The Responsibility to Protect and the Problem of Military Intervention Author(s): Alex J. Bellamy Source: International Affairs (Royal Institute of International Affairs 1944-), Vol. 84, No. 4 (Jul., 2008), pp. 615-639 Published by: Wiley on behalf of the Royal Institute of International Affairs Stable URL: http://www.jstor.org/stable/25144868 Accessed: 09-02-2017 10:33 UTC JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://about.jstor.org/terms Royal Institute of International Affairs, Wiley are collaborating with JSTOR to digitize, preserve and extend access to International Affairs (Royal Institute of International Affairs 1944-) This content downloaded from 147.251.237.97 on Thu, 09 Feb 2017 10:33:14 UTC All use subject to http://about.jstor.org/terms The Responsibility to Protect and the problem of military intervention ALEX J. BELLAMY From inauspicious beginnings, the 'responsibility to protect' (R2P) has come a long way in a relatively short space of time. The principle was endorsed by the United Nations General Assembly in 2005 and unanimously reaffirmed by the Security Council in 2006 (Resolution 1674). * Ban Ki-moon has identified the challenge of translating R2P 'from words into deeds' as one of the cornerstones of his Secretary-Generalship.2 The principle has also become part of the working language of international engagement with grave humanitarian crises: the head of the Human Rights Council's mission to Darfur, Jodie Williams, used it to evaluate the government of Sudan's performance, finding that it had 'manifestly failed' in its responsibility to protect its citizens;3 the Security Council referred to the principle in mandating the UN?African Union (AU) hybrid mission for Darfur (UNAMID) (Resolution 1706, 2006); and both Kofi Annan and Ban Ki-moon used R2P in relation to their diplomatic efforts to resolve the post-election conflict in Kenya.4 Yet evidence of international disquiet with R2P abounds. The Williams Report was denounced by Arab and Asian members of the Human Rights Council, and it took six months to persuade the Security Council to reaffirm a principle to which its members had given their assent in 2005 ;5 several governments have argued that they did not in fact endorse the principle in 2005 and have committed themselves only to further deliberation;6 and members of the Fifth Committee of the General Assembly (Administrative and Budget) resisted the appointment of a special adviser 1 On a failed Canadian attempt to secure a technical General Assembly resolution on R2P in 2002, see Maria Banda, 'The responsibility to protect: moving the agenda forward', paper prepared for the UN Association of Canada, March 2007, p. 10; World Federalist Movement, 'Civil society meeting on the responsibility to protect', final report, Geneva, 28 March 2003, p. 9. See also UN General Assembly, '2005 summit outcome', A/60/L.1, 20 Sept. 2005, paras 138?9. On the impact of the war in Iraq, see Gareth Evans, 'When is it right to fight?' Survival 46: 3, 2004, pp. 59?82; Alex J. Bellamy, 'A responsibility to protect or a Trojan horse? The crisis in Darfur and humanitarian intervention after Iraq', Ethics and International Affairs 19: 2, 2005, pp. 31-54. 2 Ban Ki-moon, 'Annual address to the General Assembly', 25 Sept. 2007, SG/SM/11182. 3 The Williams Report, Report ofthe High-Level Mission on the situation of human rights in Darfur pursuant to Human Rights Council decision S-4/101, A/HRC/4/80, 7 March 2007, paras 19-20. 4 See Roger Cohen, 'African genocide averted', New York Times, 3 March 2008. 5 These cases are examined at length in Alex J. Bellamy, 'Realizing the responsibility to protect', International Studies Perspectives, forthcoming 2009. See the positions of Algeria and Egypt, S/PV.5319 (Resumption 1), 9 Dec. 2005, pp. 3 and 6. See also Gareth Evans, 'The responsibility to protect: an idea whose time has come . . . and gone?', lecture to the David Davies Memorial Institute, University of Wales, Aberystwyth, 23 April 2008. International Affairs 84: 4 (2008) 615-639 ? 2008 The Author(s). Journal Compilation ? 2008 Blackwell Publishing Ltd/The Royal Institute of International Affairs This content downloaded from 147.251.237.97 on Thu, 09 Feb 2017 10:33:14 UTC All use subject to http://about.jstor.org/terms Alex J. Bellamy mandated only to develop the 'concept' of R2P and build consensus around it.7 In the end, the committee agreed to the appointment of Edward Luck, but insisted that the phrase R2P be removed from his job title. Luck will be paid $1 a year for his services and still has neither a telephone nor an email account at the UN.8 So difficult has it proved to forge consensus on R2P that the principle's supporters are concerned lest its inclusion on the agenda of the 2008 General Assembly damage rather than advance its prospects.9 For very different reasons, some advocates of R2P have themselves taken to criticizing the principle that emerged from the 2005 World Summit, labelling it 'R2P lite'.10 These problems originate from a common source: confusion about the relationship between R2P and non-consensual military intervention. On the one hand, there is a common belief among governments (especially members of the Non-Aligned Movement) that R2P is simply a more sophisticated way of concep tualizing and hence legitimizing humanitarian intervention.11 These concerns have been expressed by governments for as long as the concept has been around but have been fuelled by events.12 For instance, since 2005 it has been widely suggested that R2P 'legalizes' or 'legitimizes' non-consensual intervention potentially without the sanction of the UN Security Council.13 Stephen Stedman, a senior adviser to Kofi Annan on UN reform, argued that Annan's agenda had included 'a new norm, the responsibility to protect, to legalize humanitarian intervention' and then claimed that the 2005 World Summit had succeeded in establishing 'a new norm to 7 See Fifth Committee of the General Assembly, GA/AB/3837, 4 March 2008. It should be noted that Edward Luck was not obliged to seek the support of the Fifth Committee because his is not a salaried position. Luck wisely chose to expose himself to scrutiny in the committee in order to win crucial political support for his mandate. See Gareth Evans, 'The responsibility to protect'; Fifth Committee of the General Assembly, GA/AB/3837, 4 March 2008. 9 This sentiment has been expressed to the author by foreign affairs officials from two member states that are prominent supporters of R2P. 10 This phrase was first suggested to the author by Thomas Weiss, who served as director of the ICISS research directorate. Weiss argued that the R2P concept emerged from the World Summit 'relatively intact' but that because the summit closed off the possibility of intervention not authorized by the Security Council it could be considered 'R2P lite'. See Thomas G. Weiss, Humanitarian intervention: ideas in action (Cambridge: Polity, 2007), pp. 116?17. The term is also used by Don Hubert, who worked with Weiss in the ICISS research directorate. 11 This position has been very commonly expressed by governments. When the Security Council first discussed R2P, at its 2002 retreat, the Chinese insisted that all questions relating to the use of force defer to the Security Council and Russian diplomats argued that by countenancing unilateral intervention, R2P risked undermining the UN Charter. See Yevgeny Primakov, 'UN process, not humanitarian intervention, is world's best hope', New Perspectives Quarterly, 2 Sept. 2004, wwwdigitalnpq.org/global_services%20viewpoint/02-09 04Primakov.html; Jennifer M. Welsh, 'Conclusion: humanitarian intervention after 11 September', in Jennifer M. Welsh, ed., Humanitarian intervention and international relations (Oxford: Oxford University Press, 2004), pp. 176?88. Likewise, in 2005 India's ambassador to the UN insisted that 'we do not believe that discussions on the question [of R2P] should be used as a cover for conferring any legitimacy on the so-called "right of humanitarian intervention" or making it the ideology of some kind of "military humanism'": statement by Mr Nirupam Sen, permanent representative, at the informal thematic consultations of the General Assembly on the Report of the Secretary-General entitled In larger freedom: towards development, security and human rights for all (A/59/2005) (on Cluster III issues: 'Freedom to live in dignity'), 20 April 2005. 12 See the reports on 'Regional roundtables and national consultations' in Thomas G. Weiss and Don Hubert, The responsibility to protect: research, bibliography, background (supplementary volume) (Ottawa: International Development Research Centre, 2001), pp. 349?98. 13 Examples of this include Alicia L. Bannon, 'The responsibility to protect: the UN World Summit and the question of unilateralism', Yale Law fournal 115: 5, 2006, pp. 1156?65. 6i6 International Affairs 84: 4, 2008 ? 2008 The Author(s). Journal Compilation ? 2008 Blackwell Publishing Ltd/The Royal Institute of International Affairs This content downloaded from 147.251.237.97 on Thu, 09 Feb 2017 10:33:14 UTC All use subject to http://about.jstor.org/terms The Responsibility to Protect and the problem of military intervention legalize humanitarian intervention'.14 Governments have themselves tried to use R2P to win support for coercive interference since 2005. The most obvious recent example was the French attempt to use R2P to persuade the Security Council to authorize the forcible distribution of humanitarian assistance in the wake of Cyclone Nargis in May 2008. This prompted discussion about the potential for humanitarian intervention in Myanmar/Burma and attracted criticism from China, Indonesia, Vietnam and South Africa.15 In addition, several prominent figures associated with R2P argue forcefully that the principle is primarily concerned with non-consensual intervention and that its other elements (such as the preven tion of genocide, war crimes, ethnic cleansing and crimes against humanity) are secondary.1 Given all of this, it is not hard to see why many governments continue to suspect that R2P is simply a 'Trojan horse' for the legitimization of unilateral interven tion. On the other hand, some supporters of R2P argue that the principle that emerged from the 2005 World Summit was inadequate because it did not provide clear guidance about the circumstances in which coercive military intervention might be justified or about the appropriate decision-making process in situations where the Security Council is deadlocked. They argue that the set of criteria proposed by the International Commission on Intervention and State Sovereignty (ICISS) in 2001 to guide international decision-making in times of major humani tarian emergencies was an important casualty of pre-summit diplomacy in 2005 and should be put back on the international agenda.17 Achieving a deeper consensus on R2P and making progress towards the Secre tary-General's goal of translating it from 'words into deeds' is unlikely while confu sion remains about what the principle's implications are for military intervention. Sceptics will continue to see R2P as a 'Trojan horse' for unilateral intervention and supporters will focus on finessing and applying criteria to guide intervention. The principal aim of this article, therefore, is to clarify what R2P says about military intervention and what it can contribute in practice. To do this, I proceed in three stages. The first seeks to clarify the meaning of R2P. It identifies the principle's roots in two related but quite different contexts: the notion of sovereignty as responsibility, as developed by Francis Deng and Roberta Cohen in the 1990s, and the debate about unilateral humanitarian intervention sparked by NATO's 1999 intervention in Kosovo.1 The tensions between these two distinct roots 14 Stephen John Stedman, 'UN transformation in an era of soft balancing', International Affairs 83: 5, Sept. 2007, PP-933, 938. 