility to Protect’ doctrine that emerged in the wake of that conflict. The second event, though it did not have a large impact on the literature at the time, has in retrospect exercised perhaps an even greater influence on all subsequent discussion of humanitarian intervention and the Responsibility to Protect. I refer to the case of Rwanda in 1994, which has been described as perhaps the clearest case of genocide since the Holocaust.¹ These two events are notable in several respects, but the one that concerns me in this chapter is how their treatment in the philosophical literature often manifests two ideal-theoretic assumptions that are absolutely central to the ethics of humanitarian intervention. Despite this central importance, these assumptions are not usually made explicit; in most philosophical discussions they are adopted tacitly or near-tacitly. The first assumption, reflected particularly in the literature on Rwanda, is that in target states where humanitarian intervention is being considered, there are two distinct and easily identified groups: ‘bad guys’ committing serious human rights abuses, and innocent civilians against whom the abuses are being committed. The second assumption, reflected particularly in the large literature on Kosovo, is that external to the target state in question there are suitably qualified ‘good guys’—prospective interveners who possess both the military power and the moral integrity to carry out a successful humanitarian

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

intervention.² A corollary of these two assumptions, it may be unnecessary to add, is that the good guys do not support the bad guys. The relevant prospective interveners either firmly oppose those carrying out the abuses (NATO, in Kosovo) or, at worst, are guilty of a failure to oppose them (the UN Security Council, in Rwanda). It may sound odd to call these assumptions ‘ideal-theoretic’, since in an ideal world there would be no need for humanitarian intervention at all. Nevertheless, the assumptions are ideal relative to possible worlds—including the actual one— in which severe human rights abuses do occur and would, ideally, be halted by a morally suitable intervener. They are ideal assumptions in that the prospects for a successful and truly humanitarian intervention are incomparably greater when they hold than when they do not. Indeed, whether or not the assumptions hold is generally decisive with respect to whether a given intervention is (likely to be) morally justified. This is easily seen when we imagine cases where they do not hold. Consider a civil war pitting several factions against one another, each of which is committing serious human rights abuses against different groups of civilians. Suppose, further, that the prospective intervener is itself implicated in comparably serious abuses, including (through its military support) the abuses committed by one or more of the factions in the civil war. In short, consider any case in which there are not one but many ‘bad guys’ and in which even the supposed ‘good guys’ are supporting some of the ‘bad guys’. Evidently, the conditions for a truly humanitarian intervention in such a case are very far from ideal. Its prospects for success may be close to nil, and it simply may not be possible to improve the humanitarian situation through the use of armed force. It is an interesting methodological question, virtually unexplored in the relevant literature, whether such decidedly non-ideal assumptions are any less relevant to philosophical discussions of humanitarian intervention than the two ideal assumptions that philosophers have tended to adopt. If anything, I believe, the non-ideal assumptions are more relevant. The reason, just suggested, is straightforward. The fact that we live in a messy, decidedly non-ideal world has significant implications for the ethics of humanitarian intervention. It should make us sceptical about the likelihood that such intervention will be justified in particular cases, and sceptical about the alleged benefits of relaxing international norms that govern its use. To be clear, I am not claiming that humanitarian intervention could never be justified, or even that there has never actually been a justified intervention. I am merely suggesting that non-ideal assumptions that approximate our actual world are far more likely to be relevant and useful to the ethics of humanitarian intervention than ideal-theoretic assumptions that bear little resemblance to reality.

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

NON - IDEAL THEORY



I have said that philosophical treatment of these cases often manifests tacit or near-tacit adoption of the two ideal assumptions mentioned. No philosopher, of course, believes that international affairs is populated by pure ‘good guys’ and ‘bad guys’. Nevertheless, philosophical discussion of such cases often proceeds as if these assumptions were accurate, or at the very least, as if they were (far) more accurate than their non-ideal counterparts.³ Some examples may be useful. David Luban, in the course of arguing that NATO’s Kosovo intervention was, in the end, morally justified, explicitly adopts the second assumption regarding the moral suitability of prospective interveners. He asserts, but does not argue, that ‘even if America’s interventions have often been purely self-interested, there is no reason to doubt that the Kosovo intervention (like the intervention in Somalia) was fundamentally a humanitarian effort’.⁴ More commonly, the assumption that there is a morally suitable prospective intervener is tacit. Terry Nardin, for instance, finds it ‘regrettable’ that NATO’s intervention in Kosovo lacked any sound legal basis, implying that its intervention was nonetheless morally justified. Like many others he also deplores the world’s failure to act to stop the Rwandan genocide, again implying that such an intervention would have been justified. However, he provides no support for the claim that NATO (in Kosovo) or the UN Security Council (in Rwanda) was a morally suitable intervener in those cases.⁵ Finally, Heather Widdows provides an example of tacitly making the first assumption, in the context of her claim that, ‘If ever humanitarian intervention is justified, it was justified in Darfur.’⁶ She provides plausible evidence regarding the scale of the humanitarian crisis in Darfur (c. 2005), but says nothing about the multi-dimensional character of that conflict, which bore many of the characteristics of a civil war, with serious abuses being committed on all sides.⁷ In my view, adopting such assumptions—whether explicitly or tacitly—is what makes such discussions of humanitarian intervention essentially ideal-theoretic. By contrast, to the extent that an author argues for or against one of these assumptions, her work counts (to that extent) as non-ideal-theoretic. The aim of this chapter is to present and partly argue for a non-ideal-theoretic approach to the ethics of humanitarian intervention. Given what has just been said, this does not involve starting from non-ideal assumptions rather than ideal ones (though as indicated, such an approach may well be more appropriate given the actual world). Rather, it involves empirically scrutinizing the two ideal assumptions, testing them against the available evidence. For instance, I examine some empirical considerations relevant to the designation of certain parties as ‘the bad guys’, which inevitably involves comparison to other parties to the conflict. I also look at the moral suitability of potential humanitarian interveners such as the United States and other Western powers—not only their motives and intentions

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

but also (and not unrelated) their realistic prospects for effecting humanitarian ends in particular cases. The idea is that if the two ideal assumptions look to be badly mistaken when measured against the available evidence, then it is highly unlikely that the relevant intervention is morally justified. In presenting and arguing for such a non-ideal theoretic approach, my strategy is basically to look closely at some real-world cases that have attracted the most philosophical attention, with an eye toward testing the accuracy of the ideal assumptions. As suggested, what we find when we look is that the designated ‘bad guys’ are not the only bad guys in the vicinity, and sometimes not even the worst. Furthermore, the ‘good guys’ (prospective interveners) are rarely if ever as good as they are typically assumed to be. Often, they have played a key role in contributing to the build-up of the humanitarian crisis through their military backing of one of the belligerent parties. In short, what we find when we look at the cases that have attracted the most attention is that the conditions for a successful humanitarian intervention are very far from ideal. Although I shall not argue for the claim here, I believe that a review of these cases supports a far more restrictive norm of permissible intervention than has been advocated in the literature.⁸ These claims are exemplified well in the case of Rwanda, to which I devote the most attention. They are also reflected in the case of Kosovo, which I consider more briefly and from a somewhat different angle. To be sure, there is not room here for a comprehensive discussion, and I don’t pretend to have reached settled conclusions about either of these cases. However, in highlighting empirical considerations that I believe to be central to the ethics of humanitarian intervention in these cases, my aim is not to resolve such questions definitively, but rather to present a distinctive theoretical approach to such questions, one in which normative conclusions—particularly insofar as they depend on the accuracy of the two ideal assumptions—are to a large extent ‘driven’ by relevant empirical considerations. I focus on Rwanda both because it is widely believed to represent the recent real-world case that comes closest to meeting the two ideal assumptions, and because it has received relatively little detailed discussion of this sort in the philosophical literature.⁹ I make no great claim to originality for the approach I present and defend here. There is certainly non-ideal-theoretic work in the literature, even if it is not usually identified as such.¹⁰ If there is anything new in this chapter, it is that I try to draw attention to the central ethical importance of the two ideal assumptions identified, and that I use them as the basis for distinguishing ideal-theoretic from non-ideal-theoretic approaches to the ethics of humanitarian intervention.

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

NON - IDEAL THEORY



The remainder of the chapter is structured as follows. In the following section I spell out my account of the distinction between ideal and non-ideal theory. I then examine the case of Rwanda, reviewing some recent studies that severely complicate the familiar claim that it was a prime candidate for humanitarian intervention. Next I look at the case of Kosovo, highlighting data that support a similarly sceptical stance. Although these data were identified at the time of NATO’s intervention, they have often been overlooked in subsequent discussion. However, in examining the case of Kosovo I devote the most attention to the much discussed ‘inconsistency’ or ‘selectivity’ objection to humanitarian intervention. After rejecting what I believe is an implausibly ideal-theoretic formulation of that objection, I consider an acceptably non-ideal-theoretic form of it. In the final section I draw some tentative conclusions and make some brief comparisons to the current crisis in Syria.

Ideal vs. Non-Ideal Theory I will be arguing for a non-ideal-theoretic approach to humanitarian intervention, and this requires at least a rough account of the distinction between ideal and non-ideal theory. According to the relatively straightforward account I favour, the distinction essentially concerns the extent to which a normative theory includes as a provisional starting point relevant empirical considerations about the social world and/or human nature. A normative theory that contains fairly extensive relevant empirical considerations is relatively non-ideal; a theory that includes fewer such considerations is relatively ideal. Whether a normative theory is ‘ideal’ or ‘non-ideal’ is thus comparative and a matter of degree. The word ‘relevant’ indicates that this empirical (descriptive/explanatory) component of a normative theory can be expected to vary according to context and normative purpose. A non-ideal theory about healthcare reform in the United States can be expected to focus on empirical considerations rather different from those that concern a non-ideal theory about fishery protection in New Zealand. As I see it, deciding which considerations are relevant to the case at hand is largely a matter of (reasoned) judgement, and it is probably not possible to make very general recommendations about which should be included and which left out.¹¹ Nevertheless, for any given case it is the provisional inclusion of relevant empirical considerations that determines whether a theory counts as relatively ideal or non-ideal. Two clarifications are in order. First, although non-ideal theory places greater emphasis on relevant empirical considerations, the central, defining claims of non-ideal theory are normative in just the way that the central, defining, claims of

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

ideal theory are.¹² An ideal theory of justice might be guided by the question: ‘What would a perfectly just society look like?’ A non-ideal theory of justice might be guided by the question: ‘What would a considerably more just society (than the one we currently inhabit) look like?’¹³ The fact that the second question includes empirical considerations about current society does not make it any ‘less normative’ than the first question, which needn’t include such considerations. Second, in saying that non-ideal theory includes relevant empirical considerations about the social world and/or human nature as a provisional starting point, the word provisional is crucial. Non-ideal theories are perfectly able to critically evaluate these starting points. As long as they respect the principle that ‘ought implies can’, non-ideal theories are not committed to taking these starting points as fixed constraints on their central normative claims.¹⁴ This is not to say, of course, that all such non-ideal theories do so evaluate. There are better and worse non-ideal theories, which do a better and worse job of criticizing (or accepting as legitimate) various aspects of current political, social, or human reality. The point is that it is not essential to non-ideal theory that it accepts the current state of the world as legitimate. What characterizes such theories as nonideal is the extent to which they try to provide (or explicitly presuppose) a descriptive or explanatory account of relevant empirical considerations, not the extent to which they accept such considerations as fixed points, immune to moral criticism. This approach to the ideal/non-ideal distinction is less fine-grained than others in the literature, and it may be useful to situate it in relation to a recent survey of work on the topic. Laura Valentini has distinguished three different ways in which the ideal/non-ideal distinction may be understood, all of which have their roots in John Rawls’s A Theory of Justice. The first is the distinction between full and partial compliance theory, where the focus is on the extent to which agents can be expected to comply with the principles of justice. The second is the distinction between utopian and realistic theory, where the focus is on the extent to which normative principles are constrained by empirical facts. The third is the distinction between ‘end-state’ and ‘transitional’ theory, where the focus is on the extent to which a theory sets out a long-term goal for institutional reform, as opposed to addressing the question of how (usually in gradual steps) that goal is to be achieved.¹⁵ The present approach to the ideal/non-ideal distinction effectively encompasses the first two of these distinctions, though it is perhaps closest to the ‘utopian vs. realistic’ reading. One way to understand the critique of the previous section is that the literature on humanitarian intervention often manifests assumptions that are too unrealistic or utopian. However, even for a ‘realistic’

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

NON - IDEAL THEORY



theory the idea that normative principles are constrained by empirical facts must be qualified. As noted, although non-ideal theories are characterized by the fact that they take into account relevant empirical considerations, they nevertheless are perfectly able to critically evaluate those considerations (consistent with respecting ‘ought implies can’). Again, non-ideal theory is no ‘less normative’ than ideal theory. This point also bears on Valentini’s third reading of the ideal/non-ideal distinction, that between ‘transitional’ and ‘end-state’ theories. Like Amartya Sen, I reject the idea that we need to know what a perfectly just world looks like in order to know what would make our existing world more just.¹⁶ Indeed, this probably understates the case. As I see it, we have very little idea about what a perfectly just world would look like, yet it is nevertheless the case that we have some quite good ideas about how to make our existing world more just. However, if we have very little idea about what a fully just ‘end-state’ would look like, we probably needn’t worry too much about which of the various ‘more just’ alternatives will put us on the correct path or transition to the fully just end-state.¹⁷ Finally, regarding the issue of compliance, this is best understood as one (important) relevant empirical consideration among many that must be taken into account in non-ideal theory. In other words, the issue of compliance should not be treated separately from other relevant empirical considerations. This is not to downgrade its significance. With respect to theorizing about humanitarian intervention, for instance, we have already seen that a major question is whether prospective interveners can be expected, broadly speaking, to follow the rules. It is worth emphasizing that the present understanding of the ideal/non-ideal distinction is a non-moralized one. There is nothing intrinsically ‘better’ about non-ideal (or ideal) theory, and there is little point to the question whether normative political theory should be ‘more ideal’ or ‘more non-ideal’ simpliciter. As suggested, this is a question that can be assessed only with reference to a particular purpose or case. Nevertheless, for theories that purport to be relevant to real-world concerns—for theories that focus centrally on real-world cases and draw conclusions evidently intended to have real-world implications—there are fairly obvious reasons why such theories should strive to be relatively non-ideal. Do theories of humanitarian intervention qualify as such theories? It is pretty clear that many if not most of them do: most philosophers writing on humanitarian intervention have been interested in assessing it in the real world. We are interested not only whether humanitarian intervention should be pursued in particular cases, but also whether relevant norms should be enshrined in international law. To this extent, it would seem uncontroversial that philosophical

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

assessment should be based on an accurate picture of the real world.¹⁸ I now turn to a case where recent evidence suggests that our understanding of this picture stands in need of revision.

Rwanda The Rwandan genocide of 1994 is generally taken to be a case where humanitarian intervention clearly would have been justified, if only the international community had mustered the will to act. It bears emphasis that the international community is usually thought to have been guilty of a culpable omission, rather than having played any active role in contributing to the genocide. As far as I can tell, philosophical discussion has not challenged this assessment and typically proceeds as if it were true.¹⁹ In this section I sketch the standard account of what happened in Rwanda in 1994, picking out the elements of the picture that conform to the two ideal-theoretic assumptions already discussed. I then present some recent empirical findings that cast doubt on this standard account, particularly when viewed in the proper context, namely, that Rwanda was in the midst of what was effectively a foreign invasion funded and supported by Uganda and ultimately its superpower patron, the United States. Viewed in this context, the recent findings undermine the accompanying standard view of what the international response to the crisis should have been. Needless to say, there is not space to make this case definitively here. My aim, rather, is to outline a distinctively non-ideal-theoretic approach to the question of humanitarian intervention in Rwanda, focusing on relevant empirical considerations that in my view have been mostly overlooked. According to the standard and familiar narrative, the ethnic violence that engulfed Rwanda during the middle of 1994 constitutes a relatively clear case of genocide: a widespread, ethnically motivated mass slaughter that was partly planned beforehand by high-ranking members of the ethnic majority Rwandan Hutu, targeting members of the minority (but historically dominant) Rwandan Tutsi. In particular, following the assassination of President Habyarimana (a Hutu) on 6 April, extremist Hutu groups launched widespread attacks that killed up to one million Tutsis, as well as some moderate Hutus, in the space of about 100 days. Much of the killing was carried out with low-tech weapons such as machetes. During this time, the rebel Tutsi army, the Rwandan Patriotic Front (RPF)—which had been fighting a civil war since October 1990—gradually drove the Hutu government forces from power by August 1994. Despite its military victory, however, the RPF was unable to halt the Tutsi slaughter and inherited a country whose Tutsi population had been decimated.²⁰

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

NON - IDEAL THEORY



The corresponding standard narrative about the global response (or lack thereof) is also familiar. The UN Security Council, though aware of the situation in Rwanda and easily able to stop the low-tech killings, simply failed to act. In particular, US President Bill Clinton, wary of repeating the previous year’s (1993) humanitarian debacle in Somalia, failed to muster the political will necessary to send a UN peacekeeping force that could have halted the killings, an omission he later singled out as the greatest regret of his presidency. Expressing the consensus view in a recent article, Larry May writes, ‘Indeed, omissions by the international community are now well-recognized as crucially allowing for the genocide to gain steam in Rwanda.’²¹ Several of the moral claims implied by this narrative are either constituted by or crucially rely upon empirical claims, prominently including the claim of genocide and the claim that the international community’s main failing was one of omission not commission. The accuracy of these latter claims bears directly on the relevance of the two ideal assumptions already discussed, namely, that there were clearly identifiable ‘bad guys’ in Rwanda (who were committing gross abuses against innocent civilians), and that there were suitably qualified ‘good guys’ outside the country who opposed the bad guys and could have stopped them. Hence their accuracy bears on the question whether Rwanda constitutes a relatively clear case where humanitarian intervention would have been justified, perhaps obligatory.²² We shall see that the claim of genocide, even if correct, is far from the whole story in Rwanda. Filling in the details undercuts the familiar claim that the international community was guilty (merely) of an omission and that it should have responded to the genocidal violence with an armed ‘humanitarian intervention’. Rather, there is a good case to be made that what the international community, and above all the United States, should have done is to stop arming and supporting the Rwandan Patriotic Front, which it had been doing (through Uganda) from at least the late 1980s. The RPF’s 1990 invasion and terror against the largely Hutu population was a major contributing factor to Hutu–Tutsi ethnic tension and hatred. And the RPF’s likely role in the assassination of President Habryarimana, and its subsequent military advances throughout Rwanda, was arguably the chief factor behind the subsequent genocidal violence. At any rate, it was the factor most within the international community’s ability to control, even if the RPF was not the main perpetrator of that subsequent violence. I shall work my way toward this tentative conclusion by considering the most pertinent empirical claims in the standard narrative about Rwanda. These would appear to be as follows. First, the vast majority of persons killed during the ‘100 days of genocide’ were Tutsis. Second, the vast majority of killers were

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

Hutus. Third, the mass killings of Tutsis by Hutus were not spontaneous killings or acts of violence that occurred in the context of war, but rather were the carrying out of a prior Hutu plan, borne of ethnic hatred, to exterminate Tutsis. Fourth and related, the triggering event for genocide was the assassination of President Habyarimana by extremist Hutus. Fifth, there was a suitably qualified agent outside Rwanda—one with the requisite military ability and moral integrity—to stop the genocidal killing, saving tens (perhaps hundreds) of thousands of innocent lives. In combination, the first four considerations lend support to a claim of genocide; adding the fifth lends support to a claim of justified humanitarian intervention. I shall argue that the current evidence suggests that all five claims are deeply problematic. Setting aside for now the final claim, of the first four claims only the second is more likely to be true than false. Let us consider them in turn. The first, that most of the victims of the Rwandan genocide were Tutsis, has been called into question on several grounds by the authors of arguably the most comprehensive research on the killings, Christian Davenport and Allen Stam.²³ They tentatively suggest that more than half the persons killed in Rwanda during 1994 were Hutus. Their suggestion is based on a comparison of census data from 1991, which listed the Tutsi population at 600,000, a Tutsi survival figure of 300,000, and a total number killed at between 800,000 and 1 million.²⁴ Davenport and Stam’s tentativeness on this question stems from the difficulty of being at all confident about the number of Hutus versus Tutsis killed. First, there are very few ethnic or physiognomic features that distinguish most Hutus from most Tutsis.²⁵ Within Rwanda at that time, an individual’s ethnic identity would be known mainly by local knowledge. However, RPF military advances and outright terror from 1990 to 1993, along with anti-Tutsi government propaganda, had caused about 1.3 million (mostly) Hutu Rwandans to flee their homes, hence from areas where they could be easily identified. To this huge number of internally displaced persons must be added several hundred thousand Hutus who had fled from neighbouring Burundi, following the assassination of its Hutu president by Tutsi soldiers six months earlier, in September 1993. In short, at the start of 1994 some 2 million Hutus and a smaller number of Tutsis—perhaps a seventh of the population of Rwanda—were displaced, living in areas where their ethnic identity could not be easily ascertained. It is likely that many of these persons were killers and also that many of them were killed, points to which we return shortly. What of the second assumption, that most of the killers were Hutus? As just noted, there is little doubt that many of them were. Arguably the dominant strand of the mainstream narrative, amply documented, is that Hutu groups killed tens and perhaps hundreds of thousands of innocent persons, Hutus as well as

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

NON - IDEAL THEORY



Tutsis.²⁶ This is not, however, the whole story of such large-scale killing. An October 1994 UNHCR briefing, known as the Gersony Report, claims that from April to July 1994 the RPF was systematically killing perhaps 10,000 Hutu civilians per month, just in those areas of southern Rwanda newly under its control (that is, the figure does not include the number of Hutu civilians the RPF undoubtedly killed elsewhere in Rwanda during this period). According to the report, the manner of killing is strikingly similar to the manner in which Hutus are alleged to have killed Tutsis, for instance, with hoes or machetes.²⁷ Nor can these be dismissed merely as ‘reprisal’ killings by the RPF. According to a US State Department memo discussing the report, ‘the UNHCR team speculated that the purpose of the killing was a campaign of ethnic cleansing intended to clear certain areas in the south of Rwanda for Tutsi habitation’. This rationale would be consistent with RPF terror directed at Hutus in northern Rwanda over the previous few years (1990–3), previously discussed. In short, even if it is true that most of the killers during the 100 days of genocide were Hutus, the RPF also engaged in mass killing, and its motivation for doing so was hardly better, morally speaking, than what Hutus are accused of in the standard narrative. Let us now consider the third claim, which is that the killings were not spontaneous acts or ordinary killing in the context of war, but rather preplanned for the purpose of exterminating Tutsis for reasons of ethnic hatred. The familiar claim is that there were plans on the part of some high-ranking Hutus to eliminate Tutsis, and that these plans guided bands of killers, including the notorious Interahamwe. As it turns out, there are reasons to doubt that there was ever any ‘master plan’ to eliminate Tutsis, and the main suspects, including the alleged ringleader, Colonel Bagasora, were acquitted of the charge of genocide at the ICTR trial. Much more important from the perspective of humanitarian intervention, there is strong evidence that the vast majority of Hutu killing of Tutsis does not fit this ‘pre-planned’ mould. In what Davenport and Stam identify as the ‘most shocking’ result of their research, they found that mass killings in territories nominally under FAR (Government of Rwanda) control occurred largely in response to RPF military advances: ‘When the RPF advanced, large-scale killings escalated. When the RPF stopped, large-scale killings largely decreased.’²⁸ As far as I can tell, this claim has not been challenged anywhere in the literature on Rwanda. Assuming it is true, what could explain this pattern, and what is its moral significance? Even if most of the killing was carried out by Hutu groups affiliated with the FAR, the pattern of killing identified by Davenport and Stam—one in which mass killing increases with RPF military advances—is in tension with, and arguably inconsistent with, the claim that the killing was

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

carried out primarily with genocidal intent. At the very least, the two scenarios have very different moral implications, including implications for prospective humanitarian interveners. Killing a person because she belongs to an ethnic group that you are motivated to destroy out of ethnic hatred is very different, morally speaking, from killing her because she belongs to an ethnic group that you are motivated to destroy out of a reasonable fear that its leaders plans to subjugate your own people. A good case can be made that the latter description is closer to the truth in Rwanda: if Hutu groups were hell-bent on genocide, why wait until the RPF advanced on them to launch their killings? The more likely explanation would seem to be that most of the killing was a frenzied and terrified response—by Hutus already driven from their homes by RPF terror—to renewed RPF military advances. Again, it bears emphasizing that over a million, mostly Hutu, civilians had fled their homes in areas of northern Rwanda under RPF control from 1990 to 1993.²⁹ Although the exact cause of this flight is debatable, the most likely explanation involves RPF terror and threats against the civilian population, of which there are credible reports. People do not generally flee their homes en masse without good reasons to fear for their safety. And, having fled from their homes in the north, it is not hard to imagine that these displaced Hutu masses would strike out in terror in response to RPF military advances. This suggestion finds some support in the work of Alan Kuperman, who argues that the RPF persisted in its military campaign to seize control in Rwanda fully expecting that its invasion would trigger a violent backlash against Tutsi civilians in Rwanda.³⁰ The fourth claim is that extremist Hutus were behind the assassination of President Habyarimana, ostensibly because he was cooperating too closely with the RPF and Tutsis generally. This claim, it appears, is the least likely of the five to be true. A considerable amount of evidence suggests that the RPF assassinated Habyarimana and that this fact was covered up with the key assistance of the United States. In particular, an inquiry into the assassination was carried out by Australian investigator Michael Hourigan, under the auspices of the ICTR, which at the time considered its mandate to include events during any part of 1994. After nearly a year of fruitless attempts to tie the assassination to Hutu groups, Hourigan received credible leads that the RPF was behind it. After initial words of encouragement from ICTR Chief Prosecutor Louise Arbour, she changed her tune abruptly one week later, telling him, somewhat incredibly, that the assassination did not fall within the ICTR mandate. Hourigan resigned soon thereafter, speculating that a high-ranking US official, in his view probably Secretary of State Madeleine Albright, informed Arbour that there was to be no investigation of any RPF role in the assassination.³¹ The assassination, it may be

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

NON - IDEAL THEORY



worth repeating, is widely agreed to be the triggering event behind the genocidal violence that followed. This leads us to the fifth and final empirical claim in the standard narrative about Rwanda, which is that there was a suitably qualified prospective intervener outside the country, one that was capable of stopping the slaughter. The usual candidate, of course, is the United Nations Security Council. However, the case of Rwanda illustrates a common and much discussed problem with that body as a potential intervener, namely, that at least one of its permanent members can usually be counted upon to have a vested interest in the country targeted for intervention. In the case of Rwanda, at least three members had such an interest: France, Britain, and the United States. Much has been written about the role of France,³² but I focus here on the lone superpower, which provided crucial military, logistical, and diplomatic support for the Rwandan Patriotic Front beginning with its invasion from Uganda (another close US ally) in October 1990 and continuing throughout 1994 and beyond. Again, space prevents any detailed discussion, but a few points are worth noting. First, the US, through its Ugandan ally, bore an important measure of responsibility for the transformation of the RPF from a poorly organized force of about 4,000 in October 1990, into the formidable army of perhaps 50,000 that was able to take the entire country in less than three months in 1994.³³ Second, during this period, the US not only turned a blind eye to probable RPF terror against the domestic, largely Hutu population, causing them to flee en masse from areas of the country under RPF control, it also ensured that a harsh programme of ‘structural adjustment’ was imposed by the IMF and World Bank, effectively gutting Rwanda’s social services and further impoverishing its population while it weakened the Rwandan government.³⁴ Third, as already noted, there is strong circumstantial evidence that the US was instrumental in ensuring that the assassination of President Habaryimana was never properly investigated; there have also been suggestions (including by former UN Secretary General BoutrosGhali) that the CIA was involved.³⁵ In summary, the culpability of the US role in Rwanda, up to and including in April 1994, cannot sensibly be described merely as a culpable omission. To the contrary, the US appears to have played a quite active role in numerous and decisive ways. Even its actions in April 1994 cannot properly be described as a mere failure to act. President Clinton did not fail to authorize peacekeeping troops after 6 April; rather, he insisted—over the objections of the Rwandan ambassador to the UN—that most of the UNAMIR troops on the ground be withdrawn.³⁶ Throughout the crisis, the US systematically blocked any military intervention in Rwanda, even without US participation.³⁷ In short, it is far from clear that the UN Security Council was an appropriately

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

qualified prospective intervener in Rwanda in 1994, given the vested interests of its most powerful member, the United States. A brief summary of the argument presented in this section is in order. The political violence that engulfed Rwanda during 1994 was far more complicated than the simple genocide narrative suggests. Without denying that hundreds of thousands of Tutsi were killed at the hands of Hutu groups, there is good reason to believe that much of this killing was motivated not by ethnic hatred but was rather a frenzied response to RPF military advances³⁸ and was motivated more generally by a fear of Tutsi domination that was not at all unfounded. Before, during, and after this period, the RPF received crucial military, financial, and diplomatic support from the United States, and it apparently persisted in its military campaign in the early 1990s fully expecting that this would provoke a violent backlash against Tutsis. It is important to record that there may well have been more Hutus killed in Rwanda during the 100 days of genocide than Tutsis. There is also compelling evidence that the RPF was responsible for a great deal of this killing itself and was behind the assassination of Hutu President Habyarimana, the event that all sides agree triggered the subsequent genocidal violence. Finally, potential interveners, France and Britain, but above all the United States, had far closer ties to the fighting parties than has generally been recognized in the philosophical literature. These ties rendered the UN Security Council a dubiously appropriate ‘humanitarian’ intervener, and its actions strongly suggest that its dominant member had a clear interest in seeing an RPF military victory. This suggestion receives further support from its actions over the past twenty years, in which it has continued to lend support to Kagame’s regime, despite its central role in gross abuses in the Democratic Republic of the Congo, a country in which killings exceed those in Rwanda by perhaps a factor of ten.³⁹ In the end, the question: ‘Should the international community have launched a humanitarian intervention in Rwanda in April 1994?’ is clearly not the only, and perhaps not even the main, question for political philosophers interested in that part of the world to be asking.

Kosovo The literature on the ethics of NATO’s humanitarian intervention in Kosovo in 1999 is considerably more extensive than the comparable literature on Rwanda. Nevertheless, there is nearly as much ethical agreement about the former case as there is about the latter. The consensus view on Kosovo is that NATO’s decision to intervene in March 1999 was justified, but that its exclusive use of high-altitude bombing, in an effort to minimize NATO casualties (there were no NATO

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

NON - IDEAL THEORY



combat deaths) but at the expense of civilian life, was not. It is easy to agree with the latter point, but my interest here is with the former claim, that the decision to intervene itself was justified. Again, I make no attempt at giving a conclusive argument. Rather, I focus on central empirical claims that bear on the two ideal assumptions above, hence on the legitimacy of NATO’s intervention. Specifically, I consider three relevant empirical claims, focusing on the third. First, NATO’s insistence at the Rambouillet Conference of March 1999 that 30,000 of its troops be provided full and unfettered access throughout FRY territory, as well as full immunity to FRY law, was contrary to the spirit of good-faith negotiation; indeed, it was tantamount to a demand for unconditional surrender.⁴⁰ Second, it was reasonable to expect—and apparently was expected by NATO’s Commander Wesley Clark—that NATO’s bombing of Kosovo/Serbia (which also led to the withdrawal of the OSCE verification monitors on the ground) would lead to the escalation of FRY abuses, particularly mass expulsion of ethnic Albanians from Kosovo.⁴¹ Third, the human rights abuses committed by the Milosevic regime up until March 1999, though real and surely condemnable, were quite comparable to abuses occurring elsewhere in the world at the time, including Colombia; and they paled in comparison to those committed in recent years (mid-1990s) by NATO member Turkey.⁴² The first two of these claims, if true, provide relatively straightforward bases for objecting to NATO’s intervention. The first claim seems blatantly to violate the ‘last resort’ criterion for a just war. If NATO was sincerely interested in a nonviolent solution to the conflict, why insist on such unreasonable terms? The second claim arguably violates the ‘reasonable chance of success’ criterion, and in any case suggests that NATO acted in ways that could be expected to make the situation in Kosovo worse, not better. Granted, it could be argued that the claim fits into the common criticism that NATO’s intervention (merely) violated jus in bello criteria with its high-altitude bombing, rather than showing that its intervention wasn’t justified to begin with. Though I can’t argue the point here, I would suggest that, when considered with the rest of NATO’s actions and omissions, it is better seen as part of a jus ad bellum violation. With a few exceptions,⁴³ neither of these first two claims has been widely discussed as potential objections to NATO’s Kosovo intervention in the philosophical literature on humanitarian intervention. My main concern in this section, however, is with the third claim—that there were comparable abuses being committed elsewhere in the world that NATO could have halted much more easily than the abuses in Kosovo. This has been interpreted as a version of the ‘consistency’ or ‘selectivity’ objection to humanitarian intervention.⁴⁴ After considering the claim in a bit more detail, I consider and reject an

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

implausibly ideal-theoretic form of that objection and some responses to it. I then offer a more plausible, non-ideal-theoretic version and suggest that it indeed constitutes a potent objection to NATO’s intervention in Kosovo. Consider, then, the claim that FRY/Serbian abuses against Kosovar Albanians, though real enough, were quite comparable to human rights abuses being committed elsewhere in the world that NATO could have halted more easily. A bit of context is useful. The Kosovar struggle for independence from FRY, which had been mostly non-violent until 1995, had become significantly more violent in early 1998, with the Kosovo Liberation Army (KLA) not only engaging FRY army and police, but also attacking Serbian civilians identified with the regime, such as mail carriers. It is worth noting that as late as February 1998 the KLA was regarded by the US as ‘without any question a terrorist group’.⁴⁵ By the middle of 1998 it controlled 40 per cent of the territory of Kosovo. Faced with an increasingly successful armed insurrection, Milosevic responded with a major military offensive, in the process committing significant human rights abuses. By the end of the year, about 2,000 people had been killed in Kosovo and several hundred thousand had become internal refugees.⁴⁶ Such numbers are of course bad enough. However, they are no worse than comparable (1998) figures for Colombia and Turkey, and in some important respects they are not as bad.⁴⁷ For instance, the casualty and refugee toll in Kosovo were essentially new in 1998, but had been occurring at the same level for several years in Colombia. In Turkey the toll had been far worse in previous years, with the government conducting a scorched earth campaign against a Kurdish insurgency, one that destroyed over 2,000 villages, killed nearly 20,000 and produced 2 million internally displaced persons.⁴⁸ Most significantly, both of these countries were close allies of NATO’s dominant member, the United States. Given this fact, it seems that the US could have halted these comparable or worse abuses much more easily, arguably non-violently.⁴⁹ I now wish to consider and reject two common ways of interpreting this objection. The first is that the US-led intervention in Kosovo was unjustified simply because it exemplified an inconsistent pattern; in particular, the US/NATO could have (but chose not to) halt comparable human rights abuses in Turkey and Colombia. This claim is plainly a non-starter. It invites Bill Clinton’s easy response: ‘We cannot respond to such tragedies everywhere, but when ethnic conflict turns into ethnic cleansing where we can make a difference, we must try, and that was clearly the case in Kosovo.’⁵⁰ The second and more common way of interpreting the claim, which moves beyond the mere charge of inconsistency, is as follows. Although the US could have intervened to halt comparable abuses elsewhere (Turkey and Colombia),

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

NON - IDEAL THEORY



it chose to intervene in Kosovo ultimately because of self-interest (perhaps it felt it needed to preserve the credibility of NATO). But ‘humanitarian’ interventions may not be undertaken for self-interested reasons, so NATO’s intervention in Kosovo was unjustified. Although this version of the objection is more persuasive, there is again a plausible response to it, which is that it is not impermissible to intervene for reasons of self (or national) interest, so long as such reasons function merely as a ‘tiebreaker’.⁵¹ And that seems reasonable. If a prospective intervener can halt only one of several concurrent and equally bad humanitarian crises, why should it matter, morally speaking, if it uses its own interests as a ‘tiebreaker’ in choosing where to intervene? It is important to see, however, that the force of this ‘tiebreaker’ response assumes that the cases in which humanitarian intervention is being considered are relevantly similar. In particular, it assumes not only that the severity of human rights abuses occurring in the various prospective target states are roughly comparable, but also (and arguably more important) that the intervener’s prospects for bringing about a successful humanitarian intervention are roughly the same in each case where intervention is being considered. It is only under these assumptions that an appeal to tiebreakers makes moral sense; otherwise it will be too easy to classify as humanitarian interventions that are predominantly selfinterested. Yet these are (relatively) ideal assumptions that often fail to hold in the real world. In the case at hand it is the latter assumption that apparently fails to hold, at least if the third empirical claim is correct. For the point is precisely that, if NATO wished to halt serious human rights abuses, it could have done so much more easily in Turkey or Colombia than in Kosovo. The former were close allies of the US, highly dependent on it for weapons and diplomatic cover; hence its prospects for a successful humanitarian intervention were much higher in the former two countries than in the latter. Now consider a harder line against the consistency/selectivity objection, namely, that interventions must be judged on their own merits, without regard to comparison cases. For instance, James Pattison argues that ‘when considering whether an intervener is legitimate in a particular case, we should concentrate on the details of that case and, crucially, on whether it will be effective . . . [W]hat the selectivity objection overlooks is this: what an intervener does or does not do in one state should not change the judgement of the legitimacy of its intervention in another, unless it will actually affect this intervention.’⁵² Applying this reasoning to the present case yields the claim that, even if NATO should have intervened in Turkey or Colombia, this should not change our judgement about the legitimacy of its intervention in Kosovo. What matters crucially to the latter question is whether NATO’s intervention was likely to be effective.

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

This response to the consistency/selectivity objection has an initial ring of plausibility to it. However, it is not hard to find an ethical basis for the claim that facts about a prospective intervention in B are potentially relevant to questions about whether one should intervene in A. According to standard formulations of act consequentialism, for instance, a person ought to perform that act, of those available to her, that promotes the greatest good. To do this, she must of course consider the various alternatives available to her. Moreover, it is easy to see the ethical relevance of these alternatives when we consider a case of ‘humanitarian intervention’ on an individual level: James, an able-bodied man, is taking his daily walk in the park. Coming around a bend, he sees two separate altercations taking place. In one, quite near to him, two ten-year-old boys are delivering a beating to a much smaller boy of perhaps 5. The second altercation is taking place across a wide river that James could cross with some difficulty; it involves two boys in their mid-teens beating a boy of perhaps 10. Now consider two questions. First, what should James do? Second, should James cross the river to intervene on behalf of the ten-year-old? In response to the first question, it seems pretty clear that the first thing James should do is to help the five-year-old, which he can accomplish much more easily than helping the ten-year-old across the river. It may be that after doing this he should then try to cross the river to help the second boy. However, as we have seen there are good consequentialist reasons for James to focus his energies where they are likely to do the most good. It would be ethically unmotivated to insist—absent any further information—that the altercation across the river deserves ethical priority and must be addressed ‘on its own merits’, independently of considerations regarding what James should do about the altercation right in front of him. To strengthen the point, alter the case so that not only is James less likely to promote the good by trying to help the boy across the river, he is more likely to promote his self-interest by doing so. (Perhaps James has an arrangement with the boys beating the five-year-old closest to him such that, in exchange for his turning a blind eye toward their transgressions, they will give him a portion of whatever they are able to steal from their younger victims.) In this case, there is even less ethical reason to treat the question ‘Should James help the boy across the river?’ independently of his alternatives. Were James nevertheless only to try to help the boy across the river, and then try to justify his action by claiming, ‘At least I was trying to do some good’, we would not be very impressed. The parallel to the case of Kosovo should be clear. In considering whether NATO’s intervention there was justified, there is no ethical reason to exclude any

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

NON - IDEAL THEORY



consideration related to what it should have done elsewhere, in particular, in Turkey and Colombia. Of course, there are obvious reasons of realpolitik why these other cases were excluded in most ethical and other commentary on NATO’s intervention in Kosovo. In the ‘real world’ of the late 1990s, NATO sanctioning Colombia, much less Turkey, was extremely unlikely. However, as was argued earlier in this chapter, a non-ideal-theoretic approach to humanitarian intervention, though it may and perhaps should take this aspect of realpolitik into account, is by no means committed to treating it as a fixed point, that is, as immune to ethical criticism. As we have seen, there are good reasons to criticize it. In short, a charitable version of the consistency/selectivity objection to humanitarian intervention must be suitably non-ideal-theoretic. Its force should depend on the relevant details of the case in question, in particular, on the relative prospects for promoting humanitarian ends versus self-interest in different candidate states. Stated abstractly, the objection may be formulated as follows: If there are several candidate states where humanitarian intervention would be appropriate, and a prospective intervener’s prospects for humanitarian success and promotion of self-interest vary considerably amongst these candidate states, then it would be objectionably inconsistent or selective to target for intervention a state in which the prospects for humanitarian success are low but promotion of self-interest is high. Given the third empirical claim, this non-ideal-theoretic version of the consistency/selectivity objection does seem to pose a serious problem for NATO’s intervention in Kosovo. Assuming that Kosovo, Turkey, and Colombia all qualify as ‘candidate states where humanitarian intervention would be appropriate’, there appears to be a good case that NATO selected for intervention a state (Kosovo) in which the prospects for humanitarian success were relatively low but promotion of self-interest relatively high. At a minimum, the comparison suggests that humanitarianism was a very minor concern of NATO in Kosovo and that its motives were largely selfinterested—a suggestion that is reinforced if indeed NATO did not negotiate in good faith at Rambouillet and rushed into a bombing campaign that reasonably could have been expected to make things worse. Of course, the claim about NATO’s relative prospects for a successful intervention, along with the first two empirical claims discussed in this section, could be wrong. My aim, again, has not been to assess the ultimate plausibility of those claims, but rather to argue that their truth is highly relevant to the question whether NATO’s intervention in Kosovo was justified.

