home       biography       publications       speeches       organisations       images       @contact

The Responsibility to Protect: After Libya and Syria

Presentation by Professor the Hon Gareth Evans, Former Australian Foreign Minister, President Emeritus of the International Crisis Group and Chancellor of The Australian  National University, to Directors Seminar, Belfer Center, Harvard Kennedy School of Government,  12 November 2015

This year marks the 10th anniversary of the unanimous resolution of the UN General Assembly, following the 2005 World Summit, endorsing the principle of the international community’s “Responsibility to Protect” those at risk of genocide, war crimes, ethnic cleansing and crimes against humanity, recognizing that this was a matter of international peace and security concern even when such crimes are committed wholly within the boundaries of a single sovereign state.

The resolution identified three separate dimensions to the responsibility involved: the responsibility of a state to its own people not to either commit such mass atrocity crimes or allow them to occur (now referred to as Pillar One); the responsibility of other states to assist those lacking the capacity to so protect (Pillar Two); and the responsibility of the international community to respond with “timely and decisive action” (including ultimately with coercive military force if that is authorised by the Security Council)  if a state is “manifestly failing” to meet its protection responsibilities (Pillar Three).

The resolution had its origins in the report of the 2001 International Commission on Intervention and State Sovereignty which I had the privilege of co-chairing, which was a response to international community’s failure to act effectively or at all, or with the legal authority of the Security Council when it did act, in the 1990s horror cases of Rwanda and the Balkans. The global North liked in principle the idea of “humanitarian intervention”, the South hated it, and by and large nothing happened: the mass murders went on as they had been going on from time immemorial.
My commission laid the ground for a new consensus by changing the language of the debate from “the right to intervene” to “the responsibility to protect”, and the substantive focus of the debate away from single-minded preoccupation with coercive military intervention to a much more nuanced spectrum of preventive, reactive and rebuilding strategies. 

“R2P”, as we in the Commission conceived, and as the UN member states adopted it –  after  four years of protracted diplomatic wrangling –  really was designed for pragmatists rather than purists. Its intended contribution was not to international relations theory but political practice. It was designed not to create new legal rules but rather a compelling new sense of moral and political obligation to apply existing ones.  It was to generate a reflex international response that genocide, other crimes against humanity and major war crimes happening behind sovereign state walls were everybody’s business, not nobody’s. The bottom line was always to change behaviour: to ensure that global policymakers would never again have to look back, in the aftermath of yet another genocidal catastrophe, and ask themselves – with a mixture of anger, incomprehension and shame – how they could possibly have let it all happen again.

So, looking back over the last decade, how well did we succeed in this very ambitious aspiration? Looking at the present catastrophe in Syria, where R2P gained no traction at all,  and the horrible aftermath of the initially-successful R2P-based military intervention in Libya, it would be easy to be cynical – as many critics are – and say that the whole enterprise has been a complete waste of time, or worse.
Let me offer you my own rather more positive stocktake, using as benchmarks the four big things that R2P was designed to be: a normative force; a catalyst for institutional change; a framework for preventive action; and a framework for effective reactive action when prevention has failed.

R2P as a Normative Force. 

The British historian Martin Gilbert wrote two years after the 2005 World Summit, that acceptance of the responsibility to protect is “the most significant adjustment to sovereignty in 360 years”. That’s a large call, but it is certainly true to say that R2P has gained over the last decade much more worldwide normative traction than most observers had thought possible, and has done so in a way that remains unimaginable for the concept of “humanitarian intervention” which it has now almost completely displaced (Although “humanitarian intervention” language does linger on especially in US academic discourse –  not helped by the Obama administration’s unwillingness, evidently for domestic political reasons, to itself use “R2P” language in any of its non-UN discourse.)

The best evidence of the growing normative traction of R2P is in the annual debates on R2P in the General Assembly, even in the aftermath of the strong disagreements over the Libyan intervention in 2011which have had many sceptics pronouncing its death rites.  Certainly there is more general comfort with its first and second pillars than the third, and there will always be argument about what precise form action should take in a particular case, but the basic principles are under no threat. In the most recent annual General Assembly debate on R2P in September, statements expressing overwhelming support for all of them were made by or on behalf of 89 states from every regional group, with no other serious dissent evident.

Further evidence of the acceptance achieved by R2P lies in the record of the Security Council itself. Notwithstanding again the continuing neuralgia about the Libyan intervention and the paralysing impact of that on its subsequent deliberations on Syria, this has not stopped the Security Council continuing to refer to, and apply, the R2P doctrine. Between 2005 and 2011 it had in fact passed only four resolutions mentioning R2P, but after its March 2011 decisions on Libya (and the accompanying case of Cote d’Ivoire), it had – at my last count to October this year – endorsed 31 other resolutions directly referencing the responsibility to protect, including measures to confront the threat of mass atrocities in Yemen, Libya, Mali, Sudan, South Sudan and the Central African Republic.