15 'World fears for plight of Myanmar cyclone victims', New York Times, 13 May 2008. 1 See Weiss, Humanitarian intervention, p. 106. 17 One of the principal aims of the Global Centre for the Responsibility to Protect, created in February 2008, is to clarify the circumstances in which non-consensual military force might be legitimate. See also Gareth Evans, 'Responsibility to protect in 2007: five thoughts for policy makers', presentation at a panel discussion at the UN, New York, 13 April 2007; Marc Saxer, 'The politics of responsibility to protect', Friedrich Ehert Stiftung briefing paper, Berlin, April 2008, pp. 5-6. 1 A third root, recently invoked by France in relation to the cyclone in Myanmar/Burma, is the 'right to interfere' developed by Bernard Kouchner from the late 1980s. See B. Kouchner and M. Betatti, Le Devoir d'ingerence: peut-on laisser mourir? (Paris: Denoel, 1987). Although Kouchner's ideas are innovative and interesting, there is little evidence to suggest that they influenced either Deng and Cohen or the ICISS. 617 International Affairs 84:4, 2008 ? 2008 The Author(s). Journal Compilation ? 2008 Blackwell Publishing Ltd/The Royal Institute of International Affairs This content downloaded from 147.251.237.97 on Thu, 09 Feb 2017 10:33:14 UTC All use subject to http://about.jstor.org/terms Alexf. Bellamy go some way towards explaining the confusion about the relationship between R2P and military intervention. I then set out what world leaders signed up to in 2005?2006 and what this means for military intervention. The second section focuses on criteria for decision-making and argues that R2P should be dissociated from the criteria put forward by the ICISS. Although they undoubtedly consti tuted an innovative proposal, there is little likelihood of international consensus on criteria, and its proponents overstate their practical and political utility. The final part of the essay explores what R2P can contribute to the practice of military intervention. R2P is well placed to make at least three important contributions, I argue. First, it can help minimize the problem of 'moral hazard' identified by Alan Kuperman.19 Second, it can reduce the temptation for policy-makers to focus exclusively on military responses to grave humanitarian problems.20 Finally, by establishing a political commitment to protection it creates a mandate for progress in thinking about the capacities and doctrines needed to increase the effectiveness of protective forces once deployed. R2P and military intervention In order to clarify what R2P has to say about military intervention and understand the tensions described above, we need to understand its roots and to grasp what the General Assembly and Security Council agreed and why they agreed to it. Sovereignty as responsibility In 1993, the then UN Secretary-General Boutros Boutros-Ghali appointed Francis Deng, a well-respected former Sudanese diplomat, as his Special Representative on Internally Displaced People (IDPs).21 In appointing Deng and highlighting the problem of IDPs, Boutros-Ghali was responding to both urgent humanitarian need and a vexing political dilemma. As wars became less a matter of conflict between states and more a struggle between forces within states, so the number of internally displaced expanded. When Deng was appointed there were some 25 million IDPs globally, compared to a little over 1 million a decade earlier.22 Remaining within national borders, IDPs were afforded no special international protection of the kind offered to refugees and so they remained critically vulner able to the whims or failings of their home state. Unsurprisingly, therefore, a combination of violence, disease and deprivation contrived to make mortality rates among IDPs higher than among the general population, sometimes by as much as 50 times.23 19 Alan J. Kuperman, 'The moral hazard of humanitarian intervention: lessons from the Balkans', International Studies Quarterly 52: i, 2008, pp. 49?80. 20 A problem identified by Alex de Waal in relation to Darfur. See Alex de Waal, 'Darfur and the failure of the responsibility to protect', International Affairs 83: 6, Nov. 2007, PP- J039-54 21 For an excellent account of the IDP problem and the work of Deng and Cohen, see Thomas G. Weiss and David A. Korn, Internal displacement: conceptualization and its consequences (New York: Routledge, 2006). 22 Weiss, Humanitarian intervention, p. 90. 23 Francis M. Deng, 'The impact of state failure on migration', Mediterranean Quarterly 1514, Fall 2004, p. 18. 6i8 International Affairs 84: 4, 2008 ? 2008 The Author(s). Journal Compilation ? 2008 Blackwell Publishing Ltd/The Royal Institute of International Affairs This content downloaded from 147.251.237.97 on Thu, 09 Feb 2017 10:33:14 UTC All use subject to http://about.jstor.org/terms The Responsibility to Protect and the problem of military intervention The principal challenges confronting Deng and his colleague, Roberta Cohen from the Brookings Institution, were how to persuade host governments to improve protection for IDPs and how to work around the denial of assistance by sovereign authorities.24 As Deng himself put it, 'the internally displaced are paradoxically assumed to be under the care of their own governments despite the fact that their displacement is often caused by the same state authorities'.25 To argue their way around the use of sovereignty to deny international assistance for IDPs, Deng and Cohen devised the notion of 'sovereignty as responsibility'. The concept's starting point was recognition that the primary responsibility for protecting and assisting IDPs lay with the host government.2 No legitimate state, they argued, could quarrel with the claim that it was responsible for the well being of its citizens?and in practice no government did in fact quarrel with this proposition. Where a state was unable to fulfil its responsibilities, they went on to assert, it should invite and welcome international assistance.27 Such assistance helped the state by enabling it to discharge its sovereign responsibilities and take its place as a legitimate member of international society.2 During major crises, troubled states faced a choice: they could work with international organizations and other interested outsiders to realize their sovereign responsibilities; or they could obstruct those efforts, and thereby sacrifice their good standing and sover eign legitimacy.29 To translate 'sovereignty as responsibility' into protection for IDPs, Deng and Cohen worked with legal experts to define IDPs, identify the rights they already enjoyed under existing human rights instruments and place those rights into the context of displacement, presenting the results of this work in the form of 'guiding principles', which were set out in 1998.3? The principles recognized that primary responsibility for displaced people rested with the local authorities but asserted that consent to international aid should not be 'arbitrarily withheld', especially when the local authorities were unable or unwilling to provide the necessary assistance.31 The principles were adopted by the UN's Inter-Agency Standing Committee (IASC), the OSCE and the AU. ECOWAS called upon its members to disseminate and apply them. In addition, several countries (Burundi, Colombia, the Philip pines and Sri Lanka) have incorporated the principles into national law and others 24 Deng, 'Impact of state failure on migration', p. 20. 25 Francis M. Deng, 'Divided nations: the paradox of national protection', Annals of the American Academy of Political and Social Science 603, Jan. 2006, p. 218. 2 Roberta Cohen and Francis M. Deng, Masses inflight: the global crisis of internal displacement (Washington DC: Brookings Institution, 1998), p. 275. 27 Deng, 'Impact of state failure on migration' , p. 20. 2 Francis M. Deng, Sadikiel Kimaro, Terrence Lyons, Donald Rothchild and I. William Zartman, Sovereignty as responsibility: conflict management in Africa (Washington DC: Brookings Institution, 1996), p. 1. 29 Deng et al., Sovereignty as responsibility, p. 28. 30 See Roberta Cohen, 'The guiding principles on internal displacement: an innovation in international standard setting', Global Governance 10: 3, 2004, pp. 459?80; Roberta Cohen, 'Developing an international system for internally displaced persons', International Studies Perspectives 7:1, 2006, pp. 87?101. 31 Principles 3 and 25 of the 'Guiding principles on internal displacement'. For a detailed explanation and commentary, see Walter Kalin, 'Guiding principles on internal displacement: annotations', American Society of International Law Studies in Transnational Legal Policy, 32, 2000. 619 International Affairs 84:4, 2008 ? 2008 The Author(s). Journal Compilation ? 2008 Blackwell Publishing Ltd/The Royal Institute of International Affairs This content downloaded from 147.251.237.97 on Thu, 09 Feb 2017 10:33:14 UTC All use subject to http://about.jstor.org/terms Alexf. Bellamy are looking at following suit.32 But at what point could a state be judged to have forfeited its sovereignty, and what body has the right to make that decision? Deng and his collaborators were sketchy on these points, but suggested that sovereignty as responsibility implied the existence of a 'higher authority capable of holding supposed sovereigns accountable' and that this higher authority should place the common good ahead of the national interests of its members. Clearly, the UN Security Council comes closest to fitting the bill. 'Sovereignty as responsibility' focused on the responsibilities of host govern ments and maintained that vulnerable populations were best protected by effective and legitimate states. In practice, Deng's approach to assisting IDPs required an invitation from the host state and focused on the use of diplomacy in those cases where the host state refused assistance (especially Turkey, Myanmar/Burma and Algeria). Sometimes, persistent diplomacy paid dividends (for instance, Turkey relented in 2002).33 In other situations, however, the doctrine would point to the referral of cases to a 'higher authority'?namely, the UN Security Council. Kosovo and the ICISS The second?and more obvious?of R2P's roots is the international commission (ICISS) that produced the eponymous report. The ICISS report Responsibility to protect was primarily concerned with reconceptualizing humanitarian intervention in the wake of the Kosovo crisis and the Secretary-General's challenge to the 1999 General Assembly to resolve the tension between sovereignty and fundamental human rights.34 The commission had its genesis in early 2000, when Canadian foreign affairs officials Don Hubert, Heidi Hulan and Jill Sinclair began advocating an 'International Commission on Humanitarian Intervention' in response to events in Kosovo and Annan's challenge. Canada's foreign minister, Lloyd Axworthy, persuaded Annan to endorse the commission, though its title was revised to omit the controversial 'humanitarian intervention' label.35 Nonetheless, the alterna tive adopted?'Intervention and State Sovereignty'?indicated the body's main aim: to reconcile the occasional need for armed intervention to protect vulnerable populations with the principles of state sovereignty. The commission's recommendations are well known. They were premised on the notion that when states are unwilling or unable to protect their citizens from grave harm, the principle of non-interference 'yields to the responsibility to protect'. The report aimed to escape the irresolvable logic of 'sovereignty versus human rights' by focusing not on what interveners are entitled to do ('a right of intervention') but on what is necessary to protect people in dire need and the responsibilities of various actors to afford such protection. These responsibilities 32 Cohen, 'The guiding principles', p. 470. 