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

Summarizing this section, I have considered three empirical claims that appear to be centrally relevant to the question whether NATO’s intervention in Kosovo was justified. I have focused on the third of these claims and its relation to the ‘consistency’ or ‘selectivity’ objection to humanitarian intervention. I have argued that the objection needs to be considered in an appropriately non-ideal-theoretic form, and that when this is done it indeed poses a serious objection to NATO’s intervention in Kosovo.

Conclusion In this chapter I have focused on two ideal-theoretic assumptions that are central to the ethics of humanitarian intervention and that are often tacitly or neartacitly adopted in philosophical discussions on the topic. I have argued for a nonideal-theoretic approach to humanitarian intervention that tries to assess the accuracy of these assumptions, focusing on the cases of Rwanda and Kosovo. Although limitations of space have precluded reaching any settled conclusions, I have suggested that there are plausible reasons to doubt that the two ideal assumptions hold in either of these familiar cases, with important implications for the question whether intervention was or would have been justified in either case. In concluding, I would like briefly to address the ongoing conflict in Syria, where I believe there is far greater recognition—again, if only tacitly—that neither ideal assumption holds. Specifically, it is widely recognized that there is a bona fide civil war in Syria, in which there is not one but many ‘bad guys’— even if the Assad-led government is responsible for the greatest share of severe human rights abuses. It is also recognized, though perhaps not as widely, that world and regional powers are intimately involved in the Syrian conflict, with Russia and Iran providing arms to the Syrian government, and the United States, Britain, Saudi Arabia, and Turkey backing the Syrian rebel factions. Given that both sides have committed gross abuses against civilians, none of these states (or others closely allied to them) is in a morally suitable position to launch an armed humanitarian intervention. In short, Syria appears to constitute another case in which it is not clear that there is any military solution to the humanitarian crisis. This is not to say, it should be unnecessary to add, that there is nothing the world can do to improve the situation in that country.⁵³ The fact that neither of the ideal assumptions holds in Syria is fairly clear from detailed analyses of the current situation,⁵⁴ and can be gleaned even from shorter news reports. For instance, President Obama in 2014 called for $500 million in new military assistance to the Syrian rebels, a significant increase on what he was

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

NON - IDEAL THEORY



expected to request from the US Congress. However, in clear recognition of the fact that many within the Syrian rebel forces are (or are closely aligned with) radical Islamist groups that have committed serious human rights abuses, Obama’s request includes the proviso that prospective recipients of the funds are to be ‘vetted’ in order to ensure that such groups are excluded.⁵⁵ Finally, given that it is widely (if tacitly) recognized that neither of the ideal assumptions holds in Syria, it is in my view no coincidence that there are at present very few calls for humanitarian intervention into that country, even amongst those who have long defended the doctrine.⁵⁶ Whether or not this is the correct diagnosis, the present consensus seems to be that military intervention into Syria is unlikely to do much good and likely to do harm.

Notes 1. Mamdani, M., 2007, ‘The Politics of Naming’, London Review of Books, 8 Mar. 2007, 5. Available at: . 2. There are of course other ideal-theoretic assumptions that are typically in play. One, on which Luban has focused, is the assumption that the public in the intervening country is willing to tolerate a humanitarian intervention in which its own soldiers may be killed. D. Luban, ‘Intervention and Civilization: Some Unhappy Lessons of the Kosovo War’, in P. DeGreiff and C. Cronin (eds), Global Justice and Transnational Politics (Cambridge, MA: MIT Press, 2002), 79–115. 3. Occasionally, an author provides evidence that appears to undermine one of the assumptions, yet proceeds as if the assumption were nevertheless true. For instance, H. Frowe, The Ethics of War and Peace: An Introduction (New York: Routledge, 2012), 84–5, notes, without further comment, that in the late 1990s ‘the Serbs and the Kosovars were effectively engaged in a civil war’; she also gestures at the fact (though she does not put it this way) that the prospective intervener, NATO, insisted in negotiations, as a condition for it not to begin bombing, that its forces be provided full and unfettered access to the Federal Republic of Yugoslavia. As we shall see, this has been perceived as an (unreasonable) demand for unconditional surrender and a violation of the ‘last resort’ criterion for just war. 4. Luban, ‘Intervention and Civilization’, 91. 5. T. Nardin, ‘The Moral Basis of Humanitarian Intervention’, in J. H. Rosenthal and C. Barry (eds), Ethics and International Affairs: A Reader, 3rd edn (Washington, DC: Georgetown University Press, 2009), 85–101. Elsewhere, in the context of the war in Iraq, Nardin is much more sceptical that the US is a morally qualified humanitarian intervener. See T. Nardin, ‘Humanitarian Imperialism: Response to “Ending Tyranny in Iraq” ’, Ethics and International Affairs, 19 (2005): 21–6. Other examples where philosophers have tacitly adopted one or more of the two ideal assumptions include the following. A. Altman and C. H. Wellman, A Liberal Theory of International Justice (Oxford: Oxford University Press, 2009), ch. 5. A. Bellamy, ‘Responsibility to Protect

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



6. 7.

8.

9.

10.

11.

12.

13. 14.

CHALLENGES FOR HUMANITARIAN INTERVENTION

or Trojan Horse? The Crisis in Darfur and Humanitarian Intervention After Iraq’, Ethics and International Affairs, 19 (2005): 31–53. G. Brock, Global Justice: A Cosmopolitan Account (Oxford: Oxford University Press, 2009), ch. 7. S. Caney, Justice Beyond Borders: A Global Political Theory (Oxford: Oxford University Press, 2005), ch. 7. S. P. Lee, Ethics and War: An Introduction (Cambridge: Cambridge University Press, 2012), 110–12. E. Luck, ‘The Responsibility to Protect: Growing Pains or Early Promise?’, Ethics and International Affairs, 24 (2010): 349–65. F. Teson, ‘The Liberal Case for Humanitarian Intervention’, in R. Keohane and J. L. Holzgrefe (eds), Humanitarian Intervention: Ethical, Legal, and Political Dilemmas (Cambridge: Cambridge University Press, 2003), 93–129. M. J. Smith, ‘Humanitarian Intervention: An Overview of the Ethical Issues’, Ethics and International Affairs, 12 (1998): 63–79. M. Walzer, Arguing about War (New Haven: Yale University Press, 2004). H. Widdows, Global Ethics: An Introduction (Durham: Acumen, 2011), 94. See Mamdani, ‘Politics of Naming’, for an account of the complexity of the situation in Darfur and the slim prospects that an armed intervention would have made the situation better. See the International Commission of Inquiry Report on Darfur (2005) for evidence that both the government and rebel groups were guilty of war crimes (though the former was certainly worse in this regard). For instance, A. Buchanan and R. Keohane, ‘Precommitment Regimes for Intervention: Supplementing the Security Council’, Ethics and International Affairs, 25 (2011): 41–63. An important exception is T. Pogge, World Poverty and Human Rights (New York: Polity Press, 2002), ch. 8. Also see my n. 10. Orend runs through some of what I will call the ‘standard narrative’ on Rwanda, but nevertheless tacitly adopts the second ideal assumption, in this case that the UN Security Council was a morally qualified intervener. B. Orend, The Morality of War (Toronto: Broadview Press, 2006), 92–4. Many of the essays collected in Jokic are good examples of a non-ideal approach, as is Pogge, World Poverty. A. Jokic (ed.), Lessons of Kosovo: The Dangers of Humanitarian Intervention (Toronto: Broadview Press, 2003). In this respect, the question which empirical assumptions to provisionally include in one’s normative theory is in the same position as the question how to apply normative principles to action: both inevitably require judgement. See O. O’Neill, Toward Justice and Virtue (Cambridge: Cambridge University Press, 1996), 58. Also see L. Valentini, ‘Ideal vs. Non-Ideal Theory: A Conceptual Map’, Philosophy Compass, 7/9 (2012): 654–64. Compare L. Valentini, ‘On the Apparent Paradox of Ideal Theory’, Journal of Political Philosophy, 17 (2009): 332–55 and 337–8, who claims that a theory must presuppose an account of a perfectly just world in order to count as a normative theory. See e.g. A. Sen, The Idea of Justice (New York: Belknap Press, 2011). M. Blake makes essentially the same point with respect specifically to the policies and actions of states that currently exist: ‘Distributive Justice, State Coercion, and Autonomy’, Philosophy and Public Affairs, 30 (2001): 257–96 and 264. Also see A. Gutmann and F. Thompson, Why Deliberative Democracy? (Princeton: Princeton University Press, 2004), 57.

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

NON - IDEAL THEORY



15. Valentini, ‘Ideal vs. Non-Ideal Theory’; J. Rawls, A Theory of Justice (Cambridge, MA: Harvard University Press, 1971). 16. Sen, Idea of Justice. 17. Cf. for a different view A. J. Simmons, ‘Ideal and Nonideal Theory’, Philosophy and Public Affairs, 38 (2010): 5–36. 18. To be clear, I am not suggesting that there are no other legitimate research aims in this area. Conceivably, one might be interested in the purely abstract question whether a humanitarian intervention could ever be justified. However, very little if any work in the literature takes that form. 19. See, for instance, Buchanan and. Keohane, ‘Precommitment Regimes’; Frowe, Ethics of War and Peace; Orend, Morality of War, 92ff. Also S. Power, A Problem from Hell: America and the Age of Genocide (New York: Basic Books, 2002), 335. A partial exception is Pogge, World Poverty, ch. 8. He makes some methodological points similar to the ones I have made here, but his article was published (2002) before much of the evidence I emphasize was in print. His account of what happened basically conforms to what I have called the mainstream account. 20. A. Des Forges, Leave None to Tell the Story: Genocide in Rwanda (New York: Human Rights Watch, 1999). P. Gourevitch, We Wish to Inform You that Tomorrow We Will Be Killed with our Families (New York: Picador, 1999). G. Prunier, Rwanda: History of a Genocide (New York: Columbia University Press, 1997). 21. L. May, ‘Complicity and the Rwandan Genocide’, Res Publica, 16 (2010): 135–52. The standard view is also expressed by Tanguy, the former US Executive Director of Doctors Without Borders/Medecins Sans Frontieres: ‘Significantly, even Bill Clinton’s administration, which included people of good will and an outspoken sense of duty like [Anthony] Lake, was unable to rally the necessary political will to protect the victims of genocide in Rwanda.’ J. Tanguy, ‘Redefining Sovereignty and Intervention’, Ethics and International Affairs, 17 (2003): 141–8. 22. See J. Davidovic, ‘Are Humanitarian Military Interventions Obligatory?’, Journal of Applied Philosophy, 25 (2008): 134–44. 23. C. Davenport and A. Stam (manuscript), ‘Rwandan Political Violence in Space and Time’. Available at: . 24. The validity of the census figure has been challenged by those who claim that the number of Tutsis was likely to be an underestimate in the Hutu-ruled country. See Prunier, Rwanda, 263ff., and A. Jones, ‘Denying Rwanda: A Response to Herman and Peterson’, 2010, . 25. Davenport and Stam, ‘Rwandan Political Violence’; M. Barnett, Eyewitness to Genocide: The UN and Rwanda (Ithaca, NY: Cornell University Press, 2003), 11. 26. Des Forges, Leave None; Gourevitch, We Wish to Inform You; M. Mamdani, When Victims Become Killers: Colonialism, Nativism, and the Genocide in Rwanda (Princeton: Princeton University Press, 2001). 27. Gersony report summary, available at: . 28. C. Davenport and A. Stam, ‘What Really Happened in Rwanda?’, Miller-McCune, available at: .

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

29. R. Philpot, Rwanda and the New Scramble for Africa (Montreal: Baraka Books, 2013); Prunier, Rwanda; Barnett, Eyewitness to Genocide, 57. 30. A. Kuperman, ‘Provoking Genocide: A Revised History of the Rwandan Patriotic Front’, Journal of Genocide Research, 6 (2004): 61–84. 31. See Philpot, Rwanda and the New Scramble, ch. 6. 32. See in particular Prunier, Rwanda. 33. A. Stam, ‘Understanding the Rwandan Genocide’, public lecture given at the University of Michigan, 2009, accessed at: ; D. Peterson and E. Herman, ‘Rwanda and the Democratic Republic of the Congo in the Propaganda System’, Monthly Review (2010), . 34. See M. Chossudovsky, The Globalization of Poverty: Impacts of IMF and World Bank Reforms (London: Zed Books, 1999), ch. 5. 35. Philpot, Rwanda and the New Scramble, 114–15. 36. See Luban, ‘Intervention and Civilization’, n. 22 and sources cited there. 37. Philpot, Rwanda and the New Scramble, 18. 38. Even a harsh critic of this revisionist line such as Adam Jones has conceded this possibility. Jones, ‘Denying Rwanda’. 39. For a summary of the US role in the Democratic Republic of Congo from the 1960s until the present day, see R. W. Miller, Globalizing Justice: The Ethics of Poverty and Power (Oxford: Oxford University Press, 2010). See also C. Black, ‘The Rwandan Patriotic Front’s Bloody Record and the History of UN Cover-ups’ (2010), ; Philpot, Rwanda and the New Scramble; P. Reyntjens, The Great African War (Cambridge: Cambridge University Press, 2009); and T. Wilkinson, ‘The Spanish Indictment of High-Ranking Rwandan Officials’, Los Angeles Times, 7 Feb. 2008, . 40. N. Chomsky, The New Military Humanism: Lessons from Kosovo (London: Pluto Press, 1999), 107–10; R. Falk, ‘Humanitarian Intervention After Kosovo’, M. Cohn, ‘The Myth of Humanitarian Intervention in Kosovo’, A. Pavkovic, ‘Humanitarian Intervention in Nationalist Conflicts: A Few Problems’, all in Jokic (ed.), Lessons from Kosovo. 41. Chomsky, The New Military Humanism, 20–1; Pavkovic, ‘Humanitarian Intervention’. 42. Chomsky, The New Military Humanism, 52–62; Cohn, ‘Myth of Humanitarian Intervention’. 43. See particularly the essays by Falk, Cohn, Pavkovic in Jokic, Lessons from Kosovo, as well as Chomsky, The New Military Humanism. 44. J. Szende, ‘Selective Humanitarian Intervention: Moral Reason and Collective Agents’, Journal of Global Ethics, 8 (2012): 63–76. 45. Chomsky, The New Military Humanism, 31, quoting US Special Envoy to the Balkans, Robert Gelbard. Also see Cohn, ‘Myth of Humanitarian Intervention’, and Pavkovic, ‘Humanitarian Intervention’. 46. Chomsky, The New Military Humanism, 16, citing NATO sources. Others put the deaths in the low thousands.

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

NON - IDEAL THEORY



47. See Chomsky, The New Military Humanism, and Cohn, ‘Myth of Humanitarian Intervention’, cited above. 48. Human Rights Watch, ‘Weapons Transfers and Violations of the Laws of War in Turkey’ (2005). Available at: . 49. There is a response to this comparison, which is that because of Milosevic’s track record earlier in the decade, particularly in Bosnia, it was reasonable to expect that at some point in the near future he would commit massive human rights abuses against the Albanian Kosovars. See Pavkovic, ‘Humanitarian Intervention’, 66, for some critical discussion of this response. 50. Chomsky, The New Military Humanism, 3. 51. Luban, ‘Intervention and Civilization’, and Szende, ‘Selective Humanitarian Intervention’; J. Pattison, Humanitarian Intervention and the Responsibility to Protect: Who Should Intervene? (Oxford: Oxford University Press, 2010), 22. 52. Pattison, Humanitarian Intervention, 21, his emphasis. 53. See Bennis for some suggestions about how to improve the situation, which prominently involves halting arms shipments to both sides. In my view, this remains the most promising solution at the present time (late 2016). P. Bennis, ‘5 Concrete Steps the US Can Take to End the Syria Crisis’, The Nation, 25 May 2014. 54. United Nations, ‘Report on the Alleged Use of Chemical Weapons in the Ghouta Area of Damascus on 21 August 2013’ (2013), ; S. Hersh, ‘The Red Line and the Rat Line’, London Review of Books 36/8 (17 Apr. 2014). 55. For some criticism, see Zakaria, who points out that there are some 1,500 insurgent rebel groups in Syria, and that the most powerful ones are radical Islamist. F. Zakaria, ‘Obama Caves to Conventional Wisdom on Syria’, Washington Post, 10 July 2014. 56. See, for instance, F. Teson, ‘Syria and the Doctrine of Humanitarian Intervention’, Bleeding Heart Libertarians, . An exception is D. Postel and N. Hashemi, ‘Use Force to Save Starving Syrians’, New York Times, 10 Feb. 2014.

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

5 The Leeriness Objection to the Responsibility to Protect Marco Meyer

‘Responsibility to Protect’ (R2P) is a UN doctrine that reconceptualizes state sovereignty as conditional on the ability and willingness of states to protect fundamental interests of people living in their territory. One aspect of R2P is that the international community acquires a responsibility to intervene, in extreme cases militarily, if states fail to protect these fundamental interests of their inhabitants. In this chapter, I will propose a new objection to the Responsibility to Protect, which I call the leeriness objection. According to this objection, implementing R2P is morally unjustified because it undermines trust between states, which leads to an undermining of the norm of non-aggression. The norm of non-aggression is undermined because bystander states cannot gather sufficient direct evidence that intervening states are not waging aggressive wars under the guise of R2P, nor can they trust the testimony of intervening states. As states are conditional cooperators, in a sense to be explained, this lack of knowledge weakens the willingness of bystander states to comply with the norm of nonaggression. This previously unnoticed objection is important because it calls into question the moral status of R2P. If it does indeed undermine the trust that underpins the international order, implementing R2P could lead to more unjust wars. This indirect negative effect of R2P could thus outweigh its positive moral contributions. The leeriness objection is distinct from two other, more familiar objections to R2P. The first type of objection challenges the moral justification of the principles underlying R2P, while the second type of objection questions whether R2P can be reliably implemented in practice. The latter objection comes in two kinds: they either question the competence of intervening powers to understand the cultural, political, and religious complexities of the country or group they plan to rescue,

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

THE LEERINESS OBJECTION



or emphasize the risk that R2P could be used by intervening powers to cover up the pursuit of extrinsic motives that lead to unacceptable outcomes for the target population. In exploring the leeriness objection I assume, for the sake of argument, that these familiar objections can all be met. Thus, I assume that, pace the first objection, R2P is morally justified, and, pace the second objection, R2P can be implemented without unacceptably high levels of failure and abuse. The leeriness objection rests on a feature of international relations that philosophical discussion of R2P usually abstracts from, namely that pursuing justice in international relations can come into conflict with sustaining order in the international community, threatening the preconditions for the peaceful coexistence of states.¹ In a strong international order, states conceive of themselves as being bound by common rules in their relations with one another. Maintaining order comes into tension with pursuing justice if pushing for more just rules would weaken the willingness of states to adhere to these rules. My central contention is that R2P has the potential to make states leery of each other, and thereby to weaken the international order. Many powerful states are deeply suspicious of each other’s foreign policy as it is, especially when it comes to military intervention. Keeping the peace among states that are deeply suspicious of each other requires sufficiently simple rules that they can be easily monitored by each state. The worry is that R2P introduces a level of epistemic complexity that threatens peace. My aim in this chapter is to articulate the leeriness objection to R2P in detail, in order to prepare the ground for an assessment of its strength. The force of the leeriness objection depends on empirical claims, which I cannot settle here. Rather, I point out which empirical questions need to be answered so that the strength of the leeriness objection can be assessed, and which steps can be taken in the implementation of R2P to mitigate the force of the objection. I set the stage by providing the relevant background to R2P and the criticisms levelled against it, distinguishing the leeriness objection from other, more common objections to R2P. The following sections explore the leeriness objection in detail. First, in order to assess whether a given military intervention is justified, states typically have to rely on the testimony given by the institutions implementing R2P, and chiefly on testimony given by intervening states. Second, states will often lack sufficient reason to trust intervening states, or fail to trust intervening states despite having sufficient evidence for their trustworthiness, and will therefore often fail to come to know that R2P has not been abused, even if in fact it has not been. Third, the norm of non-aggression is epistemically less demanding than R2P, and therefore it is less vulnerable to the leeriness objection. Fourth, the lack of knowledge caused by R2P might weaken the norm of

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

non-aggression. Hence, R2P could lead to more unjust wars. The final section indicates the empirical questions that must be settled for the strength of the epistemic objection to be evaluated, and suggests ways to mitigate its force.

The Responsibility to Protect and its Critics The doctrine of state sovereignty enshrined in the UN Charter prohibits states from intervening in one another’s affairs. An ongoing re-evaluation of the accepted doctrine of state sovereignty has, however, been undertaken in light of the humanitarian catastrophes of the 1990s, such as those that took place in Rwanda, Bosnia, and Kosovo. In 1999, UN Secretary General Kofi Annan challenged UN members to find ways for the international community to respond to threats of genocide and other human rights violations. To address Annan’s call, the Canadian government sponsored a distinguished commission, the International Commission on Intervention and State Sovereignty (ICISS). In 2001, the commission issued a report that sets out to redefine state sovereignty as entailing responsibilities as well as rights.² According to this conception, state sovereignty is limited by a state’s ability and willingness to protect the people who live in its territory from certain grave violations of their basic human rights. This conception of state sovereignty underlies a novel doctrine for dealing with grave human rights violations, called the Responsibility to Protect (R2P), also set out in the report. Where a state cannot protect its people from grave human rights violations, R2P ascribes responsibility to the international community to support the respective state in meeting its obligations towards its people. R2P emphasizes the prevention of rather than reaction to human rights violations by encouraging the international community to help build a state’s own capacities to protect human rights. However, if a state fails to cooperate, thereby leaving its population at risk of grave human rights violations, the state loses its immunity to outside interference in internal affairs. In the most severe cases, R2P allows for military interventions designed to protect basic human rights. Despite some pushback, R2P has taken root in the international community. The UN World Summit of 2005 endorsed the broad outlines of the report’s recommendation,³ and in 2009, UN Secretary General Ban Ki-moon issued a report that attempted to move the focus of discussion away from the question of whether to implement R2P to the question of how to implement it.⁴ In 2011, the Security Council had recourse to R2P to justify its authorization of the military intervention in Libya.⁵ In the course of the development of R2P, the original conception as proposed by the ICISS has undergone substantial change. Most significantly,

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

THE LEERINESS OBJECTION



the understanding of what counts as a grave human rights violation has grown narrower. In the current version of R2P, only genocide, crimes against humanity, war crimes, and ethnic cleansing are considered just causes for military intervention. It is this revised version of R2P that currently commands the greatest support. Its general support notwithstanding, R2P also has vocal detractors. Miguel d’Escoto Brockmann, former president of the General Assembly, is one of its sharpest critics. In 2009, he organized a thematic dialogue hosted by the United Nations General Assembly on the future of R2P. The main point of the critics led by Brockmann was that R2P could become a tool for powerful states to interfere with the affairs of other states, not with the intention to protect human rights, but to serve ulterior interests.⁶ This challenge is usually seen to give rise to one of two objections: either, it is viewed as a challenge to the very principles R2P rests on; or, granting that R2P is morally sound, it is viewed as a challenge to the ability of UN institutions to implement R2P without failure or abuse. The leeriness objection is a third and distinct way to interpret Brockmann’s challenge. According to the leeriness objection, R2P weakens the norm of nonaggression insofar as it undermines the ability of states to know that the norm of non-aggression is generally adhered to. The general idea is as follows. Never mind whether R2P will in fact be abused. As R2P is increasingly invoked to justify military interventions, many states will come to believe that the norm of nonaggression is crumbling, if they view R2P as a mere moral cover for essentially aggressive wars. In effect, states may feel less bound by the rules that underpin the international order, in particular the norm of non-aggression. As the international order weakens, states may see less reason to abstain from aggressive wars, whether openly or under the cover of R2P. The leeriness objection is independent of the former two objections, in the sense that if either or both of the former objections can be met, the leeriness objection still stands; moreover, if either or both succeed, the force of the leeriness objection is likewise unaffected. If, however, at least one of the former two objections could decisively undermine the justification of R2P, it would be of mere academic interest to raise the leeriness objection. To further motivate exploring the leeriness objection, let me then briefly discuss the former two objections. Consider first the objection that something is wrong with the content of the principles underlying R2P, and that R2P is therefore morally unjustified. Assessing the content of these principles is a very complex task, outside the purview of this chapter. Fortunately, casting some doubt on the effectiveness of the objection is made easier by the fact that the content of the principles does not need to be correct in every detail to secure the justification of R2P. The reason is that there

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

already are laws and norms in place designed to regulate war waging. The most pressing moral question, then, is whether R2P morally improves upon the status quo. I will therefore assume throughout that the content of the principles underlying R2P can only undermine the justification of R2P if they further depart from the correct moral principles than the principles underlying the status quo. On the surface, at least, it can be argued that the principles on which R2P is founded constitute a significant step forward, serving as they do as an articulation of the main concerns animating very influential recent contributions to the ethics of war.⁷ These principles are as follows. First, only individuals are of fundamental moral concern and are thus the only bearers of moral rights. R2P meets this commitment by limiting state sovereignty by privileging the protection of individual human rights. Second, to the extent that states can be said to have moral claims, these claims derive from the rights of individuals. R2P emphasizes the importance of nation states in the protection of human rights, but does not recognize any independent moral standing of states. Third, states have moral obligations to foreigners as well as citizens. R2P respects this concern by postulating the responsibility, rather than the liberty, of the international community as a whole to protect the basic human rights of individuals in foreign states if these states fail to do so themselves. In respect to these three points, the principles that underlie R2P would seem to match very influential articulations on the morality of war better than the status quo. This gives some plausibility to the claim that R2P may be morally superior to the status quo. Consider second the objection that while the principles underlying R2P may be morally progressive, implementing R2P will lead to an unacceptable level of either failure or abuse. The difference between failure and abuse concerns the reason why a given military intervention does not achieve its proclaimed humanitarian goal. Whereas failure is a matter of lack of competence to understand the cultural, political, and religious context of a conflict sufficiently well to intervene successfully, abuse is a matter of the (intentional or perhaps unintentional) intrusion of extrinsic motives for intervention by the intervening party, leading to unacceptable outcomes for the target population. To assess this objection, we need to look at the institutions tasked with implementing R2P. Implementing institutions include the states carrying out the intervention, in particular their military; monitoring organizations, assessing the state of affairs in conflict regions and advising authorizing institutions, which will also often include the armies of intervening states, but also their security services, and journalists, NGOs, and international monitoring organizations; and authorizing institutions, such as the Security Council, charged with the task of authorizing military interventions. I label these types of institutions collectively as

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

THE LEERINESS OBJECTION



implementing institutions. While it is important to recognize that implementing organizations have a complex structure, it is even more important to see that the UN framework heavily relies on powerful states to gather intelligence, commit resources, and carry out military interventions. I assume that distrusting some of the most important states in the UN framework is sufficient to distrust implementing institutions as a whole. Remember that throughout this chapter I regard the implementation of R2P as morally justified when it morally improves the status quo. Consider a distinction between two kinds of failings authorizing institutions might succumb to, whether on the ground of lack of competence or lack of moral motivation. On the one hand, authorizing institutions might authorize military interventions which are not in fact justified by the principles underlying R2P. These failings can be considered false positives. On the other hand, authorizing institutions might fail to authorize military interventions which are justified by the standards of R2P. These failings can be considered false negatives. Since R2P is more permissive of the authorization of military interventions than the status quo, the implementation of R2P should not increase the number of false negatives. The concern, thus, is whether R2P will lead to an unacceptable rate of false positives.⁸ Since the authorizing institution for R2P is the Security Council of the UN, and since the voting structure of the Security Council has a demonstrated tendency towards inaction rather than towards excessive action, there is little reason to believe that R2P will lead to many interventions in the first place. In addition, several authors have suggested alternative institutional arrangements, which could, if adopted, reduce the likelihood of false positives. For example, Allen Buchanan and Robert O. Keohane suggest so-called pre-commitment regimes to remedy some of the epistemic failings of the Security Council. In particular, they advocate pre-commitment regimes for the protection of weak democracies.⁹ Under such regimes, a democratic government would authorize a group of other democracies to intervene in its territory in response to violence that it cannot control. An instance that would warrant intervention, for example, would be a domestic attempt to overthrow the standing government. Other authors have made further proposals for reducing the likelihood of abuse of R2P. For example, Jeff McMahan has argued in favour of an international court invested with the sole responsibility of applying just war theory to current conflicts.¹⁰ Hence, there is reason to believe that the second objection—that in spite of R2P’s sound moral foundations its implementation will lead to abuse— can be met even within the current institutional environment. Moreover, there are proposals for creating institutions that would provide even better checks against abuse.

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

These brief remarks do not seek to decisively refute the two objections. They do show, however, that these objections could well prove unsuccessful. Hence whether the leeriness objection is successful could turn out to be decisive for the overall justifiability of R2P.

The Epistemic Demandingness of R2P The macro structure of the leeriness objection is simple: P1 Implementing R2P may undermine the knowledge¹¹ of states that the norm of non-aggression is generally adhered to. P2 If states lack knowledge that the norm of non-aggression is generally adhered to, this might undermine their reasons to comply with the norm of non-aggression (and thereby lead to more unjust wars). C Implementing R2P may lead to more unjust wars. Clarifying the premises is a more complicated task. I will spend the next few sections clarifying the first premise, and then tackle the second. Suppose the UN Security Council authorizes a military intervention in some state appealing to R2P. The Security Council is essentially a place in which certain states, first and foremost the permanent five members, meet and negotiate claims.¹² To get authorization for the intervention, at least some of the states in the Security Council need to take an active interest, and all of the five permanent members need to refrain from putting in their veto. I label the group of states pushing for intervention intervening states. I am interested in the perspective not of intervening states, but of states that have no special stake in whether an intervention is authorized, whether they are represented in the Security Council or not. I label these states bystander states. In the case of the recent intervention in Libya, for example, the United States was an intervening state, whereas India was a bystander state. I will talk throughout of intervening states and bystander states simpliciter, but it is important to note that who the states in the respective category are will change depending on the intervention we consider. Indeed, the main point of the leeriness objection is that, because states fail to come to know that intervention was in fact justified in cases where they were bystanders, these same states may decide to fight aggressive wars, thus turning into intervening states (if unjust ones). Bystander states have an interest in finding out whether interventions authorized by the Security Council are in fact justified by the standards of R2P.¹³ To do this, bystander states need to establish whether material facts obtain that satisfy the criteria for military intervention set out by R2P.¹⁴ Now let us assume that the

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

THE LEERINESS OBJECTION



intervention does indeed satisfy the criteria set out by R2P, and is therefore justified. The question at stake is whether the bystander state can come to know that the intervention is justified. In support of the first premise of the leeriness objection, I will argue that, in a wide range of cases, bystander states cannot obtain this knowledge. Therefore, bystander states also cannot come to know that the intervention does not violate the norm of non-aggression—even though, ex hypothesi, the intervention in fact does not. I will argue, moreover, that the implementation of R2P can cause this distinct lack of knowledge about the moral status of military interventions. My argument for this first premise of the leeriness objection has three parts. First, I argue in this section that bystander states have to rely on testimony provided by the institutions implementing R2P. Then, I argue in the next section that in most cases bystander states do not have sufficient reason to trust the testimony given by these institutions, and even if they do have sufficient reason may fail to trust nonetheless. This lack of trust, in turn, can impede knowledge transfer via testimony from intervening states to bystander states received from these sources. Finally, I will argue that the epistemic burden on bystander states is lower if the status quo is maintained. Why would a bystander state need to rely on testimony provided by implementing institutions? R2P recognizes the ‘four great crimes’ of genocide, war crimes, crimes against humanity, and ethnic cleansing as just causes for military interventions. A just cause is necessary to justify a military intervention by the standards of R2P, but is not sufficient. In addition, so-called precautionary principles need to be met, including last resort and proportional means.¹⁵ In order for the bystander state to know that the military intervention is justified, it needs to know that the intervention has a just cause and that these precautionary principles are met. Each of these elements gives rise to distinct epistemic challenges, which I will consider in turn. Some of these challenges can potentially be mitigated within the broad institutional framework in which R2P is currently implemented. Meeting others, if possible at all, requires substantially different institutions. I come back to ways in which these epistemic challenges may be mitigated at the end of the chapter. I maintain that, given the content of R2P and the current institutional framework, implementing institutions will be the only parties capable of surmounting all of these epistemic challenges for a wide range of military interventions. Thus only they are able to know firsthand that an intervention is justified. Bystander states, therefore, will often need to rely on the testimony of implementing institutions to come to know that a given military intervention is justified.

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

Just cause The just causes R2P recognizes cover a wide range of atrocities, differing widely in seriousness and scope. For instance, war crimes generally refer to breaches of the Geneva Conventions or other violations of the customs of war. While all the conventions and customs protect important moral goods, not every possible war crime is substantial enough to provide a just cause for military intervention. Therefore, R2P needs to be supplemented with substantiality requirements designed to prevent it from becoming too permissive. Proposals for these requirements have included conditions that crimes are committed systematically, are planned, or are intended.¹⁶ Each of these proposals poses significant epistemic challenges for bystander states wanting to assess whether a just cause obtains. On the one hand, groups that plan and systematically perpetrate crimes do well to try and hide their plans and obscure their systematicity. On the other hand, groups who are victims of violence in a conflict have a vested interest in depicting the perpetrators as diligent planners carrying out war crimes systematically and with intent. Even for monitoring institutions active on the ground it can be extraordinarily difficult to ascertain whether substantiality requirements obtain, trapped as these institutions are between the claims of opposing parties with clear motivations for misrepresentation.¹⁷ It must also be emphasized that intervention cannot wait for a formal determination that a crime has been committed, let alone whether it meets the appropriate substantiality requirements; policy makers, rather, must conduct assessments as events unfold. Access to relevant information is naturally severely limited in conflict situations. Information compiled and made available by NGOs and journalists can often provide some relevant information about events as far as they can be directly observed. The most comprehensive information about observable events is, however, often collected by intervening states.¹⁸ Thus bystander states are unlikely to have direct access to the information they require to assess the material facts relevant to the conditions of just cause. For this reason, bystander states will need to rely on the testimony of implementing institutions.

Last resort The condition of last resort requires that other means lesser and milder than military intervention have been explored and have been found unsuitable to address the just cause of the intervention. Of the precautionary principles, last resort is perhaps the most difficult to evaluate outside the small diplomatic and military planning circles that are responsible for preparing an intervention.

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

THE LEERINESS OBJECTION



Identifying alternatives to military intervention typically involves non-public diplomatic channels, making it difficult in principle for bystander states to verify whether these alternatives have been explored in full. The diplomatic services of intervening states usually have the best, and sometimes only, access to the key actors in a given conflict, putting them in a unique position to judge whether the last resort condition is met. Bystander states are, therefore, unlikely to have direct access to the information they need to judge whether the condition of last resort has been met, or whether instead reasonable offers of peaceful settlement have been ignored by intervening states. Hence bystander states will still need to rely on the information provided by intervening states.

Proportionality The condition of proportionality requires that the suffering and destruction caused by the intervention is reasonable in proportion to the importance of achieving the just cause of the intervention. Thus, proportionality assessments involve an immensely complex balancing exercise between very different kinds of value: the value of achieving the just cause on the one hand, and the disvalue of the death and destruction caused on the other.¹⁹ In the parlance of R2P, proportionality also covers the condition of necessity, according to which the intervention must be calculated to minimize the harm it causes. These calculations need to be based on estimates and uncertain expectations. Given the challenges of making accurate calculations in the face of an intervention, it is easy for intervening states to misrepresent a disproportional intervention as proportional. Whether proportionality assessments have been done in good faith is often difficult to determine, even after the intervention has already taken place. Another epistemic challenge connected to proportionality is that intervening states will not reveal their detailed military and political plans in advance, since that would give opponents too much advantage, and so bystander states will necessarily be ignorant of much that would justify or fail to justify the intervention. Hence bystander states will often need to defer to implementing institutions where proportionality assessments are concerned. One might object that, while the epistemic challenges of judging whether a military intervention satisfies the conditions of R2P are formidable, they are not in all cases unsurmountable. In response, I would like to concede at the outset that there are cases where it is easy to determine that a military intervention is not justified by R2P. Consider Russia’s intervention in Georgia in 2008, which the Russian government justified by appeal to R2P.²⁰ The Russian government claimed that Georgia threatened Ossetians with atrocities equivalent to genocide, and that Russia was therefore justified to step in militarily. Yet, Russia never

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

attempted to substantiate this charge of genocide, not before, during, or after its intervention. The Russian army moved 20,000 troops and 100 tanks into Georgia, also into regions with no Ossetian population. I submit that these facts suffice to establish that the intervention lacked a just cause and was disproportionate. This information was readily available during the intervention, found in a range of sources, including reports by security services, NGOs, and newspapers. Granting that interventions that fail to meet R2P criteria can sometimes readily be identified as such does little damage, however, to the force of the leeriness objection. It is often easier to come to know that some token does not belong to a particular type, than to come to know that it does. For instance, in a murder investigation, most people can readily be ruled out as perpetrators of the crime. Establishing that a particular person is indeed the murderer, in contrast, is often much more difficult. Information establishing that a given intervention does satisfy the R2P criteria is often difficult to come by, as I have already argued, and the example of Russia’s 2008 intervention does not affect this point. Hence, in a large array of cases, bystander states will need to rely on testimony, and the only fully competent sources of testimony are implementing institutions, in particular intervening states.

