Abstract
This article engages with the debate on the efficacy of the Responsibility to Protect (R2P) in the wake of the Arab Spring by articulating a defence of its role in preventing the commission, escalation, or recurrence of atrocity crimes. Taking as its starting point the claim by UN Secretary-General (UNSG) Ban Ki-moon that prevention remains the most important aspect of the principle of R2P, the article illustrates the extent to which prevention is embedded in R2P, the means by which it can be leveraged, and the obstacles to its operationalisation. The first section outlines why and how the prevention of the four crimes identified in the 2005 World Summit Outcome Document became so important to UN member states. The second section analyses efforts to implement the commitment to prevention within the UN, regional organisations, and individual states. The final section offers an explanation for why prevention is in fact a controversial practice – despite the universal rhetorical commitment to its prioritisation – and advances a series of steps which might be undertaken to advance it.
Introduction
When analysing the impact of the Responsibility to Protect (R2P) on events related to the Arab Spring, scholars focus largely on the principle’s capacity (or incapacity) to catalyse a timely and decisive response from the international community to situations in the Middle East. This has manifested most commonly in criticisms of the international response to the crisis in Syria, such as those articulated by Aidan Hehir in this special issue. This article, by contrast, takes as its starting point the claim made by UN Secretary-General (UNSG) Ban Ki-moon, that prevention remains the most important aspect of the principle of R2P (Ban Ki-moon, 2012). 1 R2P is much more than just a means by which the international community can react – militarily or otherwise – to the commission of atrocity crimes; a central component of R2P since its inception in 2001 has been the ‘Responsibility to Prevent’ (ICISS, 2001a: xi). The preventive dimension of R2P has the advantage of being a means to forestall atrocity crimes and to reduce the costs of international engagement. It has also commanded greater political support from states than reactive intervention, and thus helps to build a firmer foundation for further evolution of the principle. The central purposes of this article are therefore twofold: to temper the negative assessments of R2P’s impact by emphasising the potential inherent in the responsibility to prevent; and to frame future discussions on how to better operationalise prevention. As the UNSG lamented in his 2014 Report to the General Assembly, there is still too little concrete commitment to prevention, and resources remain skewed towards crisis response (Ban Ki-moon, 2014: para 73).
The first section below explains why the prevention of the four crimes and violations identified in the 2005 World Summit Outcome Document (UN, 2005) became so important to both member states and the UN Secretariat, and outlines subsequent efforts to implement the commitment to prevention within the UN, regional organisations, and within individual states. The second section focuses on why prevention is in fact a controversial practice – despite the almost universal rhetorical commitment to its prioritisation – and suggests a series of measures that might be undertaken to advance it.
Prioritising prevention
It has long been clear that preventing conflict and atrocity crimes is preferable to responding to them. This is evident both in terms of relative human suffering and in terms of the comparative financial and political costs involved (Carnegie Commission on Preventing Deadly Conflict, 1997; Miall, 2004). Prevention can take many forms, from long-term structural assistance, to more short-term, pre-emptive action designed to forestall what is perceived to be an impending catastrophe (Fein, 2009: 321–322; ICISS, 2001b: 27; UN Office on Genocide Prevention and the Responsibility to Protect, 2014: 2). Both sets of measures, however, are characterised by the same underlying logic: that prevention is both possible – and hence violence is not inevitable – and more conducive to international peace and security than post-conflict reaction.
Among the key constitutive elements of the principle of R2P, prevention has been singled out by many as the most critical (Adams, 2013: 1; Evans, 2008: 79). Both scholars and policy-makers have argued that it is advantageous to act to prevent atrocity crimes from being committed, given the barriers and high costs of reacting once they are already underway (Bellamy, 2011; Global Responsibility to Protect, 2011; Stamnes, 2008).
2
In the blunt words of the ICISS:
In Kosovo, almost any kind of preventive activity – whether it involved more effective preventive diplomacy, or the earlier and sharper application of coercive preventive measures like the credible threat of ground-level military action – would have had to be cheaper than the $46 billion the international community is estimated to have committed…in fighting the war and following up with peace-keeping and reconstruction (2001a: 71).