33 Weiss and Korn, Internal displacement, pp. 74?5. 34 International Commission on Intervention and State Sovereignty, The responsibility to protect (Ottawa: International Development Research Centre, 2001), p. 9; Kofi Annan, 'Annual Report of the Secretary General to the General Assembly', 20 Sept. 1999. 35 Lloyd Axworthy, 'Human rights and humanitarian intervention', address, Washington DC, 16 June 2000; Lloyd Axworthy, Navigating a new world: Canada's global future (Toronto: Knopf, 2003), p. 191. 620 International Affairs 84: 4, 2008 ? 2008 The Author(s). Journal Compilation ? 2008 Blackwell Publishing Ltd/The Royal Institute of International Affairs This content downloaded from 147.251.237.97 on Thu, 09 Feb 2017 10:33:14 UTC All use subject to http://about.jstor.org/terms The Responsibility to Protect and the problem of military intervention were about much more thanjust armed intervention. In addition to a 'responsibility to react' to massive human suffering, international society also had responsibilities to use non-violent tools to prevent such suffering from happening, and, where it did happen, to rebuild polities and societies afterwards. Indeed, the commission argued that prevention was the most important aspect of R2P.3 Despite stressing the critical importance of prevention, the commission's main focus was on intervention. It dedicated only 9 of its 85 pages to prevention, and only 16 to the responsibilities to prevent and rebuild, whereas 32 pages were devoted to intervention. The commission's discussion of intervention centred on two questions: In what circumstances is intervention legitimate? And what insti tutions are entitled to authorize intervention? In relation to the first question, the commission proposed just cause thresholds ('large scale loss of life' and 'large scale ethnic cleansing') and precautionary principles ('right intention', 'last resort', 'proportional means' and 'reasonable prospects'), arguing that if states committed themselves to these principles, it would be easier to build consensus on how to respond to humanitarian emergencies. In addition, it would be harder for states like China and Russia to oppose genuine humanitarian intervention because they would have committed themselves to a responsibility to protect in cases of large scale loss of life and ethnic cleansing (just cause thresholds). On the other hand, it would be harder for states to abuse humanitarian justifications because it would be very difficult to satisfy all the criteria in non-genuine cases. In relation to the question of authority, the commission argued that the Security Council had the primary responsibility to act when a host state was unwilling or unable to protect its citizens. To improve the Council's decision-making, the commission suggested that the permanent members (P5) agree to refrain from casting their veto in threshold-crossing situations where no vital national interests were at stake and where a majority of the Council supported collective action. If the Council nevertheless failed to act in the face of threshold-crossing crises where the precau tionary principles indicated that intervention was appropriate, concerned states could approach the General Assembly and, failing that, relevant regional organi zations. Thus the commission outlined a hierarchy of responsibility, starting with the host state and rising through the Security Council to the General Assembly, regional organizations and coalitions of the willing to, finally, individual states. The ICISS succeeded in refraining the humanitarian intervention debate by stressing the primary responsibility that states had towards their own citizens, situating non-consensual intervention within a wider continuum of measures including prevention, rebuilding and non-forcible means of reaction, and identi fying a range of practices other than armed intervention that could contribute to the prevention and mitigation of genocide and mass atrocities. However, the commission's own regional round tables, as well as post-report consultations with NGOs and governments organized by the Canadian government and civil society organizations, all highlighted widespread hostility to 'humanitarian intervention' and a broad consensus against the idea of a so-called 'right of 3 ICISS, Responsibility to protect, p. xi. 621 International Affairs 84:4, 2008 ? 2008 The Author(s). Journal Compilation ? 2008 Blackwell Publishing Ltd/The Royal Institute of International Affairs This content downloaded from 147.251.237.97 on Thu, 09 Feb 2017 10:33:14 UTC All use subject to http://about.jstor.org/terms Alexf. Bellamy intervenion', especially where that right was associated with unilateralism.37 But this deep-seated scepticism towards intervention did not necessarily translate into a rejection of the underlying purpose of R2P?the prevention of genocide and mass atrocities, and the protection of vulnerable populations. The adoption of language focusing on the rights of endangered populations rather than the rights of interveners helped illuminate a broad constituency of states and civil society actors prepared to acknowledge that sovereignty entailed responsibilities and that international engagement might be legitimate in certain circumstances. However, the commission's focus on non-consensual intervention and apparent openness to intervention not authorized by the Security Council meant that R2P was unlikely to command consensus among world leaders without some important revisions. It is not surprising, therefore, that the R2P principle that emerged from the 2005 World Summit was different in many respects from the doctrine espoused by the ICISS, even if the name and the central idea remained the same. The key to under standing what R2P has to say about military intervention lies in recognizing that R2P as an international principle is different from the concept proposed by the ICISS and from the doctrine espoused by Deng and Cohen, even though it draws on both. International consensus on R2P When governments, regional organizations and the UN talk about R2P they mean not the concept put forward by the ICISS but the principle endorsed by world leaders at the 2005 World Summit and reaffirmed by the Security Council in 2006.3 That principle was informed by the commission's work, by Deng and Cohen's work on IDPs, and by the UN's work on the protection of civilians (including the Security Council's interest in this theme, which predates the ICISS), but is different from all of them in important respects. Although lengthy, paragraphs 138 and 139 ofthe World Summit outcome document are worth repeating in full: 138. Each individual state has the responsibility to protect its populations from genocide, war crimes, ethnic cleansing and crimes against humanity. This responsibility entails the prevention of such crimes, including their incitement, through appropriate and necessary means. We accept that responsibility and will act in accordance with it. The international community should, as appropriate, encourage and help States to exercise this responsi bility and support the United Nations in establishing an early warning capability. 139. The international community, through the United Nations, also has the responsibility to use appropriate diplomatic, humanitarian and other peaceful means, in accordance with Chapters VI and VIII of the Charter of the United Nations, to help protect populations from war crimes, ethnic cleansing and crimes against humanity. In this context, we are prepared to take collective action, in a timely and decisive manner, through the Security 37 See World Federalist Movement and International Policy Group, Civil society perspectives on the responsibility to protect, final report, 30 April 2003; Noel M. Morada, 'R2P roadmap in southeast Asia: challenges and prospects', UNISCI (Research Unit on International Security and Cooperation, Universidad Complutense, Madrid) discussion papers 11, May 2006. 38 UN Security Council Resolution 1674 (2007). 622 International Affairs 84: 4, 2008 ? 2008 The Author(s). Journal Compilation ? 2008 Blackwell Publishing Ltd/The Royal Institute of International Affairs This content downloaded from 147.251.237.97 on Thu, 09 Feb 2017 10:33:14 UTC All use subject to http://about.jstor.org/terms The Responsibility to Protect and the problem of military intervention Council, in accordance with the Charter, including Chapter VII, on a case-by-case 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 cleansing and crimes against humanity. We stress the need for the General Assembly to continue consideration of the responsibility to protect populations from genocide, war crimes, ethnic cleansing and crimes against humanity and its implications, bearing in mind the principles of the Charter and international law. We also intend to commit ourselves, as necessary and appropriate, to helping States build capacity to protect their populations from genocide, war crimes, ethnic cleansing and crimes against humanity and to assisting those which are under stress before crises and conflicts break out.39 These two paragraphs can be boiled down to four basic commitments. First, all states acknowledge that they have a responsibility to protect their citizens from genocide, war crimes, crimes against humanity and ethnic cleansing. Second, they agree to provide assistance to help other states to build the capacity they need to discharge this responsibility. Third, in situations where the host state is 'manifestly failing' in its responsibility, they agree to use all peaceful means to protect vulner able populations. Fourth, should those measures fail or be deemed inappro priate, the Security Council stands ready to use all necessary means, including non-consensual force. What are the principal differences between the concept as espoused by the ICISS and the principle as agreed by world leaders? In the latter form, R2P no longer proposed criteria to guide decision-making about when to intervene; there is no code of conduct for the use of the veto; and there is no opening for coercive measures not authorized by the Security Council. The threshold on when R2P is transferred from the host state to international society was raised from the point at which the host state proved itself 'unable and unwilling' to protect its own citizens to that at which the state was 'manifestly failing' in its responsibility to do so. Finally, the idea that R2P implied responsibilities?even obligations?on the part of international society and especially the Security Council was all but removed, with the Council committed only to 'standing ready' to act when necessary. Of course, prudence dictates a 'case-by-case' approach; but the insertion of words to that effect was a deliberate attempt to water down the Security Council's respon sibility to protect.40 We should not, however, succumb to the view that the R2P principle that emerged from the 2005 World Summit was too weak or insubstantial to contribute to the practice of non-consensual intervention for humanitarian purposes. First, the World Summit clarified the principle's scope. R2P applies to genocide, war crimes, crimes against humanity and ethnic cleansing, all of which have fairly precise legal meanings grounded in the Genocide Convention, the Rome Statute of the International Criminal Court, and the practice of the international criminal 39 United Nations General Assembly, '2005 summit outcome', A/60/L.1, 20 Sept. 2005, paras 138-9. 40 I develop this point in more detail and relate it to US Ambassador John Bolton's position on R2P in Alex J. Bellamy, Responsibility to protect: the global effort to end mass atrocities (Cambridge: Polity, forthcoming 2009), ch. 3. 623 International Affairs 84:4, 2008 ? 