Obstacles to Trust The mere fact that bystander states need to rely on implementing institutions, particularly on intervening states, for testimony as a source of knowledge does not pose a special epistemic problem. After all, all of us depend on testimony for almost everything we know. The trouble is rather that the transfer of knowledge via testimony has preconditions. One way of getting at these preconditions is to think about successful transfer of knowledge as an exchange between a trustworthy speaker and a trusting hearer.²¹ For our purposes, the assumption that hearers need to trust speakers in order for the knowledge transfer to succeed is crucial. As in the fable ‘The Boy Who Cried Wolf ’, in which a shepherd boy gives a false alarm once and is then no longer heeded by the villagers, hearers will not believe a speaker they do not trust. Note that a speaker may speak the truth, be perfectly trustworthy, and yet still hearers might not trust him. In cases where trust is lacking, the transfer of knowledge fails. In brief, the leeriness objection says that, since trust is lacking between states, bystander states may not be able to come to know that R2P interventions are justified. I have argued that implementing institutions are often the principal sources of knowledge for bystander states concerning the justification of a given military intervention. The question that follows is whether implementing institutions are

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

THE LEERINESS OBJECTION



or will be able to transfer to bystander states the relevant knowledge of material facts in cases where R2P is invoked. Knowledge requires both belief and justification. Hence I will assume here that if knowledge is to be transferred to bystander states via testimony, states factually need to trust implementing institutions to meet the belief condition, and it must be reasonable for states to trust implementing institutions, thereby ensuring the justification condition is met. I will first briefly discuss the belief condition, and then the justification condition. Assume for a moment that bystander states have sufficient reason to trust intervening states regarding the information they provide about the military intervention. Bystander states would then be perfectly justified in their belief that the intervention is legitimate. Will bystander states in fact trust implementing institutions? As bystander states, considered as legal entities, do not cognize in the first place, the question should be understood as shorthand to refer to the relevant officials in the state apparatus. I think that we cannot take for granted that relevant officials reliably detect trustworthiness and form their beliefs accordingly. The reason is many people display what Quassim Cassam calls intellectual character vices, such as gullibility, dogmatism, closed-mindedness, and negligence.²² Intellectual character vices impede effective and responsible inquiry even if all the evidence that is required to justifiably form true beliefs is available, for instance by leading inquirers to ignore important evidence or to jump to conclusions. Studies of the propensity of people to believe conspiracy theories suggests that epistemic vice is widespread in the general population. A meta-study of four surveys in the United States finds that half of the American public consistently endorses at least one conspiracy theory.²³ Given this willingness to endorse conspiracy theories when good evidence is available to discard them, it would be surprising if even excellent evidence of trustworthy behaviour by intervening states was sufficient to guarantee that people form the respective beliefs. Comparable data are not available on government officials specifically. However, at least in democracies, state action is highly responsive to public opinion. Moreover, in the light of the evidence regarding the population at large, it would be surprising if state officials were free of epistemic vice. Leaving the issue of epistemic vices to one side, let us now turn to the justification condition. When is it even reasonable for bystander states to trust implementing institutions? The issue of trust in institutions has not received nearly as much attention as trust in individuals, but Russel Hardin’s work on trust in government and other institutions lends itself to be applied to the present problem.²⁴ I suspect, however, that the points I make using Hardin’s theory could be likewise expressed in terms of competing accounts of trust in institutions.²⁵

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

According to Hardin, it is reasonable for someone to trust another person or institution with regard to some issue if she has sufficient reason to believe that the person or institution is trustworthy. Thus, Hardin’s account of trust passed the buck to his account of trustworthiness. According to Hardin’s theory of trustworthiness, Alex is trustworthy for Beryl regarding some issue if Alex has encapsulated Beryl’s interests with regard to the issue.²⁶ Hardin’s preferred reason for why Alex would encapsulate Beryl’s interests is that he values the continued relationship with Beryl. For instance, a baker might encapsulate her regular customer’s interest in excellent bread because the baker values the commercial relationship she has with the customer, and wants it to continue. If the customer realizes that the baker has thus encapsulated his interest, he has reason to trust the baker regarding the quality of the bread. In contrast, day tourists have less reason to expect that staff will have encapsulated their interests in good food at a tourist restaurant, and have therefore little reason to trust that the food will be of good quality.²⁷ I think that Hardin’s preferred reason for why Alex would encapsulate Beryl’s interests, namely that Alex values continued interaction with Beryl, cannot be the only possible reason why Alex may become trustworthy for Beryl. For suppose that Beryl comes as a stranger to Alex’s hometown. Beryl picks Alex at random from the passersby, asks him for the way, and Alex knowledgeably tells Beryl. I take it that at least in some such interactions Beryl can come to know the way, despite it being common knowledge among Alex and Beryl that they are most unlikely to ever chance upon each other again. So there must be other reasons than valuing the continuing of the relationship that make people trustworthy. In this case, Beryl might just be able to count on an expectation that locals will extend some generalized goodwill towards strangers if asked for the way.²⁸ But the generalized goodwill Beryl can reasonably expect is tightly circumscribed. If she was an investigative reporter, and asked Alex about his questionable business practices, she could not reasonably expect Alex to be trustworthy in this regard for her. The stakes for Alex being high, generalized goodwill predictably gives way. Asking implementing institutions about the facts relevant to the justifiability of a military intervention is similarly a high stakes case, so that bystander states cannot simply count on the generalized goodwill of implementing institutions to testify truthfully. Do bystander states have any good reason to believe that implementing institutions are trustworthy? Understood against the backdrop of Hardin’s theory, this is to ask if bystander states have sufficient reason to believe that implementing institutions have encapsulated their interest in accurate information about whether or not a given military intervention satisfies the conditions of

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

THE LEERINESS OBJECTION



R2P. According to Hardin’s preferred reason why implementing institutions encapsulate this interest of bystander states, implementing institutions would need to fear that they damage the continued relationship with bystander states if they fail to act trustworthily. The trouble is that implementing institutions have little reason to fear that they will damage their relationship with bystander states, since implementing institutions are often themselves the only existing epistemic authorities capable of assessing the grounds for intervention. Given the dual role served by implementing institutions, bystander states should expect them to maintain that the conditions for military intervention have been met no matter what. Bystander states therefore do not have sufficient grounds to trust implementing institutions in this regard, and so are unable to come to know from the testimony they provide that the intervention is in fact justified (as, remember, I assume throughout it is). There is even a powerful positive reason why it may be reasonable for bystander states to distrust implementing institutions. These institutions are dominated by a handful of powerful states, which have often behaved untrustworthily regarding military interventions in the past. States outside this exclusive club will be guided in their assessment of the trustworthiness of intervening states by past experiences, which are often not conducive to building trust between states. It is perhaps no surprise that former colonies are among the staunchest critics of R2P. Miguel d’Escoto Brockmann, of Nicaragua, referred in his cited statement to the experiences of his home country with the US, which supported and organized paramilitary operations in Nicaragua in the 1980s. But the US is hardly alone in inspiring distrust. The historical record of each of the five permanent members in the UN Security Council is not suggestive of trustworthiness when it comes to refraining from using war to promote national self-interest. Thus there are two powerful reasons for bystander states not to trust implementing institutions: the first rooted in the general observation that implementing institutions have an obvious interest in not revealing failure or abuse, the second stemming from the historical experience many bystander states have had with the states that exercise the most influence on implementing institutions, that is, permanent Security Council members. Are these reasons decisive? I can think of two objections. The first is that some institutions can overcome initial lack of trust by convincing performance. John Rawls has argued that for some parts of the population, for instance staunchly illiberal religious groups, the only hope is that these people will come to see, gradually and over time, that liberal institutions merit their trust.²⁹ There are, however, two considerations militating against this strategy in the case of R2P. The first is that there is seldom clear evidence available to corroborate the testimony of implementing institutions,

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

even long after an intervention has been completed. Thus, if a bystander state initially fails to trust implementing institutions, there is little reason why it should become more trusting over time. The second consideration is that, even if bystander states become more trusting over time, their trust could come too late for the norm of non-aggression to be adequately safeguarded. The second objection is that bystander states might be able to somehow bypass implementing institutions as a source of knowledge after all. This could be the case if bystander states have independent evidence that states pushing for intervention have no non-humanitarian ambitions in the target country. For instance, it is implausible that any of the permanent members of the Security Council would have wanted to mask an aggressive war against Rwanda under the veil of R2P in 1994, because it is not clear which non-humanitarian motives could have animated them. Had an intervention been authorized by the Security Council, bystander states might have been able to use this consideration as an epistemic shortcut to come to know that R2P was not abused in this case, or at least not deliberately so. Unfortunately, of course, we know in retrospect that the international community showed little interest in the fate of Rwanda’s population at the time. In cases where interventions have taken place, intervening states have often had some ulterior motive as well, or at least such an interest can be imputed to them, even if their humanitarian motive was primary. Thus, this epistemic shortcut is seldom helpful in practice. Moreover, three developments in the way powerful states fight wars have considerably decreased the risks that military interventions pose to their own soldiers. First, powerful states rely on mercenaries rather than their own soldiers, as the US did in Afghanistan and Iraq. Second, interventions are increasingly conducted through air strikes alone, as in Kosovo and Libya. Third, the use of drones enables states to intervene militarily at no risk to their soldiers. These military and technological developments lower the costs of intervention, and thus make it harder to eliminate potential ulterior motives. While it might be paranoid to believe that states would put large numbers of soldiers at risk in order to secure tenuous strategic interests, relatively minor motives might be more credible when states can minimize risk to their own soldiers. The epistemic challenges bystander states face together with the reasons undercutting trust in implementing institutions as the only available sources of knowledge explain why implementing R2P can render bystander states ignorant as to whether a given intervention is justified or is instead an act of aggression. Moreover, implementing institutions have ample possibilities to exaggerate or play down the seriousness of conflicts, and thereby to cover up mistakes or deliberate abuses of R2P. To repeat, I assume throughout that implementing

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

THE LEERINESS OBJECTION



institutions do not abuse R2P or commit major mistakes in its implementation. What I have argued is that, this highly debatable assumption notwithstanding, bystander states will find it difficult in many cases to come to know that this is indeed the case.

The Epistemic Superiority of the Status Quo I am concerned with the relative advantages and disadvantages of implementing R2P as compared to the status quo. Therefore, it is not enough for me to show that states face epistemic challenges under the R2P regime; I also need to show that the epistemic burden on states is lighter under the current policies. Under the status quo, waging war is permissible only in response to an attack on a state’s territory; all military interventions for purposes other than national self-defence are prohibited by international treaties. Accordingly, war waging is only legitimate when used to fight an external aggressor, and an external aggressor always fights unjustly. Identifying the aggressor can be an epistemically challenging task, not least because aggressors are often keen to disguise their attacks as acts of self-defence. The most famous historical example of such an attempt is that of the German invasion of Poland in 1939, when Hitler claimed in a speech broadcast on the day of the war’s outbreak that Poland had attacked Germany and that German troops had merely retaliated. If Hitler’s speech were the only information available about the start of the war, it would be difficult to come to know that it was Germany who had attacked Poland. Fortunately, states can draw on a number of other sources besides the information provided by state officials and state media services. A couple of days into the war not even the staunchest sceptic could doubt or deny that Germany had been the aggressive party, given that German troops continued to make their way into Polish territory—a fact that states could easily verify by several independent means. More generally, under the current regime there are several sources that provide evidence as to whether a party was forced to resort to aggression or acted unprovoked. Relying on the principle that states rarely fight aggressive wars they know they will lose,³⁰ it is usually possible to determine who is the aggressor in a military confrontation between states. Moreover, the prehistory of a military intervention often reveals which party has ambitions to attack. These indicators can be obtained from publicly available sources, including announcements made by the respective parties, war journalists, NGOs, and so forth. A complication must be noted. Identifying the aggressor does not settle the moral status of the war fought by the victim state. Of course the victim state has a

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

just cause, viz. national self-defence. But in order to be justified, the victim’s response needs to meet precautionary principles as well. Therefore, under the current regime, at least one party fights an unjust war, but at most one party fights a just war, and, in some cases, neither party fights permissibly.³¹ Many of the same epistemic challenges crop up regarding the victim’s military reaction, also in the case of R2P. Unjust wars of self-defence are, however, unlikely to weaken the norm of non-aggression. Bystander states can usually assume that a war fought in response to military aggression does not constitute an aggressive war that seeks to take cover under the norm of non-aggression, unless the aggressor is somehow complicit in its victim’s plan to stage a war of national self-defence—which is usually implausible.

States as Conditional Cooperators I have argued that implementing R2P can undermine the knowledge of bystander states that the norm of non-aggression is generally adhered to, and that the status quo does not suffer from the same epistemological problem. This completes my discussion of the first premise of the leeriness objection. In this section, I discuss the second assumption of the leeriness objection, according to which bystander states lacking knowledge that other states are generally adhering to the norm of non-aggression are more likely to start aggressive wars themselves. As indicated in the introduction, I think that the second assumption sits well with the account of international relations as developed by the English school of international relations.³² According to this view, pursuing justice can come in tension with sustaining order in international society. Order in international affairs is underpinned by adherence to rules that sustain elementary human goals, such as prosperity, security of contracts, and most of all, peace. Maintaining order requires that states conceive of themselves as being bound by common rules in their relations with one another. Not all ways of organizing the international order are equally just; indeed, an international order may be decidedly unjust. At the same time, not all institutional arrangements that would seem to promote justice better than the status quo are compatible with maintaining order. When order breaks down, violent conflict is a likely outcome. Against the backdrop of this account of international relations, the second premise of the leeriness objection can be understood as the claim that the epistemic challenges raised by R2P are likely to undermine the international order. I argue for this premise by sketching an admittedly very crude model of state behaviour, which I use to explain what drives a state’s behaviour to comply with norms of non-aggression. At the end of the section, I will revisit some of the

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

THE LEERINESS OBJECTION



cruder assumptions I make to keep the model simple, and explore whether they are likely to influence the result. My core assumption is that states are conditional cooperators. They are cooperators in the sense that they comply with norms if it is better for all that the norm holds than that it does not. But their compliance is conditional on their knowing that at least most other states comply with the norm. Being a conditional cooperator regarding norms is sensible for states. Consider that norms in general, and the norm of non-aggression in particular, serve the function of securing elementary human goals, such as security. We can think about states being placed in a prisoner’s dilemma situation regarding war waging. If states are mutually vulnerable to attack by every other state, every state is in a bad situation if no effective norm limiting aggressive wars holds. Call this the belligerent scenario. Compared to the belligerent scenario, each state would be better off in the peaceful scenario, in which every state refrains from waging aggressive wars. The problem is that, for each state, there is still yet another, better option, namely to wage aggressive wars when it is in the state’s interest, while every other state refrains from waging aggressive wars against it. Since each state might succumb to the temptation to fight aggressive wars for the sake of reaping the potential benefits of war, the peaceful scenario is fragile. The peaceful scenario, however, can be stabilized by a norm of non-aggression, insofar as norm-breach is observable, diligently monitored, and can be sanctioned. The strategy of conditional cooperation is a sensible strategy for each individual state, because the cooperative element of the strategy enables states to reach the peaceful scenario if most other states cooperate. At the same time, the conditional element enables states to avoid exploitation by other states. Since being exploited in this way is even worse for states than finding themselves in the belligerent scenario in which no state feels bound by norms limiting aggression, it is more sensible for each state to be a conditional rather than an unconditional cooperator. How do conditional cooperators deal with situations in which they do not know whether the norm of non-aggression is still generally complied with or not? If intervening states can successfully mask aggressive wars under the veil of R2P, conditional cooperators will only have reason to continue their cooperation if they trust intervening states not to exploit this ability. But is it reasonable for bystander states to trust intervening states not to wage aggressive wars under the cover of R2P? The same arguments presented in the previous section apply here. Access to good evidence that intervening states are trustworthy notwithstanding, bystander states might not trust intervening states due to epistemic vice. Even if bystander states are sufficiently epistemically virtuous, however, they will often

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

have little reason to trust intervening states to testify truthfully, since they have a powerful interest to mislead, and historical experience undermines the general presumption that states testify truthfully in matters of war and peace. Insofar as bystander states fail to trust intervening states, their willingness to comply with the norm of non-aggression will diminish. In other words, bystander states will be more likely to fight unjust wars themselves. I granted at the outset that the model of state behaviour developed in this section is crude. I think the crudity is excusable, because my aim is not so much to give a realistic model, but to bring out in a very stylized way how lack of knowledge can lead to a weakening of the norm of non-aggression. To do justice to this task requires, however, considering whether the abstractions and simplifications in the model are likely to drive the conclusion I wish to draw. One abstraction that might be thought to drive the conclusion is that each state is modelled as an independent unit, whereas in reality some states have very close connections to one another, be that in the form of economic and political unions, military alliances, or other forms of cooperation. It is plausible that such connections are much more powerful ways of ensuring peace between states than the norm of non-aggression. While I think that this point affects the generality of my argument, I do not think this lack of generality is damaging. After all, many states are not tied to many other states in close forms of cooperation or mutual dependencies, and for all these cases the norm of nonaggression is crucial. A simplification that might be thought to drive the result is that I assumed all states are roughly equally powerful, such that they all benefit from an effective norm of non-aggression, and can each decide whether they wish to adhere to it. But of course some states are protected from attacks at least from single other states by their sheer military power, rather than the norm of non-aggression, and some states are militarily so weak that they are no threat to any other state. Again, I think that this point affects the generality of my argument. For the effectiveness of the norm of non-aggression, the epistemic state of some states counts more than that of others. We should be most concerned about states that are powerful enough to be in a position to engage in aggressive wars, and that would yet benefit from an effective norm of non-aggression.

Mitigating the Force of the Leeriness Objection This completes my presentation of the leeriness objection. In this section, I want to discuss the force of the objection and suggest some ways in which its force can be mitigated. The force of the leeriness objection cannot be assessed by abstract

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

THE LEERINESS OBJECTION



argument alone, because it depends on empirical assumptions. Investigating these empirical assumptions is outside the purview of this chapter, and requires the skills of political scientists rather than those of philosophers. In discussing the first premise, I have made empirical assumptions about what bystander states can, and, especially, what they cannot, know about the material conditions that need to obtain to justify a military intervention according to the standards of R2P. These assumptions are presented in two parts. First, I have claimed that bystander states must rely heavily on testimony provided by implementing institutions, in particular intervening states. Second, I have claimed that bystander states often do not have sufficient reason to trust implementing institutions in regard to their testimony. In discussing the second premise, I have claimed that the willingness of states to comply with the norm of nonaggression is conditional on their knowledge that most other states also comply with the norm of non-aggression. I have also claimed that most states do not have sufficient reason to trust most other states in this matter. I see two ways that the force of the leeriness objection can be diminished. One is to make the norm of non-aggression less dependent on the compliance of states. This would require much stronger international institutions, which would effectively establish a monopoly for the use of force between states.³³ The other possibility is to provide bystander states with better reasons to trust implementing institutions. As I have argued, making implementing institutions more trustworthy does not automatically give bystander states sufficient reason to trust implementing institutions. One way to make it easier for states to know that R2P is not abused is to invoke it only in the most salient of cases, such as in clear cases of genocide. This would, on the one hand, ease the epistemic burden on bystander states to ascertain whether the conditions of R2P are satisfied, thus mitigating the arguments that support the first premise. Perhaps even more importantly, invoking R2P only in the most salient cases might also mitigate the force of the arguments that support the second premise. This is because if R2P is only invoked very sporadically, bystander states that fail to trust intervening states might still be able to come to know that the norm of non-aggression is adhered to by most parties most of the time, and that might suffice for international institutions to secure their willingness to comply. It might therefore turn out to be best if the legal status of the doctrine remains in limbo indefinitely. Perhaps it should be ignored in most, and only be invoked in the clearest of cases. Needless to say, this strategy carries large costs, because it will sometimes require that states refrain from morally important military interventions in order to safeguard compliance with the norm of non-aggression.

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

Conclusion According to the leeriness objection, implementing R2P may weaken the norm of non-aggression. The force of the objection depends on the question of whether bystander states have sufficient reason to trust the institutions implementing R2P. I have argued that bystander states have little reason to trust implementing institutions, and thus might fail to come to know that R2P is implemented without abuse, even if this is the case. If states are conditional cooperators, this lack of knowledge could weaken the willingness of bystander states to comply with the norm of non-aggression, and thus lead to morally unjustified military interventions. The leeriness objection to R2P stands even if implementing institutions can be made fully reliable, and the principles underlying R2P are morally sound. All that is required for the leeriness objection to hold is the fact that bystander states have no way of coming to know that implementing institutions in fact are trustworthy.³⁴

Notes 1. For a detailed study of the role of order in international relations and its relationship to justice, see Hedley Bull, The Anarchical Society: A Study of Order in World Politics (Basingstoke: Palgrave Macmillan, 2012). 2. International Commission on Intervention and State Sovereignty (ICISS), The Responsibility to Protect: The Report of the International Commission on Intervention and State Sovereignty (Ottawa: International Development Research Centre, 2001). 3. UN General Assembly 2005 (A/RES/60/1, §§138–40). 4. UN General Assembly 2009 (A/63/677). 5. S/RES/1970, S/RES/1973, S/RES/2016, S/RES/2040, S/RES/2095. 6. Interactive Thematic Dialogue of the United Nations General Assembly on the Responsibility to Protect, 23 July 2009, United Nations Headquarters, available at

(accessed Aug. 2014). See in particular the contributions of Jean Bricmont, Noam Chomsky, and Miguel d’Escoto Brockmann. 7. Cecile Fabre, Cosmopolitan War (Oxford: Oxford University Press, 2012); Cecile Fabre, ‘Cosmopolitanism and Wars of Self-Defence’, in Cecile Fabre and Seth Lazar (eds), The Morality of Defensive War (Oxford: Oxford University Press, 2014), 90–114; Jeff McMahan, Killing in War (Oxford: Clarendon Press, 2009); David Rodin, War and Self-Defence (Oxford: Clarendon Press, 2002). 8. I ignore here the problem of selective intervention, which Ned Dobos has convincingly dealt with. See Ned Dobos, Insurrection and Intervention: The Two Faces of Sovereignty (Cambridge: Cambridge University Press, 2012), ch. 6. 9. Allen Buchanan and Robert O. Keohane, ‘Precommitment Regimes for Intervention: Supplementing the Security Council’, Ethics and International Affairs, 25/1 (2011): 55.

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

THE LEERINESS OBJECTION



10. Jeff McMahan, ‘The Prevention of Unjust Wars’, in Yitzhak Benbaji and Naomi Sussmann (eds), Reading Walzer (London: Routledge, 2013), 233–55. See also Rodin, War and Self-Defence, ch. 7, and for criticism Yitzhak Benbaji, ‘Against a Cosmopolitan Institutionalization of Just War’, in Yitzhak Benbaji and Naomi Sussmann (eds), Reading Walzer (London: Routledge, 2013), 256–76. 11. I use knowledge in Hawthorne’s sense as the appropriate epistemic standard for using a proposition as a premise in practical reasoning. See John Hawthorne, Knowledge and Lotteries (Oxford: Clarendon, 2004). 12. David L. Bosco, Five to Rule Them All: The UN Security Council and the Making of the Modern World (New York: Oxford University Press, 2009). 13. Note that we are interested in the moral assertion made by the Security Council, to the effect that the military intervention it authorizes meets the criteria set out by R2P. In authorizing the intervention, the Security Council also renders an authoritative legal judgment, which creates the very fact to which it attests. It is the truth value of the moral assertion we are interested in, not the performative act of rendering a legal judgment. 14. Alternatively, the bystander state could simply trust the moral assertion of the Security Council. I assume here, however, that moral testimony is either not possible, or, more plausibly, that the conditions for moral testimony do not obtain in this case. On the possibility of moral testimony, see Paulina Sliwa, ‘In Defense of Moral Testimony’, Philosophical Studies, 158/2 (2012): 175–95. 15. ICISS, The Responsibility to Protect, 35ff., 47ff. 16. Tarun Chhabra and Jeremy B. Zucker, ‘Defining the Crimes’, in Jared Genser, Irwin Cotler, Desmond Tutu, and Vaclav Havel (eds), The Responsibility to Protect (New York: Oxford University Press, 2011), §A. 17. The controversy surrounding the intentions of the Serbian army regarding Albanians in Kosovo in 1999 is still debated. See Georg Meggle, ‘NATO-Morality and the Kosovo-War: An Ethical Commentary—Ex Post’, in Georg Meggle (ed.), Ethics of Humanitarian Interventions (Berlin: Walter de Gruyter, 2004), 293–318. Similar disagreements persist regarding most military interventions, most recently in the case of the intervention in Libya begun in 2011. 18. Morten Bergsmo (ed.), Quality Control in Fact-Finding (Florence: Torkel Opsahl Academic EPublisher, 2013), chs. 3 and 4. 19. Thomas Hurka, ‘Proportionality in the Morality of War’, Philosophy and Public Affairs, 33/1 (Jan. 2005): 34–66. 20. Gareth Evans, ‘Russia, Georgia and the Responsibility to Protect’, Amsterdam Law Forum, 1/2 (2009): 25–8. 21. This is the assurance view of testimony, as developed by Angus Ross, ‘Why do we Believe What we are Told?’, Ratio, 1 (1986): 69–88, and Richard Moran, ‘Getting Told and Being Believed’, in Jennifer Lackey and Ernest Sosa (eds), The Epistemology of Testimony (Oxford: Oxford University Press, 2006), 272–306. Faulkner and Fricker have presented the approach in terms of trust and trustworthiness. See Paul Faulkner, Knowledge on Trust (Oxford and New York: Oxford University Press, 2011), and Miranda Fricker, ‘Group Testimony? The Making of a Collective Good Informant’, Philosophy and Phenomenological Research, 84/2 (Mar. 2012): 249–76.

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

22. Quassim Cassam, ‘Vice Epistemology’, The Monist, 99 (2016): 159–80. 23. J. Eric Oliver and Thomas J. Wood, ‘Conspiracy Theories and the Paranoid Style(s) of Mass Opinion’, American Journal of Political Science, 58/4 (Oct. 2014): 952–66. 24. Russel Hardin, Trust and Trustworthiness (New York: Russell Sage Foundation, 2004). Hardin is sometimes criticized for offering an account of mere reliance, rather than of full-fledged trust. Whether that is true depends on how rich a relationship needs to be to count as a trust relationship. If Hardin’s critics are right, then the assurance view of testimony should perhaps be rephrased in terms of (Hardin-style) reliance rather than trust. I do not think that anything substantial hangs on this, and stick to Hardin’s own terminology here. 25. Except for Hardin, Trust and Trustworthiness; Fricker, ‘Group Testimony’; and Onora O’Neill, A Question of Trust: The BBC Reith Lectures 2002 (Cambridge: Cambridge University Press, 2002). 26. Hardin, Trust and Trustworthiness, 3ff. 27. There might well be other reasons why people encapsulate one another’s interests. Hardin himself discusses the case that they fall in love with each other (Hardin, Trust and Trustworthiness, 142ff.). I will assume here that there is no equivalent to love at the level of relationships between states, that would take the form of shared values or close political and trade relationships, which holds between enough states to affect my argument. 28. Annette Baier, ‘Trust and Antitrust’, Ethics, 96/2 (Jan. 1986): 231–60. 29. John Rawls, Political Liberalism: Expanded Edition (New York: Columbia University Press, 2013), lecture IV, §6. 30. This principle does not hold for non-state actors such as terrorist and revolutionary groups, because they often pursue aims that are compatible with losing militarily, see Ivan Arreguín-Toft, How the Weak Win Wars: A Theory of Asymmetric Conflict (New York: Cambridge University Press, 2005). 31. See discussions of bloodless aggression in Fabre, ‘Cosmopolitanism and Wars of SelfDefence’. 32. Bull, Anarchical Society. 33. Rodin, War and Self-Defence, 163ff. 34. I am grateful to Michael Becker, Kenzie Bok, Boudewijn de Bruin, Clare Chambers, Tony Coady, Cécile Fabre, Joachim Helfer, Jens van t’ Klooster, Alex Leveringhaus, Alex Oliver, Mike Robillard, Cheyney Ryan, Rudolf Schüssler, Thomas Simpson, Shyane Siriwardena, Chris Thompson, Patrick Tomlin, and Kate Vredenburgh for comments and discussion.

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

6 On the Uses and ‘Abuses’ of Responsibility to Protect Ned Dobos

An enduring concern about the Responsibility to Protect (R2P) is that it is open to abuse. To say that a doctrine can be abused implies that it has legitimate uses from which these abuses deviate in one way or another. So what, exactly, distinguishes the proper use of R2P from its abuse? The present chapter surveys a few possible answers to this question. The first says that R2P is abused when it is invoked to publicly justify a military intervention that is not genuinely motivated by the desire to protect. These are cases where ulterior economic or political motives are concealed behind humanitarian rhetoric. The second says that any unilateral enforcement of R2P is abusive, given that only UN-sanctioned intervention is acceptable according to the statement of R2P that member states formally committed themselves to in 2005. The third interpretation says that R2P is abused when it is employed as a pretext for cultural imperialism—for instance where it is cited to justify the forcible imposition of democratic institutions on a non-democratic country, rather than being limited to the prevention of massacre or ethnic cleansing. These definitions of ‘abuse’ are by no means exhaustive, but they strike me as some of the more obvious, and those worried about the misappropriation of the R2P norm usually have one of the three in mind. I hope to show that these concerns are somewhat overblown. First, we have no reason to assume that an intervention motivated purely by humanitarian concern will produce better outcomes than an intervention where motives of national self-interest are in the mix. Given this, ‘abuses’ of the first kind are no more inherently problematic than the uses against which they are compared. Second, although there are some respects in which UN-sanctioned interventions are morally preferable to unilateral ones, there are other respects in which the latter have stronger moral

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

credentials. And finally, I will argue that most, if not all, humanitarian interventions, not just those aimed at liberalizing illiberal regimes, involve the imposition of alien values on people who do not share them. Therefore, the concept of cultural imperialism cannot be relied upon to satisfactorily distinguish legitimate R2P interventions from abusive ones.

Ulterior Motives The ICISS report that introduced us to the Responsibility to Protect idea states that any armed intervention carried out in its name should be motivated purely by humanitarian concern.¹ In other words, R2P interventions should be disinterested: ulterior economic or geopolitical motives should be absent. This gives us one possible interpretation of what it means for a state of coalition to ‘abuse’ R2P. Instances of legitimate R2P enforcement, we might say, are those sincerely motivated by the desire to protect the victims of human rights violations, while abuses of the norm are those that help themselves to its language but are motivated—at least in part—by things other than humanitarian concern.² The notion that humanitarian intervention ought to be disinterested has a long history. In 1880, over a century before the publication of The Responsibility to Protect, English jurist Sheldon Amos wrote that ‘so far as [humanitarian] intervention is concerned, it is above all desirable that the purity of the motives should be conspicuous’.³ In a more recent statement, Bhikhu Parekh asserts that humanitarian intervention should be ‘wholly or primarily guided by the sentiment of humanity, compassion or fellow feeling’.⁴ Wil Verwey goes further, arguing that ‘relative disinterest’ on the part of the intervener is necessary for an intervention to even qualify for the designation ‘humanitarian’. A state is ‘relatively disinterested’, Verwey explains, when ‘its overriding motive is the protection of fundamental human rights, without important secondary motives of a political, economic, ideological or other self-interested nature being involved’.⁵ Oliver Corten follows suit, asking rhetorically: ‘How can one claim an action is humanitarian if it clearly arises from considerations of Realpolitik, which is the only possible explanation why some states that violate the most basic human rights are let off the hook?’⁶ But why should we even care about the motives that animate intervening states? Imagine that a billionaire donates a large amount of money to charity, and as a result many people that would have died of starvation or easily preventable disease, do not die. Surely the billionaire has done a good thing. Now suppose we discover later that the billionaire had an ulterior motive all along. His donation was motivated not by sympathy or fellow feeling, but purely by a selfish desire to appear on the Forbes Magazine list of most generous

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

ON THE USES AND ‘ ABUSES ’ OF R  P



philanthropists. This may lead us to think less of the philanthropist as a person, but would we think less of the charitable act itself? Surely it remains the case that the billionaire has done a very good thing. Selfish motives taint the agent, not the actions that stem from them.⁷ If motives have no direct bearing on the moral status of actions, then why should we condemn an intervention as ‘abusive’ simply because it is not motivated purely by sympathy or humanitarian sentiment? One possible answer is that if an intervention is contaminated by ulterior motives, then it is likely that the means used to prosecute it will not be conducive to a positive humanitarian outcome. In other words, selfish motives impede the achievement of humanitarian objectives.⁸ NATO’s campaign in Kosovo has been cited as a case in point. The late Marxist historian Ellen Meiksins Wood admits that opponents of this intervention tended to oscillate between principled opposition on the one hand, and practical opposition on the other—opposition to the war only because it was ‘bungled’ or poorly executed. According to 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’ (emphasis added).⁹ The real purpose of the war, says Wood, was to put on display NATO’s military muscle, and to thereby intimidate the Milosevic government into surrendering to NATO unconditionally. This is why NATO relied so heavily on high-tech and highly destructive bombing. This, however, proved disastrous for the very people that the war was ostensibly waged to protect. It was meant to put a stop to the ethnic cleansing of the province’s ethnic Albanian population, but it only made things worse for them. On 23 March, the day prior to the commencement of the NATO 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 Kosovo entirely. Prior to the intervention, 2,500 people had been killed in the civil war between the Serbs and the Kosovo Liberation Army. During the intervention approximately 10,000 were killed, most of them Albanian civilians killed by the Serbs, but there was also some collateral damage caused directly by NATO.¹⁰ These statistics led Noam Chomsky to suggest that ethnic cleansing was in fact the consequence of NATO’s intervention, not its cause.¹¹ Australian philosopher Peter Coghlan argues that American interests had a similar effect on the recent 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 September 11. 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 humanitarian motive. The first and second motives, says Coghlan, undermined the fulfillment

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

of the third. US forces wanted to help the Iraqi people, but largely failed in this objective because they were so determined ‘to kick some ass’ along the way.¹² The implication here is that disinterest is instrumentally important: if an intervening state is motivated by economic or political self-interest, then the intervention cannot be expected to achieve a positive humanitarian outcome, because self-interested motives adversely influence conduct, and this affects consequences. The trouble is that there are plenty of counter-examples to this generalization. Sometimes it is disinterest that seems to impede the achievement of humanitarian outcomes. One reason for this is that a state acting purely out of humanitarian motives will often find itself politically hampered by its citizens’ unwillingness to suffer military casualties for the sake of the operation. For example, the humanitarian crisis in Somalia in 1992 had little perceived bearing on US national interests, but the situation attracted much media coverage so that when President Bush Sr. resolved to take action, a New York Times/CBS poll showed that he had the support of 81 per cent of the American public.¹³ That was until October 1993, when US soldiers were ambushed in Mogadishu and eighteen Rangers were brutally killed. Pictures of their mutilated corpses were broadcast back into American households. The public demanded withdrawal and the government obliged. The unwillingness to tolerate military casualties for the sake of defending human rights abroad cut the intervention drastically short, and as a consequence of the premature withdrawal Operation Restore Hope is usually to be found on lists of unsuccessful humanitarian operations.¹⁴ This is not the only way that disinterest can impede positive humanitarian outcomes. Recall Ellen Wood’s explanation for why the Kosovo intervention was such a disaster: the campaign was limited to high-altitude bombing because this is what best served the selfish ulterior motives of NATO’s member states. Yet an equally plausible explanation for the reliance on high-altitude ‘riskless’ warfare is that this method 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. On this occasion, then, it was arguably the purity of the humanitarian motive that undermined the humanitarian outcome, this time not by causing premature withdrawal, but by leading to the adoption of methods of war that were not well suited to achieving a humanitarian end. I do not mean to suggest that pure humanitarian motives invariably impede humanitarian outcomes, or even that they typically have this effect. My point is simply that they can and sometimes do. What this means, however, is that we

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

ON THE USES AND ‘ ABUSES ’ OF R  P



cannot make any valid generalizations about how self-interested interventions have worse consequences than those that are motivated purely by the desire to protect vulnerable populations. There may be some reason to believe that disinterest enhances prospects of a positive humanitarian outcome, but there is also reason to believe—and in my estimation equally compelling reason to believe— that disinterest can interfere with the achievement of humanitarian objectives. We therefore have no good reason to condemn all R2P interventions motivated by self-interest as ‘abusive’. Or, if we insist on calling these interventions abusive, we should be prepared to accept the peculiar conclusion that abuses of R2P can be morally justified sometimes.

Unilateralism In the original 2001 statement of R2P, UN Security Council authorization for armed intervention was characterized as an ideal rather than an imperative. On this view the Security Council must always be the first port of call, but should that body be unable or unwilling to act in the face of atrocities, the responsibility to protect defaults to regional organizations, ad hoc coalitions, or even individual states. By the time the UN formally adopted R2P in 2005, however, this position had been amended. In the World Summit outcome document for that year member states agreed ‘to take collective action, in a timely and decisive manner, through the Security Council . . . should peaceful means be inadequate and national authorities are manifestly failing to protect their populations from genocide, war crimes, ethnic cleansing and crimes against humanity’ (para. 139).¹⁵ This time, there were no qualifications or exceptions added to allow for unilateral R2P enforcement in case of Security Council deadlock. Presumably this change was partly in response to the hugely unpopular decision by the US and its allies to ‘liberate the Iraqi people’ in 2003 without United Nations approval. This gives us a second possible interpretation of what it means to ‘abuse’ R2P. We might say that R2P is properly used when the UN Security Council invokes it to justify a multilateral effort to defend human rights. An example might be the intervention into Libya under Resolution 1973. This was the first time in its history that the UNSC had authorized the use of force to protect a population without the consent of the de jure authorities. On the other hand, we can say that an ‘abuse’ of R2P has occurred when a state or coalition appeals to it in order to justify a military intervention that is not authorized by the UN Security Council. Morally speaking, why is it so important for the UN Security Council to approve military interventions? There are broadly two kinds of answers to this

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

question. An instrumentalist answer says that UNSC approval is important because it tends to contribute to the achievement of desirable outcomes. An intrinsicalist answer, by contrast, says that UN authorization is morally necessary for reasons that are independent of whether this makes an intervention more or less likely to produce good consequences. For instance, one might say that every UN member state has implicitly agreed not to engage in humanitarian intervention unilaterally. To intervene unilaterally therefore constitutes a broken promise or breach of trust against other members of the international community.¹⁶ Let us interrogate each kind of answer in turn. The instrumentalist argument suggests that multilateralism helps to remove or mitigate one of the chief impediments to the success of any humanitarian intervention, namely resistance from within the target society. As Michael Walzer observes, some measure of resistance from the local population is always to be expected. But it is likely to be all the more fierce, widespread, and protracted when the local population is hostile to, or suspicious of, the intervening power.¹⁷ ‘When you don’t trust the messenger, you don’t trust the message, even if it’s a good one’.¹⁸ Now since there is a great deal of antagonism towards those states most inclined to go it alone—such as the US—this can be expected to aggravate the local resistance to any unilateral action they might embark upon, which of course increases the risk of mission failure. On the other hand, the animosity towards interveners is arguably reduced when a genuinely diverse coalition is involved; a multilateral effort will not arouse the same level of suspicion or resentment. In this respect, at least, multilateralism might be thought to improve an intervention’s prospect of delivering a positive humanitarian outcome, by mollifying the resistance to it somewhat. In this connection UN Security Council authorization is held up as the fullest realization of the ideal, and perhaps for good reason. The Security Council is comprised of five permanent members (the P5) and ten non-permanent elected members (E10), always including representatives from Europe, Africa, Asia, and the Americas. All of the major races and religions, large and small countries, capitalist and non-capitalist economies, and democratic and non-democratic states are represented. Plausibly, local populations are likely to be more receptive to a humanitarian intervention that is carried out by an organization that is in this way representative of humanity, rather than a state or coalition with its own parochial character and interests, and chequered history. Further to this, multilateralism helps to distribute the costs and risks associated with carrying out a humanitarian intervention, which makes it more politically and economically sustainable for those involved. This is important because intervention is unlikely to achieve anything lasting if the interveners are compelled

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

ON THE USES AND ‘ ABUSES ’ OF R  P



by domestic political or budgetary pressures to withdraw in a hurry. If the underlying issues that triggered the rights abuses to begin with are not dealt with, there is a good chance that the abuses will resume as soon as the interveners move out.¹⁹ The violations will not have been prevented or ended, as per the aim of armed humanitarian intervention. They will merely have been suspended or postponed. A lasting humanitarian outcome will typically demand some form of ongoing presence, and multilateralism helps make this possible. These claims for why UNSC authorization is instrumentally important are plausible enough, but once again we have some considerations pulling in the opposite direction here. First, a rapid response is crucial for the success of almost any humanitarian intervention. During the Rwandan genocide, approximately one million Tutsis and moderate Hutus were killed in the space of 100 days. That amounts 10,000 murders every day, 400 per hour, 7 per minute.²⁰ Since a UN mandate comes via negotiation and deliberation, it invariably brings delays that often cannot be afforded. In cases where the UN seeks to prosecute intervention under its own command the delays are further exacerbated. This is partly because the UN does not have a standing army. It needs to raise a military force from scratch on every occasion, cobbling together whatever troops and equipment its member states are willing to offer at the time.²¹ In addition to this, Taylor Seybolt suggests that soldiers provided to the UN as part of any such multilateral operation will tend to be undertrained and poorly equipped. The UN Security Council is described as being ‘extremely sensitive’ to maintaining political balance and fair representation of different states and regions of the world. I just suggested that this can enhance prospects of success by reducing suspicion of and hostility towards the intervening force, but at the same time it can prevent the UN from selecting mission personnel purely on the basis of their qualifications and abilities. Seybolt writes: ‘Many UN mission personnel are highly dedicated and good at what they do, but many are not . . . [making] UN operations militarily weaker than [those of] voluntary coalitions’.²² A commitment to multilateralism can also run contrary to basic military understandings of unified command.²³ During the Yugoslav civil war an agreement was reached to have NATO provide air support to UN peacekeepers on the ground protecting the Bosnian population inside the designated safe areas. Each air strike required approval not only from NATO—which has its own sometimes cumbersome collective decision-making procedures—but also from the civilian leadership of the UN. There were effectively two chains of command, each wielding veto power over the other. This hindered the effective use of NATO airpower. What happened next was one of history’s most shameful episodes: a ‘safe area’ was overrun by Serb forces and thousands of innocent civilians were

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

massacred. While the legitimacy conferred by the multilateral decision-making arrangements may have rendered the intervention more politically acceptable to international society, on this occasion it seems to have compromised the humanitarian outcome that was the objective of the operation.²⁴ In several respects, then, multilateralism can undermine an intervention’s prospect of success. On account of this, the instrumentalist argument for the necessity of UNSC authorization runs into the same difficulties as the argument for the importance of pure humanitarian motives, to wit: we have no reason to generalize, or even to presume, that a UN-authorized multilateral effort will produce better consequences than a unilateral one. This suggests that we must ultimately take recourse to some intrinsicalist argument for why UNSC authorization is morally necessary—one that does not rely on any assumptions or generalizations about how UN approval or lack thereof impacts on the consequences of R2P enforcement efforts. There are at least two intrinsicalist arguments worth distinguishing. The first appeals to the rights of the state targeted by a unilateral humanitarian intervention, while the second appeals to the rights of other states in the international community.

The target state What is it that justifies a government in issuing commands and compelling its citizens to obey? How is this to be reconciled with the moral equality of persons? A common answer is consent. If I consent to your promulgation and enforcement of law, your exercise of this authority is consistent with my autonomy and equality. This gives us some insight into what makes vigilantism objectionable. Insofar as he does not enjoy the consent of the people that he coerces into compliance with the law, the vigilante treats them in a way that is not consistent with their moral status. On the other hand, it is acceptable for the police and courts to enforce the law and administer punishment because the state does enjoy the consent that the vigilante lacks. From this we might extrapolate that unilateral intervention—the international analogue of vigilantism—constitutes a wrongdoing against the target state. If we take it that the 2005 R2P commitment empowers the Security Council alone to engage in or to authorize humanitarian intervention, then we can say that every UN member state, being party to that commitment, has consented to the Security Council’s right to intervene. But no state has consented to unilateral law enforcement, or to enforcement by ad hoc coalitions and regional organizations for that matter. Hence unilateralism wrongs the target state in the same way that vigilantism wrongs its victim. This argument is faulty. To see where it goes wrong we need to consider more carefully law enforcement at the domestic level. On closer inspection it appears

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

ON THE USES AND ‘ ABUSES ’ OF R  P



that whether or not the law-breaker is wronged by the vigilante depends on the nature of his offence. Suppose that your neighbour, an ordinary citizen, spots you jaywalking or not wearing a seatbelt while driving, and takes it upon himself to force you to comply with the law. There is a definite sense in which you are wronged here. But what if your neighbour were to stop you from beating or killing your spouse? Surely anyone can legitimately prevent an assailant from murdering or maiming his victim. No special authority is needed. The difference between the two cases seems to be this: it is morally wrong to assault people independently of its being illegal. Jaywalking, on the other hand, is not something that is independently proscribed by morality. In legal parlance, it is malum prohibitum: wrong merely because prohibited by statute, as opposed to malum in se, or wrong in and of itself. It seems to me that a ‘victim’ of vigilantism has grounds for complaint only if he is forced by the vigilante to refrain from behaviour that is malum prohibitum. The same can be said for law enforcement in the international arena. Some international treaties, especially those dealing with commerce and communications, require acts and forbearances that are morally neutral. With respect to these laws one could plausibly say that unilateral enforcement wrongs the target state. But respect for human rights is not a convention-dependent obligation which has no force outside of international law. Rather, like an individual citizen’s obligation not to commit assault, a state’s obligation to honour human rights is pre-institutional, and these are requirements that any party may enforce. Before moving on I should make one important qualification. Nobody would deny that it is permissible for an ordinary citizen to intervene in a domestic dispute in order to prevent his neighbour from beating or killing his spouse. But this example presupposes that there is no time to call the proper authorities, such that private law enforcement is necessary to prevent the crime. But what if there is time to call the police? I take it that, because vigilantism is unnecessary in this case, it is also unjustified, and this is despite the fact that the offence being committed is malum in se. By analogy, where there is time to engage the UN in dealing with a humanitarian crisis, a state or coalition that bypasses the UN and acts unilaterally is rightly condemned. This, however, is not because the state targeted by the unilateral action is wronged in any sense. Rather, it is because the international community is wronged. This brings me to the second intrinsicalist argument for why UN authorization is morally important.