From a more conceptual standpoint, the significance of the responsibility to prevent derives from two considerations: the logical priority of prevention within the original framework set out by ICISS (i.e. preventive efforts should be pursued before the responsibility to react and the responsibility to rebuild are considered) (2001a: xi); and the capacity of preventive efforts to reinforce the entire spectrum of actions associated with R2P – a point I return to later.
When it elaborated on the ‘responsibility to prevent’, ICISS outlined three essential conditions for effective prevention: knowledge of the situation at risk (which the Commissioners believed required a more effective early warning system); an understanding of the policy measures available (what the Commissioners referred to as the ‘preventive toolbox’); and sufficient political will (which ICISS argued depended upon effective leadership and the power of ‘good arguments’) (ICISS, 2001a: 20). ICISS also identified some of the challenges associated with development of a strengthened preventive capacity, which I will address in more detail below. These include the lack of funds available for preventive efforts, the dangers of exacerbating domestic tensions through increased international involvement, and the difficulty of mobilising political will before a crisis becomes apparent (ICISS, 2001a: 19–29). In the years following the publication of the ICISS recommendations, some of these challenges served to marginalise the preventive dimension of R2P. Furthermore, within academic debates and policy circles, more attention was initially paid to the responsibility to react (particularly when it involved the use of military force).
Developments after the endorsement of the 2005 Summit Outcome Document, however, have served to re-establish the prominent place of prevention on the R2P agenda. Within individual states, the prevention of atrocity crimes is gaining momentum, exemplified by initiatives such as the Atrocities Prevention Board (APB) in the United States. 3 Situated within the National Security Council and run directly out of the White House, the APB signals that atrocity prevention and response has become a core objective of US foreign policy, and that the prevention of atrocity crimes in particular is becoming a more conscious and consistent focus. A second example is the creation of ‘national focal points’ for R2P within governments, designed to enable or strengthen efforts to improve atrocity crime prevention and response. 4
Regionally and internationally, prevention is also attracting more focused policy attention. Within Africa, for example, the Economic Cooperation for West African States (ECOWAS) has significantly enhanced its early warning infrastructure to confront impending atrocity crime situations. Significant strides have also been made through the International Conference on the Great Lakes Region, with the adoption of national committees for the prevention and punishment of genocide, war crimes, crimes against humanity, and all forms of discrimination. To date, such committees have been established in Kenya, South Sudan, Tanzania, Uganda, and Zambia. Within Latin America, a network has been established for preventing genocide and mass atrocities, co-organised by the governments of Argentina, Bolivia, Brazil, Chile, Colombia, Costa Rica, Ecuador, El Salvador, Guatemala, Honduras, Mexico, Nicaragua, Panama, Paraguay, Peru, Uruguay, and Venezuela.
Finally, within the UN, there is a greater emphasis upon the prevention of conflict and crises – symbolised by UNSG Ban Ki-moon’s designation of 2012 as the ‘year of prevention’ and emphasis on the preventive dimension of the responsibility to protect. More concretely, the UNSG’s 2010 and 2013 reports to the General Assembly focused specifically on prevention and early warning, and in 2014 the UN Joint Office of the Special Advisers on Genocide Prevention and the Responsibility to Protect published the ‘Framework of analysis for atrocity crimes: A tool for prevention’, detailing the means by which risk factors can be identified and addressed prior to the commission of atrocity crimes (UN Office on Genocide Prevention and the Responsibility to Protect, 2014).
This concerted emphasis on prevention within the UN is driven in part by the internal review panel (IRP) established by the UNSG to investigate UN actions during the final stages of the 2009 Sri Lanka conflict. The panel’s report, issued in late 2012, described a ‘systemic failure’ of UN action, extending across the organisation’s various departments and agencies (at both field and headquarters levels), and reaching to individual member states. According to the IRP report, during the final stages of the conflict, when protection challenges were most acute, member states did not hold a single formal meeting on Sri Lanka in the Security Council, General Assembly, or the Human Rights Council (Report of the Secretary-General’s Internal Review Panel on United Nations Action in Sri Lanka, 2012). In response to the panel’s conclusion that these events marked a grave failure of the UN to adequately respond to early warnings of impending crisis, the UNSG initiated the ‘Human Rights Up Front Action Plan’, a system-wide effort to ensure that the UN strengthens the link between early warning and early action, and decisively responds to the risk of or on-going serious violations of international human rights and humanitarian law (UN, 2014).