2008 The Author(s). Journal Compilation ? 2008 Blackwell Publishing Ltd/The Royal Institute of International Affairs This content downloaded from 147.251.237.97 on Thu, 09 Feb 2017 10:33:14 UTC All use subject to http://about.jstor.org/terms Alexf. Bellamy tribunals for former Yugoslavia and Rwanda. This is a much clearer formulation than that offered by the ICISS. Second, the World Summit has clarified relevant roles and responsibilities. In line with the doctrine put forth by Deng and Cohen, all states have a primary responsibility towards their own citizens. All other states have a responsibility to assist their peers in fulfilling this primary responsibility. Should a state manifestly fail in its responsibility, the Security Council in partner ship with relevant regional organizations has a responsibility to use whatever means it determines necessary and appropriate. Finally, there is no such thing as an 'R2P event or crisis', in that there is no moment at which something becomes relevant to R2P. To suggest that such a thing exists is to revert to the old language of intervener's rights. A state's responsibility to its citizens does not appear and then evaporate; nor does the world's responsibility to assist and support that state, or the Security Council's responsibility to take all necessary means when appro priate. In other words, it is not the nature of the responsibility that changes, but the most appropriate means of preventing genocide, war crimes, crimes against humanity and ethnic cleansing, and of protecting vulnerable populations in any given situation. R2P on military intervention As Edward Luck has argued, it is important not to confuse what we would like the R2P principle to be with what it actually is.41 R2P sets out responsibilities that states have to their own citizens (the primary responsibility to protect), responsi bilities that all states have as members of the international community (responsi bilities to help build capacity and use peaceful means to prevent and protect) and responsibilities that certain institutions have (the Security Council's responsibility to use all appropriate means when necessary, in partnership with relevant inter national organizations). Contrary to much contemporary writing on the subject, R2P does not set out criteria for the use of force, suggest that there are 'just causes that justify the use of force beyond the two exemptions ofthe UN Charter', offer pathways for intervention not authorized by the Security Council, amend the way the Council does business, apply more widely than to the four specific crimes listed in the extract reproduced above, or promise intervention in every case.42 My task in the remainder of this article is twofold. In the next section I seek to show that the rejection of criteria was politically inevitable and practically inconsequential. Then, in the final section, I will set out three ways in which R2P can make an important contribution to the prevention of genocide, war crimes, crimes against humanity and ethnic cleansing, and the protection of vulnerable populations. 41 See Edward C. Luck, 'The responsible sovereign and the responsibility to protect', Annual review of United Nations affairs 2006/2007 (New York: Oxford University Press, 2008), vol. 1, pp. xxxiii?xliv. 42 The quote is from Saxer, 'The politics of responsibility to protect', p. 6. 624 International Affairs 84: 4, 2008 ? 2008 The Author(s). Journal Compilation ? 2008 Blackwell Publishing Ltd/The Royal Institute of International Affairs This content downloaded from 147.251.237.97 on Thu, 09 Feb 2017 10:33:14 UTC All use subject to http://about.jstor.org/terms The Responsibility to Protect and the problem of military intervention The limits of criteria The ICISS criteria (just cause thresholds and precautionary principles) to guide decisions about military intervention were intended to fulfil three primary functions. First, in an attempt to avoid any future cases like that of Rwanda, where the world stood aside as 800,000 people were butchered in genocidal violence, the just cause thresholds were intended to create expectations about the circumstances in which the international community?primarily the UN Security Council? should become engaged in major humanitarian catastrophes, consider intervening with force and constrain permanent members from casting pernicious vetoes for selfish reasons.43 Second, responding to a need to avoid future situations like that of Kosovo, where the Security Council was blocked by veto, the criteria provided a pathway for legitimizing intervention not authorized by the Security Council.44 Finally, Ramesh Thakur, one of the most prominent commissioners, argued that criteria should be viewed as constraining governments' ability to 'abuse' R2P and limiting the scope of potential Security Council interventionism.45 According to Thakur, the criteria would both 'make it more difficult for coalitions of the willing to appropriate the language of humanitarianism for geopolitical and unilateral interventions' and make the Security Council's deliberations more transparent.4 Consensus on criteria, he insisted, would make it more, not less, difficult for states to claim a humanitarian mantle for armed intervention.47 Before we assess the extent to which the criteria would be able to fulfil these functions, it is important to begin by stressing how little political support they received. Most of the P5 were sceptical about them from the outset. At the Security Council's annual retreat in May 2002 the United States rejected them outright on the grounds that permanent members should not constrain their right to cast their veto whenever they saw fit.4 Russia and China expressed concern that the criteria could be used to bypass the Security Council.49 Although the British government had earlier presented its own version of criteria to guide decision-making and circumvent a Security Council veto, along with France it worried that agreement on criteria would not necessarily deliver the political will and consensus required for effective responses to humanitarian crises.50 Negative attitudes towards criteria were 43 Nicholas J. Wheeler, 'Legitimating humanitarian intervention: principles and procedures', Melbourne fournal of International Law 2: 2, 2001, p. 566. 44 See e.g. Independent International Commission on Kosovo, Kosovo report: conflict, international response, lessons learned (Oxford: Oxford University Press, 2000), opening summary. This line of reasoning is also developed later in the report. 45 Ramesh Thakur, The United Nations, peace and security: from collective security to the Responsibility to Protect (Cambridge: Cambridge University Press, 2007), p. 260. 4 Ramesh Thakur, 'A shared responsibility for a more secure world', Global Governance 11: 3, 2005, p. 284. 47 Ramesh Thakur, 'Iraq and the responsibility to protect', Behind the Headlines 62: 1, 2004, pp. 1?16. 4 Welsh, 'Conclusion', p. 180. This led prominent observers to write in 2004 that 'the Bush administration does not and will not accept the substance of the report or support any formal declaration or resolution about it': S. Neil Macfarlane, CarolinJ. Thiekling and Thomas G. Weiss, 'The responsibility to protect: is anyone interested in humanitarian intervention?', Third World Quarterly 25: 5, 2004, p. 983. The United States did not change its position on criteria. See Sir Adam Roberts, 'The United Nations and humanitarian intervention', in Welsh, ed., Humanitarian intervention, p. 90. 49 Primakov, 'UN process, not humanitarian intervention, is world's best hope'. 50 Welsh, 'Conclusion', p. 204, n. 4. 625 International Affairs 84:4, 2008 ? 2008 The Author(s). Journal Compilation ? 2008 Blackwell Publishing Ltd/The Royal Institute of International Affairs This content downloaded from 147.251.237.97 on Thu, 09 Feb 2017 10:33:14 UTC All use subject to http://about.jstor.org/terms Alexf. Bellamy only hardened by the US-led invasion of Iraq in 2003. Fearing that criteria might be used to justify the invasion, a forum of social democratic governments rejected a British proposal to endorse the idea.51 In the post-invasion context, the Canadian government recognized that a full-scale effort to persuade the General Assembly to endorse criteria could 'backfire terribly', destroying potential consensus on R2P.52 To be fair, there was some international support for a limited role for criteria. The proposal was endorsed by the UN's High Level Panel, convened by Kofi Annan, and in Annan's personal blueprint for reform.53 Significantly, however, Annan separated the cornmitment to R2P from the proposed criteria, placing the former in a section on the rule of law and leaving the latter in a section on the use of force. He did this to reinforce the view that R2P was not only about the use of force and to protect R2P from the almost inevitable rejection of criteria.54 The AU's 'Ezulwini Consensus' on UN reform endorsed the High Level Panel's criteria for guiding the Security Council but, at the insistence of South Africa, observed that these guidelines 'should not undermine the responsibility of the international community to protect'.55 It was clear from the outset of the negotiations preceding the 2005 World Summit that there would be no consensus on criteria. Whereas several African states endorsed the view that criteria were essential to making the Security Council's decisions more transparent, accountable (to the wider membership) and hence legitimate, the United States, China and Russia opposed them?though for very different reasons: the United States because it believed that criteria would limit its freedom of action, the others because they feared that criteria might be used to circumvent the Council. Many other influential states, most notably India, shared this latter view; although it was publicly expressed by only a few states, Canada's regional consultations had revealed that it was a significant under lying concern in many parts of the world, especially Asia. In consequence, the recommendation for criteria was watered down into a commitment to continue discussing criteria, in order to keep the Americans, Chinese, Russians and Indians on board.5 Ultimately, however, the diplomats charged with selling R2P to the world recognized that criteria would be a 'bridge too far' for the Americans and the proposal was never seriously put on the table.57 From this brief overview it is clear that it was always unlikely that members of the General Assembly or Security Council would be persuaded to adopt criteria, and that there was a real danger that persisting with the linkage of criteria to R2P would have prevented the endorsement of R2P in 2005 and its reaffirmation the 51 See 'British PM urges tougher stance against brutal regimes', Agence France-Presse, 14 July 2003; Kevin Ward, 'Process needed so countries know when to intervene to protect human rights', CBS (Canada), 13 July 2003. 52 'Civil society meeting on the responsibility to protect', final report, Ottawa, 8 April 2003, p. 9. 53 UN High Level Panel on Threats, Challenges and Change, A more secure world: our shared responsibility, A/59/565, 2 Dec. 2004, para. 203; Kofi Annan, 'In larger freedom: towards development, security and human rights for all', A/59/2005, 21 March 2005, para. 126. 54 William R. Pace and Nicole Deller, 'Preventing future genocides: an international responsibility to protect', World Order 36: 4, 2005, p. 25. 55 African Union Executive Council, 'The common African position on the proposed reform of the United Nations', ext/EX.CL/2(VII), Addis Ababa, 7-8 March 2005, sec. B (i). 5 Pace and Deller, 'Preventing future genocides', p. 28. 57 Memo from Allan Rock to the author, 12 Nov. 2007. 626 International Affairs 84: 4, 2008 ? 2008 The Author(s). Journal Compilation ? 