The international community We have seen that the very first incarnation of the R2P doctrine in 2001 allowed for military intervention without a UN mandate in some circumstances, but the

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

later version that member states formally committed themselves to in 2005 made no such allowances. Arguably, by formally adopting the 2005 version, each member state of the UN implicitly agreed to relinquish henceforth any right to enforce R2P unilaterally. It follows that a member state breaches this commitment—and thus violates the legitimate expectations of its co-signatories—by engaging in R2P intervention without UN approval. This argument is I think compelling where (1) the intervention proceeds without a UN mandate having been sought, and (2) the intervening state/coalition cannot plausibly claim that there was no time to engage the UN, or that authorization would certainly have been denied. Matters are different where authorization is sought but refused, or where the intervening state can plausibly claim that there was either no time to pursue a mandate, or no chance of winning one. Consider a familiar thought experiment: You make a promise to meet a friend at a certain time in a certain place for a trivial reason. On your way there you come across the scene of an accident that has left a person injured and in need of urgent attention on the side of the road. If you stop to render assistance, you will not be able to meet your friend, and therefore you will break your promise. Nevertheless, most would agree that this is permissible, if not obligatory. The explanation, crudely put, is that obligations arising out of one’s promises are pro tanto and need to be weighed against countervailing moral considerations— including conflicting duties—before any all things considered moral judgement can be made. In the thought experiment the countervailing considerations are so weighty that they justify you in defaulting on your pro tanto obligation to your friend. As with obligations arising out of interpersonal promises, I submit that obligations arising out of a state’s international commitments are pro tanto. To be sure, these commitments should not be seen as trivial, and so a very strong justification needs to be provided for their transgression. Nevertheless, in some cases there will be sufficiently weighty countervailing considerations. For example, suppose that a state must break with its international commitments in order to ensure that egregious human rights abuses beyond its borders are brought to an end. This is precisely the predicament of a state or coalition that seeks a UNSC mandate for a humanitarian intervention but does not receive one. It is also the predicament of a state that does not have time to appeal to the UN, or that has strong grounds for believing that a mandate will not be forthcoming even if the case is made to the Security Council. The state can prevent a great evil, but it must breach the terms of its international agreements to do so. Unilateral intervention might in these cases

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

ON THE USES AND ‘ ABUSES ’ OF R  P



constitute some kind of infringement against the international community, but it is not a violation. That is, it is not an unjustified infringement.²⁵ The imperative to prevent the violation of basic human rights simply trumps fidelity to international agreements, just as the imperative to render urgent assistance at the scene of an accident takes moral priority over one’s promise to meet a friend. So, if the ‘abuse’ of R2P is understood to mean its unauthorized enforcement or implementation, again I think we must conclude that the abuse of the norm is not an especially troubling prospect. Humanitarian interventions that are ‘abusive’ in this sense can be morally justified. On occasion they may even be morally preferable to UN-sanctioned interventions, all things considered.

Cultural Imperialism Some will argue that any armed intervention into a foreign country, if it is aimed at enforcing particular standards of conduct within that country, amounts to cultural imperialism: the imposition of alien values on a community that does not share those values. A more moderate view is that only some interventions are open to this charge. Michael Walzer’s is the best-known argument for the latter position, so that is what I propose to concentrate on. A state that is oppressive by Western, liberal standards might be governing its people ‘in accordance with their traditions’ says Walzer. In such cases there is a bond or ‘fit’ between the rulers and the ruled that outsiders are forbidden to upset.²⁶ To compel a government that enjoys some such ‘fit’ to change the way that it governs, or to violently depose it, is to deny its subjects their right to a state that embodies their customs and values, and that they can truly ‘call their own’.²⁷ In fact, foreign intervention is prohibited as long as we can presume that a ‘fit’ obtains according to Walzer, and this is a presumption that he thinks we must make in all cases. Ordinarily, foreigners are in no position to judge the reality of that union [fit], or rather, they are in no position to attempt anything more than speculative denials. They don’t know enough about its history, and they have no direct experiences and can form no concrete judgments, of the conflicts and harmonies, the historical choices and cultural affinities, the loyalties and resentments, that underlie it.²⁸

For this reason, Walzer says we must always presume that an illiberal government fits with its people, and accordingly we must honour the principle of nonintervention vis-à-vis that government. To intervene would be to force our political customs and values on people who do not share them.

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

In the event of genocide or ethnic cleansing, however, Walzer says that the absence of fit becomes ‘radically apparent’.²⁹ We can be sure that this is not simply a state governing its people in accordance with their traditions; it is compelling evidence of deep discord between the rulers and the ruled. The presumption of fit is thus defeated. In these cases, and only in these cases according to Walzer, intervention against the state does not constitute a form of cultural imperialism. It does not involve denying people their right to be governed in accordance with their traditions, since the government targeted by the intervention is patently not performing this function any longer. This gives us our third possible interpretation of what it means for R2P to be abused. Let us say that R2P is properly used when it is invoked to justify intervention against a state engaged in genocide or ethnic cleansing, since intervention in these cases is not imperialistic (given that the presumption of fit is overturned). On the other hand, if R2P is cited to justify intervention against a state engaged in ‘ordinary’ oppression, like the denial of free speech and democracy, then the doctrine is abused.³⁰ In these circumstances the presumption of fit still stands, so R2P is being used as a pretext for cultural imperialism. These abusive interventions are objectionable because they do not make allowances for cultural differences, effectively requiring all states to take the form dominant in the West, namely that of a free-market liberal democracy. I am not convinced that uses and abuses can be satisfactorily distinguished this way. To see why, let us flesh out in a little more detail Walzer’s notion of ‘fit’. At times he seems to be saying that a state fits with its people as long as it governs them in accordance with their customs and traditions, period. But this cannot be all there is to it. It is after all possible for the members of a historic nation to grow averse to their inherited culture, or at least to its political manifestation. A government that continues to force it down their throats regardless cannot gain any kind of moral standing by so doing, irrespective of how intrinsically valuable Westerners happen to find the culture being perpetuated. For there to be a morally significant union between the rulers and the ruled the former must also enjoy the allegiance of the latter. Walzer does acknowledge this, at least implicitly. The presumption of fit, he says, is based on the presumption that a people will defend their state against foreign encroachment, proving that they value it enough to fight and to die for it.³¹ ‘Fit’ should thus be understood as a state expressing the inherited culture of its people and enjoying their allegiance because of it. This is the kind of relationship that is to be presumed to exist in all illiberal states. Next we need to determine precisely whose allegiance is to be presumed. Is it that of every citizen unanimously, or simply that of a significant proportion of the

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

ON THE USES AND ‘ ABUSES ’ OF R  P



citizenry? If a state enjoys the unanimous allegiance of its people then it clearly has a right against outside interference. But this is just not something that can be reasonably presumed, ever. National cultures and traditions are almost always contested. Where some citizens see an expression of the ‘national soul’, others will see nothing but oppression and corruption, or even a perversion of their inherited culture. Those who feel that their government accurately embodies their national identity may very well remain loyal to it despite being abused and neglected. (Many of those tortured to death at the behest of Ivan the Terrible are said to have died praying for the Tsar. Centuries later prisoners incarcerated in Communist Russia’s gulags would become frantic with grief upon hearing the news of Stalin’s death.³²) But it is safe to say that, in every single illiberal society, one will also find disgruntled citizens who deny that the state expresses their culture, who feel no allegiance towards their government, and who would welcome foreign troops that arrive to depose it, or at least to force a change in its behaviour. By ‘fit’, Walzer can therefore only mean the allegiance of a significant proportion of the citizenry; this is the most that can ever be reasonably presumed. But if that is all that ‘fit’ consists of, then even regimes that engage in genocide or ethnic cleansing might ‘fit’ with their people. In Hitler’s Willing Executioners, Daniel Goldhagen paints a picture of Nazi Germany in which the people and the party ‘fit’ one another more snugly than most historians have let on.³³ Goldhagen claims that an ‘eliminationist’ antiSemitism had evolved into ‘an axiom of German society’ by the time Hitler came to power—eliminationist because it was conducive to a particular kind of solution to ‘the Jewish problem’.³⁴ His claim is essentially that the Nazi persecution and attempted extermination of the Jews was a policy that mirrored and expressed the regnant culture of the day. Goldhagen may well have overstated the extent and the intimacy of the fit between the German people and the Nazi regime, as many critics of his work have since charged. But the important point for my purposes is just that the occurrence of genocide failed to make it ‘radically apparent’ that the state did not enjoy the allegiance of substantial numbers of its citizens. The Nazi regime ‘fit’ the German population despite engaging in genocide. The same can be said for China between the years 1966 and 1969, during which scores of people were tortured, displaced for the purposes of ‘re-education’, killed, and reportedly even cannibalized in the name of the Cultural Revolution.³⁵ Popular involvement in the state-sanctioned atrocities was unprecedented. Jonathan Glover goes so far as to suggest that ‘the crushing repression which drove so many to mental breakdown or suicide was carried out

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

by nearly everyone’.³⁶ Far from exposing a rupture in the relationship between the Chinese people and their state, the mass basic rights violations that characterized these years only served to confirm the public’s loyalty towards Chairman Mao. Research conducted in what was Yugoslavia in the mid-1990s suggests that substantial numbers of Serbs probably remained loyal to their government throughout its attempts to expel the ethnic Albanian population from Kosovo. A 1993 survey revealed that more than 50 per cent of Serb respondents believed that ‘all Albanians are primitive and uncivilized’. The following year a poll showed that only 52 per cent of Serbs found the idea of sharing a country with Albanians tolerable, some of whom seem to have responded with a very minimal conception of sharing in mind: only 33 per cent said that they were willing to socialize with Albanians. Yet as little as 1.9 per cent favoured autonomy for the province, with 41.3 per cent supporting ‘strict state measures’ to bring Kosovo into line.³⁷ The upshot is this. Even states that engage in genocide or ethnic cleansing sometimes ‘fit’ their people. And so even interventions aimed at preventing these crimes are open to the charge of intruding on a ‘fit’ and imposing values on people who do not share them. Walzer is mistaken to think that the cultural imperialism objection is one that applies exclusively to interventions against ‘ordinarily oppressive’ regimes. It applies to every intervention, and cannot therefore be relied upon to separate the proper uses of R2P from its abuses.

Conclusion The worry that R2P is open to abuse remains ubiquitous, but just what constitutes an abuse of the norm has not been given much rigorous philosophical attention. By way of trying to remedy this neglect I have outlined three possible understandings of what it means to abuse R2P. The first identifies abusive R2P interventions as those that are not motivated purely by altruistic concern. The second identified abusive interventions as those that are not duly authorized. And the third identifies abusive interventions as those that are culturally imperialistic, the argument being that only a subset of all interventions are open to this objection. I have argued that, if we accept any of these definitions of ‘abuse’, then abusive interventions can still be morally justified. Indeed, they are often no more morally suspect than the instances of non-abusive R2P enforcement against which they are compared. Having said this, I have acknowledged that these conceptions of abuse are not exhaustive. There may be other forms of R2P abuse not surveyed here that do warrant grave moral concern. If so, I hope that this short chapter will motivate others to be more explicit in identifying and interrogating them.

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

ON THE USES AND ‘ ABUSES ’ OF R  P



Notes 1. International Commission on Intervention and State Sovereignty, The Responsibility to Protect (Ottawa: International Development Research Centre, 2001). 2. For a fuller discussion of ulterior motives in humanitarian intervention, see Ned Dobos, ‘Consistency in the Armed Enforcement of Human Rights Norms’, Journal of Moral Philosophy, 8 (2011): 92–109, reprinted in Thom Brooks (ed.), Just War Theory (Leiden: Brill, 2013). 3. Amos, Political and Legal Remedies for War (New York: Harper, 1880), 159. Quoted in Saban Kardas, ‘Humanitarian Intervention: A Conceptual Analysis’, Alternatives: Turkish Journal of International Relations, 2/3–4 (Fall and Winter 2003): n. 51. 4. Quoted in Helen Frowe, ‘Judging Armed Humanitarian Intervention’, in Don E. Scheid (ed.), The Ethics of Armed Humanitarian Intervention (Cambridge: Cambridge University Press, 2014), 96. 5. Wil. D. Verwey, ‘Humanitarian Intervention in the 1990s and Beyond: An International Law Perspective’, in Jan Nederveen Pieterse (ed.), World Orders in the Making (London: Macmillan Press, 1998), 201. 6. Oliver Corten, ‘Humanitarian Intervention: A Controversial Right’, UNESCO Courier, 52/7 (July–Aug. 1999): 59. 7. Ned Dobos, Insurrection and Intervention: The Two Faces of Sovereignty (Cambridge: Cambridge University Press, 2012), 66–7. 8. I explore this question in ‘Idealism, Realism, and Success in Humanitarian Intervention’, Philosophia, 44/2 (2016): 497–507. 9. Ellen Meiksins Wood, ‘Kosovo and the New Imperialism’, in Tariq Ali (ed.), Masters of the Universe? NATO’s Balkan Crusade (London and New York: Verso, 2000), 196. 10. Michael Mandelbaum, ‘A Perfect Failure: NATO’s War Against Yugoslavia’, Foreign Affairs, 78/5 (Sept./Oct. 1999): 3. 11. Noam Chomsky, ‘A Review of NATO’s War over Kosovo’, Z-Magazine (Apr.–May 2001). 12. Peter Coghlan, ‘War and Liberation’, in Raimond Gaita (ed.), Why the War was Wrong (Melbourne: Text Publishing, 2011), 143–4. 13. See William J. Durch, ‘Introduction to Anarchy: Humanitarian Intervention and ‘State Building’ in Somalia’, in William J. Durch (ed.), UN Peacekeeping, American Policy, and the Uncivil Wars of the 1990s (New York: St Martin’s Press, 1996), 320. 14. Richard Falk writes that ‘the higher value attached to the lives of the intervening side . . . led to a callous disregard for the lives of the indigenous people, especially civilians’, who were killed indiscriminately by American forces. The battle that killed eighteen US Rangers claimed the lives of several hundred Somalis, a statistic that history forgot. Richard Falk, ‘Intervention Revisited: Hard Choices and Tragic Dilemmas’, The Nation (20 Dec. 1993): 760. 15. United Nations General Assembly, 2005 World Summit Outcomes, paragraph 139, online at , accessed Oct. 2016. 16. I have introduced these arguments elsewhere. See ch. 7 of Ned Dobos, Insurrection and Intervention.

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

17. Michael Walzer, ‘The Moral Standing of States: A Response to Four Critics’, Philosophy and Public Affairs, 9/3 (1980): 209–29. 18. Shibley Telhami, ‘Double Blow to Mideast Democracy’, Washington Post (1 May 2004). 19. Edward Luttwak, ‘Give War a Chance’, Foreign Affairs, 78/4 (July/Aug. 1999): 36–44. 20. Veronique Mistiaen, ‘Offering Dignity to Rwanda Genocide Victims: Survivor Confronts the Horror to Honor the Dead and Living’, San Francisco Chronicle (9 Apr. 2004). 21. Seybolt, Humanitarian Military Intervention: The Conditions for Success and Failure (New York: Oxford University Press, 2007), 273. 22. Seybolt, Humanitarian Military Intervention, 273. 23. Eric Heinze, Waging Humanitarian War: The Ethics, Law, and Politics of Humanitarian Intervention (Albany, NY: State University of New York Press, 2009), 119. 24. Heinze, Waging Humanitarian War, 120. 25. The distinction between ‘infringement’ and ‘violation’ was introduced by Judith Jarvis Thomson in ‘Some Ruminations on Rights’, Arizona Law Review, 19 (1977): 45–60. 26. Walzer, ‘Moral Standing of States’, 212. 27. Walzer, ‘Moral Standing of States’, 226. 28. Walzer, ‘Moral Standing of States’, 212. 29. Walzer, ‘Moral Standing of States’, 214. 30. The term ‘ordinary oppression’ comes from Walzer himself. ‘Moral Standing of States’, 218. 31. Walzer, ‘Moral Standing of States’, 213. 32. Daniel Myerson, Blood and Splendor: The Lives of Five Tyrants, from Nero to Saddam Hussein (New York: Harper Collins, 2000), 84–5 and 168. 33. Daniel J. Goldhagen, Hitler’s Willing Executioners: Ordinary Germans and the Holocaust (New York: Knopf, 1996). 34. Goldhagen, Hitler’s Willing Executioners, 29. 35. Marcus Mabry, ‘Cannibals of the Red Guard: Smuggled out of China, Evidence of Atrocities’, Newsweek (18 Jan. 1993). 36. Jonathan Glover, Humanity: A Moral History of the Twentieth Century (London: Jonathan Cape, 1999), 292. Emphasis added. 37. L. Nikolic, ‘Ethnic Prejudices and Discrimination: The Case of Kosovo’, in Florian Bieber and Zidas Daskalovski (eds), Understanding the War in Kosovo (London and Portland, OR: Frank Class, 2003), 55–6.

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

7 Scrutinizing Intentions Chrisantha Hermanson

Though R2P has gained international support as a developing norm since its debut, the principle has been, from its start, accompanied by a worry that if we accept R2P we may open the door for abuse at the hands of international actors acting under the guise of R2P to pursue self-interested rather than genuinely humanitarian ends (for ease of exposition I term this the abuse-worry).¹ The purpose of this collection is to critically address the abuse-worry and, in particular, the possibility that states, motivated by geopolitics or power politics, will misuse R2P to pursue their own interests. This necessarily puts the motivations/ intentions of potential interveners under scrutiny. Accordingly, the overarching aim of this chapter is to provide a normative analysis of intervention in cases where interveners meet the just war requirement of just cause (and, we will assume, all other just war requirements) but fail, in whole or in part, to meet the requirement of right intention. To begin, consider the following:² Abuse-worry₁/Military Power-1: Mass-atrocities are occurring in State-B. State-A militarily intervenes and stops the mass-atrocities. State-A effectively ends the mass-atrocities and cites R2P as its justification for intervention; however, State-A’s primary motivation during the intervention is to put on a show of military power within the region surrounding State-B and intimidate its, State-A’s, potential competitors (States C and D). Abuse-worry₂/Military Power-2: Same as Military Power-1 except that while State-A intends to show off its military power primarily by intervening and stopping the mass-atrocities, it also plans to build a permanent military base inside State-B in order to cement its power within the region. Abuse-worry₃/Military Power-3: Mass-atrocities have just ended in State-B but post-atrocity conditions remain volatile. Without significant military and economic aid from the international community, it is a near certainty that

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

State-B will suffer economic collapse and the resumption of atrocities and, as a result of both, the lives of the target-population remain in grave danger. StateA agrees to intervene, providing both military and economic support on the condition that State-B allows State-A to build a permanent military base inside its territory. Though State-A cites R2P as its reason for intervening and for building the military base (it can provide State-B continued protection this way), its true motivation for both is to make a statement of military power to other states and gain military dominance in the area. In all three scenarios, there are ongoing/impending mass-atrocities which State-B is unwilling/unable to stop, and State-A (and any other state) has a prima facie just cause for intervention.³ However, in each case State-A fails to meet the just war requirement of right intention (as it is commonly understood) and, therefore, might be thought to abuse R2P because it intervenes for reasons of power politics rather than the humanitarian principles enshrined in R2P. Each scenario presents us with a different way by which an actor might fail to meet the requirement of right intention (henceforth abbreviated RI) and we will discuss each in turn. Before we begin, let us pause and define some key terms. International community is herein understood in a broad sense to mean states (and their individual members), coalitions of states, and supra-state entities such as the AU and the UN. Target-state is herein defined as the state in which mass-atrocities are or have taken place and target-population is understood as the population living in the target-state. Next, the term secondary aim is used to indicate that an agent/ intervener has a tangible objective other than/in addition to obtaining the just cause, the achievement of which requires that agent to take actions outside of those necessary for the just cause. The use of the term secondary, however, is not meant to indicate that this aim is of less importance to the agent; whether the secondary aim is of primary or secondary priority to the agent is not taken up here. Mass-atrocities are defined as those instances of atrocities (killings, torture, rape, etc.) in which at least 1,000 people are killed annually.⁴ Finally, for our purposes, fundamental human rights are understood to include: (a) the right to physical integrity (especially the right not to be killed, tortured, etc.); (b) the right to autonomy (the right to a significant degree of freedom of thought and action); and (c) the right to community (the right of individuals to form, shape, and maintain valuable relationships with others—to include the right to family and national self-determination). Admittedly, this is a very abridged account of fundamental human rights which I cannot defend here, but for our purposes herein it will prove sufficient.⁵

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

SCRUTINIZING INTENTIONS



Abuse-Worry₁ I have placed my discussion of the potential for abuse of R2P, motivated by geopolitics and/or power politics, within the framework of just war theory. Just war theory provides fairly clear guidelines with which to determine whether a war/intervention is just or unjust (in whole or in part) and, in so doing, also provides observers with a relatively reliable way of detecting the abusive use of force. It is no wonder, then, that the 2001 ICISS report adopts many just war requirements as requirements of R2P. By incorporating just war criteria within the principle of R2P, the drafters of R2P have also given us—the international community—a fairly clear way to detect and/or prevent abuse of R2P. The ICISS report specifically lists just cause, right intention, last resort, ‘proportional means’, likelihood of success, and right authority, as requirements of R2P.⁶ For the purpose of this chapter, let us grant that an actor has a just cause for intervention when mass-atrocities are ongoing or imminent.⁷ Let us further grant (for the sake of limiting and focusing our discussion) that the actors in each scenario have met the requirement of just cause and all of the jus ad bellum requirements except for RI. Having claimed that abuse-worry₁, abuse-worry₂, and abuse-worry₃ are (at least) violations of RI, we need to develop an idea of what this requirement entails. RI requires that the objectives one aims to achieve in war/intervention be limited to defending and securing one’s just cause.⁸ So, while just cause tells us what causes are grounds for war, RI demands that one limit one’s actions in war/ intervention to achieving only those causes. The purpose of RI is to prevent agents from engaging in such things as territorial expansion, conquest, economic aggrandizement, purely self-interested regime change, etc. Understood in this rather broad form—the objectives one aims to achieve in war/intervention must be limited to defending and securing one’s just cause—RI seems relatively unproblematic. However, when we focus on the concept of intention, a serious complication arises. An actor’s intention can be understood in (at least) two ways. First, what I term the ‘action-component’, we might say an actor intends X if his main objective/aim is to do X. Second, what I term the ‘motivation-component’, we might say that an actor intends Y if his underlying motivation/reason for doing X is to achieve Y, where X and Y may be the same or different. The focus of X being what is done, while the focus of Y being why it is done. To see this distinction, imagine that Freddy sees an elderly woman, Wilma, trip and fall in the cross-walk of a busy street. He helps her up and assists her to safety. He helps her—X—because he wants to help her and make sure she is safe—Y. In this case what Freddy actually does—X—and why he does it—Y—are

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

the same. Now suppose that Freddy sees Wilma fallen in the cross-walk and plans to walk by her when he also notices Vivica, who he longs to impress, watching from inside a stopped bus. Freddy now runs to Wilma’s aid. Here Freddy helps Wilma, X, because he wants to impress Vivica, Y, and though Y leads to X, they represent two distinct intentions; the action-component and the motivationcomponent. The intervention by India in East Pakistan in 1971 also illustrates this distinction. A common understanding of this intervention (supported by the claims made by the Indian government at the time) is that India intervened in East Pakistan with the purpose/objective of stopping the mass-atrocities being committed there—this was their tangible military goal, X. Thus, India met the action-component of RI. However, India’s underlying motivation/reason for pursuing this objective was to stop the flow of Pakistani refugees into India, Y. Thus, India did not meet the motivation-component of RI. When many theorists speak of RI, they include both the action-and the motivation-component under the umbrella of RI. However, there is a significant problem with the motivation-component. This component is absolutely reliant on self-reflection and self-regulation; outsiders are unlikely to know the underlying motivations of an actor prior to a war/intervention (though there may be indicators of likely motivations). Accordingly, when it comes to the ability to judge the justice of other people’s interventions, the motivation-component will usually have to be assessed retrospectively based upon what the actor actually did during the intervention (and may not even be clearly discernible then) and is, thus, not a reliable criterion for observers before a war/intervention. Moreover, when we are scrutinizing the motivations of actors within the international community we will certainly find a multitude of motivations which cannot always be easily untangled. Most importantly, upholding the motivation-component as a requirement of war/intervention may unnecessarily (perhaps even unjustly) prohibit otherwise just wars/interventions. Though India did not meet the motivation-component of RI, I think most of us will agree that the intervention was just; the intervention (arguably) met all other just war requirements, saved hundreds of thousands of people from extermination, and allowed millions of refugees to return home.⁹ So, while we may hold that it is preferable for an actor to have only humanitarian motivations (thereby meeting the motivation-component of RI), I submit that the requirement of RI must be limited to the action-component.¹⁰ Let us examine this claim and its implications further. Consider again the abuse-worry₁ scenario. Here State-A’s objective is to intervene and stop the mass-atrocities occurring in State-B. State-A’s motivation for intervening is to make a statement of military power to neighbouring states. If we grant that RI must be limited to the

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

SCRUTINIZING INTENTIONS



action-component, then we must also grant that State-A meets this requirement and (so long as it meets the other just war requirements) its intervention is just. The case becomes even stronger when we consider a very real alternative—that State-A is denied ‘permission’ to intervene on the grounds that it does not meet the motivation-component, State-A does not intervene, no other agents intervene, and millions of people in State-B continue to be killed. Of course this does not mean that we cannot criticize State-A’s underlying motivation—it might be preferable if State-A’s motivation was simply to end the atrocities taking place in State-B. Given these conclusions, in order to more accurately describe scenarios such as abuse-worry₁ and the Indian case, we must not attach the term ‘abuse’ to these interventions; rather we might say that these interventions are justified though, perhaps, not ideal. Before moving on, let us push the conclusion that the motivation-component of RI is preferable but not required and ask whether there may be cases where an agent’s motivation is objectionable to such a degree that intervention would, in fact, become unjust. Consider the following: Political Influence-1: Mass-atrocities are occurring in State-B. State-B is predominantly made up of population-Y and population-Z, politically and ethnically divided groups. The mass-atrocities are largely being carried out by militias from population-Y against civilians from population-Z. State-A militarily intervenes, ends the mass-atrocities, helps establish post-atrocity stability, and soon withdraws its forces from State-B. State-A cites R2P as its justification for intervention; however, State-A’s primary motivation during the intervention is to shift the position of dominance within State-B from population-Y to population-Z, with whom it shares ethnic/historical ties. By protecting population-Z and thereby increasing its power, State-A also hopes that it will have more political influence over State-B. Political Influence-2: Same as Political Influence-1, except that to achieve its objectives State-A establishes a post-atrocity government dominated by population-Z (without holding fair elections) and trains and arms a strong military manned by members of population-Z. In both scenarios, State-A has two motivations: stop the atrocities and shift the balance of power from population-Y to population-Z. More strongly, State-A hopes to create conditions whereby population-Z can dominate population-Y. This additional motivation is clearly an unjust/abusive reason for intervention. However, even in this case it seems to me that we must maintain that where this unjust motivation is not accompanied by actions outside of those needed to

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

achieve the just cause, the intervention, on the whole, remains justified though not ideal. Accordingly, in Political Influence-1, though we may certainly criticize State-A’s motivation and correctly label it as unjust, because State-A does not take any actions outside of those necessary to stop the atrocities, its intervention remains just.¹¹ By contrast, in Political Influence-2, because State-A takes additional actions outside of the pursuit of its just cause to achieve its unjust motivations, its intervention is unjust—it has violated the action-component of RI. Having concluded that interventions which fall under the category of abuseworry₁ do not actually abuse R2P, though they may not be ideal, let us now explore abuse-worry₂.

Abuse-Worry₂ Abuse-worry₂ presents us with an obvious question. Does an agent who has and pursues secondary aims through actions outside of those needed to achieve the just cause necessarily violate the action-component of RI and, if so, does this render the intervention unjust/abusive? In this section we will break down this question into three parts. First, is it possible for a secondary aim to fall outside of the achievement of a just cause and yet still be justifiable? Second, is it possible to separate the pursuit of secondary aims from the use of force? Third, what role does consent of the target-population play in the justness of the pursuit of secondary aims? With these questions in mind, consider the following: Education-1: Mass-atrocities are occurring in State-B. State-A militarily intervenes and stops the mass-atrocities. As a secondary aim, State-A begins construction on a network of schools within State-B, with the intention of enabling a larger portion of the target-population to receive basic elementary education. We will also be returning to and discussing Military Power-2 and Political Influence-2. These scenarios show the normative complexity of abuse-worry₂ and highlight the three parts of the core question just listed. First, let us consider the possibility that a secondary aim may fall outside the achievement of a just cause and yet still be justifiable. Unlike the secondary aim in Political Influence-2 which is clearly unjust, the secondary aim in Education-1 suggests that secondary aims, as such, are not necessarily unjust or abusive. In Education-1, State-A’s secondary aim is to expand education opportunities for the target-population and this is an objective that I think most people would agree, when taken in isolation, is a prima facie justifiable pursuit. This does not mean that State-A’s actions in Education-1 are fully justifiable (as we will

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

SCRUTINIZING INTENTIONS



discuss), it simply means that if we are to find fault in this scenario we must look beyond the simple fact that State-A has a secondary aim. Thus the answer to the first component of our core question is that, while some secondary aims are clearly unjust, there are also some secondary aims which are, prima facie, justifiable. Of course, this forces us to ask what separates an unjust/abusive secondary aim from a justifiable one. Though answering this question demands more consideration than we have the space for here, the comparison of Political Influence-2 with Education-1 points to a strong indicator. I submit that when a secondary aim violates or hinders individuals’ fundamental human rights, as in Political Influence-2, it is unjust/abusive. By contrast, when a secondary aim protects or encourages the fulfilment of individuals’ fundamental human rights, as in Education-1, it is prima facie justifiable. Military Base-1 is not as straightforward; building a military base in a foreign country seems, on its own, to be a neutral action (even when done to promote one’s own military power) and so we must look further to determine whether and why this scenario is justifiable or abusive. The Military Base-1 scenario brings to focus the second component of our core question. Taken in isolation building a military base in a foreign country is neither obviously justifiable nor abusive. What is important for our purposes is that this neutral secondary aim is being pursued in conjunction with the use of military force. In the foregoing paragraph our question was whether the secondary aim was unjust, justifiable, or neutral. Here we need to know whether the pursuit of a neutral or justifiable secondary aim through the use of force is justifiable or abusive. The short answer, as the requirement of just cause dictates, is that military action is only justifiable when it is used to protect fundamental human rights. This gives us a place to start, but still leaves many questions. At first thought this would seem to render the pursuit of the secondary aim in Military Base-1 unjust/ abusive, while also suggesting that Education-1 is justifiable. However, this conclusion is too quick. Though enabling individuals to get an education surely promotes those individuals’ rights to autonomy, receiving structured/formal education is not necessary for the realization of the right to autonomy. One can be fully or, at least, significantly autonomous without having ever attended formal elementary school. The problem here is the tricky distinction between, on the one hand, those things which are absolutely necessary for the protection and realization of fundamental human rights and, on the other hand, those things which help promote or enhance the realization of those rights but are not necessary. It seems to me that the focus of just cause is on the former—the protection of those things absolutely necessary for the realization of fundamental

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

human rights. If this is so, then we may have to conclude State-A’s pursuit of its secondary aim in the Education-1 case—building schools—is unjust because building schools is not necessary for the protection/promotion of fundamental human rights and, therefore, this is an unjust use of force.¹² The conclusion that the pursuit of secondary aims which do not directly protect fundamental human rights is an unjust use of force not only adheres to the limits of just cause, it is also in line with the dictates of RI which demands, recall, that in order for a war/ intervention to be justified the use of force must be limited to fulfilling the just cause. The argument that it is unjust to use force to achieve secondary aims which do not directly protect fundamental human rights may lead us to the stronger claim that it is unjust for an agent to have secondary aims not relating to the direct protection of fundamental human rights during an intervention. Before we reach this further conclusion, however, we need to ask whether it is possible to separate the use of force from the pursuit of some secondary aims? Consider the following: Education-2: Same as Education-1 except that the task of building schools is made difficult by resistance from a large portion of the target-population who oppose Western-style systems of learning and, as a result, State-A’s intervening forces are used to secure the school construction sites. Education-3: Same as Education-1 except that, while State-A maintains post-atrocity security, State-C sends civilian contractors into State-B and begins construction on a network of schools within State-B, with the intention of enabling a larger portion of the target-population to receive basic education. There is no resistance from the target-population against the schools and, as a result, there is no use of military force associated with the construction of the schools. While it may not always be readily apparent whether an intervener’s secondary aims are tied to the use of force, the case of Education-2 shows that sometimes an agent’s secondary aim is clearly dependent upon the use of force. Here State-A employs force not only in relation to the just cause, but also to carry out its secondary aim of building schools; without the use of force, these schools would not be built. By contrast, the case of Education-3 suggests that it is possible for an agent to have secondary aims and pursue these separate from the use of force even when force is being used in the same place (the target-state) in relation to the just cause. I have introduced a third party—State-C—in order to stress the separateness of the pursuit of the just cause from the construction of schools and show that it is conceptually possible for these two activities to take place simultaneously and yet apart from each other to a significant degree. It may very well be that the construction of schools by State-C is only made possible because

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

SCRUTINIZING INTENTIONS



State-A’s military forces have halted the atrocities and have thereby created an environment of stability without which State-C could not build schools. We can grant this and still accept the further claim that the construction of schools is not dependent upon the use of force insofar as force is not being used in direct relation to the schools’ construction. If we agree on this point, then, removing the third party, it seems possible for State-A acting alone, as in Education-1, to pursue secondary aims concurrently with but significantly separate from the use of force used to obtain/secure the just cause. That is, it is possible that State-A’s secondary aim of building schools may be justifiable as long as its achievement does not involve the direct use of force. Note, however, that I have said that it may be justifiable. While Education-1 does not seem blatantly abusive, left unqualified it may be unjust even though State-A’s secondary aim is justifiable and even when/if it is pursued without the use of force. To see this, let us turn to the third component of the core question—the matter of target-population consent. Consider again Education-1 and Education-3. Even in these cases where the agents seem to have justifiable secondary aims which they can pursue separate from the use of force, it is not clear that this pursuit is justifiable because both cases fail to consider what the target-population wants. As noted, we are working under the assumption that, in all our cases, the interveners meet all just war requirements apart from RI. This includes, of course, the requirement of legitimate authority. Now, admittedly there is a great deal of debate and disagreement among just war theorists over what exactly legitimate authority dictates, but two points of relative agreement are that legitimate authority holds that, in order for a war/intervention to be just, the agent, first, must be acting on behalf of a group of people whose fundamental human rights are threatened and, second, must be acting with that group’s consent.¹³ Some theorists have argued that when mass-atrocities are occurring it is reasonable for an agent to assume that the target-population consents to intervention, and therefore, in these cases it is not necessary for interveners to obtain explicit consent for the actions they carry out in order to stop the atrocities. The assumption is deemed reasonable on two counts. First, most people facing slaughter or torture would want to be rescued from this fate and, accordingly, in cases of mass-atrocity it is fair to conclude that the target-population would consent to being protected. Second, given the urgency of the need for aid during mass-atrocities, it seems unreasonable to demand that interveners gain explicit consent from the target-population—a task that would certainly prove time consuming and probably, given the volatile environment, impossible.¹⁴ So there is some sense that, in the cases we are discussing, there is a level of consent among the target-population in relation to the actions of the interveners.

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

However, secondary aims not related to the direct protection of fundamental human rights do not enjoy this same assumption of consent. One might argue that the requirement of legitimate authority (its consent aspect) need not apply to these secondary aims because, as we have established, they are only justifiable when undertaken without the use of force and the requirement of legitimate authority, as a condition of just war theory, pertains to the use of force. On a surface level this is true; however, I have brought the requirement of legitimate authority into our discussion for two reasons. First, it calls our attention to the importance of consent. Second, it reminds us that these secondary aims, though not carried out by force, are, nevertheless, carried out in association with the use of force—they are being carried out within the context of an intervention. Given this second point we may want to demand that the interveners meet the requirement of legitimate authority in regard to obtaining their secondary aims even when they do not pursue these aims with the direct use of force. Furthermore, the requirement of consent is certainly not limited to just war theory and the use of force. It seems to me that the right of national self-determination also demands that agents obtain explicit consent of the target-population in relation to their, the interveners, secondary aims. Though there is much debate over the concept of national self-determination and what this right/principle entails, for the purpose of this discussion let us grant that national self-determination is the right of nations, to some significant degree of political autonomy, through which they can collectively determine—decide on, shape, protect—those things that are essential to their collective way of life. There is further disagreement over what constitutes a nation—some hold that state borders define a nation, others argue that there may be supra-state, nation-state, and sub-state nations. We do not have the space to engage this debate here, so let us grant that the target-population constitutes a nation (even when/if it is broken into smaller nations—the important point here being that they are a separate political community from the interveners). By contrast, there is relative agreement over what national self-determination prohibits; according to this right, ‘outsiders’ (agents not part of a nation and/or state) are prohibited from oppressing, attempting to control, or unjustly interfering in the collective life of the nation. Though still somewhat vague, when taken together this broad understanding is enough for our purpose because it demands that interveners have the consent of the target-population before carrying out their secondary aims, when and if these secondary aims affect the collective life of the targetpopulation.¹⁵ It is hard to think of many secondary aims which would not fall into this category. Certainly, the secondary aims in Education-1 and Military Base-1 demand the consent of the target-population based on the dictates of national self-determination.¹⁶

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

SCRUTINIZING INTENTIONS



Let us return to our core question. Does an agent who has and pursues secondary aims necessarily violate the action-component of RI and, if so, does this render the intervention unjust/abusive? In answering this question I have argued, first, that the fact that an intervener has secondary aims does not, on its own, render an intervention unjust. An agent’s secondary aim is unjust when it involves the violation or hindrance of individuals’ fundamental human rights. A secondary aim may also be neutral or justifiable. Second, I argued that it is unjust for an agent to pursue a neutral or justifiable secondary aim (when this secondary aim is not necessary for the protection of fundamental human rights) through the use of force; such pursuit violates the action-component of RI. However, I further argued that it may be possible for interveners to separate their activities involving the use of force from those involved with their secondary aims and, in these cases, the pursuit of these secondary aims is not, necessarily, unjust. Finally, I argued that in those cases where interveners meet the previous two requirements they must obtain the explicit consent of the target-population in order for their actions in relation to their secondary aims to be justifiable.