Three factors help to explain the development of these institutionalised variants of prevention nationally, regionally, and internationally. The first is the particular way in which policy-makers, especially in the West, learned from past instances of atrocity crimes. The lesson of cases such as Rwanda and Srebrenica was that the signs of crisis had been present, but that outside actors had failed to look for them and/or to pay sufficient attention to them (ICISS, 2001a: 21). As the UNSG put it in his first report on the Responsibility to Protect:
[In] each case there were warning signs. Violence of this magnitude takes planning and preparation, as well as a contributing political, social and economic context … the signals of trouble ahead were, time and again, ignored, set aside or minimised by high-level national and international decision makers with competing political agendas. (Ban Ki-moon, 2009: para. 6)
Many in the scholarly community drew a similar lesson, but framed it in terms of the observation that atrocity crimes are not ‘random’ events, but usually reflect a complex interaction of different risk factors over an extended period of time (Asia-Pacific Centre for R2P, 2009; Reike et al., 2015). While various schools of thought on genocide and atrocity crimes put different weight on various explanatory factors, the most commonly cited risks include: the presence of war and armed conflict, which creates both a motive and enabling environment for mass killing; economic and/or social instability and crisis, which can both generate both motives for violence and weaken the capacity of state actors to respond; an exclusionary ideology, which facilitates the creation of group-identities along hierarchical lines; an authoritarian government, in which deference towards leaders and elites erodes normative checks on orders to perpetrate violence; and a history of previous atrocity crimes, which heightens perceptions of grievance and threat (Bellamy, 2012; Chirot and McCauley, 2006; Harff, 2003; Kiernan, 2003; Mann, 2005; Midlarsky, 2005; Shaw, 2007; Staub, 1989; Valentino, 2004; Weitz, 2003). However these factors are arranged and measured, the common message that emerges from their analysis is that atrocity crimes are part of a dynamic continuum, along which outside actors could engage preventively – if only they could mount the political will and capacity to do so.
The second factor explaining the renewed emphasis on prevention was the nature of the political climate preceding and following the 2005 World Summit. As noted by Edward Luck, former Special Advisor to the Secretary-General on the Responsibility to Protect, the concern among many states in the developing world that R2P might be misused by powerful states to justify coercive intervention undertaken for other purposes – a fear fuelled by some of the arguments forwarded for the use of force against Iraq in 2003 – was a prominent feature of the debates in the run-up to the Summit. Accordingly, Luck noted, ‘there was little reason to be optimistic about the prospects for achieving the Summit’s endorsement of R2P’ (2009: 17). Yet, according to Luck, the focus on prevention in Article 139 – assisting states under stress before crises break out – is one of the reasons why the text of the Summit Outcome Document was able to garner unanimous endorsement. More broadly, emphasis on R2P’s non-coercive means was seen by many as a wise strategy for enhancing its political legitimacy. This view also played into the comprehensive implementation plan set out by the UNSG in 2009, which articulated R2P’s now broadly accepted three pillars (Ban Ki-moon, 2009: 8–9).
The third and perhaps least recognised explanation for the prioritisation of prevention was the belief of Ban Ki-moon that preventive action can help to ensure that individual states and international organisations have a better understanding of the dynamics of the particular country in question and how they might be affected by various courses of action – at both an early and a later stage. In other words, early and non-coercive action is valuable not only because it may be less costly, or because it may be perceived as more legitimate, but because it can enhance the effectiveness of all forms of third-party involvement. A key assumption of the UNSG’s report, as Luck explains it, was that ‘early engagement and preventive efforts would provide a better feel for the situation at hand, as well as providing the local actors with a keener sense of the normative and political concerns of some key international actors’ (Luck, 2014). Such ‘in-country efforts’ could therefore ‘lay the political groundwork for the consideration of pillar three responses’ if signs of ‘manifest failure’ began to materialise (Luck, 2010: 182). This particular understanding of the role of prevention is clearly reflected in the passages of the 2009 report that stress a ‘continuum of graduated policy instruments’ across the spectrum of the three pillars, and the need to understand how conditions, circumstances and needs vary from country to country (Ban Ki-moon, 2009: paras 9, 47).