2008 Blackwell Publishing Ltd/The Royal Institute of International Affairs This content downloaded from 147.251.237.97 on Thu, 09 Feb 2017 10:33:14 UTC All use subject to http://about.jstor.org/terms The Responsibility to Protect and the problem of military intervention following year.58 Moreover, a diplomatic effort to persuade states to adopt criteria in the future would require the investment of a significant amount of political capital with little chance of success and a heightened likelihood that such advocacy would create a backlash resulting in a retreat from the principle endorsed in 2005.59 Finally, it is not at all clear that the R2P principle itself would be strengthened by the addition of criteria. In what remains of this section, I will examine the three putative functions of criteria. The first function of criteria?creating expectations?is most easily dispensed with, because although the 2005 World Summit did not endorse criteria, it did identify the crimes from which governments had a responsibility to protect populations and the circumstances in which that responsibility ought to be taken up by international society. The summit, it will be recalled, insisted that states have a responsibility to protect their populations from genocide, war crimes, crimes against humanity and ethnic cleansing, and that this responsibility should be taken up by the Security Council in cases where a government was 'manifestly failing' to provide such protection. There is broad agreement that the Security Council should be engaged in such circumstances. For example, the Chinese government's 2005 position paper on UN reform agreed that 'massive humanitarian' crises were 'the legitimate concern of the international community'. ? It is important to recognize, however, that agreement on thresholds does not guarantee agreement on whether the thresholds have been breached or on what is the most appropriate response in actual cases. This problem was raised throughout the ICISS consultation process and has been aired many times since.61 It has also been evident in practice: in relation to Kosovo, the disagreement between NATO and Russia boiled down to judgements about whether the conflict there was sufficiently grave to warrant armed intervention; 2 more recently, in relation to Darfur, governments more or less agreed on the gravity of the threat but disagreed about the most appropriate course of action and the responsibility of the Sudanese government. 3 Indeed, even advocates of R2P disagreed on whether the just cause thresholds and precautionary principles justified armed intervention in this case.64 On the other hand, agreement has been reached in less high-profile cases, such as those of the Democratic Republic of Congo (DRC) and Burundi, without the need for thresholds.65 58 This is a good example of a 'moral limit' in international politics. For a thorough investigation of moral limits, see Richard M. Price, ed., Moral limit and possibility in world politics (Cambridge: Cambridge University Press, 2008). 59 On the post-2005 'revolt against R2P' see Bellamy, 'Realizing the responsibility to protect'. 0 'Position paper of the People's Republic of China on the United Nations reforms', 8 June 2005. 1 See Jeffrey Boutwell, 'Report on the Pugwash Study Group on Intervention, Sovereignty and International Security meeting', Venice, Italy, 10?11 Dec. 1999; Weiss and Hubert, The responsibility to protect, supplementary volume, pp. 351?2; Welsh, 'Conclusion'; Wheeler, 'Legitimating humanitarian intervention'. 2 See Simon Chesterman, fust war or just peace? Humanitarian intervention and international law (Oxford: Oxford University Press, 2001), p. 221. 3 Alex J. Bellamy, 'A responsibility to protect or a Trojan horse?' 4 Gareth Evans's International Crisis Group argued not. See International Crisis Group, 'Getting the UN into Darfur', Africa Briefing 43, 12 Oct. 2006, pp. 15-17. Others, such as Eric Reeves and Samantha Power, disagree with this perspective. 5 Susan C. Breau, 'The impact of responsibility to protect on peacekeeping', fournal of Conflict and Security Law 11: 3, 2007, pp. 450?52; Victoria K. Holt and Tobias C. Berkman, The impossible mandate? Military preparedness, 627 International Affairs 84:4, 2008 ? 2008 The Author(s). Journal Compilation ? 2008 Blackwell Publishing Ltd/The Royal Institute of International Affairs This content downloaded from 147.251.237.97 on Thu, 09 Feb 2017 10:33:14 UTC All use subject to http://about.jstor.org/terms Alexf. Bellamy Nor is there much evidence to suggest that the thresholds could constrain the use of the veto. It has been suggested, for example, that Russia and China might have been 'compelled' into abstaining on a vote authorizing intervention in Darfur had such a resolution been tabled in the Council and backed by the argument that intervention would be the only means of relieving the humanitarian catastrophe. But China's actual performance in the Council suggests that it would be more than willing to use its veto in such cases. In relation to Darfur, China threatened vetoes on measures far less intrusive than non-consensual military intervention, such as comprehensive targeted sanctions and no-fly zones. 7 Given that China's position on Darfur enjoyed the support of a significant chunk of the Non-Aligned Movement, the League of Arab States and the Organization of the Islamic Conference, it is not clear where the pressure to abstain in such a vote would have come from. What, then, of the second function of criteria?to provide a way of legitimizing armed intervention without Security Council authorization? We should note at the outset that this has been the most oft-cited function of criteria since they were first mooted in the 1970s.6 Interest in criteria was reignited by the Security Council's failure to reach a consensus over Kosovo. In the wake of the storm over Kosovo, Tony Blair called for five tests to guide decisions on intervention and the Foreign Office circulated a draft paper on the subject among the P5. 9 Blair's view that criteria would provide guidelines for when regional organizations and coalitions of the willing might legitimately intervene without the sanction of the Security Council was endorsed by the Independent International Commission on Kosovo (IICK). Working towards its finding that the intervention in Kosovo was 'illegal but legitimate', the IICK lent support to the idea of using criteria as thresholds for determining whether or not to use force to alleviate humanitarian emergencies. It recognized that while the UN Charter's restrictions on the use of force contributed to international peace and security by prohibiting aggressive war, there might be circumstances?as in Kosovo?where intervention was needed as a last resort but was not likely to be authorized by the Security Council because of a threatened veto. Criteria, the commission reasoned, might create pathways for states to intervene legitimately in the most extreme emergencies without Council authorization.70 There was never much likelihood that the UN membership would endorse guidelines providing a pathway to intervention not authorized by the Council. Moreover, it is not altogether clear what such a pathway would contribute. After the responsibility to protect and modern peace operations (Washington DC: Henry L. Stimson Center, 2006), pp. 201?224. Nicholas J. Wheeler and Justin Morris, 'Justifying the Iraq War as a humanitarian intervention: the cure is worse than the disease', in Ramesh Thakur and Waheguru Pal Singh Sidhu, eds, The Iraq crisis and world order (Tokyo: UN University Press, 2006), p. 460. 7 Bellamy, 'A responsibility to protect or a Trojan horse?'; Paul D. Williams and Alex J. Bellamy, 'The responsibility to protect and the crisis in Darfur', Security Dialogue 36: 1, 2005, pp. 27?47. See e.g. Richard B. Lillich, ed., Humanitarian intervention and the United Nations (Charlottesville, VA: University Press of Virginia, 1973). 9 Tony Blair, 'Doctrine of the international community', speech to the Economic Club of Chicago, Hilton Hotel, Chicago, 22 April 1999. See John Kampfner, Blair's wars (London: Free Press, 2003), pp. 50-53; Wheeler, 'Legitimating humanitarian intervention', p. 564 and n. 51. 70 IICK, Kosovo report, opening summary. 628 International Affairs 84: 4, 2008 ? 2008 The Author(s). Journal Compilation ? 2008 Blackwell Publishing Ltd/The Royal Institute of International Affairs This content downloaded from 147.251.237.97 on Thu, 09 Feb 2017 10:33:14 UTC All use subject to http://about.jstor.org/terms The Responsibility to Protect and the problem of military intervention all, states already have moral arguments for armed intervention in the worst cases. For example, the US used moral rather than legal language to justify its partici pation in Operation Allied Force, arguing simply that the moral imperative to protect people?in this case, the Albanian population of Kosovo?from ethnic cleansing overrode the legal ban on the use of force.71 Critics did not argue in response that massive ethnic cleansing did not, in some circumstances, provide grounds for intervention. Instead, as mentioned earlier, they quibbled over the gravity of the threat, the prudence of intervention and the appropriate source of authority for it.72 It is difficult to see how criteria would help in a case like this. Interveners would argue that the criteria were satisfied and that their actions were thus legitimate; critics would argue that they were not. In the end, international society is left with borderline judgements about the legitimacy of armed inter vention and individual states are left making up their own minds on the basis of their perception of the facts of the case and the relative importance of sovereignty, non-intervention, the protection of human rights and prudential calculations.73 International law relating to the use of force and crimes such as genocide already provides a common language for this debate. It is not clear what criteria would contribute in addition. This brings us to the third putative function of criteria: restricting abuse. This concern has become somewhat redundant in the wake of the World Summit's adoption of R2P. The danger of abuse is raised whenever there is a pathway to legitimate intervention that circumvents a deadlocked Security Council. Paragraph 139 of the outcome document, reproduced above, clearly declares that it is for the Security Council to determine whether enforcement measures are necessary in the event of states manifestly failing to protect their citizens. Consequently, the constraining function of criteria would apply only to Security Council decision making, and the Council already contains mechanisms for guarding against abuse?not least the requirement for a majority vote and the veto.74 The closest historical case we have of Council-sanctioned 'abuse' was its endorsement of the French Operation Turquoise at the end of the Rwandan genocide. The French intervention was widely regarded as 'abusive' because France's primary aim was not humanitarian and the intervention could have done more to save lives.75 The problem in that case was not that France intervened, but that it did not do enough to protect Rwandans. Given that this is the best case we have of Council 'abuse', it seems safe to conclude that the Council's own operating procedures are sufficient guard against potential future 'abuse'. 71 Nicholas J. Wheeler, Saving strangers: humanitarian intervention in international society (Oxford: Oxford University Press, 2000), p. 279; Michael Byers, War law: international law and armed conflict (London: Atlantic Books, 2005), p. 101. 72 For the full range of views on Kosovo, see Albrecht Schnabel and Ramesh Thakur, eds, Kosovo and the challenge of humanitarian intervention: selective indignation, collective action and international citizenship (Tokyo: UN University Press, 2000). 