Abuse-Worry₃ In the last section I argued that in order for the pursuit of a secondary aim to be justifiable/non-abusive, interveners must, among other things, obtain the consent of the target-population before they carry out this aim. In some mass-atrocity and post-atrocity cases it may be difficult to distinguish genuine consent from coercion. Societies suffering from and/or recovering from mass-atrocities are in dire need of help and, given this, it is possible for questions regarding secondary aims to be posed or interpreted as conditions or ultimatums. A question like ‘Can we build a military base along your border?’ might carry with it the implicit threat ‘And if you say no, we will not provide you with the military and/or economic aid you desperately need.’ Abuse-worry₃ focuses on this sort of conditionality— either implicit or explicit. Abuse-worry₃ is particularly interesting because it forces us to re-examine a principle which has enjoyed a long history of support within the field of political theory. The principle, which I call the duty to aid, holds that that if V (a victim) is in grave danger and R (a rescuer) has the ability to rescue/aid V without unreasonable risk to himself (herein termed the ‘excessive risk proviso’), then R has a duty to aid V. Four decades ago, Peter Singer illustrated the duty to aid with the case of the drowning child, stating: ‘if I am walking past a shallow pond and see a child drowning in it, I ought to wade in and pull the child out’.¹⁷ Abuseworry₃ adds a new element to the case. Imagine that a man is walking by a

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

shallow pond and sees a child drowning in it. Rather than jumping in immediately to save the child, first he says to the child that he will rescue him on the condition that the child agrees to mow his (the man’s) lawn every Sunday as payment for his (the man’s) services. The child agrees and the man jumps in and rescues the child. Such is the case with abuse-worry₃; State-A is offering State-B desperately needed aid, but makes this aid conditional. Both scenarios present an interesting question regarding the duty to aid: granting that R is morally obligated to provide aid when/if V is in grave danger and when R can do so without excessive risk to himself, is it ever morally legitimate for R to place conditions upon the provision of this aid? Or, within the context of R2P, is it morally legitimate for interveners to put conditions on the provision of desperately needed aid? Three points before we begin. First, as we proceed, the reader will note that, unlike the preceding sections, there will be little mention of right intention within this section. This is so because I am carrying the arguments made in the previous sections forward into this section. Let me explain. When State-A places a condition on its aid to State-B, that condition represents a secondary aim. Accordingly, we can assume that State-A has at least two motivations and two objectives tied, in some way, to the intervention. Going back to the first section, we can conclude that, whatever State-A’s secondary motivations are (so long as State-A does not carry out actions outside of those needed to obtain the just cause), these secondary motivations, however unjust or non-ideal, do not render State-A’s intervention unjust even though they do not meet the motivation-component of RI. Further, applying the arguments from the second section, so long as State-A’s secondary aims are (a) either justifiable or neutral, (b) not directly pursued via force (unless these aims involve the direct protection of fundamental human rights), and (c) have been consented to by the target-population, undertaking these secondary aims does not constitute a violation of the action-component of RI or abuse of R2P. However, and this is the issue we take up in this section, when consent is coerced and where that coercion is unjust, State-A acts unjustly in relation to the target-population and abuses R2P. Second, I have hinted that there is a distinction between consent and coercion and suggested that in cases where State-A places conditions on its aid it coerces State-B. Though we do not have the space to dive into the rich literature on coercion, a few points of clarification are helpful here. Like most complex concepts, there is much debate over the exact meaning and boundaries of the word ‘coercion’, but there is relative agreement over a few important points. First, when we say that R coerces V, we are saying that R ‘intentionally uses a credible

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

SCRUTINIZING INTENTIONS



and severe threat of harm’ in order to get V to act in a certain way.¹⁸ Second, when V is coerced, he is effectively forced by R to act in a certain way, X, because R’s threat makes doing X the only realistic or reasonable option for V. Third, a coercive threat can take (at least) three forms: Coercive-threat₁: R causes some harm to V and threatens to continue this harm until V does X. Coercive-threat₂: R threatens to cause some harm to V unless V does X. Coercive-threat₃: R owes V something, B, and V is in desperate need of B, but R threatens to withhold B unless V does X. The cases we will discuss fall under the category of coercive-threat₃. Finally, the fact that R forces V to act in a certain way does not, by virtue of this fact alone, make R’s coercive threat unjust (this point enjoys less consensus, but I include it because we will discuss a scenario where this clearly appears to be the case). With these four points in mind, let us assume that in all the conditional aid cases we discuss R/State-A coerces V/State-B.¹⁹ Thirdly, I must outline four aspects of the conditional aid scenarios we are concerned with herein. First, the scenarios are those in which V is in desperate need of aid—either V or someone in V’s care will die or otherwise permanently lose the ability to lead a minimally decent life—and R has the ability to aid V. So, within the abuse-worry₃ scenario, this means that mass-atrocities are imminent, ongoing, or have just ended (and the resumption of atrocities remains a serious threat), State-B is in dire need of aid, and State-A has the ability to provide this aid. Second, the excessive risk proviso has been met; StateA has the ability to aid State-B without significant risks to its own population.²⁰ Third, R’s condition of aid is such that V must choose between agreeing to R’s condition and receiving aid or losing the ability to lead a minimally decent life and/or having someone in his care lose the ability to lead a minimally decent life. We are assuming, then, that if State-B does not agree to State-A’s condition, no aid will be given to State-B and the atrocities will either continue or, in postatrocity cases, atrocities will resume and/or hunger and disease (typical to postatrocity situations) will lead to numerous preventable deaths. Fourthly, R could provide aid to V unconditionally; R’s condition is not physically unavoidable given the circumstances of the case. Throughout this discussion, let us assume that these four criteria are met. Keeping these points in mind, we are ready to analyse various forms of conditional aid and identify the different moral wrongs that these different conditions may entail. We begin with the condition most often discussed in the coercion literature (W1). Consider the following scenarios:

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

Boating-1: A boater comes upon a man who is about to drown in the middle of an otherwise empty lake. The boater offers to rescue the man on the condition that the man pays him, the boater, $10,000.²¹ In this scenario, R’s condition of aid takes advantage of V’s dire circumstances for his own excessive gain. In Feinberg’s words, R’s condition of aid in such cases is ‘parasitical, exploitative . . . unconscionable . . . opportunistic’.²² So, what (W1) captures is the fact that R coerces V for reasons of greed and/or opportunism and this is clearly unjust. In a number of atrocity and post-atrocity cases international actors have given military and economic aid on the condition that they, the donors, are granted business contracts/concessions in the target-state’s petroleum, diamond, and/or other lucrative industries.²³ Such conditions, designed to advance the economic interests of the intervening agents, are clear instances of (W1); these conditions are motivated by greed. Another case of (W1), called ‘procurement-tied aid’, occurs when international actors give aid on the condition that the target-state uses the aid to buy material and/or services from the donor state’s companies.²⁴ The case of Military Base-1 would also fall under the category of (W1) if State-A made aid conditional upon permission to build its military base because while this would not provide State-A with monetary gains, the conditions are driven by greed based on power-oriented gains. A second unjust condition of aid (W2) is a condition by which R tries to exert influence over some aspect of V’s life which has no necessary connection to the deliverance of aid and, therefore, constitutes unjust interference in V’s special relationships. Consider the following scenario: Millionaire-1: A boy will die unless he undergoes surgery. A well-intentioned millionaire, who doesn’t like the boy’s mother’s parenting style, says that he will pay for the surgery on certain conditions. The mother must agree to read the boy a book every night, take the boy to the park at least once a week, and spend at least one hour a day with the boy with her mobile-phone, computer, and television turned off.²⁵ The reader may agree that the millionaire’s conditions, if implemented, would help the mother be a better parent. It is not the content, per se, of the conditions which is morally wrong in (W2). Rather, the conditions are wrong because they constitute an unjust interference into the parent–child relationship; the millionaire does not simply advise the mother, he forces her, by his condition, to change her parenting style. While it may be advisable that the mother reads to her child, whether she does this is a matter that must be worked out between the mother

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

SCRUTINIZING INTENTIONS



and child. (W2), then, is a violation of V’s (both mother and child) right to community (family) by way of unjust interference. This, of course, is not the only wrong R commits. Millionaire-1 is interesting because it involves multiple victims which R victimizes in different ways. Though R fails in his duty to aid the child, this failure is not meant to, and does not, coerce the child. Rather, the millionaire coerces the mother. Millionaire-1 (and subsequent Millionaire scenarios) is also unique in that, because it involves multiple victims, it presents us with a moral wrong that is not apparent in the single victim cases. I suggest that in cases where there is more than one victim and these victims are joined in a special relationship, R also violates V’s right to community. In Millionaire-1, the millionaire, by failing in his duty to aid the child, threatens to end the parent–child relationship and, in so doing, violates both the child’s and the mother’s rights to community— in this case their right to family. This point is especially important and interesting when applied to mass-atrocity cases, as it suggests the possibility that denial of aid to the target-population may (among other things) violate their right to community—in this case their right to national self-determination. As we move on through variations of this scenario, it is helpful to keep these points in mind. So the distinctive feature of (W2) is that R’s condition unjustly interferes in a special relationship which V has with another person (who might also be seen as a victim). Though I have not seen (W2) discussed in the coercion literature, it is important because one of the claims often made against conditional military and/ or economic aid is that it gives the international community undue influence in the affairs of the target-state and, thereby, violates the target-population’s right to national self-determination. Conditions which might fall under the category of (W2) within the context of R2P include aid made conditional upon democratization of the target-state or, as in Education-1, the construction and expansion of formal education. Though the content of these conditions may be highly beneficial to the target-population if implemented, making the provision of desperately needed aid dependent upon agreement to these conditions is unjust. As already discussed, such conditions constitute an unjust interference into a special relationship—in this case the relationship between members of a nation (thereby violating national self-determination).²⁶ Another morally wrong condition of aid (W3) is a condition which R could legitimately ask for in separate circumstances, but which cannot justifiably be made in connection with aid. To illustrate this, consider the following: Boating-2: A boater comes upon a man drowning in the middle of an otherwise empty lake. The boater recognizes the man and realizes that the

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

man owes him $10,000 (a loan the boater made to the man on fair terms). The boater offers to rescue the man on the condition that he pays him the $10,000 owed within three days. Doctor-1: V needs life-saving surgery. R, the doctor, says to V that he can and will do the surgery, but only for a fee. The fee is fair in so far as it contributes to the expenses involved with the surgery, helps to pay the debt the doctor owes for his training, and provides the doctor with part of his reasonable standard of living. It is certainly less clear that there is a moral wrong being committed in the above cases, but I submit that there is. We can readily grant that the boater has a legitimate claim that V repay the loan and the doctor has a legitimate claim to charge for his services. Moreover, a case could be made that, in Doctor-1, V even has a moral duty to pay the doctor (if he is able) in order to contribute to R’s continued ability to practise medicine and aid others in dire need. However, keeping in mind that the excessive risk proviso is met, it seems morally inappropriate for R to make such payments a condition of aid. The morally required thing for R to do, it seems, is to aid V and then try to settle the debt. One might object that if R is morally justified in requesting payment for his services, why is he not able to make his services conditional upon payment, especially where such payment will help ensure the availability of aid to others. There are two problems. First, in many of these desperate aid situations time is a critical factor. If a fireman insists on standing outside a burning house until the people within agree to pay him, several people may die in the process. Secondly, there may be cases in which a victim cannot afford to pay. It seems to me that, when the excessive risk proviso is not a factor, denying a victim aid because he cannot afford to pay for one’s services is unjustifiable. To reiterate, R may justly request/seek payment after aid is given; my point is that the actual provision of aid ought not be made conditional upon this payment. Interestingly, the argument that (W3) is an unjust condition becomes stronger when we place it into the context of R2P. One particularly controversial form of conditional aid is that which makes aid dependent upon the payment of debts. In post-atrocity Bosnia, international actors lent Bosnia $45 million dollars in aid on the understanding that this so-called aid was to be used to repay Bosnia’s portion of the former Yugoslavia’s Cold War debt.²⁷ In this case the conditional ‘aid’ did very little to actually help the Bosnians who desperately needed aid. However, most cases in which international actors make aid conditional on the repayment of debts are morally worse than (W3) because it is far from clear that the payment being asked for would be a legitimate request outside of the aid scenario (like the

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

SCRUTINIZING INTENTIONS



request of the doctor for payment in Doctor-1). When atrocity-committing regimes are given loans—without question or condition—by international actors, there seems no tenable argument to support the claim that post-atrocity societies ought to repay those loans, especially since they were very likely used to fund the atrocities committed against them. Much less tenable is the idea that desperately needed aid ought to be made conditional upon repaying such debts.²⁸ The overarching point here is that when people suffering/recovering from massatrocities desperately need aid, it is unjust to make this aid conditional on repayment either of the aid itself or past due debt. Whether it is justifiable for donors to request repayment after giving aid, rather than as a condition of aid, is a separate and important question but one which we do not have the space to take up here. The final unjust condition of aid (W4) occurs when R’s condition, though wellintentioned, makes aid to V unnecessarily conditional upon the actions of some other person, P. Consider the following: Millionaire-2: A boy will die unless he undergoes surgery. His mother, P, is a drug addict. A millionaire offers to give the mother money for the life-saving surgery on the condition that she spends the money on the surgery not drugs. Doctor-2: A boy is in need of life-saving surgery as the result of a beating by P, his parent. The doctor tells P that he will perform the life-saving surgery on the condition that P agrees not to beat the boy again. These scenarios are interesting because R is actually trying to act in a morally just way—he is trying to protect the child—but is misguided in his attempt. R’s condition is misguided because it makes saving V’s life conditional on P agreeing not to act unjustly towards V (this is especially misguided when P is the cause of V’s predicament in the first place). When the fulfilment of R’s duty to V is not literally dependent upon P’s fulfilling his/her duty to V, then R’s duty to V ought not to be made dependent upon P’s fulfilling his/her duty. In Millionaire-2, the millionaire ought simply to bypass the drug-addicted mother and pay the doctors to perform the surgery. Similarly, in Doctor-2, the doctor ought to perform the life-saving surgery for the child and report the parent to the appropriate child-protective services. So (W4) is a moral wrong because it makes aid to V unnecessarily conditional on the actions of some other party P. R’s condition in (W4) is certainly not an egregious wrong like (W1), but it is, nevertheless, morally wrong (perhaps it is fitting to call it ‘morally mistaken’). (W4) is especially pertinent in the context of R2P and we will return to this shortly.

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

Now that we have identified morally wrong conditions of aid, let us return to the matter of morally justifiable conditional aid. Consider the following: Doctor-3: V₁ needs a life-saving drug. R, the doctor, has the drug. R knows that V₁ intends to unjustly kill V₂ as soon as he receives the drug. R says that he will give V₁ the drug only if V₁ agrees to hand over his gun and not commit murder.²⁹ This scenario presents R with a moral dilemma: he has a duty to aid V₁ but he knows that if he aids V₁ unconditionally he will be enabling V₁ to commit a moral wrong against V₂. Intuitively, the doctor’s condition seems morally legitimate because he coerces V₁ in order to save V₂’s life. If we accept that it is sometimes just to kill an attacker in order to save a victim (as we do when we accept that it is sometimes just to wage wars of defence or carry out humanitarian intervention), then it seems reasonable to also accept the claim that it is sometimes just to coerce an attacker in order to save a victim. This conclusion requires us to qualify the duty to aid beyond the standard excessive risk proviso. Such a qualification is made by Peter Singer in his seminal article, ‘Famine, Affluence, and Morality’, in which he states the duty to aid as follows:³⁰ [I]f it is in our power to prevent something bad from happening, without thereby sacrificing anything of comparable moral importance, we ought, morally, to do it. By ‘without sacrificing anything of comparable moral importance’ I mean without causing anything else comparably bad to happen, or doing something that is wrong in itself, or failing to promote some moral good, comparable in significance to the bad thing that we can prevent. Singer’s account of the duty to aid goes beyond the excessive risk proviso which applies to the risk incurred by R, and adds that aid is also conditional upon that aid not causing/allowing some other foreseeable moral wrong to occur (I term this the Singer proviso).³¹ Accordingly, a justifiable condition of aid (J1) is a condition which is made to prevent a moral wrong that would likely occur as the result of R’s unconditional aid to V. There may be some difficult cases which fall somewhere between (J1) and (W4), but many cases will be distinguishable. One useful distinction is made evident by a comparison of Doctor-2 (an instance of (W4)) and Doctor-3 (and instance of (J1)). In Doctor-3, R’s unconditional aid to V will foreseeably enable V to commit a moral wrong. By contrast, in Doctor-2, there is no reason to believe that if R gives unconditional aid to V (the child), V will then go and commit some moral wrong. The main difference is whether the moral wrong R is trying to prevent is one that V will likely commit if he receives unconditional aid.

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

SCRUTINIZING INTENTIONS



Another crucial point about (J1) is that, where R’s unconditional aid to V will likely lead to some serious moral wrong, conditional aid is not only morally justifiable, it is (in my view) morally obligatory. In Doctor-3, R has a duty to aid V₁ and V₂. V₂’s life, however, is only in danger if R aids V₁ unconditionally. This creates a very interesting dynamic. V₁, by intending to murder V₂, has placed R in a forced choice in which R must choose between the following: Option-1 (unconditional aid): R aids V₁ unconditionally and thereby enables him to unjustly kill V₂. Option-2 (no aid): kill V₂.

R refuses aid to V₁ and, thereby, ensures that he does not

Option-3 (conditional aid): R aids V₁ on the condition that he hands over his gun and agrees not to kill V₂. Option-1 is morally unjust because, if chosen, R would be partly to blame for the unjust killing of V₂. Where Option-3 is not available, Option-2 seems to be the only morally justifiable course. However, I suggest that where Option-3 is available and reasonably reliable it is morally obligatory. A less optimistic reader may argue that Option-3 will very rarely be reasonably reliable and, therefore, R ought to choose Option-2 in all cases on the grounds that V₁ will renege on his promise to R and kill V₂ once his own life is saved. However, if we agree that in cases of self- and other-defence, the death of the perpetrator ought to be the option of last resort (a requirement of just war), then it seems that, where viable, Option-3 is morally required. Furthermore, when we apply this type of scenario to more complex cases of R2P (as we will shortly), even the less optimistic reader may agree that Option-3 is the most morally sound option. The claim that morally justifiable conditions of aid are also morally obligatory leads to the following addition to the duty to aid. If V is in grave danger and R has the ability to rescue V without (a) unreasonable risk to himself (the excessive risk proviso) and without (b) thereby enabling/contributing to some other serious moral wrong (the Singer proviso), then R has a duty to rescue V. In cases where the Singer proviso is not met and R could provide some condition to avoid the other serious moral wrong, R must provide this condition. In cases where the Singer proviso is not met and R cannot provide some condition to avoid the other serious moral wrong, R must not provide aid. Thus, in the converse of (J1) we have another type of unjust aid: (W5) aid is unjust when R gives it unconditionally despite having reasonable knowledge that such aid will lead to a ‘comparably bad’ moral wrong. Now let us put our discussion of (W4), (W5), and (J1) into the context of R2P. When aid is given unconditionally to individuals or groups who are likely to

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

carry out or resume atrocities or use aid for otherwise unjust purposes, international actors are guilty of (W5). After the 1994 Rwandan genocide, for example, international actors sent $1.3 billion in unconditional aid to the Hutu refugee camps inside the DRC despite knowing that a significant portion of this money was being controlled and used by the Hutu militants responsible for the Rwandan genocide. These same extremists, sustained by aid, subsequently helped ignite and carry out the atrocities in the DRC.³² The implication of (W5) is that when international actors are aware that their aid, given unconditionally, will likely be used for unjust purposes, conditional aid is just and obligatory, in line with (J1). There is a danger that international actors wanting to avoid (W5) will commit (W4) rather than implement conditions in accordance with (J1). Returning to the Rwanda-DRC example, a few international actors, not wanting to supply the Hutu militants controlling the refugee camps, cut aid to these camps all together. This had little effect because most donors continued to pour in aid. Imagine, for the sake of argument, that all donors had cut off aid to the camps and made their aid conditional upon the militants agreeing to demilitarize (thereby rendering the militants incapable of committing further atrocities). This would be an instance of (W4) because it would make the aid to innocent refugees (who were in desperate need of food, shelter, clean water, and medical supplies) dependent upon the actions of the militants—the very same militants that forced many of these individuals into the refugee camps in the first place. Instead, international actors ought to have deployed military/police forces to the camps to ensure that aid was being delivered to non-militants and that militants were not controlling the camps. If deployment was not possible (though I think it was and that in most cases it would be), then just and obligatory conditions of aid, in this case, could have specifically targeted militants such that those who relinquished their weapons would receive aid (while also providing aid to the refugees). Barring this and other possible mechanisms to reduce the provision of aid to militants, the decision of whether or not to deliver aid may turn on a calculation of proportionality (another requirement of just war theory). In such a case, interveners would have to ask whether the overall good achieved by the provision of aid (where ‘good’ equates to the protection of fundamental human rights) outweighs the harm caused by this same provision (where ‘harm’ equates to the violation of fundamental human rights). This, admittedly, would be a difficult calculation to make. Comparing all the possible harms of pouring aid into camps largely controlled by atrocity-committing militants against all the possible harms of forgoing aid to camps entirely is quite a tall task and I am inclined to lean on the side of providing aid. However, given the global military and NGO capabilities, I am

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

SCRUTINIZING INTENTIONS



also inclined to think that such painful choices will not have to be made very often (though an unnecessary undercommitment of resources may require such choices). The crucial point here is that, where there is the option of utilizing such capabilities, unconditional aid is unacceptable. It is necessary to pause here for a moment and reiterate and clarify the reach of the arguments I have made herein regarding conditional aid. In this section I identified four types of unjust conditional aid (W1–W4), one type of unjust unconditional aid (W5), and one type of just and obligatory conditional aid (J1). Before we began our discussion I specified that the situations with which we were concerned were those cases in which V/State-B was in desperate need of aid— without such aid V/members of the target-populations’ lives would be lost—and R/State-A had the ability to provide this aid without excessive risk to him-/itself. In these cases, I argued, conditional aid constituted coercion and, further, in many cases, this coercion was unjust (as in (W1)–(W4)). When we step outside of the parameters of these particular cases, it is not necessarily the case that conditional aid that falls under the categories of (W1)–(W4) remains unjust. One would need to present a different, or at least, expanded argument, for example, to show that aid made conditional on the expansion of education or the holding of democratic elections was unjust in cases where that aid was not needed to save the lives of the target-population, and I’m not sure that such an argument would succeed. To see this, consider the following: Education-4: State-B is a developing and relatively stable state which applies for international aid in order to reinforce its economic growth. State-A offers to provide general budgetary aid to State-B on the condition that State-B earmarks a portion of this aid for the expansion of its education system and allows State-A to help with the physical construction of hundreds of schools. Education-5: Mass-atrocities have recently ended in State-B. State-B has elected a new government and, with the continued help of aid from State-A (and other donors), is providing necessary public services to its population and has begun the process of economic recovery. State-A’s military forces, who intervened and stopped the mass-atrocities, remain in State-B to provide necessary post-atrocity security. Further, State-A is providing State-B with unconditional economic aid which helps provide food and water to the targetpopulation. State-A has also proposed to provide additional aid to State-B on the condition that this aid be earmarked for the expansion of education. In the Education-4 scenario State-A does not coerce State-B, nor does State-A’s condition fall under the category of (W2), because State-B can genuinely choose

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

to refuse State-A’s conditions without suffering unacceptable consequences (as it would in Education-1 if aid was made conditional upon the construction of schools). Turning to Education-5, it seems to me that, despite the fact that State-B is still in a position where it desperately needs aid (therefore falling into the parameters we set out earlier), State-A does not coerce State-B in this scenario, nor is its condition an instance of (W2). This is so because State-A has separated this education-oriented conditional aid from the deliverance of desperately needed aid and, in so doing, has afforded State-B the ability to choose, without facing dire consequences, whether to accept this aid or not. It is important to keep these different cases in mind when determining whether conditional aid, more generally and within the context of R2P, is abusive or unjust.

Conclusion The aim of this chapter was to critically examine the requirement of RI within the context of R2P. We focused on three types of potential abuse. First, we looked at abuse-worry₁ in which the interveners had secondary, sometimes even unjust, motives for intervening but did not attempt to attain these secondary motives through actions outside of those needed to achieve the just cause. In these cases, I concluded, the intervener’s actions did not constitute an unjust use of force or abuse of R2P, even in those cases where the secondary motives were unjust. Next we looked at abuse-worry₂. In these cases interveners had secondary aims and, in order to attain them, took actions beyond those necessary for achieving the just cause. Here I argued that the actions of the interveners could still be justifiable or non-abusive so long as: (a) the intervener’s secondary aims were not unjust, (b) the secondary aims were not attained through the use of force, and (c) the interveners gained the consent of the target-population regarding their secondary aims. The converse of this, of course, is that when/if interveners (a) pursue unjust secondary aims, (b) carry out neutral or justifiable secondary aims via the use of force, and/or (c) fail to obtain the consent of the target-population before performing their secondary aims, those actions are unjust and abuse the principles of R2P. Finally, we discussed abuse-worry₃. Here I argued that when a society is facing or recovering from mass-atrocities it is unjust for interveners to place conditions on their aid, except in those cases where the conditions are necessary to prevent some other moral wrong (as required by (J1)). The converse of this argument is that, within the context of R2P situations, conditional aid which does not fall under the category of (J1) is abusive (and barring conditions such as those in Education-5). The next step in this discussion is to explore how

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

SCRUTINIZING INTENTIONS



we—the international community—can work to prevent these abuses. This is a discussion for another time, though we have already made some headway in this direction, for identifying what constitutes abuse of R2P is clearly an essential step in preventing that abuse.

Notes 1. ICISS, The Responsibility to Protect (Ottawa: International Development Research Centre, 2001). 2. In all the scenarios discussed herein, I assume that State-B government is unwilling/ unable to stop the atrocities independently. 3. A just cause for war/intervention is generally understood as one which aims to defend individuals’ fundamental human rights when those rights are under attack. For discussion on just cause and the other just war requirements, see: A. J Coates, The Ethics of War (Manchester: Manchester University Press, 1997); James T. Johnson, Morality and Contemporary Warfare (New Haven: Yale University Press, 1999); Jeff McMahan, ‘Just Cause for War’, Ethics and International Affairs, 19/3 (2005): 1–21; Michael Walzer, Just and Unjust Wars (New York: Basic Books, 2006). 4. For a full list of what constitutes atrocities, see: ICC, Elements of Crime (The Hague: ICC, 2011); ICC, Rome Statute of the International Criminal Court (The Hague: ICC, 2011), arts. 6–8; USIP, Report of the Chilean National Commission on Truth and Reconciliation (Notre Dame, IN: United States Institute of Peace and University of Notre Dame Press, 2002), 51–6. I have chosen 1,000 deaths as the threshold for massatrocities because (among other reasons) this is the threshold used on most massatrocity databases and also falls in line with the conventional war threshold. 5. For discussion on fundamental human rights, see: Joseph Raz, The Morality of Freedom (Oxford: Oxford University Press, 1998) and ‘Rights-Based Moralities’, in J. Waldron (ed.), Theories of Rights (Oxford: Oxford University Press, 2009); Henry Shue, Basic Rights (Princeton: Princeton University Press, 1996); Jeremy Waldron, Liberal Rights (Cambridge: Cambridge University Press, 1993). 6. ‘Proportional means’ as it is described in the report falls more in line with the jus in bello requirement of necessity than proportionality, though there are also some descriptions of proportionality as it is traditionally understood. ICISS, Responsibility to Protect, p. xii. 7. This definition is consistent with the just war tradition and matches the R2P ‘just cause threshold’. See ICISS, Responsibility to Protect, p. xii. 8. Johnson, Morality and Contemporary Warfare, 32–3, and McMahan, ‘Just Cause for War’, 5. 9. For a detailed account of the Indian intervention, see: Onkar Marwah, ‘India’s Military Intervention in East Pakistan, 1971–1972’, Modern Asian Studies, 13/4 (1979): 549–80. 10. For further discussion on the distinction between objectives and motivations, see: James Pattison, Humanitarian Intervention and the Responsibility to Protect (Oxford: Oxford University Press, 2010), 153–68.

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

11. This conclusion may not hold in those cases where either State-A or State-C is capable of carrying out the intervention and where State-C does meet the motivation-component of RI (or at least does not have unjust motivations). The basic argument is that where there are more than one potential interveners, the most ideal (the agent that best meets the just war criteria and is the most capable) has the duty to intervene. For further discussion, see: Pattison, Humanitarian Intervention. 12. I realize that the conclusion that receiving basic education is not, itself, a fundamental human right is a controversial claim which will strike some as incorrect. Since I cannot defend this claim here, I ask those who disagree to substitute the building of schools for the building of roads or reconstruction of infrastructure—secondary aims which benefit the target-population without also greatly benefiting the interveners (as this is what the Education scenarios are meant to capture). 13. Some theorists stress that statehood is also an essential part of legitimate authority, though when pushed with cases like the Rwandan genocide, most of these theorists are willing to put aside statehood as a requirement of legitimate authority. For further discussion on this point and on legitimate authority generally, see Coates, Ethics of War, 127–9; Cecile Fabre, ‘Cosmopolitanism, Just War Theory and Legitimate Authority’, International Affairs, 84/5 (2008): 963–76; Jeff McMahan, ‘The Morality of Military Occupation’, Loyola International and Comparative Law Review, 31 (2009): 112; Darrel Moellendorf, Cosmopolitan Justice (Oxford: Westview Press, 2002), 121. 14. See Fabre, ‘Cosmopolitanism’, at 973, and F. M. Kamm, ‘Failures of Just War Theory: Terror, Harm, and Justice’, Ethics, 114 (2004): 652. 15. Target-population consent may be gained through various forms, to include (though not limited to) referendum, permission by elected government, and consultation with local government and civil society groups. The finer details of consent—e.g., what percentage constitutes consent—are important, but require much more time than can be discussed here. 16. For definitional clarity and further reading on national self-determination, see: Avishai Margalit and Joseph Raz, ‘National Self-Determination’, Journal of Philosophy, 87/9 (1990): 439–61; David Miller, On Nationality (Oxford: Oxford University Press, 1999) and National Responsibility and Global Justice (Oxford: Oxford University Press, 2007); Yael Tamir, ‘The Right to National Self-Determination’, Social Research, 58/3 (1991): 565–90. 17. Peter Singer, ‘Famine, Affluence, and Morality’, Philosophy and Public Affairs, 1 (1972): 231. 18. Tom Beauchamp, ‘Autonomy and Consent’, in F. Miller and A. Wertheimer (eds), The Ethics of Consent (Oxford: Oxford University Press, 2009), 69. This discussion of coercion, as the reader will note, is limited to coercive threats rather than coercive threats and offers. Though I do believe that offers can also be coercive, this point is widely debated in the literature. For our purposes, here, we do not need to take a stance one way or another and this is why I have limited our discussion to coercive threats. That said, some may consider coercive-threat₃ to be an offer rather than a threat—however, I (and many others) disagree here. If I work at my job all week and at the end of the week my employer says he will not pay me unless I agree to wash his

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

SCRUTINIZING INTENTIONS

19.

20.

21. 22. 23. 24. 25.

26. 27. 28.



car, this, in my view, constitutes a threat rather than an offer because my employer is withholding something—my pay—which is rightfully mine. If instead he promised to give me a weekly bonus if I washed his car, this would be an offer. For discussion of these points, see: Ian Carter, Matthew Kramer, and Hillel Steiner, Freedom: A Philosophical Anthology (Oxford: Blackwell Publishing, 2007), 249–315; Joel Feinberg, The Moral Limits of the Criminal Law, iii. Harm to Self (Oxford: Oxford University Press, 1989); Harry Frankfurt, ‘Coercion and Moral Responsibility’, in T. Honderich (ed.), Essays on Freedom of Action (London: Routledge & Kegan Paul, 1973), 63–86; Vinit Haksar, ‘Coercive Proposals [Rawls and Gandhi]’, Political Theory, 4/1 (1976): 65–79; John Kleinig, ‘The Nature of Consent’, in F. Miller and A. Wertheimer (eds), The Ethics of Consent: Theory and Practice (Oxford: Oxford University Press, 2009), 3–24; and Jeffrie Murphy, ‘Consent, Coercion, and Hard Choices’, Virginia Law Review, 67/1 (1981): 79–95. There is room for disagreement over what constitutes a significant risk to an intervener’s population, but for this discussion we are assuming that no such risk exists. That said, I understand an intervention to be significantly risky when/if carrying out the intervention puts the fundamental human rights of the intervener’s population at risk. This may be, among other possibilities, because the costs of war cut too deeply into the budget of the intervening state’s public services or because the intervention will somehow trigger physical threats to the lives of the population (perhaps retaliatory attacks). I have argued elsewhere that the calculation of risk does not usually include the risks incurred by intervening soldiers; see: Chrisantha Hermanson, ‘Duties in the Wake of Atrocity: A Normative Analysis of Post-Atrocity Peacebuilding’ (University of Oxford thesis, 2013), 143–59. Robert Nozick, ‘Coercion’, in I. Carter, M. Kramer, and H. Steiner (eds), Freedom: A Philosophical Anthology (Oxford: Blackwell Publishing, 2007), 266. Feinberg, Moral Limits, iii. 198. See e.g. James Boyce, ‘Aid Conditionality as a Tool for Peacebuilding: Opportunities and Constraints’, Development and Change, 33 (2002): 1035. For further details, see: Clair Apodaca, ‘Global Economic Patterns and Personal Integrity Rights After the Cold War’, International Studies Quarterly, 45 (2001): 593. This is a spinoff of the classical mistress scenario presented in Feinberg, Moral Limits, iii. 230. I realize that this case is a bit controversial, as some readers may object that we do not generally say that millionaires have a duty to pay for dying children’s medical care. I see no reason, however, why this should not be so, since, so long as the excessive risk proviso is met, this seems to be required by the duty to aid. I cannot argue this case fully here, so for those unconvinced, but who nevertheless hold that R is acting unjustly by way of a coercive or exploitative or otherwise unjust offer, we can agree to disagree on terms, for the distinct moral wrongs discussed regarding these cases do not hinge upon these terms. This does not preclude interveners from trying to promote these things, just from making aid conditional upon them. Boyce, ‘Aid Conditionality as a Tool for Peacebuilding’, 1041. For more discussion, see: James Boyce and Madalene O’Donnell, ‘Policy Implications: The Economics of Postwar Statebuilding’, in J. Boyce and M. O’Donnell (eds), Peace

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



29. 30. 31.

32.

CHALLENGES FOR HUMANITARIAN INTERVENTION

and the Public Purse: Economic Policies for Postwar Statebuilding (Boulder, CO: Lynne Rienner Publishers, 2007), 292–3. This assumes that R cannot prevent V₁ from murdering V₂ in some other less harmful way (e.g. helping V₂ flee). For a similar scenario, see: Haksar, ‘Coercive Proposals’, 72. Singer, ‘Famine, Affluence, and Morality’, 231. Whether Singer meant his account of the duty to aid to be interpreted in this way is unclear because throughout the remainder of the article he concentrates exclusively on the standard excessive risk proviso, so I may be taking some liberty here. Accordingly, my claim is not that Singer argues for (J1). My claim, is simply that (J1) is an appropriate understanding of the duty to aid and Singer’s statement expresses (J1) whether he intended this or not. For a moral critique of the DRC case regarding the complicity of international actors who provided aid to refugee camps, see: Chiara Lepora and Robert Goodin, On Complicity and Compromise (Oxford: Oxford University Press, 2013), 130–50. For details on the Rwanda-DRC case, see: Alex De Waal, ‘Democratizing the Aid Encounter in Africa’, International Affairs, 73 (1997): 629; Sarah Lischer, ‘Collateral Damage: Humanitarian Assistance as a Cause of Conflict’, International Security, 28 (2003): 79–80 and 107–8. For an instance of (W5) in Bosnia, see: Boyce, ‘Aid Conditionality as a Tool for Peacebuilding’, 1039.

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

8 ‘Words Lying on the Table’? Norm Contestation and the Diminution of the Responsibility to Protect Aidan Hehir

Introduction In this chapter I argue that the Responsibility to Protect (R2P) has failed to meaningfully influence the behaviour of states. Though still vaunted by some as a transformative concept, I argue that the evidence supplied to support these claims is inherently weak. Given R2P’s putative status as a norm rather than a law, its efficacy is premised on it being adopted by states through a process of socialization and rhetorical entrapment which, it is claimed, is impelled by states’ growing desire to avoid the ‘shame’¹ and/or ‘social exclusion’² incurred through the violation of R2P’s precepts. Yet, while this theory of norm diffusion has an empirical basis with respects to other norms, the evolution of R2P suggests that, rather than it having transformed the behaviour of states, states have transformed the character of R2P. The steady process by which R2P has been transformed into a largely ineffective empty signifier has been impelled to a significant degree by the rise of ‘new powers’ who have influenced the evolutionary trajectory of R2P. I illustrate this through an analysis of Russia’s engagement with R2P since the first General Assembly debate on the concept in 2009; this analysis demonstrates, I argue, both that Russia’s position has remained unchanged—if not in fact hardened—and that its interpretation of R2P has become the generally accepted version of the concept. Russia’s influence on the evolution of R2P has manifest in both the way R2P has come to be employed by the Security Council, and the extent to which Pillar III has been rendered moribund.³ Since 2011 the Security Council has regularly invoked R2P in Resolutions; while this has been lauded by some as

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

evidence of R2P’s growing traction,⁴ the manner in which R2P is used, I argue, coheres with the Russian—and indeed also the Chinese—position on R2P. In this sense, R2P’s increased usage, far from constituting evidence of its ‘progress’, actually constitutes evidence that it has been neutered by those emerging powers keen to reaffirm principles related to sovereignty and intervention that are diametrically opposed to the original ethos of R2P. Additionally, as Russia’s power has grown, so too has its willingness to overtly act in ways that manifestly run counter to the original ethos behind R2P; this is most evident with respects to Russia’s stance on Syria since 2011, which illustrates the limited traction of Pillar III.

R2P: Norm or Law? The origins of R2P are well known and not especially contentious; its efficacy, however, is a matter of some debate. R2P’s affirmation at the 2005 World Summit provided the concept with official recognition; it did not, however, mean R2P constituted a new law.⁵ The four crimes cited as being within R2P’s purview in the World Summit Outcome Document were proscribed long before 2005, as was the principle that the international community had a role to play in regulating states’ adherence to these laws domestically. R2P was, therefore, a means by which pre-existing laws could be invoked succinctly. As noted by Francis Deng— credited by many⁶ as the first to explicitly link sovereignty and responsibility through his work on internally displaced peoples in the 1990s—R2P was ‘nothing new’. Yet, he argued, R2P does have value because it is sharpening our consciousness of the problem and rallying forces together with a sharper focus on how to act to resolve, to address these problems, to protect populations. It’s almost a restatement of resolve and perhaps a rearrangement of our tools or a mobilization of the relevant tools and relevant actors, to be alert to what needs to be done and to be mobilized to do it.⁷

In this respect, R2P is held to constitute a source of pressure on states; a means by which states can be extolled both to behave responsibly internally and to act to prevent, and if necessary halt, mass-atrocity crimes abroad. Thus, rather than being a law, R2P is invariably portrayed as a norm.⁸ The role of norms became an increasingly popular issue within the discipline of International Relations in the post-Cold War era. In contrast to those who argued that states singularly behave according to narrowly defined national interests, constructivist accounts argued that prevailing norms within the society of states shaped state action.⁹ While these norms often did not have legal expression, they exercised a constraining influence on states by virtue of the fact that they

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

‘ WORDS

LYING ON THE TABLE ’?



delineated standards of appropriate behaviour.¹⁰ Acting in violation of these standards, it is argued, while not necessarily incurring automatic legal censure, isolates the violator and thus negatively impacts on their international reputation.¹¹ This view is based on more than just academic theory; numerous examples of extant norms that exercised a constraining influence on the behaviour of states have been advanced to add empirical weight to this proposition.¹² The constructivist account was naturally seized upon by those who believed in R2P’s transformative potential. With institutional reform deemed impossible, R2P ostensibly constituted ‘the only show in town’;¹³ thus the only viable means by which the behaviour of states could be altered was through the proliferation of R2P which, it was—and still is—claimed, would consolidate its status as a norm and thus enable it to exercise an influence over state behaviour.¹⁴ Therefore, drawing in particular on the norm life cycle model,¹⁵ proponents sought to increase the visibility of R2P; in particular great emphasis was placed on embedding the term in international political discourse so that states could not shirk their 2005 commitment to R2P.

Success in Theory: Failure Where it Matters There is no shortage of reflections extolling the transformative influence of R2P. According to some it has made ‘tremendous progress’¹⁶ and ‘begun to change the world’.¹⁷ These often effusive reflections on R2P are invariably premised on two claims; first that R2P commands international consensus and second that the term has become firmly embedded in international political discourse. Superficially, neither claim is false. The consensus on R2P is said to be evident in the fact that states do not tend to reject the concept. Instances when states have publicly disavowed R2P are very rare; in fact, invariably when states discuss the concept they express support for it.¹⁸ This has been particularly evident in the nine General Assembly debates on R2P held since 2009. While some states have at various stages expressed reservations about certain aspects of R2P—discussed in greater detail later in the chapter—it is difficult to find examples of outright public hostility to the principles underpinning the concept; it is certainly impossible to divine anything approximating a significant consensus amongst states that R2P should be abandoned. Given the normative evolution of norms as outlined in the norm life cycle model, this absence of opposition and prevalence of support is naturally seen by many as significant. Having emerged in 2005, R2P can, it is claimed, be seen by many to have gone through the various stages in the model with state support constituting

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

evidence of the ‘norm cascade’ phase. Had R2P—once advanced in 2005—been thereafter either ignored or rejected by states, naturally it would be impossible to claim that it was exercising an influence on state behaviour. The prevalence of rhetorical support for R2P, therefore, certainly coheres with the normative progress outlined in the life cycle model. Additionally, not only has R2P become a routinely avowed concept at the General Assembly, it has also become a term increasingly invoked in Security Council resolutions.¹⁹ To date, R2P has been mentioned in sixty-eight Security Council resolutions, sixty-four of which have been passed since 2011.²⁰ This proliferation of resolutions that mention R2P since 2011 has—again perhaps unsurprisingly—been heralded as evidence that R2P is making progress and exercising real-world influence.²¹ Additionally, the Security Council’s engagement with R2P is not unique, other international and regional organizations have demonstrated a similar increase in references to the concept since 2011.²² When one adds to this the increase in state membership of R2P-related groups such as the ‘R2P Focal Points’²³ and the ‘Group of Friends of R2P’,²⁴ it is not difficult to construct a narrative which holds that R2P is ‘making progress’ towards becoming a fully established norm that is increasingly exercising a meaningful influence on state behaviour.²⁵ Though widely espoused, this narrative is, however, ultimately, fundamentally flawed; it is based on a superficial interpretation of progress and, most obviously, doesn’t equate with contemporary international relations. If R2P was as transformative as some have claimed, we would naturally assume that this would be reflected in the behaviour of states, specifically with respect to their willingness to prevent and/or halt mass atrocity crimes. In fact the opposite trend is all too apparent; in recent years, myriad reports from NGOs, think-tanks, and humanitarian organizations have noted a pronounced degeneration in international respect for human rights, an increase in state oppression, a greater unwillingness on the part of the ‘international community’ to take action in response to looming or actual intra-state mass-atrocities, and a greater prevalence of actual mass-atrocities.²⁶ In his 2016 report on R2P, UN Secretary General Ban Ki-moon noted that the international community had consistently ‘fallen woefully short’ in its response to intra-state crises, and observed that the ‘frequency and scale of atrocity crimes have increased’.²⁷ The idea that R2P is making ‘tremendous progress’ clearly appears less convincing in light of these gloomy reflections. And yet, the evidence proffered in support of the ‘R2P is changing the world’ narrative is unquestionably true; states do routinely affirm R2P, and the Security Council has increasingly invoked the concept. More broadly, there is significant evidence to back up the claims that norms influence state behaviour and that

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

‘ WORDS

LYING ON THE TABLE ’?



R2P coheres with the norm life cycle model. What then explains the pronounced disjuncture between R2P’s emergence as a norm and its apparent inability to meaningfully influence the behaviour of states? This is the focus of the following section.