While none of these have authorized a Libyan-style military intervention (and a great many references are just in Pillar One terms, referring to states bearing the primary responsibility to protect their own populations) they make clear that the Council is comfortable with both the language and substance of the doctrine in all its dimensions. 
With the weight behind it of a unanimous General Assembly resolution at head of state and government level, and with all the further UN member-state acceptance it has acquired since, I believe that R2P can certainly now be described in moral and political terms as a new international norm – and, moreover, not just an “emerging” one. It does not create more legal obligations than already exist under international law in relation to genocide, other crimes against humanity and war crimes, and there is a long way to go before we could begin to describe the whole package as new customary law, but it does amount a new standard of behaviour, and a new guide to behaviour, generally accepted as such, for every state.

R2P as an Institutional Catalyst. 

All the normative consolidation in the world will not be of much use if R2P is not capable of delivering protection in practice. That means for a start the continued evolution of institutional preparedness, at the national, regional and global level, particularly at the crucial stages of early prevention, and early reaction to warning signs of impending catastrophe.

Although much more needs to be done, the story in this respect so far has been reasonably encouraging. Particular effort is going into the creation of “focal points” within key national governments and intergovernmental organizations – of which the US Atrocities Prevention Board, administered out of the NSC, is a key example – namely high-level officials, or groups of officials, whose designated day-job it is to analyse mass-atrocity risk situations and to energise an appropriately swift and early response within their own systems and in cooperation with others. The global network of these focal points, organised by the Global Centre for the Responsibility to Protect, the New York based NGO whose advisory board I chair, now has over 50 states signed up, from every region of the world, although Asian countries has been slower than those in other regions to sign on.

Probably the most crucial institutional need for the future is to create a culture of effective support for the International Criminal Court and the evolving machinery of international criminal justice, designed to enable not only trial and punishment for some of the worst mass atrocity crimes of the past, but potentially providing an important new deterrent for the future.

In the civilian sphere, more institutional response capacity is needed in the form of the organization and resourcing of civilian capability able to be utilized, as occasion arises, for diplomatic mediation, civilian policing and other critical administrative support for countries at risk of atrocity crimes occurring or recurring: commitments to develop that capability have to date been more often rhetorical than real.

In the military sphere, the main need is to have in place properly trained and capable military resources available both for rapid ‘fire-brigade’ deployment in Rwanda-type cases, and for long-haul stabilization operations like those in the Congo and Sudan, not only in no-consent situations, but where vulnerable governments request this kind of assistance. And although the establishment of effective military rapid reaction forces on even a standby basis remains more an aspiration than a reality, key militaries – again with the US playing a prominent role (thanks to the work of the HKS Carr Centre’s Sarah Sewall with the US Army) are devoting serious time and attention now to debating, and putting in place, new force configuration arrangements, doctrine, rules of engagement and training to run what are now described as “Mass Atrocity Response Operations” (MARO).

Here as elsewhere, regional organizations can be expected to play an ever more important role, exercising the full range of the responsibilities envisaged for them in Chapter VIII of the UN Charter. So far, although both the European and African Unions have shown occasional willingness to act collectively, only ECOWAS in West Africa has so far shown a consistent willingness to respond with a full range of diplomatic, political, economic and ultimately military strategies in response to civilian protection crises. Regional and sub-regional organizations in Latin America, and above all in Asia, have lagged a long way behind.

R2P as a Preventive Framework. 

The credibility of the whole R2P enterprise has depended from the outset on giving central importance to prevention, in three different contexts. First, long before any atrocity crime has occurred or been threatened, but when ethnic or religious or other tensions, unresolved economic or other grievances, or manifest governance inadequacies, or all of the above, suggest there may be a serious problem in the making unless these underlying issues are systematically addressed. Second, when warning signs – like overt hate propaganda – begin to accumulate, and more rapid and focused preventive responses have to be mounted if catastrophe is to be averted. And third, in a post-violence situation, where the crucial need is to rebuild the society in a way which seriously addresses all the underlying causal issues, and ensures that the whole ugly cycle does not recur.

The good news about prevention is that the toolbox of relevant measures at all preventive stages – across the whole spectrum of political and diplomatic, economic and social, constitutional and legal, and security strategies – is well known, and as experience accumulates, and lessons-learned literature proliferates, there is an ever more detailed and sophisticated understanding by professionals of the detailed strategies that are likely to be most effective, and cost-effective.