73 For a discussion of borderline legitimacy judgements in relation to intervention, see Ian Clark, Legitimacy in international society (Oxford: Oxford University Press, 2005), pp. 199?205. 74 Thakur, The United Nations, peace and security, p. 260. 75 Wheeler, Saving strangers, pp. 208?41. 629 International Affairs 84:4, 2008 ? 2008 The Author(s). Journal Compilation ? 2008 Blackwell Publishing Ltd/The Royal Institute of International Affairs This content downloaded from 147.251.237.97 on Thu, 09 Feb 2017 10:33:14 UTC All use subject to http://about.jstor.org/terms Alexf. Bellamy The argument that criteria are to be valued because they make it harder to put forward humanitarian justifications for intervention is plausible only in either of two circumstances: first, if criteria are connected to a pathway for legitimizing intervention not authorized by the Security Council?and, by specifying that coercive measures must be authorized by the Security Council, R2P clearly does not offer such a pathway; second, if we believe that the Security Council has become too proactive and requires limitation?an argument not often aired by either academics or governments, and with good reason. It is difficult to see, therefore, what the just cause thresholds and precautionary principles would add to R2P or contribute to decision-making about armed inter vention. As it stands, R2P clearly identifies its scope, its thresholds (genocide, war crimes, crimes against humanity and ethnic cleansing) and the international bodies responsible for discharging the responsibility. Although R2P thresholds are unlikely to generate political will by themselves in relation to particular cases, the endorsement of R2P by the General Assembly and Security Council demonstrates a broad consensus that international society should be engaged in protecting populations from grave harm. Beyond this basic admission of respon sibility, criteria are unlikely to foster consensus on how to act, deter the use of vetoes, provide anything other than a self-serving pathway to the legitimization of intervention not authorized by the Security Council, or?although they may be able to constrain interventions not authorized by the Council?add anything to the Council's mechanisms for preventing 'abuse'. If this analysis is correct, then advocates of R2P should not invest political capital in persuading governments to endorse criteria. Endorsement in the medium term is in any case highly unlikely; but my argument here is that even if the campaign were successful, criteria would not actually improve decision-making about the use of force. Rather than trying to amend R2P by the addition of criteria, there fore, advocates should instead focus on operationalizing the principle as it is. In the final part of this article, I will identify three practical ways in which R2P can make a positive contribution to the problem of military intervention. R2P's contribution to the problem of military intervention This section identifies three important contributions that R2P can make to the problem of military intervention. First, by replacing old debates about 'humani tarian intervention' with a broad continuum of measures aimed first and foremost at preventing genocide and mass atrocities and, if those fail, protecting vulnerable populations, R2P can contribute to reducing the 'moral hazards' associated with intervention. Second, by incorporating political and diplomatic strategies along side legal, economic and military options, R2P points towards holistic strategies of engagement that can overcome the temptation to visualize complex problems in exclusively military terms.76 Third, by turning attention to the protection of civil ians from genocide and mass atrocities, R2P provides a stimulus for new thinking 7 A problem identified by de Waal in 'Darfur and the failure of the responsibility to protect'. 630 International Affairs 84: 4, 2008 ? 2008 The Author(s). Journal Compilation ? 2008 Blackwell Publishing Ltd/The Royal Institute of International Affairs This content downloaded from 147.251.237.97 on Thu, 09 Feb 2017 10:33:14 UTC All use subject to http://about.jstor.org/terms The Responsibility to Protect and the problem of military intervention about the practicalities of protection. If translated 'from words into deeds', these three contributions could deliver better protection to vulnerable populations. Moral hazards In the 2000 Millennium Report, Kofi Annan noted concerns that humanitarian intervention 'might encourage secessionist movements deliberately to provoke governments into committing gross violations of human rights in order to trigger external interventions that would aid their cause'.77 This problem has been described by Alan Kuperman as a 'moral hazard'. It refers to the phenomenon whereby the provision of protection against risk encourages or enables risk-taking behaviour. In this context, the promise of international intervention encourages groups to use violence in order to provoke reprisals and attract international support for their cause.7 For example, Kuperman argues that talk of military intervention in Kosovo in 1998 emboldened the Kosovo Liberation Army, encouraging it to use violence to provoke Serbian reprisals and take an uncompromising political position to secure NATO intervention. The reality is often more tragic: in most circumstances, having inadvertently encouraged violent rebellion by promises of intervention, international society does not deliver on its promise, leaving civilian populations more vulnerable to attack.79 While there is certainly room to quibble about the explanatory power of this moral hazard, and more research is needed, Kuperman has performed an important service in deepening our understanding of the problem identified by Annan in 2000. ? Kuperman proposes four sensible policy measures to reduce the threat of moral hazard. First, there should be no foreign intervention unless a government's actions are 'grossly disproportionate'. Second, external actors should 'expend substantial resources' to persuade states to address the legitimate grievances of non-violent movements. Third, there should be no intervention to force regime change or 'surrender of sovereignty' without robust military deployments to protect civil ians against violent backlashes. Fourth, humanitarian relief should be delivered in ways that minimize benefits to the rebels. l Bearing in mind the fact that the promise of protection is often not backed up with the actual provision of protec tion, we might add a fifth proposal: that governments should only promise to do that which they are actually prepared to deliver on. While Kuperman describes his position as a 'deviation' from R2P, his proposals actually help highlight an impor tant contribution that the principle can make. First of all, in relation to the threshold for intervention, rather than promising intervention in many cases, R2P is reserved for only those cases involving genocide, 77 Kofi Annan, 'We the peoples': the role of the United Nations in the twenty-first century, A/54/2000, 27 March 2000, para. 216. 7 Kuperman, 'The moral hazard'. 79 See Roberto Belloni, 'The tragedy of Darfur and the limits of the "responsibility to protect'", Ethnopolitics 5: 4, 2004, pp. 327-46. 0 A call issued by Ramesh Thakur and Thomas G. Weiss, 'R2P: from idea to norm?and action?', Global Responsibility to Protect 1:1, forthcoming 2009. 1 Kuperman, 'The moral hazard', p. 73. 631 International Affairs 84:4, 2008 ? 2008 The Author(s). Journal Compilation ? 2008 Blackwell Publishing Ltd/The Royal Institute of International Affairs This content downloaded from 147.251.237.97 on Thu, 09 Feb 2017 10:33:14 UTC All use subject to http://about.jstor.org/terms Alexf. Bellamy crimes against humanity, war crimes and ethnic cleansing. Indeed, R2P sets the bar substantially higher than the Security Council, which has proven willing to authorize peace operations under Chapter VII in cases that do not cross the R2P thresholds. 2 According to the Rome Statute of the International Criminal Court, to count as a crime against humanity a particular crime must be committed on a 'widespread or systematic' basis and there must be evidence that links the perpe trators' acts to a state or organizational policy. 3 While Kuperman's threshold of 'grossly disproportionate' action is not as clearly defined as R2P's 'crimes against humanity', it is hard to see how crimes against humanity could be described as anything but grossly disproportionate. Governments are given considerable leeway to use force against rebels before triggering R2P. As such, rather than encour aging rebels, R2P provides precisely the sort of disincentives to rebel action that Kuperman is seeking. Kuperman's other proposals are similarly consistent with R2P. The second calls for governments to encourage their peers to address the legitimate grievances of non-violent groups. This is precisely what is called for by R2P. Recall that in paragraph 138 of the World Summit outcome document, member states pledged to 'encourage and help States to exercise this responsibility [to protect]'. This is a short phrase with heavy import that implies a policy agenda focusing on encour aging (i.e. with incentives) and helping states to build the capacities they need to prevent genocide and mass atrocities. 4 Chief among those capacities would be the capacity to identify and resolve genuine political grievances. Kuperman's third proposal calls for a commitment to the protection of civilians, which is a core aspect of the operationalization of R2P and will be discussed at length below. My main point here is that, far from encouraging rebellions and leaving endan gered populations high and dry, R2P properly understood and translated into practice would act as a damper on moral hazards which have the potential to increase the risks to which civilian populations are exposed. Overcoming the military focus All too often, military intervention is the first port of call in international debates about how to respond to massive humanitarian emergencies, irrespective of the viability or utility of the military option. At least one commentator lays the blame for this squarely at the door of R2P.85 Referring to the international response to the crisis in Darfur, Alex de Waal argued that R2P contributed to a naive obsession 82 For instance, MINUSTAH and UNOCI in Haiti and Cote d'lvoire respectively. See Thomas Weiss, 'The sunset of humanitarian intervention? The responsibility to protect in a unipolar era', Security Dialogue 35: 2, 2004, pp. 135?53; Michael Byers, 'High ground lost on UN's responsibility to protect', Winnipeg Free Press, 18 Sept. 2005, p. B3. 3 See Asia?Pacific Centre for the Responsibility to Protect, 'Burma briefing: Cyclone Nargis and the responsibility to protect', 17 May 2008, p. 7. 4 Luck, 'The responsible sovereign'. Gareth Evans argues that the sorts of capacities needed are those that well functioning states use as a matter of habit. See Gareth Evans, Cooperating for peace: the global agenda for the lggos and beyond (St Leonards, NSW: Allen & Unwin, 1993). 5 De Waal, 'Darfur and the failure of the responsibility to protect'. 632 International Affairs 84: 4, 2008 ? 2008 The Author(s). Journal Compilation ? 2008 Blackwell Publishing Ltd/The Royal Institute of International Affairs This content downloaded from 147.251.237.97 on Thu, 09 Feb 2017 10:33:14 UTC All use subject to http://about.jstor.org/terms The Responsibility to Protect and the problem of military intervention with the deployment of military forces without much serious thinking about what international forces would actually do once deployed or how, exactly, they would contribute to building stable peace in the troubled Sudanese province. 