Explaining R2P’s Failure While norms are widely—though of course not universally—accepted as being influences on state behaviour, in recent years a new body of literature on the efficacy of norms has challenged the life cycle model. These ‘post-positivist constructivist’ accounts have noted that, once norms emerge, the means by which they are then proliferated has profound implications for their ultimate efficacy.²⁸ The repeated affirmation of a norm is not in itself evidence that the norm is effecting change; during the process of affirmation and the building of a consensus on the norm, the actual meaning and contours of the norm can alter.²⁹ This can be driven by the process of argumentation whereby the norm is moulded and reshaped, but not fundamentally altered. It can also, however, comprise a process whereby those hostile to the norm exercise a mendacious influence over the evolutionary trajectory of the norm. In this respect, those keen to blunt the impact of the norm—or just certain aspects of the norm—steer its development in particular ways without ever actually openly rejecting the norm.³⁰ Thus, rather than publicly disavowing the norm—and potentially incurring reputational damage and/or the hostility of others—‘creative resisters’ maintain a veneer of support and acquiescence whilst actually working to change the norm’s character so that it is shaped to cohere with their own interests.³¹ Fundamentally, the process by which a norm is consolidated is by definition influenced by the relative distribution of power amongst states at a given time.³² Norm diffusion clearly cannot happen in a vacuum and thus, given that the development of a norm is mediated through argumentation and debate amongst states, this process is invariably influenced by those states whose position carries significant weight. As has been widely noted, since at least 2008 a number of states have increased their relative power in international society; the rise of the ‘BRICS’ has come to be widely recognized and as having influenced the conduct of contemporary international relations. It stands to reason, therefore, that if these powers have increasingly influenced the prevailing norms in international politics, their influence must be evident with respect to the evolution of R2P. Indeed, the evolution of R2P is particularly linked to the emergence of these new powers as the concept was first debated at the UN General Assembly just as the influence of the new distribution of power is said to have begun to be felt.

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

To gauge the nature of the influence exercised by all the new powers is beyond the scope of one chapter; the following analysis looks in particular at the Russian position on R2P.

Russia’s Position on R2P Russia has never been an enthusiastic supporter of either humanitarian intervention or R2P. Its vocal opposition to the intervention in Kosovo in 1999, its policies in Chechnya, and its more general hostility to norms stemming from the ‘West’ predated the emergence of R2P. Russia has traditionally adhered to a particularly conservative interpretation of international law, preferring a literal reading of positive law to the various normative, natural law-oriented interpretations and reform agendas that emerged in the post-Cold War era.³³ In particular, Russia has consistently opposed intervention—both military and non-military forms—in the affairs of sovereign states without government consent.³⁴ Russia has, therefore, sought to resist the—invariably Western originated—challenges to positive law that have emerged in the post-Cold War era by seeking to consolidate and support the principles of ‘sovereignty, territorial integrity and non-interference in the internal affairs of states’.³⁵ It has advanced a relativist position on human rights, conceding only very limited grounds to the idea of ‘universal human rights’ which it has invariably characterized as ‘Western’ rather than universal. Russia has long believed that humanitarian principles are secondary to those laws governing inter-state relations, particularly with respect to the authorization of the use of force.³⁶ Russia’s stance was, for many years however, of limited relevance; given that the country’s economic and political woes had forced it to rely on Western aid, its capacity to meaningfully resist new ideas and practices relating to human rights, conditional sovereignty, and intervention was severely limited.³⁷ This was particularly apparent with respect to NATO’s intervention in Kosovo in 1999; though Russia tried to exercise influence over the resolution of the situation in Kosovo—then a province within the Federal Republic of Yugoslavia—its perspective was ultimately ignored.³⁸ Despite the Russian Foreign Minister Igor Ivanov warning at the time that any bombing of Yugoslavia would be treated as ‘a strike against Russia’, NATO simply went ahead with their military intervention.³⁹ Such an overt dismissal of Russian concerns is almost inconceivable today. By any standards, Russia’s power, and willingness to exercise this power, has increased dramatically since 2008; since then Russia has invaded Georgia, annexed Crimea, and engaged in sustained air strikes in Syria. It has also become increasingly belligerent at the UN, most notably through its repeated vetoing of

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

‘ WORDS

LYING ON THE TABLE ’?



resolutions designed to resolve the crisis in Syria.⁴⁰ If norms evolve through a process of discussion and debate, then clearly the influence of the dominant powers involved in this process must be felt. Given this, it would be curious if Russia’s new capabilities and disposition had not influenced the evolution of R2P especially as Western power has declined. As Ramesh Thakur notes, ‘Westerners have lost their previous capacity to set global standards and rules of behaviour’,⁴¹ and thus other actors—most obviously the BRICS countries—have a greater influence on the prevailing norms in contemporary international politics. Russia has had a particularly keen interest in shaping the evolution of R2P given its sensitivities towards the issues within R2P’s purview; as Derek Averre and Lance Davies note, ‘the R2P norm is widely contested in a pluralist international system and Russia’s political elite naturally seeks to shape its evolution’.⁴² Given Russia’s position prior to the emergence of R2P, it is not surprising that it greeted the concept with a large degree of scepticism. At the first Security Council meeting convened to discuss R2P, Russia characteristically advanced a narrowly positivist analysis of international law reiterating the primacy of state sovereignty and the illegality of humanitarian intervention.⁴³ Given that the version of R2P incorporated into the 2005 World Summit Outcome Document did not in any way alter existing international law, Russia’s consent to R2P’s affirmation at this point should not be interpreted as especially significant; indeed, one of the key concerns for the P5 in 2005—the US in particular—was ensuring that R2P’s inclusion in the Outcome Document did not affect their capacity to choose when, where, and how to respond to intra-state mass-atrocities.⁴⁴ Therefore, while for some 2005 was a breakthrough for R2P, a strictly positivist reading of the Outcome Document would interpret it as no more than a restatement of the existing system and thus of no immediate threat to Russia’s interests. Since the first General Assembly meeting on R2P in 2009, Russia has consistently issued statements which, though not overtly hostile to R2P, reiterate certain principles that certainly restrict its applicability. In 2009 Russia stressed that R2P ‘is in line with the provisions of the United Nations Charter’ and noted that no action should ever be taken unless it is ‘fully compliant with international law, in particular the United Nations Charter’.⁴⁵ This typically literal orientation cannot necessarily be cast as overt hostility to R2P, but the emphasis on abiding by the Charter is a conscious effort to ensure that two principles enshrined in the Charter remain unchanged; the inviolability of state sovereignty and the Security Council’s monopoly over the authorization of remedial responses to intra-state crises. Likewise, in its 2012 statement at the General Assembly debate on R2P, Russia again reiterated the primacy of state sovereignty, noting ‘the paramount obligation for protecting the responsibility of one’s own population lies with the

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

state. The role of the international community amounts, first and foremost, to providing the necessary assistance to the state in implementing this duty.’ With respects to Pillar III Russia declared that this aspect of R2P was ‘contradictory’ and in warning against external intervention noted that the response to Libya had ‘harmed the image of this concept around the world’. After further iterating a series of concerns regarding Pillar III, the statement concluded, ‘We do not have the confidence that in the world a consensus has been achieved on these issues’.⁴⁶ Russia’s 2015 statement on R2P again describes it as comprising several ‘timely and correct ideas’ but the statement goes on to note that these relate exclusively to Pillars I and II, namely those aspects of R2P that outline measures undertaken with the consent of the host state.⁴⁷ The 2016 Russian statement on R2P was perhaps the most critical to date; emphasizing again the damage inflicted by the 2011 intervention in Libya, the statement declared that the concept cannot be advanced ‘without an honest, multilateral analysis of what took place in Libya’, noted that it is ‘very difficult’ to envisage any further progress, and concluded by warning that the future of R2P is ‘not very clear’.⁴⁸ Russia’s hostility to Pillar III has been routinely justified by reference to the intervention in Libya in 2011. While Russia abstained on the vote on Resolution 1973—thereby facilitating its passage—thereafter it vocally criticized the means employed by NATO and in particular their focus on regime change once the intervention began. The Libya intervention, Russia has argued, illustrated the extent to which Pillar III of R2P is fundamentally compromised by the absence of clarity regarding when and where such intervention is legitimate; reflecting on the Libya intervention Foreign Minister Sergei Lavrov stated, ‘The question to be asked about the responsibility to protect is, if we discuss it seriously: is it a right or an obligation . . . if we say it is an obligation what are the criteria for interference?’⁴⁹ Russia argued that NATO used R2P as a pretext to engage in an interestled intervention which ultimately had a negative impact on peace and stability in Libya and the wider region; the post-intervention violence in Libya has been routinely cited by Russia as evidence of the destruction Pillar III ‘humanitarian’ interventions can cause. This indeed, has been one of the justifications advanced by Russia for its stance on Syria; while Russia’s willingness to repeatedly veto resolutions on Syria has been widely condemned, Russia has argued that Libya illustrates how interfering in sovereign states can exacerbate existing crises.⁵⁰

Russia and the Nature of R2P Today Evidence of Russia’s influence on the evolution and contemporary character of R2P is certainly discernible. As discussed above, Russia has essentially sought to

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

‘ WORDS

LYING ON THE TABLE ’?



circumscribe the development of R2P, specifically by limiting it to Pillar I and II, while opposing Pillar III. That this is Russia’s interpretation of R2P is, therefore, quite clear; whether this has become a more widely held interpretation, of course, is a more important matter. The evidence suggests this is indeed the case. Rather than constituting one norm, R2P is better understood as a collection of norms cohering with the three-pillar conception routinely advanced.⁵¹ It is possible, therefore, that consensus on R2P may be limited to only some of the norms/pillars. This certainly appears to be the case; opposition to Pillar III and the interpretation of the Libyan intervention as both a violation of Resolution 1973 and ultimately a disaster, are certainly not positions unique to Russia. Amongst the BRICS countries, these sentiments can be seen to have manifested in Brazil’s idea of ‘Responsibility While Protecting’⁵² and the Chinese notion of ‘Responsible Protection’,⁵³ while South Africa voiced similar criticisms both of NATO’s actions and Pillar III more generally.⁵⁴ The divisive manner in which NATO engaged in the intervention in Libya, and Libya’s subsequent descent into violent civil war, certainly added greatly to the appeal of those arguments advanced by those states sceptical about Pillar III of R2P. As Gareth Evans admitted, in the wake of the Libyan intervention, ‘consensus has simply evaporated’ on all but the most limited aspects of R2P,⁵⁵ a point also acknowledged by UN Secretary General Ban Ki-moon⁵⁶ and former UN Secretary General Kofi Annan.⁵⁷ The much vaunted ‘consensus’ on R2P also reflects the Russian position; while, as noted earlier, it is undeniably true that states have invariably expressed their support for R2P at the General Assembly debates held on the concept since 2009, the nature of this consensus is revealing. Very obviously, state consensus on R2P orientates almost exclusively around Pillars I and II; while some—primarily Western—states have at times iterated their support for Pillar III, the R2P states express their support is invariably limited to, first, the principle that states have a responsibility to protect their populations from the four crimes listed in the 2005 World Summit Outcome Document and second, the permissibility of providing international assistance to states actively seeking support to meet their internal responsibilities.⁵⁸ Regarding the first point of consensus, Jennifer Welsh notes that this evidences a desire amongst states to preserve ‘legal egalitarianism’ whereby paragraphs 138 and 139 of the World Summit Outcome Document are interpreted and employed in such a way as to reiterate the principles of sovereign inviolability and sovereign equality.⁵⁹ Indeed, Welsh has noted that many states have expressed their support for R2P, ‘precisely because it was not seen as transformational’.⁶⁰ The restriction of the meaning of R2P, therefore, explains the growing ‘consensus’ on R2P and is not necessarily illustrative of a new disposition amongst previously oppressive states. Indeed, engagement by states

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

with Pillar III has steadily diminished to the point that it today hardly features in international political debates on R2P. So divisive has Pillar III become that some have suggested it should be formally removed from R2P to minimize ongoing contention.⁶¹ Yet, without Pillar III—which essentially constitutes the means by which R2P is enforced—declaring support for Pillars I and II of R2P alone essentially imposes minimal constraints on states; as Bellamy acknowledged in a speech to the General Assembly in 2012, ‘without the use of force in the RtoP toolkit, the international community would effectively need to rely on the perpetrators to deliver protection’.⁶² The fact that the R2P affirmed today is one shorn of Pillar III therefore explains why many states—such as Bahrain, Sudan, and Qatar— actively engaged in committing systemic human rights violations domestically, have felt able to express their support for the concept; if declaring support for ‘R2P’ means affirming only that states should protect their only people and the international community may (if asked) help them in so doing then why wouldn’t states support this concept? Supporting this circumscribed rendering of R2P in no way exposes states to external censure and is, therefore, essentially cost-free. Therefore, the disjuncture between the rapid growth in state support for R2P and its actual influence on the behaviour of states becomes more understandable; prior to the emergence of R2P the international legal system was widely criticized on the basis that states were essentially entitled to self-regulate the commitments they made to international human rights law.⁶³ By the time R2P emerged there was certainly no shortage of laws proscribing human rights violations and yet R2P emerged precisely because these laws were being routinely flouted owing to the means by which they were enforced. States were, therefore, evidently comfortable with iterating their support for human rights and signing related treaties, because they knew that the regulation of their compliance with these commitments was inherently weak. While R2P emerged in response to this systemic failing, it—paradoxically—affirmed this very system. It can hardly be a surprise, therefore, that state behaviour has not changed since R2P emerged, given that it has not in any way altered the means by which state commitments to international human rights law are regulated. States are naturally quite willing to express their support for R2P provided it is the circumscribed Pillar I and II rendering, as this merely enables them to maintain their pre-existing disposition towards their commitments to international human rights law. The significance of the ‘consensus’ on R2P is, therefore, diminished and the lack of any meaningful impact of this consensus on the actual behaviour of states is understandable. The other key basis for the claims made about R2P’s ‘tremendous progress’ stems from the Security Council’s increased use of the term in resolutions. Again,

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

‘ WORDS

LYING ON THE TABLE ’?



however, while the numbers look impressive, an analysis of the manner in which R2P is actually employed reveals a very clear trend, namely that the references to R2P in Security Council resolutions have been consistently orientated to Pillars I and II. The resolutions passed since 2011 which refer to R2P do so in the context of reiterating Pillar I; that the host state is primarily responsible for intra-state affairs. Pillar III is not mentioned in any of the Security Council resolutions passed to date. R2P has, therefore, been used only to reaffirm the legal principle— which predates R2P—that states have the primary responsibility to prevent or halt atrocities crimes within their territory.⁶⁴ This is, as noted previously, a position which Russia—and China—have consistently held since R2P was first affirmed in 2005. Their positions have not changed and clearly their preferences have found expression in the Security Council resolutions on R2P. Beyond just the manner in which R2P has been employed by the Security Council, Russia’s influence on the evolution of R2P in practice has been readily evident with respect to its stance on Syria. While R2P was established as a means to improve the capacity and willingness of the international community—and especially the Security Council—to respond in a timely and effective manner to ongoing intra-state mass-atrocities, Russia has consistently impeded international efforts by repeatedly vetoing draft resolutions seeking to impose sanctions against President Assad’s regime. Quite obviously, Russia’s determination to support a key ally has in this instance trumped its commitment to R2P. This can hardly been deemed to constitute a surprise; many warned long before the crisis in Syria that R2P constituted little more than rhetoric and ‘ear candy’ and that it would always be supplanted by the P5’s narrow national interests when it really mattered.⁶⁵ It must be remembered that Russia’s vetoing of the draft resolutions on Syria is perfectly legal; one may (reasonably) question the morality of these vetoes, but given that R2P has not in any way changed the means by which the international response to intra-state crises are authorized, veto-wielding powers like Russia are legally empowered to impose their interests on the manner in which the ‘international community’ responds to any situation. The P5 are constitutionally entitled to choose whether, and how, to enforce international laws proscribing atrocity crimes. Their power to act thus constitutes merely a ‘discretionary entitlement’.⁶⁶ The P5—Russia and China in particular—have been widely condemned for their stance on Syria; respected international actors⁶⁷ have blamed divisions at the Security Council for the loss of hundreds of thousands of lives, and in August 2012 the General Assembly condemned the Security Council’s inaction.⁶⁸ In May 2014 a coalition of NGOs and think-tanks decried Russia and China’s vetoing of a draft resolution on Syria as ‘shameful’.⁶⁹ Yet, this was the fourth time these

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

states had cast a double veto on Syria, and this consistent—and seemingly implacable—opposition to remedial efforts suggested that neither state was especially concerned about incurring ‘shame’. Given that the efficacy of norms is to a great extent dependent on social shaming rather than legal censure, this lack of concern has grave implications for the efficacy of R2P.

Conclusion R2P’s profile has undeniably risen dramatically since the 2005 World Summit. It is repeatedly affirmed by states at the General Assembly, the Security Council regularly invokes the concept, states have in ever increasing numbers joined R2P groups and campaigns, and a plethora of think-tanks and NGOs have emerged singularly devoted to promoting the concept. Within academia, R2P has spawned its own mini-industry, becoming a source of countless books, articles, special issues, and research grants. Yet, the aim of R2P’s architects wasn’t to generate rhetorical support or widespread coverage; rather the goal was ‘delivering practical protection for ordinary people’.⁷⁰ Even a cursory glance at the state of human rights protection in the world today will illustrate that this goal has yet to be attained. To understand why requires an engagement with the literature on norms which examines the means by which these ideas and concepts can be manipulated during their evolution. While Bellamy’s claim that R2P commands ‘genuine and resilient international consensus’⁷¹ may on one level appear to be correct, it is imperative to examine both the nature of the rendering of R2P around which there is consensus, and identify the actors who have sought to mould this rendering. I have argued that the R2P states routinely affirm and employ—at both the General Assembly and the Security Council—is merely a restatement of principles regarding sovereignty and the permissibility of consent-based international assistance; in this respect, R2P has been manipulated to the point that it is impotent. The process by which R2P has come to be reduced to Pillars I and II illustrates Kowert and Legro’s warning that states involved in shaping the evolution of a norm may engage in the ‘instrumental manipulation of norms’ to further selfish ends.⁷² Naturally, those actors with greater leverage will exercise more influence on the evolution of a norm; Mona Lena Krook and Jacqui True’s analysis of norm development thus notes that debate conducted amidst participants where power asymmetries are evident invariably reflects the influence of power, with the powerful ultimately ‘determining what can and cannot be said’.⁷³

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

‘ WORDS

LYING ON THE TABLE ’?



This process is evident, I have argued, with respects to the Russian position on R2P.⁷⁴ Given that it is widely recognized that Russian power has increased since 2008, it would be odd indeed if Russia’s perspective on R2P hadn’t influenced its evolution since that time. An examination of Russia’s stance demonstrates that ‘Moscow’s own position has changed little’⁷⁵ but also that this position coheres neatly with the circumscribed rendering of R2P around which there is the championed ‘consensus’. Russia’s stance on R2P has consistently constituted what Averre and Davies describe as ‘a strict interpretation of the 2005 World Summit Outcome Document’ whereby its support for R2P is narrowly focused only on Pillars I and II.⁷⁶ In itself, the fact that Russia continues to hold this position is significant given that it is a member of the P5; beyond this, however, it is clear that the more general disposition of states—the source of the ‘consensus’ on R2P—is also orientated towards just reaffirming Pillar I and Pillar II. While this is not necessarily regressive, it cannot plausibly be deemed progressive given that this consensus predated R2P and was manifestly insufficient to guard against the occurrence of mass-atrocity crimes. R2P has, therefore, like other norms in the past, suffered ‘co-optation’.⁷⁷ Thus, the fact that R2P is invoked ‘habitually’⁷⁸ cannot be held to illustrate its influence; indicatively R2P has recently been described by Paddy Ashdown as merely ‘an interesting collection of words lying on the table’.⁷⁹ Rather than evidence of its efficacy, R2P’s repeated invocation is illustrative of the extent to which the concept has been moulded and circumscribed—by emerging powers like Russia—so that it now constitutes a largely ineffectual rhetorical device.

Notes 1. Gareth Evans, ‘R2P: Looking Back, Looking Forward’, Keynote Address, Phnom Penh, Cambodia, 26 Feb. 2015, . 2. Alex Bellamy, The Responsibility to Protect: A Defence (Oxford: Oxford University Press, 2015), 61. 3. In his 2009 report on R2P, UN Secretary General Ban Ki-moon outlined three ‘Pillars’ of R2P: Pillar I: ‘the enduring responsibility of the State to protect its populations’; Pilar II: ‘the commitment of the international community to assist States in meeting those obligations’; Pillar III: ‘the responsibility of Member States to respond collectively in a timely and decisive manner when a State is manifestly failing to provide such protection’. Ban Ki-moon, ‘Implementing the Responsibility to Protect’, UN Secretary General Report, A/63/677, 12 Jan. 2009, , 8–9. 4. Jess Gifkins, ‘R2P in the UN Security Council: Darfur, Libya and Beyond’, Cooperation and Conflict, 51/2 (2016): 148–65; Simon Adams, ‘The Responsibility to Protect: Ten Years On’, OpenCanada, 8 May 2015, ; Bellamy, The Responsibility to Protect, 26; Tim Dunne and Katherine Gelber, ‘Arguing Matters: The Responsibility to Protect and the Case of Libya’, Global Responsibility to Protect, 6/3 (2014): 326–49; Luke Glanville, ‘Does R2P Matter? Interpreting the Impact of a Norm’, Cooperation and Conflict, 51/2 (2016): 184–99; Global Public Policy Institute, ‘Effective and Responsible Protection from Atrocity Crimes: Towards Global Action’, Policy Paper, Apr. 2015, , 10. Ban Ki-moon, ‘Implementing the Responsibility to Protect’, 2; Bellamy, The Responsibility to Protect, 15; Jennifer Welsh, ‘Conclusion: The Evolution of Humanitarian Intervention in International Society’, in Jennifer Welsh (ed.), Humanitarian Intervention and International Relations (Oxford: Oxford University Press, 2006), 210; Carsten Stahn, ‘Responsibility to Protect: Political Rhetoric or Emerging Legal Norm?’, American Journal of International Law, 101/1 (2007): 120. Gareth Evans, The Responsibility to Protect: Ending Mass Atrocity Crimes Once and for All (Washington, DC: Brookings Institution Press, 2008), 35–7. Francis Deng, ‘JISB Interview: The Responsibility to Protect’, Journal of Intervention and Statebuilding, 4/1 (2010): 86. Bellamy, The Responsibility to Protect, 61; Jennifer Welsh, ‘Norm Contestation and the Responsibility to Protect’, Global Responsibility to Protect, 5/4 (2013): 395; Alexander Betts and Phil Orchard, ‘Introduction: The Normative Institutionalization-Implementation Gap’, in Alexander Betts and Phil Orchard (eds), Implementation and World Politics: How International Norms Change Politics (Oxford: Oxford University Press, 2014), 1. Jeffrey Checkel, ‘The Constructivist Turn in International Relations Theory’, World Politics, 50/2 (1998): 324–48. Peter Katzenstein, ‘Introduction: Alternative Perspectives on National Security’, in Peter Katzenstein (ed.), The Culture of National Security (New York: Columbia Press, 1996), 5. Paul Kowert and Jeffrey Legro, ‘Norms, Identity and their Limits’, in Peter Katzenstein (ed.), The Culture of National Security (New York: Columbia Press 1996); Thomas Franck, The Power of Legitimacy among Nations (Oxford: Oxford University Press, 1990), 24. James March and Johan Olsen, Rediscovering Institutions (New York: Free Press, 1989); Thomas Risse, ‘International Norms and Domestic Change: Arguing and Communicative Behaviour in the Human Rights Era’, Politics and Society, 27/4 (1999): 529–59; John Ruggie, ‘Territoriality and Beyond: Problematizing Modernity in International Relations’, International Organization, 47/1 (1993): 139–74. Alex Bellamy, ‘The Responsibility to Protect After the 2005 World Summit’, Policy Brief No. 1, Carnegie Council on Ethics and International Affairs, 2006, 15. Bellamy, The Responsibility to Protect, 72; Glanville, ‘Does R2P Matter?’, 184–99; Jennifer Welsh, ‘Implementing the Responsibility to Protect: Catalyzing Debate and Building Capacity’, in Alexander Betts and Phil Orchard (eds), Implementation and World Politics: How International Norms Change Politics (Oxford: Oxford University Press 2014), 125; Dunne and Gelber, ‘Arguing Matters’, 326–49; Cristina Badescu and

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

‘ WORDS

15.

16. 17. 18.

19.

20. 21.

22. 23. 24.

25.

26.

LYING ON THE TABLE ’?



Thomas Weiss, ‘Misrepresenting R2P and Advancing Norms: An Alternative Spiral?’, International Studies Perspectives, 11/4 (2010): 354–74. Martha Finnemore and Kathryn Sikkink, ‘International Norm Dynamics and Political Change’, International Organizations, 52/4 (1998): 887–917; see also Jeffrey Checkel, ‘Norms, Institutions, and National Identity in Contemporary Europe’, International Studies Quarterly, 43/1 (1999): 83–114. Adams, ‘The Responsibility to Protect: Ten Years On’. Bellamy, The Responsibility to Protect, 111. Gareth Evans, ‘R2P: The Next Ten Years’, in Alex Bellamy and Tim Dunne (eds), Oxford Handbook of the Responsibility to Protect (Oxford: Oxford University Press, 2016), 2; Adams, ‘R2P at Ten’; Ban Ki-moon, ‘A Vital and Enduring Commitment: Implementing the Responsibility to Protect’; Gifkins, ‘R2P in the UN Security Council’; Global Public Policy Institute, ‘Effective and Responsible Protection from Atrocity Crimes’. Adams, ‘The Responsibility to Protect: Ten Years On’; Bellamy, The Responsibility to Protect, 26; Dunne and Gelber, ‘Arguing Matters’, 326–49; Thomas Weiss, ‘Military Humanitarianism: Syria hasn’t Killed it’, Washington Quarterly, 37/1 (2014): 10. Global Center for R2P, ‘UN Security Council Resolutions Referencing R2P’, 2016, . Gifkins, ‘R2P in the UN Security Council’; Bellamy, The Responsibility to Protect, 26; Glanville, ‘Does R2P Matter?’, 184–99; Global Public Policy Institute, ‘Effective and Responsible Protection from Atrocity Crimes’, p. 10. Ban Ki-moon, ‘A Vital and Enduring Commitment’, 3–4. Global Centre for R2P, ‘R2P Focal Points’, 2016, . Group of Friends of the Responsibility to Protect, ‘Statement by the Group of Friends of the Responsibility to Protect in Geneva at the Informal Interactive Dialogue with Under-Secretary-General Mr. Adama Dieng, Special Adviser to Secretary-General on the Prevention of Genocide’, 3 Mar. 2016, . Simon Adams, ‘Statement of the Global Centre for the Responsibility to Protect at the 2016 UN General Assembly Informal Interactive Dialogue on the Responsibility to Protect’, 6 Sept. 2016, . UNHCR, ‘World at War: UNHCR Global Trends’, 2015, , 3–5; Uppsala Conflict Data Program, ‘Organized Violence in the World 2015’, , 1; Human Rights Watch, World Report 2015, , 1; Amnesty International, ‘Annual Report 2014/2015’, ; Jean-Marie Guéhenno, ‘The World’s Fragmenting Conflicts, International Crisis Group: The Future of Conflict’, 26 Oct. 2015, ; Freedom House, ‘Freedom in the World: 2016’, .

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

27. Ban Ki-moon, ‘Mobilizing Collective Action: The Next Decade of the Responsibility to Protect’, Report of the Secretary-General, A/70/999–S/2016/620, 22 July 2016, 3. 28. Amitav Acharya, ‘The R2P and Norm Diffusion: Towards a Framework of Norm Circulation’, Global Responsibility to Protect, 5/1 (2013): 466–79; Mona Lena Krook and Jacqui True, ‘Rethinking the Life Cycles of International Norms: The United Nations and the Global Promotion of Gender Equality’, European Journal of International Relations, 18/1 (2010): 108; Jeffrey Checkel, ‘Norms, Institutions, and National Identity in Contemporary Europe’, International Studies Quarterly, 43/1 (1999): 83–114; Wayne Sandholtz, ‘Dynamics of International Norm Change’, European Journal of International Relations, 14/1 (2008): 103; Shareen Hertel, Unexpected Power (Ithaca, NY: Cornell University Press, 2006). 29. Antje Wiener, ‘Enacting Meaning-in-Use: Qualitative Research on Norms and International Relations’, Review of International Studies, 35/1 (2009): 175–93; Alan Bloomfield and Shirley Scott, Norm Antipreneurs and the Politics of Resistance to Global Normative Change (Abingdon: Routledge, 2016); Krook and True, ‘Rethinking the Life Cycles’, 103–27; Kees Van Kersbergen and Bertjan Verbeek, ‘The Politics of International Norms’, European Journal of International Relations, 13/2 (2007): 217–38. 30. Jack L. Goldsmith and Eric A. Posner, ‘Introduction’, Journal of Legal Studies, 31 (2002): 104. 31. Alan Bloomfield, ‘Norm Antipreneurs and Theorising Resistance to Normative Change’, Review of International Studies, 42/2 (2016): 331; see also Patrick Quinton-Brown, ‘Mapping Dissent: The Responsibility to Protect and its State Critics’, Global Responsibility to Protect, 5/3 (2013): 260–82. 32. Krook and True, ‘Rethinking the Life Cycles’, 108. 33. Roy Allison, Russia, the West, and Military Intervention (Oxford: Oxford University Press 2013), 2. 34. Roy Allison, ‘Russia and Syria: Explaining Alignment with a Regime in Crisis’, International Affairs, 89/4 (2013): 818; Derek Averre and Lance Davies, ‘Russia, Humanitarian Intervention and the Responsibility to Protect: The Case of Syria’, International Affairs, 91/4 (2015): 817–18. 35. Averre and Davies, ‘Russia, Humanitarian Intervention’, 824. 36. Averre and Davies, ‘Russia, Humanitarian Intervention’, 825. 37. Allison, Russia, the West, and Military Intervention, 44–70. 38. Oleg Levitin, ‘Inside Moscow’s Kosovo Muddle’, Survival, 42/1 (2000): 131; James Hughes, ‘Russia and the Secession of Kosovo’, Europe-Asia Studies, 65/5 (2013): 993. 39. Quoted in Alex Bellamy, Kosovo and International Society (Basingstoke: Palgrave 2002), 166. 40. Justin Morris, ‘Libya and Syria: R2P and the Spectre of the Swinging Pendulum’, International Affairs, 89/5 (2013): 127. 41. Ramesh Thakur, ‘R2P after Libya and Syria: Engaging Emerging Powers’, Washington Quarterly, 36/2 (2013): 62. 42. Averre and Davies, ‘Russia, Humanitarian Intervention’, 813. 43. Welsh, ‘Conclusion: The Evolution of Humanitarian Intervention’. 44. Theresa Reinhold, ‘The Responsibility to Protect: Much Ado about Nothing?’, Review of International Studies, 36/S1 (2010): 67; Aidan Hehir, The Responsibility to Protect:

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

‘ WORDS

45. 46. 47. 48. 49. 50.

51. 52. 53.

54. 55. 56. 57.

58.

59. 60. 61.

62.

LYING ON THE TABLE ’?



Rhetoric, Reality and the Future of Humanitarian Intervention (Basingstoke: Palgrave Macmillan, 2012), 80. Russia, ‘Statement at the July 2009 GA Debate on RtoP’, . Russia, ‘Statement at the 2012 GA Debate on RtoP’, 2012, . Russia, ‘Statement at the 2015 GA Debate on RtoP’, 2015, . Russia, ‘Statement at the 2016 GA Debate on RtoP’, 2016, . Quoted in Averre and Davies, ‘Russia, Humanitarian Intervention’, 819. J. Thornhill, N. Buckley, and C. Clover, ‘Interview with President Dmitry Medvedev’, Financial Times, 19 June 2011; Y. Homsy, ‘Russia Rules Out Libyan Scenario in Syria’, RIANovosti, 9 Dec. 2012, . Bellamy, The Responsibility to Protect, 72; Glanville, ‘Does R2P matter?’, 184–99; Welsh, ‘Implementing the Responsibility to Protect’, 125. Brazil, ‘Responsibility While Protecting: Elements for the Development and Promotion of a Concept’, UN Document A/66/551-S/2011/701, 11 Nov. 2011. Ruan Zonze, ‘Responsible Protection: Building a Safer World’, China Institute of International Studies, 15 June 2012, . Alex Beresford, ‘A Responsibility to Protect Africa from the West?’, International Politics, 52 (2015): 288–304. Gareth Evans, ‘R2P and RWP After Libya and Syria’, Keynote Address to Stanley Foundation, 23 Aug. 2012, . Ban Ki-moon, ‘A Vital and Enduring Commitment’, 12–13. Natalie Nougayrède, ‘Interview with Kofi Annan: “On Syria, It’s Obvious, We Haven’t Succeeded” ’, Le Monde, 7 July 2012, . Ban Ki-moon, ‘A Vital and Enduring Commitment’, 12–13; Morris, ‘Libya and Syria’, 1265–83; Roland Paris, ‘The “Responsibility to Protect” and the Structural Problems of Preventative Humanitarian Intervention’, International Peacekeeping, 21/5 (2014): 1–35. Welsh, ‘Norm Contestation and the Responsibility to Protect’, 394. Welsh, ‘Norm Contestation and the Responsibility to Protect’, 373. Morris, ‘Libya and Syria’, 1265–83; Adrian Gallagher, ‘The Responsibility to Protect Ten Years on from the World Summit: A Call to Manage Expectations’, Global Responsibility to Protect, 7/3 (2015): 254–74; Paris, ‘The “Responsibility to Protect” and Structural Problems’; Jason Ralph, ‘R2P at 10: Looking Beyond Military Intervention’, OpenCanada, 21 May 2015, . Alex Bellamy, ‘Remarks to the General Assembly Informal Interactive Dialogue on the Responsibility to Protect’, New York, 5 Sept. 2012, .

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

63. Fitzmaurice Malgoviaa, ‘The Practical Working of the Law of Treaties’, in Mark Evans (ed.), International Law (Oxford: Oxford University Press 2006), 205–6. 64. Simon Chesterman, ‘Leading from Behind: The Responsibility to Protect, the Obama Doctrine, and Humanitarian Intervention After Libya’, Ethics and International Affairs, 25 (2011): 79–285. 65. Rhoda Howard-Hassman, ‘A Truth Commission for Africa?’, International Journal (Autumn 2005): 999–1032; Aidan Hehir, ‘Responsibility to Protect: Sound and Fury Signifying Nothing?’, International Relations, 24/2 (2010): 218–39. 66. Frank Berman, ‘Moral versus Legal Legitimacy’, in Charles Reed and David Ryall (eds), The Price of Peace (Cambridge: Cambridge University Press 2007), 161. 67. Navi Pillay, ‘UN Human Rights Chief Criticises UN over Global Conflicts’, Guardian, 22 Aug. 2014, ; UN News Centre, ‘Kofi Annan Resigns as UN-Arab League Joint Special Envoy for Syrian Crisis’, 2 Aug. 2012, . 68. General Assembly, 124th Plenary Meeting, 3 Aug. 2013, . 69. Global Center for R2P, ‘Joint NGO Statement on Use of the Veto’, 22 May 2014, . 70. International Commission on Intervention and State Sovereignty, The Responsibility to Protect (Ottawa: International Development Research Centre, 2001), 11. 71. Bellamy, The Responsibility to Protect, 2. 72. Kowert and Legro, ‘Norms, Identity and their Limits’, 493. 73. Krook and True, ‘Rethinking the Life Cycles’, 108. 74. Though China too has a similar understanding of R2P, see Aidan Hehir, ‘Assessing the Influence of the Responsibility to Protect on the UN Security Council during the Arab Spring’, Cooperation and Conflict, 51/2 (2016): 170–1. 75. Averre and Davies, ‘Russia, Humanitarian Intervention’, 832. 76. Averre and Davies, ‘Russia, Humanitarian Intervention’, 822. 77. Goldsmith and Posner, ‘Introduction’, 104. 78. Bellamy, The Responsibility to Protect, 161. 79. Tom Esselmont, ‘Analysis: As Syrian Deaths Mount, World’s “Responsibility to Protect” Takes a Hit—Experts’, Thompson Reuters News Foundation, 24 Oct. 2016, .

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

9 Responsibility to Protect, Polarity, and Society R2P’s Political Realities in the International Order Robert W. Murray and Tom Keating

The conclusion, then, is not that solidarism is inevitably limited, because international society is characterized by a deep and abiding tension between its pluralist and solidarist conceptions. Rather, the solidarist project is ‘premature’, because state leaders have not taken the moral risks that would create in international law a doctrine that would give practical substance to Kofi Annan’s claim that we are finally becoming a ‘humanity that will do more’ for fellow humans in danger. A key part of this moral transformation is an acceptance by governments in the West that humanitarian intervention is both morally permitted and morally required in cases of supreme humanitarian emergency.¹

In his 2000 call to arms, Nicholas Wheeler identified two of the most daunting barriers to state action when confronted with humanitarian emergency—the need to grapple with the realities of states as moral agents and the challenge of implementing a solidarist normative agenda in the society of states that would lead to a substantive body of international law that obligated states to act in cases of atrocity crimes. Wheeler’s comments came after a decade of the post-Cold War international order that was expected to see an international society grounded in Western values and embody the perception that Western liberalism had won the Cold War and the society of states would no longer be hamstrung by the bipolar conflict that limited state action throughout the Cold War era. Instead, post-Cold War international society was confronted with the realities of multiple failures to act in the face of egregious human rights abuse in cases like Somalia, Rwanda, Bosnia, and Kosovo.

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

It was not supposed to happen this way. American unipolarity in the wake of the Cold War finally meant that the normative framework for international society could and would include human rights as a priority for all states, and that action could take place now that the UN Security Council could overcome the stalemate of the Cold War years. The rise of the human security agenda coupled with a seeming willingness to speak of human rights as more than a Western ideal led to the belief that a new international order had emerged with the collapse of the Soviet Union that would see the triumph of Western ideas. After a decade of abject failure, new efforts were made to address the shortcomings of states taking any meaningful action, as Wheeler put it, to ‘save strangers’. The culmination of this redoubled effort was the publication of the International Commission on Intervention and State Sovereignty’s (ICISS) 2001 report titled The Responsibility to Protect (R2P). The ICISS proposed a significant change not only in how states perceive their role in enforcing the international human rights regime, but more, fundamentally sought to transform one of the most enduring institutions of international society—national sovereignty. R2P, according to the ICISS, would reframe sovereignty as conditional upon a state protecting its own people and moved from ‘sovereignty as control to sovereignty as responsibility in both internal functions and external duties’.² While heralded as a victory rhetorically, the ICISS proposal for R2P failed to gain traction, particularly after the 9/11 attacks, and thus another effort to reframe rights protection culminated in the 2005 iteration of R2P that eroded the transformative efforts of the ICISS version of the doctrine, and instead saw states reaffirm their existing commitments to conventions that called for action in four specific instances—war crimes, crimes against humanity, ethnic cleansing, and genocide. The World Summit Outcome document had the support of 192 heads of state and claimed: We are prepared to take collective action, in a timely and decisive manner, through the Security Council, in accordance with the Charter, including Chapter VII, on a case-bycase basis and in cooperation with relevant regional organizations as appropriate, should peaceful means be inadequate and national authorities manifestly fail to protect their populations from genocide, war crimes, ethnic cleaning and crimes against humanity.³

Over a decade has now passed since the World Summit Outcome proclamation, and since that time, there has been a growing rhetorical consensus around affirming R2P, yet there has been no successful implementation of a consistent framework for enforcing R2P’s tenets. This chapter seeks to explain why R2P implementation has largely been a failure to date and, further, why the ongoing evolution from a unipolar international system to a multipolar structure is likely

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

R  P , POLARITY , AND SOCIETY



to further erode the prospects of successful R2P enforcement. The chapter will take an English School theory approach to examine the failures of international society to successfully negotiate a post-Cold War order that would have seen human rights as a fundamental institution of the society of states, and will argue that international societies are created, and their normative frameworks are determined, by the order states negotiate, which is inextricably linked to the polarity of the international system. The chapter will go on to contend that the emerging multipolar system will place strain on negotiating an order that can incorporate Western and non-Western values without states sacrificing security based on who the great powers will be, and this will have potentially grave consequences for R2P and human rights protection more broadly.

Polarity and Negotiating Order When theorizing about a particular international order, it is imperative to first identify the number of dyads in the system that represent the great powers of a particular historical era. Great powers dominate the character of international politics from one era to another, and the structure of the international system is based on the number of great powers interacting with each other at once. The structure of the system can be unipolar, meaning there is a single superpower or hegemon dominating the system; bipolar, meaning there are two main powers interacting in the system at once; or multipolar, where more than two great powers are interacting at once. The structure of the international system determines not only how great powers will interact, but how other middle or minor powers will choose their respective strategies in terms of allying with great powers, forming blocs of power, and also the normative framework that will influence international society. In an anarchic international system where distrust, misperception, and the quest for survival dominate the motives of states, the fear of war is a powerful motivator and an ever-present concern. Martin Wight summarizes the international system by arguing: ‘In such a situation, mutual mistrust is fundamental, and one power can never have an assurance that another power is not malevolent. Consequently, with the best will in the world no power can surrender any part of its security to another power.’⁴ The material nature of anarchy, described by Herbert Butterfield as the ‘absolute predicament and irreducible dilemma’ of international politics, is what, in many ways, determines state action and forms the basis for their decisions.⁵ Among the most important decisions states have made throughout history is to willingly consent to forming a society of states in an effort to mitigate the impact of anarchy and prevent the outbreak of war.