The less good news is that while there is a long tradition of regular lip-service being paid to the need for effective prevention, in both national and international debates, the record of practical delivery is not stellar. Part of the problem of getting sufficient resources to engage in successful atrocity, or conflict, prevention is the age-old one that success means that nothing visible actually happens: no-one gets the kind of credit that is always on offer for effective fire-fighting. It’s not easy to get any politician excited about supporting something for which he or she is unlikely to get any recognition.

All that said, especially in the context of post-crisis prevention of recurrence, there are well documented cases of successful, and since 2005 explicitly R2P-driven, preventive strategies being implemented, particularly in post-crisis prevention of recurrence situations - notably in Kenya after 2008, the West African cases of Sierra Leone after 2002, Liberia after 2003, Guinea after 2010, and Cote d’Ivoire after 2011; and probably Kyrgyzstan after 2010.  It also needs to be recognized that 10 of the 16 current UN peacekeeping operations (involving 95 per cent of the 122,000 peacekeepers on active duty) now have protection of civilians mandates – built on R2P’s sister concept of Protection of Civilians in Armed Conflict (POC)) and most of the time those operations are succeeding in keeping the lids on some often very simmering pots. And there are current cases like Burundi, constantly on the verge of volcanic ethnic conflict, where R2P-driven international diplomatic engagement has – so far anyway – helped contain further eruption.

R2P as a Reactive Framework. 

This is where the rubber hits the road. What do we do if a state, through incapacity or ill-will, has failed to meet its Pillar One responsibilities? What do we do if prevention has manifestly failed, and mass atrocity crimes are actually occurring or imminently about to occur?

R2P from the outset has involved a whole continuum of both non-coercive and coercive responses, and is absolutely not about coercive military interventions alone, notwithstanding that these have taken over so much of the ongoing debate. Those reactive responses include diplomatic peacemaking (of the kind that was so successful in Kenya in early 2008, led by Kofi Annan), political incentives as well as political sanctions, economic incentives as well as economic sanctions, offers of amnesty as well as threats of criminal prosecution, the jamming of radio frequencies by non-forceful means, arms embargoes as well as the use of arms, and various kinds of peacekeeping falling short of full scale peace enforcement. And the application of coercive military force can of course take the form of Pillar Two assistance rather than invariably more controversial Pillar Three intervention – when done at the invitation of the government unable to deal alone with a mass atrocity situation not of its own making. The Congo Force Intervention Brigade, established by the Security Council in 2013 as part of the MONUSCO peacekeeping operation, with a strongly proactive mandate to “neutralize armed groups” – both advancing M23 forces and the retreating DRC military – has been an innovative example of international coercive force being accepted, if not necessarily initiated, by a sovereign government.

Because of the degree of sensitivity and difficulty involved in any decision to use coercive military force wholly against the will of the government of the state concerned, it has been assumed from the outset by most R2P advocates, certainly me, that it would only be in the most extreme and exceptional circumstances that it will be authorised by the Security Council. And so it has proved to be, with only the Cote d’Ivoire and Libya cases in 2011 giving rise to such a mandate.

Taking into account all these different available response mechanisms, if one is to do an honest checklist here of R2P’s successes and failures since 2005 in reacting to actual outbreaks of mass atrocity crime, one has to acknowledge that the record has been at best mixed.  The clear success stories have been Kenya in 2008, and both Cote d’Ivoire and initially in Libya in 2011(I will return in a moment to how Libya went off the rails).

The clear failures have been Sri Lanka in 2009 (where the government succeeded in defending its indefensible behaviour as a legitimate response to domestic terrorism) and Syria since 2011, as well as the continuing slow-burning ugliness of North Korea and, one should probably add the troubling plight of the Rohingya in Myanmar. We also have to add Sudan to this list. Although the original crisis in Darfur predates R2P, there has been serious deterioration there over the last two years with displacement now reaching catastrophic levels, and there have been extensive atrocities committed in South Kordofan and Blue Nile since 2011. Khartoum has agreed to two major UN peacekeeping missions, in Abyei and Darfur, but undermines, restricts, and sometimes attacks them, politically and through proxies physically. Bashir remains untouched by the ICC indictment and has complied with almost none of the 60 UNSC resolutions that have been passed on political violence, conflict and atrocities in Sudan, doing just enough to keep the Russians and Chinese close and avoid any real action bythe Council to hold the state accountable. It has all become just too politically difficult and intractable for everyone.

There has been a mixed record in South Sudan and the Central African Republic since 2013, where R2P-motivated peacekeeping operations have been mounted – quickly in the case of South Sudan, less so in the CAR, though in both cases in immense contrast to the inaction over Rwanda in the 90s – but only partially successful in curbing continuing violence.