6 This line of thinking created 'wildly inflated' expectations of what UN troops would do?including disarming the Janjaweed and providing protection to both displaced populations and those returning home. 7 According to de Waal, many activists and some political leaders simply assumed that an international force could succeed in the Herculean task of providing physical protection to Darfurian civil ians in the middle of continuing hostilities'. In the crucial period between 2004 and 2006 international actors focused on four issues relating to the deployment of peacekeepers (Who would command them? How many would be deployed? What would their mandate be? Who would pay?) and ignored much more important questions about the strategic purpose ofthe operation. This R2P-inspired focus on military peacekeepers drew attention away from the political process, which?de Waal argued?was a necessary precursor for the deployment of military forces. De Waal is right to argue that in the mindset of diplomats and commentators there remains a pervasive connection between R2P and military intervention. A clearer example of this mindset in action was the recent debate about the interna tional humanitarian response to Cyclone Nargis in Myanmar/Burma. 9 On 2 May 2008 this storm devastated the Irrawaddy delta region, leaving much of the area under water. Around 133,000 people were killed and approximately 2.5 million were directly affected. Despite the massive scale of the humanitarian catastrophe confronting Myanmar/Burma and the government's obvious inability to respond in an effective and timely fashion, the country's military regime permitted only very limited humanitarian access.90 Frustrated by this lack of progress, on 7 May the French foreign minister Bernard Kouchner proposed that the UN Security Council invoke the 'responsibility to protect' to secure the delivery of aid without the consent of the Myanmar/Burma government. This proposal was reiterated by the French ambassador to the UN and repeated by commentators, analysts and politicians, primarily in Europe and North America. Wherever it was aired, however, it tied R2P with proposals for the use of military force. As a model, some pointed to the international relief efforts in Iraqi Kurdistan in 1991, when the UK, France and United States established 'safe havens' to protect Kurds from Saddam's army. One Australian academic went further, pointing to Kosovo as an example and The principal example pointed to by de Waal was a report by the International Crisis Group: ICG, To save Darfur, ICG report 105, 17 March 2006. 7 De Waal, 'Darfur and the failure of the responsibility to protect', p. 1043. De Waal, 'Darfur and the failure of the responsibility to protect', p. 1044. In 2004 de Waal argued that foreign troops could make a 'formidable difference' to the lives of Darfuri civilians, writing that 'The immediate life and death needs of Darfur's people cannot wait for these negotiations to mature. A British brigade could make a formidable difference to the situation. It could escort aid supplies into rebel-held areas, and provide aerial surveillance, logistics and back-up to ceasefire monitoring, helping to give Darfurian villagers the confidence to return to their homes and pick up their lives': Alex de Waal, 'Darfur's deep grievances defy all hopes for an easy solution', Observer, 25 July 2004. 9 The following passage draws on Asia?Pacific Centre for the Responsibility to Protect, 'Burma briefing no. 2: Cyclone Nargis and the responsibility to protect', 17 May 2008. 90 'World fears for plight of Myanmar cyclone victims', New York Times, 13 May 2008. 633 International Affairs 84:4, 2008 ? 2008 The Author(s). Journal Compilation ? 2008 Blackwell Publishing Ltd/The Royal Institute of International Affairs This content downloaded from 147.251.237.97 on Thu, 09 Feb 2017 10:33:14 UTC All use subject to http://about.jstor.org/terms Alex J. Bellamy arguing that western countries should invoke R2P to bypass the Security Council and fight their way into Myanmar/Burma, just as NATO had fought its way into Kosovo.9I Unsurprisingly, Kouchner's proposal was rejected out of hand by China, along with the two other Asian members of the Security Council (Indonesia and Vietnam) and South Africa.92 John Holmes, the UN's Under-Secretary-General for Humanitarian Affairs, described Kouchner's call as unnecessarily confronta tional. The British minister for international development, Douglas Alexander, rejected it as 'incendiary' and Britain's UN ambassador, John Sawers, agreed with the Chinese view that R2P did not apply to natural disasters.93 The problem highlighted by both these cases?Darfur and Myanmar/Burma? seems to be that there is something inherently militaristic about R2P that diverts attention away from non-military solutions. On closer inspection, however, this is a problem produced by serious misunderstandings about what R2P says (and does not say) and about its potential to harness a wide range of measures?military and non-military?to the prevention of genocide and mass atrocities and the protec tion of populations from them. As noted earlier, the use of military intervention is only one of four key commitments associated with R2P as conceived by the World Summit. The other three?especially the commitments to encourage and help states to fulfil their responsibility, and to use a range of non-coercive measures to prevent and protect vulnerable populations?have not attracted the attention they deserve and remain under-conceptualized. Indeed, as the UN Secretary-General's special adviser commented, we have not yet begun to scratch the surface of the immense policy agenda associated with these two commitments.94 A comprehensive global policy agenda based on the mandate handed down by the General Assembly in 2005 would include (but not be limited to) measures to improve the capacity of the UN and regional organizations to provide better early warning of genocide and mass atrocities and better briefings for the UN's decision makers ; measures to help states build the necessary capacity to prevent these crimes; measures to improve international capacity to dispatch teams of peace negotiators with adequate international support; measures to enhance human rights reporting and capacity-building through the UN's Human Rights Council; measures to improve the deterrence capability of the International Criminal Court; a more systematic approach to implementing Kofi Annan's action plan for the preven tion of genocide; the use of peacekeepers as preventers of, as well as reactors to, genocide and mass atrocities; and a comprehensive system for implementing and monitoring targeted sanctions.95 All of this, in addition to the other measures described in this article, is necessary if R2P is to be properly operationalized. If it were operationalized in this way, it is not difficult to see how it would actually 91 See Ivo Daalder and Paul Stares, 'The UN's responsibility to protect', International Herald Tribune, 13 May 2008. The 'Kosovo proposal' was set out in Andrew O'Neil, 'Kosovo aid the model', The Australian, 14 May 2008. 92 See Security Council report, Update report: Myanmar, no. 4, 14 May 2008. 93 Julian Borger and Ian MacKinnon, 'Bypass junta's permission for aid, US and France urge', Guardian, 9 May 2008. Britain later backtracked somewhat, indicating that it would welcome 'discussion' of the responsibility to protect: 'World fears for plight of Myanmar cyclone victims', New York Times, 13 May 2008. 94 Luck, 'The responsible sovereign'. 95 These, and other, measures are discussed in detail in Bellamy, Responsibility to protect. 634 International Affairs 84: 4, 2008 ? 2008 The Author(s). Journal Compilation ? 2008 Blackwell Publishing Ltd/The Royal Institute of International Affairs This content downloaded from 147.251.237.97 on Thu, 09 Feb 2017 10:33:14 UTC All use subject to http://about.jstor.org/terms The Responsibility to Protect and the problem of military intervention militate against the tendency to focus exclusively on the use of military force, and instead place the use of force within a broader spectrum of measures that can be used to prevent genocide and mass atrocities and to protect vulnerable popula tions. Operationalizing protection The third contribution made by R2P is to foreground the need for practical thinking about how international peacekeepers should go about protecting civil ians. Although questions about legality, legitimacy and political will are important, the ultimate test in the direst of situations is whether international engagement succeeds in protecting vulnerable populations. Questions about the way a peace operation is organized, configured, tasked and equipped are just as important as broader political and legal questions when it comes to protecting vulnerable popula tions. It should be recalled that the UN's Independent Inquiry on the Genocide in Rwanda maintained that 'a force numbering 2,500 [UNAMIR's strength at the time of the genocide] should have been able to stop or at least limit massacres of the kind which began in Rwanda' at the start of the genocide.9 Despite this, the question of how best to protect civilians from genocide and mass atrocities has received comparatively little attention. Indeed, there is still no military doctrine that provides guidance on how peacekeepers should go about protecting vulner able citizens. By foregrounding the protection of potential victims, R2P provides an important impetus for developing doctrine in this area and translating lessons learned into action. The development of R2P as an international principle has been accompanied by a transformation of the place of civilian protection in peace operations and, as noted earlier, the Security Council invoked R2P in relation to the UN?AU hybrid mission for Darfur (Resolution 1706). Traditionally, it was thought that peace keepers should remain impartial and neutral and not be proactive in the protection of civilians. Although peacekeeping operations sometimes contained human rights components, only very infrequently was the protection of civilians considered a core part of the peacekeeper's mandate.97 The Security Council has begun to take heed of R2P in its mandating of peace operations in two ways. Today's peace operations tend to be larger and therefore better able to protect civilians than their predecessors. The UN's missions in the DRC, Sudan and Darfur are all mandated to comprise in excess of 20,000 peacekeepers. Furthermore, a combination of better coordination between the Security Council and troop-contributing nations, the UN's standby forces arrangements, and closer cooperation between the UN and regional organizations has seen a progressive decline in the gap between the number of troops mandated by the Security Council and the number actually 9 Independent Commission, Report of the independent inquiry into the actions of the United Nations during the 1994 genocide in Rwanda, 12 Dec. 1999, p. 2. 97 See K. Mansson, 'Integration of human rights in peace operations: is there an ideal model?', International Peace keeping 13: 4, 2006, pp. 547?63. 635 International Affairs 84:4, 2008 ? 2008 The Author(s). Journal Compilation ? 