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

Throughout the history of the state system, states of all power levels have used the fear of war and motive of survival as motivation to form international societies. The normative framework of these societies, which are not world governments or binding organizations, but are rather loose frameworks of state cooperation, is heavily dependent on the great powers dominating the international system at the time a society is formed. Robert Jackson summarizes the theory behind the society of states as follows: The conceptual key to international society is the manner in which sovereign states associate and relate: the character and modus operandi of their association and relations. It is formal in a significant way: it involves procedural standards of conduct, an essential normative basis of which is international law. However, it is also substantive in an equally significant way as it involves the pragmatic encounters of the separate national interests of those same independent states which, although subject to international law, are still free to lay down their own foreign policies.⁶

The consensual nature of forming a society of states, either in the regional or global sense, is essential to understanding how international societies can function. Equally as important are the practices and normative institutions required for the sustainability of a society of states. Tom Keating explains the relationship between state practice and the perpetuation of international society as he argues: ‘states both constitute and are constituted by international society. This makes it important to consider the practice of states as a way of assessing the acceptance of and commitment to international society and of considering the extent to which international society has shaped the behaviour of these states.’⁷ For a group of inherently self-interested actors operating in an anarchic system to come together willingly to form a society of states that mitigates the possibility of conflict, and has, at various times throughout history, worked to uphold or implement a normative framework codified through international law, is especially noteworthy when discussing issues like humanitarian intervention and more recently the R2P. The role of international law is a core principle upon which international society is built. Wight summarizes this idea by noting that ‘The most essential evidence for the existence of an international society is the existence of international law. Every society has law, which is the system of rules laying down the rights and duties of its members.’⁸ To understand how states interpret international law, and the principles and frameworks that comprise laws among states, it is essential to understand that states themselves make international law, which becomes part of the normative character of an international society. Jackson summarizes the role of states in creating international law:

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

R  P , POLARITY , AND SOCIETY



In international law states are the primary right and duty bearing units . . . States are the subjects of international law but that does not mean that states are the creatures of international law. Rather, the opposite is the case: international law has been created by sovereign states; they are its authors. They have made themselves its subjects. States are the leading authorities of international society, their authority is what makes international law authoritative.⁹

If states, then, are responsible for creating, implementing, and enforcing international law, this requires that states exist absent to law, and therefore sovereignty becomes an a priori condition for statehood and entry into international society. Further, one of the key organizing principles of an international society is a reciprocal recognition of sovereignty,¹⁰ which thus places sovereignty as among the most important variables in comprehending how states interact with each other and in helping to form various international societies throughout history. Following this argument, sovereignty is the core organizing principle in imbuing states with the necessary status and conditions to willingly agree to be part of a society of states that is then governed by rules and institutions, often articulated through international law, that members of the society negotiate. By seeking to mitigate anarchy, states form international societies and work toward establishing order. It is in this context that the role of great powers becomes noteworthy, as it is very much those great powers who, out of a desire to preserve their independence, sovereignty, and relative power position, help to form international societies and who view themselves as the custodians of societies of states.¹¹ By working to establish order, great powers are able not only to protect themselves, but also to promote rules and norms that are consistent with their interests in international society. International order is a necessary condition for the society of states to achieve its purposes, and order negotiated among the members of international society. Evelyn Goh defines international order as: ‘a pattern or arrangement that sustains the primary goals of a society of states. It must involve limits on behaviour, the management of conflict, and the accommodation of change without undermining the common goals and values of society.’¹² Great powers play a crucial role in negotiating, establishing, and maintaining international order, and according to Bull: Great powers manage their relations with one another in the interests of international order by (i) preserving the general balance of power, (ii) seeking to avoid or control crises in their relations with one another, and (iii) seeking to limit or contain wars among one another. They exploit their preponderance in relation to the rest of international society by (iv) unilaterally exploiting their local preponderance, (v) agreeing to respect one another’s sphere of influence, and (vi) joint action.¹³

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

Bull is sure to note that the great power roles described here are what powers can do, not always what they choose to do. It is these options available to great powers that become the basis upon which the institutional and normative agendas of international society are determined. Great powers are most interested in protecting themselves, their spheres of influence, and in promoting their interests. The extent to which great powers are willing to use their relative power position in the system to promote certain norms depends on the structure of the system, and also who the other great powers are, if the system is anything but unipolar. Options iv, v, and vi listed in Bull’s description of relationship management show that great powers can opt to respect the sphere of influence of other powers or work together in an area of common interest. For the context of the debate over humanitarian action and R2P, these considerations matter a great deal. If order is established in a multipolar system that sees powers from various areas of the world without a strong sense of cultural or normative consensus, international society will have relatively little ability to implement institutions that some powers see as more reflective of others. If a system is bipolar, the dynamics of the negotiated order and society of states will depend on the same variables, but the effort to find consensus in negotiating a normative framework will be limited to two powers rather than multiple powers. The most obvious conclusion is that a unipolar system should be the most favourable to pursuing a normative agenda reflective of that power’s own cultural dynamics and interests. The next section explores the dynamics of humanitarian intervention in the postCold War order and finds that, despite having over two decades of American unipolarity, rhetorically favourable to acting in the face of atrocity crimes, the negotiated order and normative framework of the post-Cold War international society has yet to make substantive progress towards implementation of doctrines like R2P or a coherent framework for the protection of civilians.

R2P, Humanitarian Intervention, and the Post-Cold War Order The post-Cold War period provided a rare opportunity for the development of an international order comprised of more substantive norms than had been possible for much of the Cold War. The so-called ‘liberal moment’ promised the introduction of an order free from the competitive, antagonistic stasis of the Cold War and the possibility that Western states could, through their example and superior power, lead a revolution from above whereby they would fulfil the liberal democratic aspirations of people throughout the globe. Just causes would now have the support of preponderant military power unencumbered by the need to

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

R  P , POLARITY , AND SOCIETY



deal with competing worldviews or power centres. Not surprisingly, in this light the period was marked by a significant amount of deliberation over human rights, humanitarian principles and practices, and efforts to devise new standards for state behaviour and obligation in these areas. It has also been a period marked and marred by numerous interventions ostensibly to support these standards and define a new international order. Commencing with decisions to establish safe areas in Iraq in the aftermath of the first Gulf War, to send UN forces into an eroding Yugoslavia, and deploy a protection force to Somalia, the 1990s were marked by a greater proclivity to intervene for various causes including to protect civilians, support post-conflict transitions, and facilitate democratic reforms. The UN and regional organizations were actively engaged in many interventions during this period.¹⁴ Initially, the commitment to assertive multilateralism and a robust United Nations operating in defence of human rights received support from the US government in the early days of the Clinton Administration. Whatever the motivation for leading such a reformation on the part of American officials, it was short-lived, hijacked on the streets of Mogadishu and mired in the seemingly endless conflict that accompanied the dismantling of Yugoslavia. This was the first indication that interests would continue to govern where, when, and how the US would intervene, regardless of the human rights situation on the ground. The effect was felt most immediately by victims in Rwanda and Srebrenica, and became a consistent theme throughout the Clinton years and well beyond. Despite this reversal in US policy, other states including France, Canada, and the United Kingdom took up the humanitarian cause in different ways, supporting human security agendas, instituting an International Criminal Court, and undertaking selective deployments in West Africa. By the time what remained of the former Yugoslavia flared up again over the situation in Kosovo in the late 1990s, Western governments were primed for another attempt at a humanitarian intervention intended to protect civilians at risk. The decision to intervene militarily against the Serbian government in defence of Kosovars was taken by NATO member states without the consent of the United Nations, but with the approval of its then-Secretary General Kofi Annan. The rush to intervention and the dismissal of sovereignty claims by the out-gunned Serbian government suggested a willingness to disregard or perhaps redefine the norms of international order, particularly regarding the institution of sovereignty, which was and remains at the core of intervention discussions. The intervention not only ignored the procedural norms of UN Security Council approval that had been an important framework of the existing international order, but it denied any participatory role for other great powers, specifically

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

Russia, who had in the past assumed some measure of responsibility for designing and sustaining this negotiated order and who had its own interests at play in the region.¹⁵ In one respect the intervention demonstrated the apparent willingness of states to use their own military forces to protect civilians at risk from tyranny in other countries and established a precedent for post-Cold War international society that has since played an important role in perceptions and decisions of intervention, being that of ‘illegal but legitimate’. ‘This interpretation is situated in a gray zone of ambiguity between an extension of international law and a proposal for an international moral consensus. In essence, this gray zone goes beyond strict ideas of legality to incorporate more flexible views of legitimacy.’¹⁶ Kosovo marked a crucial turning point in post-Cold War humanitarian interventions and the prospects for a new international order. By ignoring the UN Security Council, the governments who supported the Kosovo intervention were undermining the credibility of the institution and the procedural norms it embodied. Given the morality of the cause both they, and the Secretary General, found this defensible. For others, it raised serious concerns. ‘If the Kosovo war is employed as a precedent for allowing states, whether singly or in coalition, to ignore or contradict the UNSC based on their own interpretation of international morality, the stabilizing function of the UNSC will be seriously imperiled, as will the effort to circumscribe the conditions under which recourse to force by states is permissible.’¹⁷ The threat to the UNSC was particularly a challenge to other powers. This was less important for Britain and France who supported the intervention, but was of much greater concern to Russia and China, as well as to aspiring powers such as India and Brazil. The implications went beyond the specific issue of Kosovo and severely undermined the support for intervention from the wider community of states, and by association for reframing principles surrounding this practice, such as R2P. Xyzema Kurowska has pointed to Kosovo’s significance in shifting Russian policy. This trajectory is perhaps best illustrated by Russia’s initial acceptance of Western normative and strategic leadership following the demise of the Soviet Union. However, the perceived slights from the West in the 1990s quickly gave rise to the revival of the great power discourse which resists liberal norms and contests humanitarian intervention. The failure to indulge this resistance has led to the (Putin) administration’s normative defiance of the liberal world order.¹⁸

The repercussions from these interventions was evident as proponents of the R2P principle sought its acceptance at the United Nations. The idea of R2P percolated through the international community and the UN for a few years before it was formally acknowledged in the UN’s 2005 Summit Outcome Document. For some,

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

R  P , POLARITY , AND SOCIETY



the inclusion of R2P in the final document represented a confirmation by the UN of this norm that could now be deployed to protect civilians at risk and hold to account those who committed atrocities including the national governments of sovereign states. Advocates hailed references to R2P as a demonstration of its acceptance, but the principle that was approved in the document added little substance to the normative framework of international society, and more, did virtually nothing to fundamentally alter the institution of sovereignty, which would have been required if R2P was to ever be meaningfully implemented as a means of protecting populations in crisis. A different reading would see the 2005 version of R2P as more of an effort to rein in the principles of sovereignty and great power politics, and reaffirm a procedural norm that had been circumscribed by practices in the name of human rights. The inclusion of references to ‘Responsibility to Protect’ in the document may obscure the degree of uncertainty and concern about this principle among many governments. Debates leading up to its adoption demonstrated a considerable degree of opposition to the principle as envisioned by its advocates, suggesting concerns about its implications for sovereignty, for the use of force, and for the position of weaker powers in international society.¹⁹ ‘The eventual agreement produced a watered-down R2P that imposed no obligations on states, refrained from mentioning armed intervention, and insisted that R2P operations require Security Council approval.’²⁰ In effect its inclusion was intended in part to reinforce the role of the UN, and more precisely the UN Security Council, as the principal forum in which decisions about intervention and decisions regarding implementation would occur, confirming that the Permanent Five would retain the final say over if, when, and how R2P was to be applied. Once again, however, the practice would subvert the intent. The systemic conditions of unipolarity that encouraged and enabled this more aggressive pursuit of interventionist practices also fostered a degree of hubris that proved counterproductive as it undermined any incentive to pursue a more inclusive body of support for such interventions among the majority of states in international society. In actual fact such efforts were modest, at best, and were plagued with contradictions that mitigated their effectiveness in rallying support behind the cause. The justification for pressing ahead without wider support rested on a view that Western governments were at the forefront of a substantive change that others would be willing to support once they had adjusted to the righteousness and efficacy of this new ethically infused international order.²¹ The self-confidence of Western governments tended to obscure their view of the legitimacy of any other opinions that might vary from their own. They also failed to recognize the negotiated character of international order. Any ability to create a legitimate and sustainable new order, whatever its content, would require a

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

wider level of support than that provided by like-minded NATO members. Whatever these governments might have thought of the limits of international law, their willingness to flagrantly violate them in the name of their ‘just’ causes only served to call into question their motives and ultimately to weaken efforts at advancing R2P and humanitarian intervention efforts rather than helping them along. Once these motives became infected by specific security concerns following the terrorist attacks of 2001, it was even more difficult to disentangle the normative aspirations from the more parochial national interests. While Darfur and the Congo waited in vain for their humanitarian rescue, Afghanistan and Iraq had their sovereignty dismantled in the name of national security. The same fate awaited Libya in 2011. Edward Newman has recounted the statement of the President of the General Assembly, HE Miguel d’Escoto Brockmann, that ‘no one could refute the basic principle of R2P, but the manner in which it’s promoted and applied are characteristic of broader pathologies in the international system related to the abuse of power, structural inequalities and unrepresentative norms and institutions’.²² The events surrounding the decision to intervene militarily in Libya in late 2011 confirmed this sentiment. For some proponents, the United Nations Security Council Resolution 1973 authorizing the no-fly zone to protect civilians at risk in Libya marked the epitome of the responsibility to protect principle’s application in the post-Cold War period and represented a tangible example of states renegotiating fundamental characteristics of international order as it related to interventionism. For others, it demonstrated much that was wrong with the principle, as subsequent developments transformed this resolution into a very different intervention resulting in yet another instance of regime change.²³ The interveners’ decision to pursue regime change in Libya was not well received, nor was the clear lack of post-regime change plan. And it was not only major powers who expressed these concerns. For example, the South African representative to the UN expressed his government’s view that ‘[a]busing the authorization granted by the Council to advance a political regime-change agenda does not bode well for the future action of the Council in advancing the protection of civilians agenda’.²⁴ The main objective of many states in supporting the inclusion of R2P into the 2005 summit document was to strengthen the UNSC’s authority over when and how the principle would be applied. NATO’s early and complete rejection of the limits imposed by UNSC Resolution 1973 shattered any confidence in the organization’s ability to manage interventions once they had been put in motion. ‘R2P’s critics, particularly China and Russia, have learned from the Libyan campaign that the intermediate non-military steps the doctrine specifies can easily be skipped, the continuum ignored, and the

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

R  P , POLARITY , AND SOCIETY



military element operationalised rapidly and expansively so that it sidelines everything else.’²⁵ This may never have been the intention of those who framed the principle or those who supported its adoption by the UN, but the Libyan intervention confirmed that intentions had little meaning or influence on the actions of dominant powers motivated more by their own interests than by enshrining particular norms and that, whatever the merits of R2P, the potential for abuse was considerable. ‘In Russian eyes, Libya was part of a continuum, one more domino falling after Kosovo, Afghanistan and Iraq. The priority was to prevent Syria from following suit.’²⁶ Rhetoric aside, the pattern of interventions has never been purely or even primarily motivated by normative considerations. National interests, especially the interests of the most powerful members of these coalitions, have always been an integral part of these humanitarian interventions. The connection between interests and more normative concerns has been made by political leaders such as Tony Blair and also by scholars such as Nicholas Wheeler. As both recount, there is nothing inherently objectionable about interests informing such operations. Problems arise, however, when viewing such practices in terms of ‘interests’ overwhelms and takes precedence over humanitarian considerations. Looking at instances where Western governments responded to crimes against humanity and where they ignored them suggests the effort to reconcile these interventions with national interests has been the most important factor in shaping the decision to intervene and the character of the intervention. As a result, the ethical standard itself has been diminished and tarnished. It is also evident that the core support for interventions and for R2P has corresponded with a collection of states who have also exhibited a broader agenda of support for liberal democratic norms and the assertion of such norms in the face of sovereignty. A second consideration that has tarnished these interventions is power. Kofi Annan once lamented that one of the problems with humanitarian interventions was that ‘weak states are far more likely to be subjected to it than strong ones’.²⁷ The truth of the matter is that weak states are the only ones to be subjected to it. The power differential that is an inherent part of these interventions further obscures the legitimacy of the principle. These conditions make it difficult for most states to accept an argument that these norms have any currency and should receive their support. The practice to date does little to question Menon’s view that ‘despite its egalitarian allure and homage to justice, in practice R2P will simply reinforce existing hierarchies’.²⁸ While this is unproblematic for those sitting atop the hierarchy, it is deeply troubling for those not so fortunate, some of whom see their own sovereignty very much at risk if such an order were to be adopted. ‘Reassurances from R2P’s proponents that force would be a rare,

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

last-ditch response have not calmed critics, who respond that R2P cannot make good on its commitment to universality. Weak states will remain at the mercy of powerful ones, while the latter will have no cause to fear, despite the magnitude of their misdeeds.’²⁹ At the same time, other major powers might be concerned where their particular interests are at stake, even indirectly. As Newman writes, these states ‘have also been associated with a broader, assertive liberal internationalism geared towards the promotion of democracy and market economics, and the containment of rogue states [and] despite attempts to define [R2P] narrowly, controversy surrounding R2P can be explained in part by its association with this broader liberal agenda.’³⁰ The combination of these other interests has discouraged a more universal response. The manner in which it has been pursued has ignored many of the core procedural norms of international society. Great power management has been set aside in a futile effort to assert a narrow, at times unilateralist, approach. The opposition that has emerged from existing and emerging powers reflects not only competing interests but competing views of world order and their own role in defining it.

Prospects for R2P in the Emerging Multipolar Order The prospects for a judiciously conceived and universally applied R2P were never very good. The principle appeared to have captured a sentiment that emerged out of the failure to prevent or subsequently end civilian atrocities that were occurring in various parts of the world. It reflected a genuine concern for the vulnerability of civilians and a view that sovereignty was a barrier that could be breached by an international community that would operate along more principled lines. This concern remains an important feature of contemporary discussions about the role of the UN and the need to respond to vulnerable populations at risk. It did not, however, represent as its proponents repeatedly sought to argue a significant shift in prevailing views towards sovereignty. Indeed, sovereignty and the principle of territorial integrity remain central to all governments and core standards for a peaceful and stable international order. Most importantly, proponents of the principle did not appreciate (i) the unwillingness on the part of states to let principles guide interests, rather than the reverse; (ii) the extent to which practice would tarnish the principle and lead reluctant states to be even more circumspect; and (iii) the negotiated nature of international order. There is some evidence of a willingness in the society of states to adopt a more principled approach to protecting civilians at risk and to intervene in the affairs of other states primarily to protect these individuals. There is no willingness to

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

R  P , POLARITY , AND SOCIETY



compromise sovereignty in any meaningful way to do this. Additionally, there is no willingness to allow selected major powers the ability to determine if, when, and how such protections are to be put in place. Any realistic effort to develop a truly effective and sustainable norm of civilian protection must address the concerns that have surrounded the discussion of the Responsibility to Protect norm and of the actual practice of interventions since the end of the Cold War. The story of R2P’s emergence from the chaotic and inconsistent practice of intervention during the 1990s has been widely recounted.³¹ Developments since the end of the Cold War suggest that interventions have failed to move very far beyond the interest-based practices of earlier periods. Yet when combined with efforts to enshrine moral standards into international practices such as UN resolutions and new organizations like the ICC, they raise concerns about where these institutions are headed and what value they might have for governing or affecting international order. For Newman, debates over R2P reflect not only differences over the content of sovereignty or human rights norms, but also broader shifts in the process by which international norms come to be adopted and diffused. Thus, the concerns of China, India, Russia, South Africa, Brazil, amongst others, about how R2P is defined and implemented reflect tensions about the legitimacy and authority of norm diffusion, collective decision making and international institutions . . . In this sense, resistance amongst rising powers to R2P is a manifestation of a broader resistance to the normative ‘rules of the game’ and the manner in which power is legitimately exercised—or illegitimately abused—in international relations.³²

There remains a fundamental problem of real inclusiveness in these efforts to devise standards that are intended to be universal. As Newman also reminds us, much of the opposition stems both from who is doing the promotion and how it is being done. ‘Many states are instinctively guarded towards the principle because it is championed by Western states, and because it is promoted through organisations—such as the UN and NATO—that are regarded in some ways as being unrepresentative or unaccountable. In this sense, therefore, R2P raises tensions in relation to the question of how, and by whom, norms are diffused.’³³ The lack of responsiveness to the position of other states was one issue that led Louise Arbour to reassess her views of such interventions. ‘Having worked in the human-rights field internationally, I now find it incredible how the West seems to be absolutely incapable of hearing what it sounds like to the rest of the world—a total disconnect, in the promotion of what it rightly believes are universal values, while being completely oblivious to the fact that others don’t take this at face value as being a good-faith pursuit of universal goods. It just doesn’t work.’³⁴

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

Most important has been the lack of any concerted effort to incorporate the interests of other great powers. Justin Morris has suggested that the great powers have colluded to some degree and have taken on the idea of R2P and have sought to manage expectations surrounding its enforcement with ongoing concerns about international order. The tension that this raises has been managed through the UN Security Council and has led, in his view to an implicit acceptance of a responsibility not to veto R2P measures. In the existential standoff of the Cold War this phenomenon might have been at its weakest and geostrategic concerns most prominent, but it suggests, nevertheless, that a lifting of the geostrategic barriers to action, combined with a loosening of the legal constraints to intervention, would be more conducive to heralding a pattern of state practice in which intervention was more frequent than would a change in material or normative conditions alone.³⁵

It is difficult to assess the likelihood of this being sustained in the future, for events in Libya brought to the fore the problems with R2P as it was being pursued by Western governments in a highly selective manner that seemed to abuse UNSC authorization in pursuit of its own agenda rather than conforming to the idea of supporting state efforts to meet their obligations. In Libya’s case, such responsibilities were twice breached by disposing the Gaddafi government rather than holding it to account and then departing while failing to support a replacement government that interveners were quick to recognize, but unwilling to support in any tangible and sustainable manner. Given the inability of international society to effectively negotiate a postCold War international order that could have included provisions for consistent mechanisms of human rights protection, the evolving multipolar order poses an even greater challenge for how states and institutions will approach the issue. Libya demonstrated that the P5 members of the UN Security Council could come together and make decisions in the name of protecting vulnerable populations, but the conduct of the mission by NATO, and the resulting instability that plagues Libya to this day, have served more as a warning against approving such action than an example of why such actions can be a positive force in the society of states. If there was any hope of implementing doctrines like R2P or urging states, especially major powers like Russia and China, to move towards fundamentally renegotiating the institution of sovereignty, it would have been during the unipolar moment when US soft power could have been focused on such efforts. Perhaps the greatest tragedy of the postCold War international society has been the squandered opportunity of the US

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

R  P , POLARITY , AND SOCIETY



to effect positive change in international society, instead focusing on numerous efforts to ineffectively project US interests on a global scale that did not include strong progress on human rights protection. The emerging multipolar system will not only see the decline of the US preponderance of power, but it also signals the rise of non-Western great powers that are already playing a much larger role in international society. The instability created by, and aggression of, Russia in Eastern Europe and of China in the South China Sea demonstrate quite clearly that the emerging order will be negotiated and predicated on self-interest, traditional Westphalian views of sovereignty, and a pluralist society of states focused more on survival and power than human rights. The lack of universal consensus on core issues relating to human rights protection and R2P referenced in this chapter will become even more pronounced in an order where anti-Westernism will be perceived as a sign of strength. R2P’s outlook throughout the era of American hegemony was never entirely optimistic, but with powers like Russia and China now playing a much larger role in determining the normative framework of international society, it becomes incredibly difficult to see a way forward for R2P that results in its successful application. The question now for R2P advocates is how to best negotiate a new order in a multipolar structure where the progress made in the last two decades relating to human rights protection is not completely lost in the larger game of great power politics. The fact that Russia and China have both, at least rhetorically, signalled a willingness to entertain such discussions, particularly through the UN, can be seen as a positive, but when both of those states, and others, are in a more powerful position to rival US interests in certain regions, will their rhetoric match reality? To date, the outlook looks bleak, particularly given Russia’s overt aggression in Crimea and Ukraine, the disregard for civilians demonstrated with the downing of MH 17, and Russia’s actions in Syria. Beyond external activities, China and Russia have shown open disregard for human rights at home, thus casting further doubt on any willingness to commit to such norms in the society of states. If progress is to be made on R2P, interventionism, the protection of civilians, or interventionism more broadly, it falls to the states, particularly the great powers, negotiating the institutional framework of international society to determine how this might be possible. History has shown that states can alter institutional meanings and create more robust solidarism, but it depends largely on the circumstances influencing a new order, and the states negotiating it. There may be hope for R2P on a much smaller or regional scale, but at a global level, the paralysis of R2P’s progress to date is unlikely to see improvement given recent trends.

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

Notes 1. Nicholas Wheeler, Saving Strangers: Humanitarian Intervention in International Society (Oxford: Oxford University Press, 2000), 310. 2. International Commission on Intervention and State Sovereignty, The Responsibility to Protect (Ottawa: International Development Research Centre, 2001), 13. 3. United Nations, World Summit Outcome 2005, para. 139. 4. Martin Wight, Power Politics (London: Continuum, 2004), 101. 5. Herbert Butterfield, History and Human Relations (London: Collins, 1951), 19. 6. Robert Jackson, The Global Covenant: Human Conduct in a World of States (Oxford: Oxford University Press, 2000), 102. 7. Tom Keating, ‘The Transition in Canadian Foreign Policy through an English School Lens’, International Journal, 69/2 (2014): 170. 8. Wight, Power Politics, 107. 9. Jackson, Global Covenant, 102. 10. See Wheeler, Saving Strangers, 22. 11. Hedley Bull, The Anarchical Society: A Study of Order in World Politics, 3rd edition (New York: Columbia University Press, 2002), 17. 12. Evelyn Goh, The Struggle for Order: Hegemony, Hierarchy, and Transition in PostCold War East Asia (Oxford: Oxford University Press, 2013), 7. 13. Bull, Anarchical Society, 200. 14. Alan Kuperman, ‘R2P: Catchy Name for a Fading Norm’, Ethnopolitics, 10/1 (2011): 125–8. 15. See Shaun Narine, ‘NATO and the New Western Imperialism’, in Robert Murray (ed.), Seeking Order in Anarchy (Edmonton: University of Alberta Press, 2016), 197–218. 16. Independent Commission on Kosovo, The Kosovo Report Conflict, International Response, Lessons Learned (Oxford: Oxford University Press, 2000), 164. 17. Independent Commission, Kosovo Report, 174. 18. Xyzema Kurowska, ‘Multipolarity as Resistance to Liberal Norms: Russia’s Position on Responsibility to Protect’, Conflict, Security and Development, 14/4 (2014): 491. 19. Edward Newman, ‘R2P: Implications for World Order’, Global Responsibility to Protect, 5 (2013): 244–6. 20. Ranjan Menon, The Conceit of Humanitarian Intervention (Oxford: Oxford University Press, 2015). 21. See e.g. Martha Finnemore, ‘Fights about Rules: The Role of Efficacy and Power in Changing Multilateralism’, Review of International Studies, 31 (2005): 187–206. 22. Newman, ‘R2P: Implications for World Order’, 249. 23. For commentaries on the Libyan intervention, see Aidan Hehir and Robert Murray (eds), Libya, The Responsibility to Protect and the Future of Humanitarian Intervention (Houndmills: Palgrave Macmillan, 2013). 24. Cited in Nathalie Tocci, ‘On Power and Norms: Libya, Syria and the Responsibility to Protect’, Global Responsibility to Protect, 8 (2016): 70. 25. Menon, Conceit of Humanitarian Intervention, 1274. 26. Tocci, ‘On Power and Norms’, 71.

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

R  P , POLARITY , AND SOCIETY



27. Kofi Annan, ‘We the Peoples’: The Role of the United Nations in the 21st Century (New York: United Nations, 2000), 48. 28. Menon, Conceit of Humanitarian Intervention, 1676. 29. Menon, Conceit of Humanitarian Intervention, 1665. 30. Newman, ‘R2P: Implications for World Order’, 242. 31. See e.g. Gareth Evans, The Responsibility to Protect: Ending Mass Atrocity Crimes Once and for All (Washington DC: Brookings Institution Press, 2008), and Alex J. Bellamy, Responsibility to Protect (Cambridge: Polity Press, 2009). 32. Newman, ‘R2P: Implications for World Order’, 236. 33. Newman, ‘R2P: Implications for World Order’, 252. 34. Louise Arbour cited in Doug Saunders, ‘Why Louise Arbour is Thinking Twice’, Toronto Globe and Mail, 28 Mar. 2015. 35. Justin Morris, ‘The Responsibility to Protect and the Great Powers: The Tensions of Dual Responsibility’, Global Responsibility to Protect, 7 (2015): 407.

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi

10 Closing the R2P Chapter Opening a Dissident Current within Philosophy of War Sagar Sanyal

This book appears at a time when I suspect philosophical attention to the Responsibility to Protect (R2P) is waning. I say this because attention to this debate is not driven by the playing out of purely intellectual dynamics. Rather it is driven in large part by the issue being a current one in political discourse in the West. The exhaustion of the R2P research programme within analytic philosophy of war is an opportunity for setting a new research agenda. There is a diversity of political views represented among philosophers of war. Many are dissidents, highly critical of their own governments, including the US, UK, Canada, and Australia, the main centres of analytic political philosophy. It is this dissident subset whom I address. I am not much interested in the rest, and I expect they will likely find unpersuasive and arbitrary the political claims I advance in the chapter, and the materialist approach I sketch at the end. As one set of issues runs out of steam for paper topics, there will be a temptation to return to the mainstream analytic approach to philosophy of war to find new ones. For the dissident subgroup, I plead that they not do so, because the problem-set and philosophical framework of the analytic approach is an ideological strait-jacket—not because it yields false answers to the questions you do consider, but because it keeps you to safe questions that won’t discomfit your masters. Your dissidence does not have full intellectual space in which to develop and flower within those confines. As an alternative to the analytic debate, I outline the Marxist approach to philosophy of war. First I present reasons why the political salience of the R2P framework is waning. The following section is an ideological critique of the analytic approach

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi

CLOSING THE R  P CHAPTER



to philosophy of war. The three strands of the critique concern (i) idealist and materialist foundations for international politics, (ii) the urgency of different tasks we can take up as philosophers, and (iii) the narrowness of perspective of the just war theory tradition. In the third section I outline a materialist foundation for philosophy of war to contrast to the analytic foundations (which I characterize as idealist). The materialist foundation hinges on the Marxist account of imperialism. The next section then applies that theory of imperialism to the broad geopolitical sweep of post-Second World War history. The section ends with a look at the escalating imperial tensions between the US, Russia, and China.

How Material Developments Occasion, and Exhaust, Ideological Debates The philosophical debate on humanitarian intervention (HI) and Responsibility to Protect (R2P) arose in response to the development of these frameworks in high policy circles in the African Union, the UN, and the states of leading NATO powers like the US and UK. It was taken up by the ‘just war theory’ (JWT) mainstream of the philosophy of war, which represents a liberal democratic tradition that is optimistic about the morality, transparency, and accountability of the leading militaries of the West, in particular the US, UK, France, and other important NATO powers. Outside this mainstream are theorists from a range of outlooks (anti-imperialist liberals, socialists and anarchists, thinkers from the former colonies), who do not share this benign view of the leading militaries and find it problematic to have the HI debate (which assumes generally trustworthy intervening militaries). They view the leading military states as having certain material dynamics. The policy framework—whether a Cold War framing, or an HI and R2P framework, or more recently a terrorist framing—is placed atop these deeper dynamics. Occasionally there may be military interventions that are on balance humanitarian, just as there might have been Cold War interventions that were on balance aiding popular struggle against a political regime backed by the opposing superpower. But these occasional instances do not justify the intellectual energy devoted to sustaining and developing these frameworks. In the critical or non-mainstream vein, Noam Chomsky’s books on HI and R2P are the most prominent condensation of what is in fact a range of dissident political analysis.¹ But these made barely a dent in the analytical philosophy of war debate on HI. The general approach is not to answer Chomsky’s empirical judgements about the trustworthiness of leading states (anecdotally, in my experience it is rare for analytic philosophers of war to have read Chomsky’s

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

books on HI), but to wave it away as non-mainstream political history. It certainly is non-mainstream: that is the point of dissident camps in any discussion. This should be no barrier to engagement in an academic discipline that professes to discuss war, which is, after all, a phenomenon with a political history. Even worse, philosophers might define Chomsky’s issues as being outside of philosophy. Philosophers think about JWT criteria, the im/permissibility of killing non-combatants, and other debates one can carry out in the philosopher’s armchair. The reality of whether any of this is relevant or useful to the actual phenomenon of war is for someone else—a political scientist, a historian, a journalist. The analytic philosophy debate could refuse to engage with such antiimperialist critiques in the heyday of the HI and R2P debate when the mainstream news media continued to present the US and its Western allies as dogooders. If they persist in this approach today, I suspect that they will appear decreasingly relevant to political reality and increasingly seen as ivory tower ideologues for ‘our side’. This is because certain political developments over the past decade make it much harder to ignore the dissident perspective on the leading states. The developments make it harder to assume that the leading states are trustworthy vehicles of the HI and R2P powers, and so make it increasingly clear to philosophers of war that those debates may suit policy circles (which are circumscribed by diplomatic and ideological boundaries of assuming noble sovereign actors in the West), but precisely because the debates come from the policy circles of leading military states, they are not a neutral, nor an insightful framework for philosophizing about war. The political developments I have in mind are the following. There are internal developments in the leading liberal democratic states so that it is not plausible any more to treat these states as the decent, benevolent, democratic states that ideally constitute the intervening powers. The HI and R2P debate would hardly have gotten going if all leading military powers were readily acknowledged to be as undemocratic and repressive as we rightly hold Russia and China to be. The second issue is external, as it is much clearer that we are at the start of an era of greater likelihood of imperial wars, including among the leading military powers, and not just small ‘rogue states’. In the era of rising imperial war, the great military powers who were to be the disinterested global cops under R2P are instead explicitly competing with one another for territory and economic advantage. Internally, the most recent development has been the Black Lives Matter (BLM) movement in the heart of the ‘benevolent empire’, the main agent of humanitarian military interventions in the R2P framework. The response of the

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi

CLOSING THE R  P CHAPTER



police and its state backers has polarized the population on whether the US is a democratic state, or a racist and authoritarian one. BLM protesters have been met with militarized police, appalling impunity of police murderers, an ideological fightback by police associations under the ‘blue lives matter’ slogan that refuses to back down and insists that it needs more resources to ‘keep us safe’. The BLM movement has greatly widened public awareness of the militarization of the police over the past couple of decades.² In Western Europe the major internal development is the rise of fascist parties. This includes parliamentary parties which have organized mobs that intimidate and bash migrants, refugees, and Muslims.³ The context for this is West European states’ doubling down on austerity in the face of enormous protest (often repressed by militarized police). Finally, the ‘War on Terror’ framework in the West has led to the repeal of basic civil liberties, the ramping up of extra-legal surveillance and infiltration of domestic minority and political groups, the specious judicial redefinition of torture, and the explicit repudiation of international treaties. Externally, rising tensions between the major military powers of the world, especially the US, Russia, and China, suggest that the wars of the near future are likely to be imperial ones. Even short of outright war between these powers, the timely issue is not the intervention of benevolent global powers to rescue benighted foreigners in regional conflicts. Especially with the Syria war, but with the quashing of the Arab Spring more broadly, the problem is the military intervention of global and regional powers in stoking and perpetuating conflicts. The Iraq and Syria wars in particular have made the covert and overt operations of the US, Russia, Iran, and Saudi Arabia much harder to ignore even in mainstream media.

The Conservatism of Analytic Philosophy of War The analytic approach generally has tended to see philosophy as foundational, and science as resting on top of this. In political philosophy this has gone along with the view that we need not begin with a basic understanding of social phenomena (from social science) before we define a problem-set for philosophy. Since philosophy is the foundation, philosophers can just define the problem-set. Issues of social science, like the nature of capitalism, imperialism, the military industry, are matters on which philosophers need take no stance, since they are working upstream of these issues, not downstream. I call this an idealist approach to political philosophy, in the sense that the foundation is not in material phenomena, but in the realm of ideas (conceptual analysis, reflection on intuitions). By contrast, a materialist approach sees social science as foundational.

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

We first need an understanding of a social phenomenon (like war) before we define a problem-set for the discipline. It turns out that this is no mere methodological divergence. If the foundation is social science, we must first establish whether, for instance, we accept a liberal or a Marxist account of states and wars. Adoption of one or the other will lead to different problem-sets for philosophy of war.

Idealist and materialist foundations for international politics John Rawls’s Law of Peoples account is a characteristically analytic approach to international politics and, by extension, to philosophy of war.⁴ Rawls notes that states pursue power, and their interests may include religious conversion of others, enlarging empire, winning territory, gaining prestige and glory, and relative economic strength (p. 28). Sovereign states have the power to go to war in pursuit of national interest and have a degree of autonomy over their subjects (pp. 25–6). A materialist approach to philosophy of war would begin from these observations, and investigate these pursuits of states, their systemic reasons, and consequences. But conveniently for a philosophy of the establishment, having noted this realworld connection between states and war, Rawls changes the discussion! In his Society of Peoples, the actors are not states (which are observed to be deficient in democracy, and having systemic need for war), but ‘peoples’ (entities he defines into existence). A liberal people has a reasonably just constitutional democratic government that is effectively under their political and electoral control, and is not an autonomous agency pursuing its own bureaucratic ambitions. It is not directed by the interests of large concentrations of private economic and corporate power veiled from public knowledge and almost entirely free from accountability. On top of this are assumptions about the material world, which flatly hush up some of the observations he makes about states. We are to assume, for instance, that larger countries with wealthier economies will not try to monopolize the market or conspire cartels, or act as oligopolies (p. 43). This would be an odd assumption if we were talking about states (which he acknowledges may go to war for relative economic strength), but is apparently fine since we are talking about ‘peoples’. A dose of neo-classical economics is also uncritically thrown in: liberal peoples assume that a free competitive market is to everyone’s mutual advantage (p. 42). No place here to consider the history that, when states try to catch up to industrially more advanced states, they do so by infant industry protection, state subsidies, and state spending on infrastructure. The neatest catalogue of this has been compiled by Ha-Joon Chang, where the liberal states considered include Britain, France, the US, and Germany.⁵

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi

CLOSING THE R  P CHAPTER



The problem for dissident philosophers of war is that liberal states (as opposed to peoples) such as the US and UK are precisely such entities directed by concentrations of economic power, seeking territory and relative economic strength. The operation of the military wing of the state outside political and electoral control was exactly the point of the Pentagon Papers in the Nixon period, as well as the revelations of covert operations under Carter and Reagan in the ‘Iran-Contra’ scandal. When dissident and anti-imperial movements began to organize, they were targeted by the political police in the COINTELPRO operations of infiltration, agent provocateurs, and assassination (directed particularly at the Black Panther Party, and other socialist and anti-war organizations in the late 1960s). Instead of a framework for thinking about actual states and dynamics of war, Rawls sweeps all this into some category of ‘non-ideal theory’. He shifts the philosophical problem-set to liberal ‘peoples’, which are free of all these dynamics. We are to ponder the problems faced by such peoples in dealing with one another and with peoples which are not decent or liberal. Consider how strange this appears from the perspective of subject classes who bear the brunt of the wars of the ruling classes. The actual phenomena of war are that concentrations of economic and political power (ruling classes) in liberal states routinely go to war for territory and economic power. But we are not to discuss this. Instead we are to assume that we are a ‘people’ where there is no such concentration of power or systemic drive to war. Only then do we start to philosophize about war. From a materialist point of view, this is not a way to think about reality, but rather an ideological obfuscation of reality.

Urgency of philosophical tasks The discussion just concluded was about the analytic approach to international politics generally. Consider its relation to philosophy of war more specifically. With its idealist foundation, there is a suggestion that we need not consult social science before choosing our initial questions. This allows the feint that, since we have not begun with politics or sociology, our choice of problem-sets is itself politically neutral. This is a feint because, while there is an effort to bracket social science, there is a dependence on some empirical stuff beyond the realm of ideas— namely on law and treaties. The concepts chosen are those in international treaties and domestic law. A tribunal judge is not interested in the same aspects of war that interest a social scientist. This juridical focus is a consequential circumscription of the agenda of philosophy of war. The tasks chosen are, say, to help a war crimes tribunal judge whether a given war was just or unjust. A social scientist, by contrast, inquires into the causes of war, why wars recur,

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

why they are more frequent under certain social conditions than others, why laws of war are routinely broken, the nature of the main agents (state militaries, proxies, militias) and industries (armaments, aerospace) and so on. The appearance of political neutrality is belied by the juridical focus. Juridical questions are happily discussed by rulers. They can profess to follow the criteria put forth by JWT, enshrine them in treaties and laws. When philosophers or political commentators think a state failed to hold to the principles, the ruling class can say it was just a bad apple, or a mistaken judgement based on faulty information. The tight connection to policy circles, appointments to teach ‘ethics of war’ in Army Colleges, and the like, also tends to affect the content of what philosophers of war discuss. Instead of studying the underlying reasons powerful states go to war, they uncritically adopt the declarations of intent of their state spokespersons: the war is for freedom and democracy, as just a cause as you’ll ever find! Instead of approaching with scepticism policy frameworks like the Cold War, HI and R2P, or the War on Terror, they approach them uncritically and limit their role to tweaking this or that element of the framework. But another perspective is available—that of the subject classes who bear the brunt of political leaders’ decisions to war. From the latter perspective, wisdom about war requires understanding why rulers so often go to war (usually unjustly, but occasionally justly) and what the subject classes can do about it. There is not much urgency to the task of spelling out tighter jus ad bellum and jus in bello criteria, because as a matter of fact these are routinely transgressed. Wars of aggression by the liberal states (like their invasions of Afghanistan and Iraq) are routinely carried out, as is the use of disproportionate force, deliberate targeting of civilian infrastructure, use of torture. Wars are seldom the last resort, but other options are neglected generally because they would mean ceding territory, or geopolitical influence. States do not play by the existing rules, let alone tighter ones devised by jurists and philosophers. From the perspective of subject classes, the priority for philosophy of war is to figure out why ruling classes go to war, and how they can be stopped.