The most troubling, and costly, case of the failure of R2P to mount an effective response has undoubtedly been Syria. The crucial lapse was in mid-2011 when the violence was still largely one-sided, perpetrated by the Assad regime against essentially unarmed domestic political dissidents, but the Security Council failed even to condemn the regime – let alone apply sanctions, an arms embargo, or the threat of ICC prosecution – all of which it had done quickly as the first stage of its reaction to Libya (UNSCR 1970), with the result that Assad undoubtedly felt off the leash as the situation deteriorated quickly into full scale civil war.

Although the policy issues now are much more complex and difficult than they were in 2011 – with a multi-faceted civil war further complicated by the emergence of ISIL – the blame for that inaction, and much of the paralysis which has continued since cannot be wholly attributed to Russian intransigence, as frustrating as that has been. The basic problem was the reaction of majority of the Security Council, whether one things it rational or not, to what went wrong in Libya in 2011, when the P3 – the US, UK and France – were perceived as having exceeded the coercive military mandate given in UNSCR 1973.

The problem with the Libyan case was not the original decision by the Security Council in March 2011, made in the context of almost universally held fears of an imminent massacre by Gaddafi forces marching on Benghazi, to authorize coercive military force: there was no opposition to that resolution, it was immediately successfully implemented, and it was widely hailed at the time – including by me – as the coming of age of R2P, demonstrating that with quick and robust collective action, the horrors of Rwanda and Srebrenica could indeed be made a thing of the past.  It is impossible to know how many thousands of lives were saved in Benghazi by that initial intervention in Libya, but certainly possible to argue that had the UN Security Council acted anything like as swiftly and robustly in the 1990s, 8000 men and boys in Srebrenica, and close to 800,000 men, women and children in Rwanda, would still be alive today.

The problem was rather what happened next – the de facto transformation by the P3 without being willing to explore alternative approaches, and without allowing any serious further debate, of the Council’s limited civilian protection mandate into an open-ended regime change one. This was deeply resented by the BRICS states in particular (Brazil, Russia, India, China and South Africa), all of whom happened to be on the Council at the time, and I do believe, having talked to nearly all the participants in this debate, unquestionably was the major factor in the failure of the Council to agree on any response at all – even just a condemnatory resolution – when the Syrian situation started to explode in mid-2011. They took the view, whether one regards it as pig-headed or not, that they were not going to concede an inch if there was any chance that the P3 would take that inch to run a mile.

A solution simply has to be found to the current post-Libya stand-off if R2P is to have a future in all the ways that it needs to – if we are not, in the face of extreme mass atrocity situations, to go back to the bad old days of indefensible inaction as with Cambodia, or Rwanda, or Bosnia, or of otherwise defensible action taken in defiance of the UN Charter, as in Kosovo. The good news, I believe, is that a solution is in sight should agreement be able to be reached on some variant of the “Responsibility While Protecting” (RWP) proposal originally put on the table by Brazil, in which Russia, China and India, among others, have all shown considerable interest.

It has two key elements: first, close attention by the Security Council to agreed prudential criteria like last resort and proportionality before granting any military mandate in atrocity crime cases; and second, close monitoring and regular review by the Council of the implementation of any such mandate during its lifetime. The P3 are dragging their feet on this – always reluctant, particularly in the case of the US, to acknowledge any constraint on complete ad hocery on peace and security issues – but I think are gradually coming to the realization that unless some concessions are made on these fronts it will be impossible not only to avoid a veto on these mandates but even to command a basic majority on the Council. None of this is likely to help much in Syria, but it does offer some hope for the longer term.

Overall, while there are certainly plenty of challenges ahead for R2P, there are also many grounds for optimism about its future of R2P over the next decade and beyond. It is important to emphasise again that the disagreement now evident in the UN Security Council is really only about how the R2P norm is to be applied in the hardest, sharp-end cases, those where prevention has manifestly failed, and the harm to civilians being experienced or feared is so great that the issue of military force has to be given at least some prima facie consideration.   There is much more to the R2P project than just these extreme late-stage situations, and much to indicate that its other preventive, reactive and rebuilding dimensions all have both wide and deep international support. 

In November 1975, seven months after the Khmer Rouge had marched into Phnom Penh and commenced its reign of genocidal slaughter, US Secretary of State Henry Kissinger famously said to Thai Foreign Minister Chatichai:  “Tell the Cambodians that we will be friends with them. They are murderous thugs, but we won’t let that stand in our way”. 
While much more needs to be done to further embed R2P principles in global practice, it is a measure of how far we have come that it is difficult to imagine that being said today by any political leader anywhere.

Policymakers now around the world do understand the stakes, and the imperative for cooperative action, much better than they used to, when appalling crimes against humanity committed behind sovereign state walls were seen by almost everyone as nobody else’s business. Fully effective implementation of R2P across the board still has some distance to go, but I see no evidence anywhere that  anyone really wants to see a return to those bad old days.