2008 Blackwell Publishing Ltd/The Royal Institute of International Affairs This content downloaded from 147.251.237.97 on Thu, 09 Feb 2017 10:33:14 UTC All use subject to http://about.jstor.org/terms Alexf. Bellamy deployed?the slow deployment of UNAMID notwithstanding.9 Despite progress in this area, it still takes 18 months on average to deploy a peace operation fully, with significant negative consequences for endangered civilians.99 Moreover, the Security Council has begun to create mandates for the protec tion of civilians more frequently, and has gradually relaxed the early restrictions it imposed on such mandates.100 Typically, the Council has demonstrated a prefer ence for limiting the scope of civilian protection mandates by attaching caveats. Examples of these limits can be found in the mandates for the missions in Sudan (UNMIS), Liberia (UNMIL) and Cote d'lvoire (UNOCI). UNMIS was mandated 'to facilitate and coordinate, within its capabilities and in its areas of deployment, the voluntary return of refugees and internally displaced persons, and humani tarian assistance, inter alia, by helping to establish the necessary security'.101 The civilian protection mandates handed down to UNMIL and UNOCI were identi cal.102 By contrast, however, MONUC in the DRC was originally given a narrow civilian protection mandate which was gradually extended over time. MONUC is currently mandated to 'ensure the protection of civilians, including humanitarian personnel, under irnrninent threat of physical violence' without any limiting clause.103 The UN's mission in Burundi was also given a wide protection mandate from the outset.104 Thanks in large part to the Security Council's interest in civilian protection (into which its affirmation of R2P was incorporated) and the pioneering work of researchers such as Victoria Holt and Tobias Berkman, we have a relatively comprehensive understanding of what the protection of civilians by peacekeepers entails in practice, though there remains little by way of doctrinal guidance.105 In short, it entails 'coercive protection'?the positioning of military forces between the civilian population and those who threaten them.10 This may involve military measures to defeat and eliminate armed groups that threaten civilians. Since 2002, for instance, the UN's standing rules of engagement for peace operations have authorized the use of force 'to defend any civilian person who is in need of protection'.107 Sometimes, coercive protection may involve measures short of force, such as erecting military barriers around civilian populations and the gradual 9 Briefing by Jean-Marie Guehenno, Under-Secretary General for Peacekeeping Affairs to the UN Security Counci). 99 Citizens for Global Solutions, 'United Nations Emergency Peace Service: one step towards effective genocide prevention' (New York: 2008), p. 68. 100 On the links between this mandating practice and R2P, see Breau, 'The impact of responsibility to protect on peacekeeping', esp. pp. 450?2. 101 UN Security Council Resolution 1590, 24 March 2005. 102 UN Security Council Resolutions 1509, 19 Sept. 2003, and 1528, 27 Feb. 2004, respectively. 103 UN Security Council Resolution 1565, 1 Oct. 2004. For a discussion see K. Mansson, 'Use of force and civilian protection: peace operations in the Congo', International Peacekeeping 12: 4, 2005, pp. 503?519. 104 Holt and Berkman, The impossible mandate?, pp. 201-224. 105 See Victoria K. Holt, The responsibility to protect: considering the operational capacity for civilian protection (Washington DC: Henry L. Stimson Center, 2005); Holt and Berkman, The impossible mandate?. 106 Thomas G. Weiss, 'The humanitarian impulse', in David M. Malone, ed., The UN Security Council:from the Cold War to the 21st century (Boulder, CO: Lynne Rienner, 2004), p. 48. 107 D. S. Blocq, 'The fog of UN peacekeeping: ethical issues regarding the use of force to protect civilians in UN operations', fournal of Military Ethics 5:3, 2006, p. 205. 636 International Affairs 84: 4, 2008 ? 2008 The Author(s). Journal Compilation ? 2008 Blackwell Publishing Ltd/The Royal Institute of International Affairs This content downloaded from 147.251.237.97 on Thu, 09 Feb 2017 10:33:14 UTC All use subject to http://about.jstor.org/terms The Responsibility to Protect and the problem of military intervention removal of threats through negotiated (and sometimes coerced) disarmament.10 In the absence of military doctrine, however, we lack a clear understanding of how these tasks should be accomplished. The final version of the UN's capstone doctrine for peace operations (rebadged 'principles and guidelines' for political reasons) limited itself to simply observing that 'most. . . peacekeeping operations are now mandated by the Security Council to protect civilians under imminent threat' and noting that this task requires 'coordination with the UN's civilian agencies and NGOs'.109 This raises difficult questions about the relative impor tance of civilian protection and other important principles of peacekeeping such as consent, impartiality and minimum force.110 Draft UN training modules report edly insist that these other principles do not justify inactivity in the face of atroci ties, but do not provide guidance on how these concerns should be reconciled.111 For more detailed guidance we have to make do with learning lessons from current and past missions?at least for the time being. One ofthe most important examples of coercive protection was the adoption of a much more robust posture by MONUC in eastern DRC. In 2005 MONUC began a process of compulsory disarmament in Ituri district around Bunia, disarming around 15,000 combatants by June. Some groups opposed forcible disarmament, and in February 2005 fighters from the Nationalist and Integrationist Front (FNI) attacked and killed nine Bangladeshi peacekeepers. In response, Nepalese, Pakistani and South African peacekeepers, supported by Indian attack helicopters, pursued the FNI and killed between 50 and 60 belligerents, neutralizing the threat they represented to civilians.112 For its part, the Security Council further strengthened MONUC's mandate and explicitly authorized 'cordon-and-search' operations against 'illegal armed groups' thought to be threatening the civilian population.113 MONUC's Pakistani contingent also adopted a robust civilian protection posture in South Kivu. Alongside Guatemalan special forces, the Pakistanis rooted out Hutu Forces Democratiques de Liberation du Rwanda (FDLR), members of a militia associated with the 1994 Rwandan genocide and subsequent abuse of civil ians in the DRC. In October 2005 MONUC issued a disarmament ultimatum to the FDLR; when the rebels refused to cooperate, it used helicopter gunships to destroy between 13 and 16 camps. Although the mission succeeded in weakening the FDLR and restricting its freedom of movement, it neither destroyed the militia nor forced it to disarm.114 As well as applying coercion to the perpetrators of attacks on the civilian population, the Pakistanis also used innovative methods to protect civilians. For example, they organized a community watch in villages in Walungu territory and taught its members to bang pots and blow whistles 108 Holt and Berkman, The impossible mandate?, p. 52. 109 UN Department of Peacekeeping Operations, United Nations peacekeeping operations: principles and guidelines, 18 Jan. 2008, para. 42. 110 The so-called holy trinity. See Alex J. Bellamy, Paul D. Williams and Stuart Griffin, Understanding peacekeeping (Cambridge: Polity, 2004). 111 Holt and Berkman, The impossible mandate?, p. 190. 112 Holt and Berkman, The impossible mandate?, p. 165. 113 UN Security Council Resolution 1592, March 2005. 114 Holt and Berkman, The impossible mandate?, pp. 166?7. 637 International Affairs 84:4, 2008 ? 2008 The Author(s). Journal Compilation ? 2008 Blackwell Publishing Ltd/The Royal Institute of International Affairs This content downloaded from 147.251.237.97 on Thu, 09 Feb 2017 10:33:14 UTC All use subject to http://about.jstor.org/terms Alex J. Bellamy when danger was imminent. Pakistani peacekeepers were kept on high alert in the vicinity to respond to such warnings. UN peacekeepers have therefore begun to operationalize R2P in a way that incorporates the protection of civilians into their core business. To date, while the Security Council hands down mandates for the protection of civilians with greater regularity and fewer restrictions, operationalization has largely relied on impro visation in the field, as the example of MONUC in Ituri district demonstrates. Despite the evident limitations of this approach, the focus on civilian protection has contributed to a marked decline in the overall number of civilians killed in sub-Saharan African wars since 2003.II5 R2P can make an important contribu tion to the further development of civilian protection by providing the core ratio nale for such operations, marshalling the political will necessary to establish peace operations and equip them with civilian protection mandates, and emphasizing the need for long-term and multidimensional approaches to civilian protection which incorporate the prevention of genocide, war crimes, crimes against humanity and ethnic cleansing, and the rebuilding of states and societies in the wake of such suffering. Conclusion The first step along the road of translating R2P from words into deeds is a proper understanding of what the principle does and does not say about the use of non-consensual military force for humanitarian purposes. As agreed by world leaders in 2005, R2P does not countenance non-consensual military force without the authorization of the Security Council and does not set out criteria for the use of force beyond the four threshold crimes and the idea that the Council should assume responsibility in cases where the host state is 'manifestly failing' to protect. Many advocates of R2P lament the loss of key elements of the recommenda tions put forward by the ICISS, and it is appropriate and legitimate to call for amendments to the R2P principle in the future to accommodate some of these recommendations. But it is important to distinguish the R2P principle from ideas, concepts and recommendations put forth by the ICISS, various governments, and individuals such as Francis Deng. Continuing confusion about what R2P has to say about military intervention helps neither the principle itself nor those charged with making difficult decisions about how best to prevent genocide and mass atrocities and protect potential victims. It is also important not to overstate the capacity of those 'lost recommen dations'?especially the proposed criteria to guide decisions about the use of force?to resolve policy dilemmas or forge consensus in actual cases. Criteria enjoyed little international support, would not generate additional political will, would in all likelihood not constrain the use of the veto (if past practice is a good guide) and would not provide an avenue for legitimately bypassing the Security Council. For these reasons, although it is legitimate to press UN member states to 115 Human Security Report Project, Human Security Brief 2007, pp. 22?30. 638 International Affairs 84: 4, 2008 ? 2008 The Author(s). Journal Compilation ? 2008 Blackwell Publishing Ltd/The Royal Institute of International Affairs This content downloaded from 147.251.237.97 on Thu, 09 Feb 2017 10:33:14 UTC All use subject to http://about.jstor.org/terms The Responsibility to Protect and the problem of military intervention adopt criteria, the energies of R2P advocates would be better spent elsewhere. Not least, they should focus on identifying and operationalizing those aspects of R2P that could make an important difference to the way international society concep tualizes and practises military intervention: mitigating moral hazards, building multifaceted engagement strategies that reduced the tendency to focus exclusively on military solutions, and developing the doctrine and capacity needed to enable peacekeepers to protect civilians better once deployed. Focusing on these aspects would help deliver on Ban Ki-moon's promise to translate R2P 'from words into deeds' and lay the foundations for a deepening consensus on the new principle. 639 International Affairs 84:4, 2008 ? 2008 The Author(s). Journal Compilation ? 2008 Blackwell Publishing Ltd/The Royal Institute of International Affairs This content downloaded from 147.251.237.97 on Thu, 09 Feb 2017 10:33:14 UTC All use subject to http://about.jstor.org/terms