Despite all of these compelling reasons for prioritising prevention, not all analysts or supporters of R2P have fully embraced its prominent place on the agenda for implementation. There are two related concerns driving this unease. The first, most clearly expressed by Rama Mani and Thomas Weiss, is that by connecting implementation of R2P to broad preventive efforts that involve economic and social development, as well as democratisation, R2P loses its distinctiveness. ‘Broadening perspectives away from reaction’, they argue, ‘has opened the floodgates to an overflow of appeals to address too many problems…if R2P means everything, it amounts to nothing’ (Mani and Weiss, 2011: 4). Eli Stamnes also worries about eroding the political utility of the Responsibility to Protect if the principle is attached to ‘the plethora of measures that constitute preventive action’ (2009: 71). The dilemma with prevention, she argues, is that it moves the practice of R2P away from its ‘exclusivity’, which was originally its greatest asset (Stamnes, 2009: 75).
An additional worry, again voiced by Weiss but shared by others, is that by elaborating on the preventive dimension, and mobilising resources and will to implement it, policy-makers will neglect what remains the core raison d’etre of R2P: timely response to unfolding atrocity crimes. ‘[I]t is preposterous to argue that to prevent is the single most important priority’, Weiss contends. ‘[T]he most urgent priority is to react better. Most of the mumbling and stammering about prevention is a superficially attractive but highly unrealistic way to try to pretend that we can finesse the hard issues’. The ‘discourse of prevention’, Weiss concludes, ‘obscures the essence of the most urgent part of the spectrum of responsibility, to protect those caught in the crosshairs of war’ (Weiss, 2007: 104; see also Hehir, 2012: 87–118). If it is true that there is a fixed amount of policy-making time and space, and that more focus on prevention necessarily means less focus on response, then such warnings should be heeded. This author remains unconvinced that the situation is quite so ‘zero-sum’. Nevertheless, the risk remains, as I have argued elsewhere, that excessive focus on prevention, and particularly the first pillar of R2P, will exacerbate the ‘paucity of thinking’ about the range of tools available under Pillar III (not just military ones) and the need for policy-makers to understand the conditions under which they are likely to be effective (Welsh, 2011: 261).
The challenges of prevention
Beyond the concerns about the place of prevention within the broader principle of R2P are challenges related to the practice of prevention itself. There is a growing list of cases where (relatively) early action has contributed to forestalling an outbreak of escalation of widespread violence. Prior to the establishment of R2P, the United Nations Preventive Deployment mission stationed on the Macedonian/Serbian border from 1992 to 1999 was a relatively incontrovertible success (Stamnes, 2005). In terms of the post-R2P era, Kenya, Cote d’Ivoire, Guinea, Kyrgyzstan, and Mali are cited as further examples of successful preventive action (Ban Ki-moon, 2009; Luck, 2014). In the case of Kenya, coordinated international diplomacy, largely orchestrated through the UN but implemented through regional instruments, quickly focused on the unfolding crisis which followed the Presidential election in December 2007 and the violence that erupted. Francis Deng, then UN Special Adviser on the Prevention of Genocide, warned the parties to remember their commitments to R2P and specifically not to engage in hate speech (Hehir, 2012: 131). Similarly, in Guinea in 2009–2010, concerted and coordinated preventive efforts by local, regional, and international actors helped to avert a recurrence of atrocity crimes following the massacre in September 2009, during which 156 peaceful protestors were killed by government forces during an opposition rally in Conakry. (A UN Commission of Inquiry later determined that the acts rose to the level of crimes against humanity.) The fact that Guinea did not experience further atrocity crimes during the lead up to the November 2010 Presidential election, and that the situation did not deteriorate into broader ethnic conflict that risked destabilising the region, can be partly traced to effective preventive action taken during this time, particularly by the Special Representative of the Secretary-General, working in collaboration with officials from the Economic Community of West African States (ECOWAS). Their targeted efforts included preventive diplomacy, arms embargos, travels bans, and threats of International Criminal Court (ICC) prosecutions. Both Kenya and Guinea provide compelling examples of how a variety of actors, using an array of primarily non-coercive measures, can operationalise prevention. In both cases it is notable that the preventive tools were largely coordinated and implemented through the UN Secretariat, framed in terms of R2P, and undertaken in collaboration with regional actors. This points towards the criteria required for future effective preventive action.