The just war theory problem-set The just war tradition adopts the perspective of the poor state planner: wanting to do good but burdened with the responsibility of deciding whether or not to go to war. It sets out to list the conditions under which going to war would be morally acceptable. For most of history this would have been an odd interest for philosophers of war. Generally speaking, it was understood that rulers do not usually go to war because they think the cause is just, but because they seek the spoils of war, territory, and resources. So it is a merely scholastic issue to specify the

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi

CLOSING THE R  P CHAPTER



conditions under which war is just. It would be of little interest to the study of war as an actual social phenomenon (rather than war conceived only as a hypothetical event with benign sovereigns). However, this interest does make sense for ‘philosophers of the court’ like Thomas Aquinas (a torch-bearer for the Catholic Church and its allied emperor) who are willing to believe that while other rulers are immoral despots who war for gain, their ruler is different. Their ruler is built of moral fibre, and would only go to war if they believe they have a just cause, who would change their mind about going to war if only we philosophers could demonstrate that they do not meet the criteria for just war. Such writers are philosophers of the court in the sense that their expected audience is policy makers in power (the royal court). Their strategy for reducing unjust war is to convince powerful people to be nicer. Oddly, the modern revival of JWT by Michael Walzer in 1977 appeared at the tail end of a vicious war by the US in Indochina.⁶ It had all the trimmings of injustice—no just cause, not the last resort, disproportionate force, targeting of civilians and civilian infrastructure, no legitimate authority (many bombing campaigns in neighbouring countries beyond Vietnam occurred without Congressional approval), use of torture, exposed lies of the executive branch of government to get the US into war, lack of accountability for military and civilian top brass. That war prompted a generation of dissidents in the US to connect racism and poverty at home with imperial war abroad (from Martin Luther King Jr. to the Black Panther Party). It prompted them to a deep scepticism of the ruling class’s professions of democracy in the face of enormous anti-war protest, and over a decade of propaganda, secrecy, and lies about the Vietnam war (revealed in the Pentagon Papers). It prompted a search for systemic drivers of wars (like imperialism and capitalism) rather than seeing each war as an isolated decision. There was a rediscovery of the Marxist theory of the state and imperialism which had been forced out of universities during the McCarthyist period. But the period pushed Walzer in a very different direction, reviving the court philosophy of Aquinas about the conditions under which ruling classes can justly go to war. Today, for the mainstream of the JWT tradition, the assumed perspective is that the leading capitalist states, the US, UK, France, Germany, and their allies such as Canada, Australia, are not like the rulers of old. They strive for justice, and do not generally use military force for geopolitical gain. That may have happened in the colonial days, but that is over, we are told. In the post-Second World War era (or perhaps the post-Cold War era) the powerful states, the US and UK in particular, are constrained by representative government, and by a combative press, as well as by a UN Security Council and ‘international opinion’.

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

In reality no empire has been bound by these principles. So why devote so much of the philosophy of war to refining one or another concept in the JW principles: just cause, proportionality, last resort, liability to harm, etc.? At most, this should be a debate for policy-planning and jurisprudence fields, where the idea is to work to improve law and policy from the inside. This must be done with an appreciation of how far reform can be pushed before it conflicts with the needs of states. Pacifism and realism, generally discussed as the only alternatives to the JWT, are nonetheless presented as addressing the same problem-set as JWT. Pacifists are presented as holding that no war is just (while JWT holds that some wars are just). This definition of pacifism in analytic philosophy may be a gross misrepresentation or simplification of pacifist protest movements, which often had an anti-imperialist politics, or a sociological outlook on the military and the arms industry. Many anti-imperialists, pacifist when it came to their empire joining a war, had no objection to anti-colonial wars of liberation, and so certainly were not the ‘never hurt a fly’ sort that JWT likes to call ‘pacifists’. This included non-Marxist socialists and liberals who were open-eyed about their own states’ imperialism. But the most sustained organized international movement against war across the core capitalist countries in the early 1900s was Marxist (though it suffered a defeat in the early years of the First World War as many of the socialist parties succumbed to nationalism). While organizing workers in anticonscription drives they argued for turning the imperial wars of the ruling classes into civil wars. If workers are to die anyway, better to die fighting to topple your ruler at home than to gain spoils for your ruler abroad. These folks, agitating as pacifists in the imperial centres before the First World War, could at the same time countenance civil war to overthrow their ruling classes, and wars of national liberation in the colonies. Analytic philosophy’s presentation of pacifism is usually shorn of this sort of political content, as the JW tradition (as a form of court philosophy more generally) is reticent to discuss the political economy of war and especially the imperialism of the leading states. Analytic philosophy of war is so narrowly bound to the problem-set of JWT that it generally relegates this politics to some other discipline, outside of philosophy of war. If pacifists have a critique of JWT, it can only be presented as a rejection of the claim that ‘some wars can be just’. Realism holds that states generally go to war for geopolitical gain, and unlike the JWT tradition it may be happy to accept that the US in the post-war era is an imperialist power despite representative government and international institutions like the UNSC. But realism remains a ‘philosophy of the court’ in the sense that it is not interested in trying to change this. It is not interested in whether

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi

CLOSING THE R  P CHAPTER



states can be constrained (by, say, protest, rebellion, sabotage, strikes, dismantling of the military-industrial components). It is interested only in helping its preferred state be more powerful than its rivals. And, concerned as it is with the task of strengthening one or another of the leading capitalist states, it is implacably hostile to movements that want to overthrow or resist these states.

Materialist Foundations for Philosophy of War I discuss wars as phenomena in class society. Class societies involve rule by some class(es) over others. Rule is a material process that requires the direction of the labour power of the subject classes, and direction of the surplus product of society. Rule is internally contested, as subject classes struggle to control their labour power, and the product of their labour. Class struggle determines a ‘subsistence level’ that allows the subject classes to reproduce themselves, above which the labour product of an individual becomes ‘social surplus’. In this internal contestation the ruling classes use some of the instruments and institutions also used in war—armies and spies—to quell internal rebellion. In the competition between rival ruling classes, differences are ultimately settled by war over territory, populations, mines, ports, commercial routes, industrial centres. Wars are resource intensive. The resources come from the social surplus. The more surplus you can accumulate, the better placed you are for war. Accumulating more surplus can mean forcing the subject classes to work more intensely, having a larger subject labour force, reducing its ‘subsistence level’ to free up more for accumulation, and technology that raises the productivity of a given amount of labour. Aside from labour dynamics, competition between ruling classes also involves dynamics of controlling specific geographical sites. More fertile arable land allows a smaller amount of farming labour to produce the same amount of food. Mines are sources of ores needed for weapons making. Timber was necessary for naval fleets. In the late 1800s steel replaced timber as the state-of-the-art for fleets, and along with it came a much greater need for coal (and later, oil) in steel-making. Control over ports is necessary not only for housing fleets and launching invasions, but also indirectly as key sites of commercial concentration where ruling classes can accumulate wealth from international trade. Centres of industrial production are needed both for the forward and backward linkages that allow a roiling pace of accumulation, and more specifically for high-tech arms manufacture in the industrial era. By the 1500s these dynamics of war involved not only the state (especially bureaucracy, tax collection, public works, and navy) but also private actors like

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

merchants and traders who accumulated wealth in the state’s jurisdiction and big money-lenders who lent to states to finance wars (the loans being repaid through the spoils of war). The era of mercantile imperialism from 1500 to the 1700s involved just this complex interplay of states, their navies, debt-financed wars (as between England and France), and merchant capital. The Portuguese, Spanish, Dutch, French, and English empires, together with the first great joint stock companies like the Dutch, British, and French East India Companies and the Hudson’s Bay Company fought one another for control over ports and sea lanes in the Americas and the Indian Ocean.⁷ Up to this point the relative importance of a site of production, or a territory, or a population had depended on existing infrastructure, technological developments, and discoveries of new land (like the Americas) and new mines. The era of the capitalist mode of production (mid-1700s onwards) introduces specific new features. The dynamism of capitalist production means a corresponding turbulence in the relative importance of territories and populations. The typical form of competition in the era of merchant imperialism was about controlling a territory from which you could source cheap commodities (often cheap because forcibly acquired) to sell dear somewhere else. Competition in production means that you produce the commodity in the first place, and so cheap raw materials and cheap labour begin to matter. Moreover, in the mercantile period your control over the source of the commodity, say spice ports in the Indian Ocean or gold and silver mines in the Americas, often meant that potential competitors did not have the same product to sell. But with capitalist production, we are generally speaking of products like textiles that can also be made by a competitor. So control over markets also becomes important. Imperial wars become increasingly driven by the competitive movements of capital in search of cheaper materials, cheaper labour, and bigger markets. Use of force to acquire markets was important for the British textile industry, at the dawn of industrial capitalism. The immense Indian market was already supplied by its domestic textile industry, but was forcibly de-industrialized by the British so that British imports could take over the market. War for the raw materials of industry was particularly clear in the late 1800s ‘Scramble for Africa’ as Britain, France, Germany, Belgium, and others carved up the continent for its mines and rubber plantations to supply their respective industries. States compete to attract production capital by offering the cheapest (usually most repressed) work force (hence the move of some production capital away from the US, Western Europe, and Japan to China and other developing countries over the past three decades).

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi

CLOSING THE R  P CHAPTER



Finally, once the capitalist mode of production is in place, a long-term trend emerges for the expansion of production. A capitalist produces in order to make profit. If successful, they have first made back all their raw materials and wage outlays and so are in a position to repeat the same volume of production in the next period (assuming costs have not changed). But the profit comes on top of this. What to do with this profit? For a start, the capitalist herself generally receives a wage (in exchange for her managerial or entrepreneurial labour) as well as a profit (in exchange for having contributed the seed capital), so the profit is more than she needs to subsist at her standard of living. Certainly over the whole economy it is simply not possible for capitalists to consume their profits. So the profits of the first period must be reinvested. The reinvestment expands production beyond the volume of the first period (putting aside complications about non-production investment such as real estate speculation, which do not affect the basic point). But the capitalist does not merely seek to expand production. She seeks to do so profitably. If the expanded production means employing more workers and takes the economy below its normal rate of unemployment, this will begin to raise wages (which cuts into profitability). If the expansion means that necessary raw materials become more scarce in the economy, this will push up materials costs (again eating into profitability). If the existing market for the product at a price that can earn a profit was already saturated, then the expanded production will leave unsold goods (eroding profitability). In various ways profitable opportunities for investment can dry up in an economy even though there are unemployed people, unused raw materials, and unsatisfied need. So another task emerges for capitalist states—finding profitable outlets for capital when opportunities diminish domestically. The connection between wars and profits should not be misconstrued. Wars are not uniformly profitable for different industries (arms industries and heavy industries generally do well of course), and may even be a net drag on profitability for the country. A war footing can make some industries fall behind because of lack of infrastructure spending relevant to them, or it can push up resource costs (as resources are diverted to war uses) and wages (as states usually create jobs during wars). The issue is one of time scales. Wars undertake a redivision of territories and restructuring of alliances and relations of subservience. The main benefits from this come after the war. Imperialist war is not (generally) undertaken ad hoc at the urging of this or that capital in this or that country. Rather, it usually results from longer term planning by the state about which regions are strategically important to its economy and major industries.

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

In sum, the basic Marxist position on imperialism is this. Imperialism as a capitalist phenomenon (rather than older empires like the Roman Empire) emerges from the dynamics of capital accumulation rather than from the dynamics of state per se. Different capitals must compete with one another by obtaining cheaper raw materials, fuels, and labour power, by accessing new markets and excluding rival capitals from them. Even without capitals actively seeking out the help of the state, states are compelled to help capitals because the level of employment, the rate of accumulation, and the composition of the economy all matter to the power of the state. Rising unemployment brings more riots and rebellions for the state to manage. A slowing rate of accumulation reduces the pool of wealth on which the state can draw in its competition with rival states, and may lead to capital flight. Military powers care about strategic parts of the economy—fuel, heavy industry, electronics among them. Imperialism alternates between periods of war and military interventions short of war, and periods of diplomatic and trade agreements. It is one and the same set of competitive dynamics that can sometimes be resolved by diplomacy (usually reflecting the military and economic power of the negotiators), and sometimes spills over into war.

Political Conflict Rooted in Economic Competition We are in a period of rising imperialist rivalry that may spark into open war— with the US, Russia, and China the major powers involved. But before getting to the current period, I want to expand on the point in the previous paragraph that imperialist dynamics do not always lead to war. In certain periods, means short of war may suffice for economic competition by state power. The brief historical excursus also gives a sense of the economic roots of imperial policy, and of the fact that economic competition does not disappear in times of peace.⁸ In the 1950s the US displaced Britain and France as the major imperial power in the Middle East. It began a policy of securing its control over the oil resources of the region. At the time its own industry was largely fuelled by domestic reserves. It sought control over Middle East oil as an instrument to keep in line Japan and Western European allies, who were dependent on imports from the region. This required opposing the nationalist and pan-Arabist movements (Mossadegh, Nasser) and propping up brutal regimes friendly to the US through arms, training, and funds (Saudi Arabia and other Gulf monarchies, Iran under the Shah, Israel). In the 1960s and 1970s, the heavy cost to the US of the war in Vietnam left the US with less infrastructural spending on other parts of the economy, and

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi

CLOSING THE R  P CHAPTER



pushed up employment levels and with them wage costs. At the same time, competitors like Germany and Japan did not have similar military costs, and were becoming more competitive, with lower waged work forces, and public spending on infrastructure to attract manufacturing. In the 1970s the US encouraged Saudi Arabia to raise oil prices because, even though the US economy would receive a shock from this, Germany and Japan were much more reliant on imported oil than was the US. This was to drive home the lesson to German and Japanese states that while their manufacturing was becoming more competitive, they were still dependent on flows of oil from the Middle East, where the US was the major imperial power. This show of military and economic dependence of the allies on the US was to help cement the alliance and forestall efforts at independent imperialism by the two rivals. The other main response to the rising competitiveness of German and Japanese economies was repression of labour (and lowering of wages and the social wage) at home under Reagan in the 1980s. China began orienting to manufacture for export in the late 1970s, but this really took off in the 1990s. Also in the early 1990s the fall of the USSR empire released many Eastern and Central European countries from a focus on trade with the USSR bloc of economies and opened them up to manufacture for export to the West instead. Both these regions offered pools of much cheaper labour than was available in the main industrial centres of the time in North America, Western Europe, and Japan. Global capital headquartered in the West and producing for a Western market increasingly located new productive investments in Eastern Europe and China, contributing to the relative slowdown in the manufacturing sector in the West. By the end of the 1990s US policy-planners urged that the state seize the ‘unipolar moment’ after the Cold War to reshape geopolitics to forestall the rise of the next challenger (expected to be either Europe or China). This included foreign policy stalwarts like Zbigniew Brzezinski (National Security Advisor to the Jimmy Carter administration), as well as think-tanks of the neo-conservative right (Project for a New American Century) and the liberal centre (Center for Strategic and International Studies).⁹ It was uniformly agreed that control over Middle East oil remained crucial, as Europe and China both depended on it. While the general plan for the Middle East has been pursued since 2001, it was not as successful as planners anticipated and has been complicated by various events. Chief among them has been the ‘Arab Spring’. While the mainstream narrative picked up only on some of the protesters’ demands (an end to authoritarianism and corruption), the left has emphasized that other critiques were also present right from the start. The Arab region had faced years of neo-liberal policy

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

that significantly lowered the standard of living of the masses and raised unemployment. From the start, significant sections of the protest movements in Egypt and Tunisia were explicitly critical of the IMF’s backing of their countries’ economic model and the US’s military and diplomatic backing of the Mubarak and Ben Ali regimes. The Arab Spring threatened the ruling regimes of the region as most countries had undergone similar neo-liberal changes, and had close ties to the US. It also threatened US political influence in the region, as ally regimes might be replaced by regimes which sought alliances with other global powers instead. Most generally the Arab Spring threatened to unsettle the role that the Middle East occupied in the international economy: primarily a source of oil and gas, with very little manufacturing and infrastructure, the resulting highly unequal societies maintained by patrimonial and authoritarian regimes.¹⁰ The other major complication is that the Afghanistan and Iraq invasions ended up much more problematic than anticipated. With the US bogged down, several regional powers began manoeuvring to fill the power vacuum in the region, to create allies among new leaders or to fund political parties and militias which hoped to take power. This included US allies like Saudi Arabia, Turkey, Qatar, and Israel, as well as non-allies like Russia and Iran (which fortuitously found itself rid of antagonistic regimes to either side—the Taliban in Afghanistan and Saddam in Iraq). In Eastern and Central Europe the inflow of global capital in the 1990s was accompanied by NATO’s push to incorporate these countries into their political/ military alliance. Russia gradually lost economic and political clout in the process. But since it remained a significant military and economic power even after losing its empire, it eventually reacted against this expansion of the NATO alliance when Putin came to power, renationalizing the energy sector as a basis for economic strength, and seeking to recover economically and militarily the countries formerly in its sphere of influence. European states have been decreasing their reliance on Russian natural gas fearing that Russia may use this dependence to leverage geopolitical power. Russia is seeking new markets instead—especially Turkey and China. Russia’s energy pact with China is also to lessen the impact of Western sanctions, while China sees the deal as a way to reduce its dependence on Middle East oil, which is heavily under control of US allies. Both the US and Russia have increased the deployment of military hardware, working with their respective allies, in Eastern Europe. In the last few years, Georgia and Ukraine have seen proxy/open conflicts between US and NATO on one side and Russia on the other. At its 2014 summit, NATO agreed on the biggest build-up since the Cold War, including the pre-positioning of equipment and

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi

CLOSING THE R  P CHAPTER



supplies at bases. The member states committed to escalate defence spending. Meanwhile budget and debt crises in Europe mean that this increase in spending will sharpen tensions at home. Competition between China and the US is taking both military and economic forms. In the Asia-Pacific region, until recently the US pursued a trade agreement that excludes China (the TPP), hoping to use its remaining economic clout and especially its military alliances to create a barrier to further expansion of the Chinese economy. In response, China is pursuing a rival trade agreement that excludes the US and that hopes to build on its increasing economic ties with several of its neighbours. The Chinese government is intent on securing stable supplies of raw materials and transport routes for its exports to help its growing economy. It has been building roads, railways, and pipelines to link Europe, Middle East, Central Asia, and Northeast Asia. In part it is trying to reduce its dependence on imported oil that is transported by shipping lanes rather than over land, as the shipping lanes are generally controlled by the US and its allies.¹¹ China has been using its economic growth to ramp up its military expenditure, and is increasingly asserting its claims over island chains which are disputed by Japan, Philippines, and Vietnam. The island chains are strategically located in shipping lanes, fisheries, and underwater oil and gas reserves and so none of the players want to give them up. The US is backing these countries in their disputes with China. In the ‘pivot to Asia’ under Obama, the US ramped up its military presence in the Asia-Pacific, encouraged its allies in the region to build up their defence budgets, and conducted provocative military exercises with its allies (which include Australia, Japan, South Korea, and Vietnam). The exercises consolidate the inter-operability of allied militaries with US command. China’s New Development Bank has been providing loans to developing countries especially for projects to extract raw materials needed in China. Using these kinds of infrastructure loans, China has become among the most important trade and investment partners for both Africa and Latin America (displacing the US in both cases). In response to the rising influence of China in Africa, the US has ramped up its string of military bases in Africa.¹²

Conclusion Having detoured through ideological critique of the analytic approach to philosophy of war, and a sketch of a rival materialist foundation, we have returned to the present moment. I repeat the judgement with which I opened the chapter: given political developments over the past few years, the likely flashpoints for

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi



CHALLENGES FOR HUMANITARIAN INTERVENTION

conflict in the near future are areas of imperial rivalry rather than isolated events of ruthless despots crushing subject populations. Moreover, ‘our side’ is among the imperial powers, and looks decreasingly liberal or democratic, and increasingly authoritarian and fascistic internally. This is not the time for theorizing about benevolent liberal democracies intervening militarily to save benighted foreign populations. Rather, it is a time for theorizing about the drive towards war in both the so-called liberal democracies and the authoritarian powers of Russia and China. The common root of the drive to war in all of these powers is economic competition on such a large scale that states (and not just firms) become involved.

Notes 1. Noam Chomsky, A New Generation Draws the Line: Kosovo, East Timor and the Standards of the West (London: Verso, 2000) and Noam Chomsky, The New Military Humanism: Lessons from Kosovo, (Monroe, ME: Common Courage Press, 2002). 2. Keeanga-Yamahtta Taylor, From #BlackLivesMatter to Black Liberation (Chicago: Haymarket Books, 2016); Lesley Wood, ‘Policing with Impunity’, in Leo Panitch and Gregory Albo (eds), The Politics of the Right, Socialist Register, 52 (London: Merlin Press, 2015), 333–46. 3. Leo Panitch and Gregory Albo (eds), The Politics of the Right, Socialist Register, 52 (London: Merlin Press, 2015). See especially the chapters by Fekete and Eley. 4. John Rawls, The Laws of Peoples (Cambridge, MA: Harvard University Press, 1999). The bracketed page numbers in the next few paragraphs are to this edition. 5. Ha-Joon Chang, Kicking Away the Ladder (London: Anthem Press, 2002). 6. Michael Walzer, Just and Unjust Wars: A Moral Argument with Historical Illustrations (New York: Basic Books, 1977). 7. Giovanni Arrighi, The Long Twentieth Century: Money, Power, and the Origins of our Times (London: Verso, 1994). Nick Robins, The Corporation that Changed the World: How the East India Company Shaped the Modern Multinational (London: Pluto Press, 2006); Alexander Anievas and Kerem Nisancioglu, How the West Came to Rule: The Geopolitical Origins of Capitalism (London: Pluto Press, 2015). 8. The general narrative of post-war political economy is based on David Harvey, A Brief History of Neoliberalism (Oxford: Oxford University Press, 2005); and Sam Gindin and Leo Panitch, The Making of Global Capitalism: The Political Economy of American Empire (London: Verso, 2012). The general narrative of geopolitical tussles between the US, Russia, and China is based on Ashley Smith, ‘The Asymmetric World Order’, International Socialist Review, 100 (2016): ; and Ashley Smith, ‘US Imperialism’s Pivot to Asia’, International Socialist Review, 88 (2013): . More specific references are given below. 9. Zbigniew Brzezinski, The Grand Chessboard: American Primacy and its Geostrategic Imperatives (New York: Basic Books, 1997); Thomas Donnelly, Rebuilding America’s

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi

CLOSING THE R  P CHAPTER



Defenses: Strategy, Forces and Resources for a New Century (Washington, DC: Project for the New American Century, 2000); Guy Caruso (Center for Strategic and International Studies), ‘The Geopolitics of Energy into the 21st Century’ (testimony before (US) Senate Energy and Natural Resources Committee, 21 Mar. 2001). 10. Gilbert Achcar, The People Want: A Radical Exploration of the Arab Uprising (Berkeley-Los Angeles: University of California Press, 2012); and Adam Hanieh, Lineages of Revolt: Issues of Contemporary Capitalism in the Middle East (Chicago: Haymarket, 2013). 11. Shuaihua Wallace Cheng, ‘China’s New Silk Road: Implications for the US’, Yale Global Online, (28 May 2015), . 12. Nick Turse, ‘America’s Empire of African Bases’, TomDispatch.Com (17 Nov. 2015), .

OUP CORRECTED PROOF – FINAL, 19/4/2018, SPi

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

Name Index Acheson, Dean 2 Albright, Madeleine 86 Amos, Sheldon 124, 137 Annan, Kofi 102, 193 Aquinas, Thomas 207 Ashdown, Paddy 177 Assad, Bashar al- 34, 39, 47, 49, 94, 175 Augustine 40 Averre, Derek 171, 177, 180–1 Bellamy, Alex J. 174–82 Blair, Tony 193 Boutros-Ghali, Boutros 87 Brzezinski, Zbigiew 213 Buchanan, Allen 96–7, 105, 120 Bush, George Sr. 126 Bush, George W. 27, 45 Bush, Laura 59, 72 Butterfield, Herbert 185, 198 Carr, E.H. 2 Carter, Jimmy 205, 213 Cassam, Quassim 111, 122 Chang, Ha–Joon 204 Chomsky, Noam 201–2 Clark, Wesley 89 Coady 1, 14 Coghlan, Peter 125, 137 Cole, Juan 30, 37 Corten, Oliver 124, 137 Curuvija, Slavko 25 Das, Ramon 10–11, 75 Davenport, Christian 84–5, 97 Davies, Lance 171, 177, 181–2 Deng, Francis 166, 178 D’Escoto Brockmann, Miguel 103, 113, 120, 192 Dobos 8, 11, 120, 123, 137 Evans, Gareth 70, 73–4, 121, 173, 177–82, 199 Fischer, Joschka 17 Gaddafi, Muammar 27 Gandhi, Mahatma 19, 27–36, 196 Gent, Stephen E. 15, 37 Goh, Evelyn 187, 198 Goldhagen, Daniel 135, 138

Habyarimana, Juvénal 82, 87 Haradinaj, Ramush 25 Hardins, Russel 111 Hehir, Aidan 12, 165, 180, 182, 198 Hermanson, Chrisantha 8, 12, 97–8, 139, 163 Hobbes, Thomas 2 Holbrooke, Richard 20, 24 Hourigan, Michael 86 Hussein, Saddam 6, 43–50, 138, 214 Idris, King 27 Ivanov, Igor 170 Jackson, Robert 186 Kathman, Reed M. 15, 37 Keating, Tom 13, 183, 186, 198 Kennan, George 2 Keohane, Robert O. 96–7, 105, 120 Khrushchev, Nikita 49 Ki-moon, Ban 12, 102, 168, 173, 177–81 King, Martin Luther 207 Kissinger, Henry 2 Klitschko, Vladimir 49, 58 Kostunica, Vojislav 25, 37 Kowert, Paul 176, 182 Krook, Mona Lena 176, 180, 182 Kuperman, Alan 8, 14, 86, 98, 198 Kurowska, Xyzema 190, 198 Legro, Jeffrey 176, 178, 182 Luban, David 77, 95, 98–9 Luther King, Martin 207 Machiavelli, Nicolo 2 MacKinnon, Catherine 62, 73 May, Larry 83 McKinley, William 3–4, 14 McMahan, Jeff 105, 120–1, 161–2 Mengzi (Mencius) 52 Menon, Ranjan 193, 198–9 Meyer, Marco 11, 100 Mill, John Stuart 66–8, 73 Miller, Richard W. 9, 38 Milosevic, Slobodan 16–25, 39, 99, 125 Morgenthau, Hans 2 Morris, Justin 180–1, 196, 199 Mossadegh, Mohammad 212

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



NAME INDEX

Mubarak, Hosni 29, 31, 214 Murray, Robert W. 13, 183, 198 Nardin, Terry 77, 95 Nasser, Abdel 212 Newman, Edward 192–5, 198–9 Niebuhr, Reinhold 2 Nixon, Richard 205 Obama, Barack 215 Parekh, Bhikhu 71, 74, 124 Pattison, James 91, 99, 161–2 Putin, Vladimir 27, 190, 214 Pyatt, Geoffrey 48 Rawls, John 80, 97, 113, 121, 163, 204–5, 216 Reagan, Ronald 205, 213 Reynolds, Henry 8, 14 Rizal, José 4 Sanyal, Sagar 8, 13, 200 Sen, Amartya 81 Smith, Ian 53

Spivak, Chakravorti 66, 73 Stalin, Joseph 40, 135 Stam, Allen 85, 97 Thaçi, Hashim 25–6 Thakur, Ramesh 171, 180 Thompson, Janna 10, 59 Tudjman, Franjo 18 Valentini, Laura 80–1, 96–7 Verwey, Wil 124, 137 Walzer, Michael 50–1, 54, 58–68, 72–3, 96, 128, 133–8, 161, 207, 216 Weber, Max 2 Welsh, Jennifer 173, 178, 180–1 Wheeler, Nicholas 183–4, 193, 198 Widdows, Heather 77, 96 Wight, Martin 185–6, 198 Wood, Ellen Meiksins 125–6, 137 Wood, Reed M. 15, 37 True, Jacqui 176, 180 Zhenmin, Liu 46, 58 Zunes, Stephen 9, 15

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

General Index action military 26, 37, 63, 70, 145 non-violent 9, 25, 31, 36 aid conditions on 156–9 duty to 149, 153, 156–7, 164 unconditional 156–9 Afghanistan 2, 10, 26, 59–69, 192–3, 206, 214 Africa 35, 46, 215 African Union (AU) 35, 140, 201 Albania, Albanians 16–26, 38, 89–90, 121, 125, 136 Amnesty International 18 Arabia, Arab 28, 33, 35 Arab Spring 203, 214 Assad regime 34, 39, 47, 49, 94, 134, 175 atrocity 6–194 passim Australia, Australian 8, 20, 207, 215 Boer war 8

Colombia 89–93 Congo 88, 192 Conservatism 203–4, 208–9, 213 Croatia, Croatian 18–22, 47 Crimea 48, 170, 197 Dayton Accords 19, 22, 24 democracy 5, 34, 56, 71, 134, 194, 204–7 pro- 24–8, 30, 35 demonstrations, peaceful 19, 31, 191, 202–3 deterrence 4 dictatorships 5, 33–5 displacement 10, 46 Doctors without Borders 18 East Timor 39 Economics, economic 94, 204, 212–15 aid 71, 139, 152–3, 159 sanctions 53 education 3, 16, 28, 59, 66–8, 135, 144–61 Egypt 29–31, 214 European Union (EU) 5, 27, 203, 214

Balkans 17, 24, 26 Ben Ali regime 29, 34, 214 Black Lives Matter (BLM) 202–3 Black Panther Party 205 Bosnia-Herzogevina 10, 20, 22, 59 peace process 20 Srebrenica 61, 189 Boykotts 19 BRICS 12–13, 171–3 Bush Administration 27

Geneva Convention 62, 108 Georgia 27, 109–10, 170, 214 Gulf War 53, 189

capitalism 203–11 Catholic Church 207 Cedras regime 46 Chechnya 170 children 3, 33, 58–62, 71, 163 China 32–4, 38–57, 135, 175, 190–216 Cultural Revolution 135 citizens 2, 6, 39, 42, 56, 61, 64, 68, 71, 126, 130–1, 134, 134–5 civilians, Civil 15, 21–3, 26, 32–4, 39, 76, 85–6, 89–90, 94, 125, 206–7 casualties 44 innocent 4, 11, 75, 83, 129 protection of 188–97 society 37 Clinton Administration 19, 23–4, 189 COINTELPRO 205 Cold War 57, 154, 183–96 post- 166, 183–96

Haiti 46 human rights 29, 53–5, 104, 124–6, 131–3, 168, 170, 176, 184, 189, 191 abuses 70, 75–6, 89–94 declarations 53 fundamental 140–58, 185 language of 53 law 174 protection 195–7 violations of 2, 62, 70–1, 102–4, 124, 174, 207 Watch 18 humanitarian intervention (HI) 1–217 passim and JW framework 6–9 bystander states 11, 100–20 debate 200–17 diplomacy 1, 5, 9–11, 19, 21, 37, 54–5, 212 ethics of 38–58, 206 ideal-theoretic assumptions 10, 75–94, 204–5 intervening states 100–24

Finland 42 France 46, 87–8, 189–90, 201, 204, 207–11

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



GENERAL INDEX

humanitarian intervention (HI) (cont.) military 1–14, 17, 40–1, 61–3, 69–72, 101–20 morality of 1–14, 16, 21, 52, 104, 131, 175, 190, 201 motives 77, 93, 101, 104, 114, 123–30, 160, 185, 192 and non-ideal theory 75–99, 204–5 objectives 4, 8, 125–7, 130, 140–4, 150, 192 philosophical discussion of 7, 10, 75–82, 89, 200–3, 206 and rape of Bosniak women 60–3 and suppression of Afghani women 63–9 rescue 5–6, 12, 46–7, 51, 59–66, 149–57 rhetoric 6, 11–13, 123–4, 165, 168, 175–7, 183, 197 ideology 200–2, 205, 215 Imperialism 32, 123–4, 201, 203, 207–13, 209–15 anti- 18, 32 and capitalism 209–15 cultural 133–6 India 19, 41, 45, 66, 106, 142–3, 190, 195, 210 independence 19 Indonesia 39 injustice 38–58 foreign 39–40, 52–5 moral response 39 systematic 40, 43, 51, 206–8 international affairs 3, 60, 77, 116, 212–15 International Commission on Intervention and State Sovereignty (ICISS) 102, 124, 141, 184 international community 29, 82–8, 100–4, 114, 131, 142, 153, 166–74, 190, 194 International Court of Justice 26 International Criminal Court 63, 189 ICTR 85–6 international humanitarian law 36 international relations 10, 209–15 Iraq 5–8, 23, 27, 48, 50, 53, 114, 125–7, 189, 192–3, 203, 206, 214 -Iran war 46 Operation Desert Storm 23 US invasion of 27 Israel, Israeli 22, 56, 212, 214 ISIS 59 Japan, Japanese 41, 210, 212–15 just war theory 6–9, 40–2, 105, 141, 148, 158, 201–8 abuse of JW conditions 8, 12, 206–9 and common sense 40–3, 207 Confucian legacy 40, 42, 52, 57 Great Test 42–3 jus ad bellum 6–8, 69, 206–7 jus in bello 8, 89, 161, 206–7

just cause 6–10, 41, 42, 61–8, 107–10, 140–6, 161, 207 last resort 7–11, 41, 107–9, 157, 207–8 legitimate authority 7–8, 147–8, 162, 207 likelihood of success 5, 41 non-combatant immunity/civilian immunity proportionality 109–10, 141, 207 prudential condition 6–8 right authority 141, 207 right intention 7–8, 12, 139–41, 150 Kagame regime 88 Kosovo, Kossovars 36–7, 38–9, 49, 75–9, 102, 114, 125–6, 136, 170, 189–93 Alliance for the Future of (AAK) 25 autonomous status 25, 27 Democratic League of (LDK) 25 Democratic Party of (PDK) 25, 27 guerrilla warfare 19 independence 16–27 Liberation Army (KLA) 16–26, 90 military campaign 20–4 NATO air attacks 20–7 NATO bombings 20–7 NATO Intervention ethics 88–94 and non-ideal theory 88–94 non-violent independence movement 19 Otpor! (‘Resistance’) 24 Kurds, Kurdish 39, 46, 90 law, international 1, 131, 170–1, 186–7, 192 Libya 9, 13, 15, 27–37, 102, 106, 114, 127, 172–3, 192–6 alternatives to war 29–31 Ansar al-Sharia 35 Benghazi 28–9, 32, 34 bombing of 16 civil war 15, 28–33 military campaign 31–3 National Transitional Council 33 and NATO intervention 29–35 non-violent struggle 28–37 opposition to intervention 28–9 poverty and wealth 28 Salafi 34 Sufi 34–5 Tripoli 27–35 Tuareg 35 Macedonia 19, 24 McCarthyism 207 Marxist analysis 8, 200–12 Middle East 5, 26, 32–3, 212–15 military alliances 26, 37, 118, 215

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi

GENERAL INDEX

intervention 5–10, 36–41, 61–72, 101–120, 127, 131, 210–13 violence 7 militias 9–10, 16, 28, 34–5, 43, 59–64, 143, 206, 214 Montenegro 22, 25 moralism, moralizing 3–6, 52, 69, 104, 131, 201 integrity 11, 75, 84 judgement 39, 41–6, 57, 132 morality 2–14 thinking 3, 205–6 Mubarak regime 214 Muslim 203 Myanmar 45 nationalism 16–17, 28, 32, 194, 208 NATO 9, 13, 17–48, 88–93, 170–3, 189, 192, 201, 214 Article 26 non-Article crisis response 26 propaganda 18 Nazism 42, 135, 216 NGOs 10, 158 Nicaragua 18, 113 oil 213 Sandinista 18 Organization for Security and Cooperation in Europe (OSCE) 20–4, 89 pacifism 17, 208 Pakistan, East 142 Palestine, Palestinian 22, 26, 56 peace 10, 17, 46, 57, 71–2, 101, 172 Pentagon Papers 205, 207 Philippines 3–4, 31, 215 philosophy analytic 200–8 Marxist approach 200–17 materialist 209–15 political 200–17 of war 203–12 politics, political community 9, 50–1, 148, 206 geo- 4, 51, 124, 139, 141, 206–14 realism 1–14, 208–9, 212–15 poverty 28, 55, 207 Rambouillet 93 Conference 89 proposal 20–1, 27 rebels, rebellions 9–10, 22–35, 44–56, 82, 94, 209, 212 religion, religious 3, 10, 23, 43, 45, 61–9, 100, 104, 119, 128, 204 Responsibility to Protect (R2P), the 1–217 passim



abuse of 11, 105, 141, 150, 160–1 abuse-worry 144–60 complicating the moral case for 15–37 critics of 11, 102–6 and cultural imperialism 133–6 epistemic demandingness of 106–10 epistemic superiority of status quo 115–16 failure 13, 103–4, 113, 167–70, 183–5, 194 implementation of 11, 13, 34, 101–7, 184 ineffectual 12, 165, 177 language of 11 and leeriness objection 100–22 moral justification 61, 63, 69, 103, 105, 132, 139, 143 and the multipolar order 194–7 norm contestation and the diminution of R2P 165–82 norm or law 166–7 Pillar I 173–7 Pillar II 173–7 Pillar III 12, 165–6, 172–5 polarity and society 183–99 and political realities and the international order 183–99, 201–3 and the post-Cold War order 188–94 rhetorical device 13, 165–7 scrutinizing R2P’s intentions 139–64 target population 12, 101, 104 target state 10, 11, 75, 91, 130–1 transformative concept 165–8 and ulterior motives 123–7 and unilateralism 127–33 uses and abuses of 123–38 and women 70–2 Rhodesia 53 Rwanda 11, 29, 59, 61, 75–98, 114, 129, 158, 183, 189 FAR 85 Hutus 41, 59, 82–8, 129, 158 and non-ideal theory 82–8 Patriotic Front (RPF) 82–8 Tutsi 41, 59, 82–8 Russia, Russians 9–13, 27, 165–82, 190–7, 201–3, 212–16 and R2P today 172–6 Saudi Arabia 33, 53–4, 94, 203, 212–14 secession 8, 10, 47–9, 51 Serbia, Serbs 16–32, 38, 47, 59–63, 75, 89–90, 125, 129, 136, 189 bombing 16–32 ethnic Serbs 16–32 FRY 89–90 International Crisis Group 26 victimhood 23 Spain 3–4 Somalia 77, 83, 126, 183, 189

OUP CORRECTED PROOF – FINAL, 24/4/2018, SPi



GENERAL INDEX

South Africa 53, 63–6, 173, 192, 195 sovereignty, sovereign 9–10, 38–57, 67, 100–4, 170–1, 176, 184–7 Soviet Union 40, 49, 184, 190 see also Russia starvation 5, 124 strikes 19, 23, 209 Syria 33–4, 38, 47, 49, 79, 94–5, 166, 170–6, 193, 197, 203 torture 18, 39, 43, 135, 140, 147, 203, 206–7 UN Conventions on 53 UN Resolutions on 53 Tunisia 29, 31, 34–5, 214 Turkey, Turks 22 tyrannies 10, 49–52, 62, 190 stable 10, 43–7 unstable 10, 47–9 Ukraine 47–9, 197, 214 Orange Revolution 49 United Kingdom, Great Britain 40, 87–8, 94, 190, 204–5, 207, 210, 212 United Nations (UN) 1, 34, 39, 102–6, 127–32, 189–96 Charter 27 Convention on Genocide 61 declarations 49 peacekeeping 83 resolutions 53, 55 and R2P 100, 201 Security Council (SC) 8, 11, 13, 76–7, 83, 87, 106, 113, 127, 184–96, 207 UNAMIR 87 UNHCR 85 War Crimes Tribunal 26, 205–6 World Summit Outcome Document 127, 166, 171, 173, 177, 184 United States 4–5, 17–35, 38–9, 45–57, 77, 82–94, 111, 204–5, 207, 212–15 war, wars alternatives to 19–20 civil 15, 29–31, 76, 82, 94, 129, 208

crimes 2, 21, 26, 37, 46–7, 62, 103, 107–8, 184, 205 crimes against humanity 46, 61, 63, 103, 107, 127, 184, 193 ethics of 75, 206 ethnic cleansing 2, 12, 18, 21–2, 46–7, 85, 90, 103, 107, 123–7, 134–6, 184 First World War 17, 208 genocide 2, 29, 32–4, 46, 59–65, 77, 82–8, 102–10, 119, 127–9, 134–6, 158, 184 guerrilla 19, 23, 207 inter-imperial 209–15 massacres 1, 5, 12, 18, 22, 28–39, 46, 49–53, 61, 64, 123, 207 mass killings 12, 15, 84–5 and patriarchy 68 philosophy of 13, 200–17 Second World War 17, 27, 201, 207 War on Terror 203 unnecessary 7–8, 207 weapons of mass destruction 6, 49 Western Sahara 26 women 1–217 passim Afghani 63–9 attacks on 10, 59–60 Convention on the Elimination of Discrimination against 53 gender apartheid 64 and humanitarian intervention 59–74 Muslim oppression of 10 rape of 10, 59–63, 70, 72, 140 rights of 60 in Saudi Arabia 54 Tutsi 59–63 in Victorian England 67–8 Yanukovych regime 49 Yugoslavia 16–33, 47, 59–63, 129, 136, 154, 170, 189 see also Kosovo, Serbia