Although the preventive dimension of R2P may have garnered greater political consensus than more coercive aspects of response, and – as detailed above – there have been notable ‘successes’, this does not mean its implementation will be easy or straightforward. As I have stressed previously, effective prevention measures can in some cases be highly intrusive, and thus may be just as likely to fuel state sensitivities about the erosion of sovereignty or to raise questions about the legitimate authority for preventive measures (Hehir, 2012: 261; Welsh, 2010: 153). Indeed, both of these concerns limited the preventive mandate of the Peacebuilding Commission when it was created following the 2005 World Summit (Berdal, 2009: Chapter 3). In what follows, I lay out more systematically some of the conceptual, political, and institutional obstacles to operationalising the ‘responsibility to prevent’. In so doing, I also suggest ways in which such barriers might be addressed.
Conceptual gaps
The underinvestment by policy-makers in prevention, compared to the investment in response, cannot be attributed solely to a lack of political will. There is also a lingering belief that too little is known about what actually ‘works’. This is particularly true of so-called structural prevention, which is aimed at addressing deeper potential drivers of atrocity crimes.
There are at least three reasons for our gaps in knowledge. 5 First, atrocity crimes are still (thankfully) relatively rare. They are high impact yet low probability events, making it challenging to fully substantiate arguments and claims about what preventive strategies are most effective. This ‘small N’ nature of the problem has bred a certain degree of ambiguity about what, exactly, preventive action is aimed at (Straus, 2011). The crimes and violations specified by R2P are treated as a set-piece, but in reality each of them has their own trajectory and manifestations, further complicating the task of creating a unified preventive approach. In the case of war crimes, there is the additional challenge that R2P was clearly not intended to address every instance of such crimes – only those that involve systematic and widespread attacks aimed at particular populations. 6 In light of these issues, efforts to implement the ‘responsibility to prevent’ should clarify the precise acts which third parties seek to forestall, and to segment the different kinds of situations that can give rise to them – each of which vary in terms of the groups targeted, the means used, and the motives for violence. Drawing from past experience, a preliminary segmentation would include: attacks on civilian groups associated with rebels; attacks on civilians associated with a government; attacks on a segment of the population associated with a rival in a political conflict; attacks against a segment of the population driven by extreme ideology; intense inter-communal violence; or post-regime change attacks on populations to consolidate power. 7
This (non-exhaustive) list helps to highlight the second reason for knowledge gaps: the continuing tendency to equate atrocity crime prevention with the prevention of armed conflict. The scenarios above indicate that some atrocity crimes can be, and have been, committed outside the context of armed conflict. The contemporary case of crimes against humanity in North Korea, documented by the UN Human Rights Council’s Commission of Inquiry Report (2014: 86), is a case in point. But so too are cases related to the Arab Spring – particularly Libya and Syria – in which atrocity crimes were committed or threatened before the existence of a formal armed conflict. 8 Yet, atrocity crime prevention, not only within particular organisations but also in scholarly writings, is still commonly approached through the lens of conflict prevention and the related field of conflict resolution. This reliance on a conflict prevention perspective was a prominent feature of the original ICISS report, which drew heavily on the framework and tools set out in the report of the Carnegie Commission on Preventing Deadly Conflict (1997), and has continued in key policy initiatives, including the report of the Genocide Prevention Task Force (Albright and Cohen, 2008) and former UNSG Kofi Annan’s Five Point Action Plan on the Prevention of Genocide set out in his April 2004 address to the Human Rights Commission (Annan, 2006).
In contrast to numerous scholarly works on the prevention of conflict, there has been comparatively less attention devoted to the prevention of the four specific crimes and violations covered by the principle of R2P. Genocide is the one exception, as it has generated several articles and book-length treatments of its causes and dynamics. Empirical studies of cases of genocide have generated two important findings for prevention: first, that episodes of genocide frequently occur in the context of violent conflict (Bellamy and McLoughlin, 2009; Human Security Brief, 2008; Krain, 1997; Straus, 2012); and second, that factors often identified as root causes of genocide are similar to those identified as root causes of conflict (Bellamy, 2011; Goldstone et al., 2010; Held, 2009). Nonetheless, as scholars of R2P have noted, it cannot be assumed that efforts to prevent or resolve conflict will simultaneously reduce the likelihood of atrocity crimes (Bellamy, 2011; Reike et al., 2015; Stamnes, 2009). While a large majority of the episodes of mass killing observed since 1945 occurred within the context of armed conflict, a third of them did not (Bellamy, 2011: Appendix I). There is also evidence that ‘non-conflict’ or peacetime deaths – particularly those falling below the full scale of genocide – are becoming relatively more significant; according to a 2011 report, The Global Burden of Armed Violence 2011, while on average just over half a million people now die annually in violent circumstances, just 10% of those die in formal conflict settings (Krause et al., 2011). As a result of these empirical observations, any framework for the prevention of atrocity crimes needs to incorporate a longer-term approach, aimed at reducing the risk of armed conflict, and a shorter-term approach with two prongs: one aimed at preventing peacetime atrocities and the other aimed at preventing the commission of atrocity crimes by those engaged in armed conflict.
The third and final conceptual issue, recently addressed by genocide scholar Scott Straus, is the lack of a clear method for understanding why some countries spiral into systematic and widespread violence, while others do not (Straus, 2014). Although the past decade has seen broad convergence on the key macro-level risk factors of genocide and other atrocity crimes (Ulfelder and Valentino, 2008; UN Office on Genocide Prevention and the Responsibility to Protect, 2014), he rightly points to two remaining shortcomings. First, we need a better understanding of why cases that current models or ‘watch-lists’ would predict should result in genocide and other atrocity crimes, do not end up moving down that path. ‘Even in some of the most sophisticated forecasting’, Straus explains, ‘the ratio between countries at risk of genocide and mass atrocity and countries that experience those outcomes in a given year is greater than 30:1’ (Straus, 2014: 1). So what are the more specific dynamics that lead both away from and toward the commission of atrocity crimes? The second shortcoming is the lack of a capacity to move beyond assessments of likelihood, to anticipating when atrocity crimes will be committed. In other words, what particular events or developments shift a country from being at risk of atrocity, to such crimes actually beginning?
To address the limitations of current approaches, analysts need to build a temporal dimension into their atrocity prevention framework that distinguishes between: (a) general and more structural risk factors (such as those outlined above); (b) early signs of upheaval and mobilisation that suggest increased risk of atrocity crimes (such as the propagation of an exclusionary ideology or the organisation of armed groups); and (c) late signs of imminent emergency (such as incitement to commit violence, the arming of militias, and small-scale attacks on segments of the population) (Reike et al., 2015).
In addition, researchers could follow Straus’ lead and pursue two more specific lines of inquiry. The first is to identify a set of ‘inhibitors’ to atrocity crimes – attitudes, incentives, institutions, mechanisms, and structures that foster moderation and work against violence 9 – and analyse their presence or absence across a variety of cases. 10 The second is to identify a set of ‘triggers’ that commonly ignite underlying risk factors to produce situations where atrocity crimes are committed or imminent. Some of these triggers (such as contested elections) can be anticipated, whereas others (such as the assassination of leading officials or significant military advances by an insurgent organisation) may be less amenable to prediction. Although the introduction of inhibitors and triggers into the analysis will not create a bullet-proof model for prevention, they could greatly enhance scholars’ understanding of when to engage preventively, and on what basis, and provide policy-makers with the key ingredients for an atrocity prevention strategy. Clearly this would constitute a significant advancement for this core goal of R2P.
Political barriers
Even if all of the conceptual issues around atrocity prevention were resolved, there remain significant political barriers to timely and effective prevention. Since these have been well rehearsed in the literature, and feature in some of the contributions to this Special Issue (see Morris and Hehir in particular), they will not be elaborated in great detail here. Let’s begin with the most obvious obstacle: strong political interests on the part of powerful states (whether regional or extra-regional) that work against early action to address risk factors in their ‘client’ states or regimes. This factor was clearly present in the case of Syria as the political crisis mounted during the summer of 2011, and continued to put a brake on external action as the civil war featured more frequent and more brutal instances of atrocity crimes. Arguably, it also played a role in the more recent instance of South Sudan, where the commitment of some western actors to the governing elites of the newest member of international society entailed an unwillingness to directly confront these local actors over escalating inter-communal hostility and serious human rights violations.
An additional political barrier is the unwillingness of states under stress, or their close allies, to subject their ‘situation’ to international discussion or engagement. This reluctance is clearly manifest through resistance to placing a certain country situation on the agenda of the Security Council, which is usually justified on the grounds that the signs of instability or crisis are an internal matter for that state, and not a legitimate concern of the international community. As a result, situations in which there was a risk of widespread atrocity crimes, such as Kenya following post-election violence in late 2007, or where such crimes were underway, such as Sri Lanka in the closing phase of its civil war in 2009, were never discussed on the formal agenda of the Security Council. The same dynamic has applied to other less dramatic cases, such as Burundi and Myanmar, where risk factors related to inter-communal violence have intensified, but where interested states and international organisations have proven reluctant to challenge the dominant narrative that the country is making progress or on a more positive path. This sensitivity extends beyond the question of what should be on the Council’s agenda, to the seemingly more straightforward task of assessing country situations. Already in 1985, then UNSG Javier Pérez de Cuéllar called for the UN to develop a ‘procedure to keep the world under continuing survey in order to detect nascent causes of tension’ (UN, 1985). Yet, the uneven trajectory of one such procedure, the Horizon Scanning briefings of the UN’s Department of Political Affairs, demonstrates just how politically sensitive such preventive efforts can be. Despite being welcomed by states when they were introduced in late 2010, they have gradually declined in frequency and popularity (Security Council Report, 2014: 19–20).
One antidote to these problems is to make a base level of outside scrutiny normal or routine for all Member States – much in the same way that the Universal Periodic Review (UPR) of the Human Rights Council touches every member state. This would help to overcome the apparent stigma of having activated the international community’s ‘concern’. Given that an entirely new system for periodic review of atrocity crime risks is unlikely to be realised in the near future, a more viable strategy could involve two steps: (1) building some reporting and consideration of atrocity crime risks into existing review processes; and (2) consciously encouraging states to request external assistance to counter-act negative dynamics and securing credible and legitimate third-party actors to help in that task.
Institutional constraints
The discussion above hints at another obstacle to effective prevention – namely, institutional constraints within the UN system. 11 To begin, it is worth remembering that one of the UN’s great virtues – its global footprint – also carries with it a high degree of decentralisation that is not always conducive to the effective implementation of principles such as R2P. Despite all of the ink spilled over the need for the UN to improve its early warning capacity, the body has continued to lack a coordinated system for gathering and assessing situations of emerging risk with common criteria. This weakness is finally being addressed head on, through the Human Rights Up Front Action Plan, but the Plan’s impact will depend upon two key conditions: first, the ability to engage field staff in the assessment of atrocity crime risks on a regular and systematic basis (as opposed to relying on personal initiative or motivation); and second, clear and robust mechanisms for moving the integrated assessment through the system to the senior leadership team, which must then be prepared to act on it.
That second condition can be compromised by another institutional constraint. The UN’s need to retain access for its officials and in-country field presences can lead to micro-level decisions that effectively turn a blind eye to worrying trends or particular actors that could become perpetrators of atrocity crimes. In practice, this means that information may be neglected or suppressed. Human Rights Up Front is also designed to counter-act this possibility, by rethinking the way that UN staff is selected, trained, and incentivised, and by offering stronger political support from Headquarters for difficult stances taken at field level (UN, 2014). Much will depend on whether the organisation’s recruitment and performance management system can deliver the kind of culture change needed to put serious human rights violations ‘front and centre’. The UN will also need to address the imbalances in its current field presences, especially in non-mission-settings, where there are still too few experts to monitor serious violations, assess which populations are vulnerable and why, and engage with national authorities to address signs of crisis.
A final set of institutional constraints accompanies those situations where a UN peacekeeping or stabilisation mission is on the ground, and where there is a heightened risk of atrocity crimes. Here, the ability to prevent the commission or escalation of such acts is often compromised by the laborious process involved in crafting mission mandates (which can also make it difficult to adjust to changing circumstances on the ground), the operational caveats that are frequently insisted upon by troop contributing countries (TCCs), the sometimes awkward relationship between the mission and humanitarian agencies, the lack of coordination between different international and regional actors (which can enable local ‘spoilers’ to play such actors off against one another), and – last but not least – the perennial problem of financing for peacekeeping and stabilisation. The recently completed review of UN peace operations has provided a set of concrete recommendations for how to both put peacekeeping on a more sustainable foundation and improve on the ability of missions to deliver on the robust protection of civilians’ mandates, but the burden is now very much on member states to provide the political and financial support to realise these reforms (High Level Panel on UN Peace Operations, 2015: viii).
All this said, the UN remains a crucial and indispensable part of the global infrastructure – both normative and operational – for implementing the preventive dimension of R2P (Luck, 2014). On the normative side, the General Assembly has unique convening power for elaborating on both the need for and the ‘how to’ of prevention, and for galvanising resources to assist states under stress. The Human Rights Council, human rights treaty bodies, and special procedures mandate holders, are increasingly being acknowledged as valuable mechanisms for encouraging states to meet their protection responsibilities under Pillar I and for identifying cases of escalating risks (Ban Ki-moon, 2015). Individual senior officials, starting from the UNSG and extending to the High Commissioner for Human Rights and the Special Advisers on the Prevention of Genocide and the Responsibility to Protect, can serve as independent voices on atrocity crime risks and draw attention to cases in a way that individual states are unable or unwilling to do. And on the operational side of the ledger, the UN’s agencies and programmes have the potential to serve as conduits for concerted efforts at atrocity prevention (even if they do not yet consciously do so) and are already indirectly building resilience to atrocity crimes when they succeed in helping to create effective, legitimate, and inclusive institutions. Many of these initiatives are not explicitly linked to the responsibility to prevent, due to the political connotations of R2P, but nevertheless constitute the very action and ethos underpinning the responsibility to prevent (Welsh, 2013).
Conclusion
The goal of this article was to both explain the centrality of prevention in the implementation agenda for R2P, and to present a clear picture of the challenges involved in making good on the rhetorical pledge to prioritise prevention. The Arab Spring has led many to doubt the efficacy of R2P, given both the critique of the intervention in Libya and the continuing failure of Security Council members to come to effective collective solutions on Syria. But while a number of the concerns raised about the concept are realistic appraisals of the limits of remedial action, this should not obscure us to R2P’s continued relevance and utility as a framework for preventive action. Although the Arab Spring has illustrated the problems associated with the reactive element of R2P, prevention remains a viable means by which the concept can continue to play a constructive and central role in international politics.
I have demonstrated the extent to which the responsibility to prevent is central to R2P and has gradually come to be embedded in the strategies and outlook of the UN, regional organisations, and member states. I have also identified the remaining obstacles to be overcome and advanced a set of recommendations to realise the goal of earlier and more effective action to forestall atrocity crimes. Nonetheless, the success of prevention is still predicated on the implementation of certain strategies and policies – especially by individual member states.
Reviewing the list of obstacles, one might be tempted to conclude that prevention can never work. Yet, as discussed, the list of cases where preventive action did forestall the outbreak of systematic and mass violence is growing. Of course, it is impossible to definitively determine that international involvement was the key variable in these non-violent outcomes. Indeed, one of the on-going challenges facing the Responsibility to Protect is the difficulty in measuring the impact of efforts at prevention. Analysts can become more systematic in how they assess different preventive tools, and the conditions under which they are effective, but it will always be hard to prove a negative. This is a challenge encountered in all areas of public policy where preventive strategies are implemented. It is perhaps particularly apparent with respect to conflict or atrocity crime prevention, however, as outbreaks of either attract infinitely more attention than less spectacular ‘non-events’ such as successful diplomacy (Akhavan, 2005). More fundamentally, even apparent success stories, such as Kenya and Guinea, must be interpreted with the caveat that there are inherent limits to what outsiders can do to change the trajectory of events in any society, and that certain leaders, if they perceive an existential threat or believe they will reap enormous gains, are unlikely to be deterred from committing acts that shock the global conscience.
Footnotes
Funding
The research leading to the results published in this article received funding from the European Research Council under the European Union’s Seventh Framework Agreement (FP/2007-2013)/ERC Grant Agreement n. [340956].