Preventing Mass Atrocities: Making the Responsibility to Protect a Reality

Keynote Address by Gareth Evans, President, International Crisis Group and Co-Chair, International Commission on Intervention and State Sovereignty, to United Nations University/International Crisis Group Conference on "Prevention of Mass Atrocities: From Mandate to Realisation", New York, 10 October 2007

Most of us participating in this conference spend our waking hours and working lives wrestling with problems of conflict prevention and resolution around the world, or problems relating to human rights violations: how to prevent them, stop them and punish them. But what this conference is about is not about conflict generally or human rights abuse generally; rather it’s about one very particular kind of conflict, and one particular class of human rights abuse: a very extreme kind — involving genocide, or other mass killing, or ethnic cleansing, or crimes against humanity, or war crimes, or all of the above, more often than not occurring within a single country without directly crossing borders.

Our subject is what we call in shorthand ‘atrocity crimes’ or ‘mass atrocities’. We’re here above all to talk about the ‘never again’ situations — the Holocaust which we said must never happen again, the Cambodian genocide, Rwanda, Srebrenica, after which we said each time ‘never again’ — looking back each time, with varying degrees of incomprehension, horror, anger and shame, asking ourselves how we could possibly have let it all happen again.

In terms of its outcomes, what I see this conference, and other like it, being about is creating an environment in which the next time one of these extreme, conscience — shocking situations comes hurling down the track at us, like Rwanda in April 1994, or Srebrenica in 1995, or Kosovo in 1999, or Darfur in 2003 — or looks as though it might come down the track in the medium or longer term if some effective remedial action is not taken — the world reacts appropriately. And that means not by asking — as was the case through the 1990s — whether it should react at all in cases where claims of national sovereignty were involved, but rather when and how it should act to maximum protective effect, and delivering that response accordingly

It doesn’t seem on the face of it such an impossible task to get at least this much right. Ridding the world completely of deadly conflict — yes, that is a big ask, although we are getting much better at it than we used to be. Ridding the world completely of human rights abuse – yes, that’s an even bigger ask, and it’s a little quixotic to think we might manage it in any of our lifetimes. But no more Holocausts, Cambodias, Rwandas, or Srebrenicas? Surely that’s not only thinkable, but doable. But we all know it hasn’t so far been doable — notwithstanding the Universal Declaration, and the Covenants and even the Genocide Convention — and that’s why we’re all here.

If ridding the world once and for all of mass atrocities is to be doable, we need three kinds of strategies: conceptual, to frame the issues involved, and to embed that framing in policymakers minds and instincts in a way that there’s no preliminary stumbling block to the kind of necessary global reflex action I have described; institutional, to create structures and processes, both in intergovernmental and national settings, which will be capable of delivering the preventive and reactive responses required; and political, to ensure that when each new atrocity or potential atrocity situation comes along the actual response is effective.

Most of the sessions today will be focusing on the institutional issues — the role of the new Office of the Special Representative for the Prevention of Genocide and Mass Atrocities, the role of other relevant elements in the UN system, and their interface with national governments and with civil society actors. And a recurring theme will be how to generate the political will to ensure that this institutional machinery is effectively used. What I want to focus on in my opening remarks is the threshold conceptual issue — have we in fact overcome that preliminary stumbling block that has been such a crucial inhibitor to effective responses over the generations, i.e. that internal matters are nobody else’s business, that — in its extreme form — state sovereignty is a license to kill?

For most of the four and a half centuries that have elapsed since the emergence of modern sovereign states in the Treaty of Westphalia it certainly has not been possible to argue that this stumbling block has been overcome — not even with the human rights provisions of the UN Charter and Declaration, not even with the Genocide Convention. Article 2(7) of the Charter, "Nothing should authorize intervention in matters essentially within the jurisidiction of any state”, was regarded as the key, and while this expressly did not override enforcement measures agreed under Chapter VII, there was a deep reluctance — which persists to this day — to construe at all broadly the language of that chapter as to what constitutes a threat to ‘international peace and security’.

The stumbling block was not overcome by Bernard Kouchner arguing for a ‘right of humanitarian intervention’, though this became the major currency of debate in the 1990s; nor by Kofi Annan arguing that the Charter protected individual sovereignty as well as state sovereignty; nor was the problem wholly solved, though it was partially, by Francis Deng developing the concept of ‘sovereignty as responsibility’. Right through the 1990s, as we can all remember, fierce arguments continued to wage, in the General Assembly and elsewhere, as to whether any kind of intervention, in particular coercive military intervention, was legitimate in responding even to the most egregious and genuinely conscience-shocking violations.

The conceptual breakthrough did however come, I think, it is now generally agreed, with the emergence in 2001 from the Canadian-sponsored Commission on Intervention and State Sovereignty of the concept of the ‘responsibility to protect’ — which built on those earlier building blocks, but in a way which cast the whole issue of intervention in a different light. The ‘right to intervene’ was turned on its head, focusing not on those with the might, but the victims who were suffering — what was involved was not anyone’s right, but a responsibility, and not to intervene so much as to protect.

Moreover the primary responsibility involved was, unequivocally, that of the sovereign state itself — it had the frontline responsibility to protect its own people. It was only when a state was manifestly incapable, or unwilling, to stop genocide, or ethnic cleansing, or other crimes against humanity, or war crimes, occurring that that responsibility shifted to the wider international community — in the first instance to assist the sovereign state to deal with the problem itself, by building appropriate capacity, but in extreme cases, if all else failed, through coercive action, including, as an absolute last resort, military intervention authorised by the Security Council.

As all the participants in this conference well know, the responsibility to protect principle, articulated pretty much in these terms, was adopted unanimously by the 150 heads of state and government meeting as the General Assembly at the UN’s 60th Anniversary World Summit. This was one of the very few significant achievements of that Summit in 2005, and on any view a rather remarkable achievement given the very short time the R2P idea had been part of the public debate, and the level of discomfort still felt in many G77 countries about acknowledging any major constraint on state sovereignty.

But as participants here will certainly also be aware, there is a long way to go before we can assert with any confidence that the R2P principle as so adopted, and indeed subsequently incorporated by reference in a couple of Security Council resolutions, has indeed won universal acceptance — to the extent that there genuinely will be the kind of reflex response to an emerging mass atrocity situation of the kind I described at the outset.

I suspect that a good part of the reason for this is that those of us who strongly, indeed passionately, support the responsibility to protect principle have not yet done a good enough job in describing exactly what it is about, and in particular what are the kinds of situations to which it applies, and what are the appropriate policy responses at different stages in the evolution of a situation of concern.

If the R2P concept is to win genuine universal consensus, and to become effectively operational in practice, there are two crucial threshold conceptual requirements.

The first is that it not be seen too narrowly, as only about non-consensual military intervention. On the contrary, it is a principle that emphasises prevention — and at the early, not just the late. It implies encouragement and support being given to those states struggling with situations that have not yet deteriorated to the point where genocide or other atrocity crimes are a reality, but where it is foreseeable that if effective preventive action is not taken, with or without outside support, they could so deteriorate.

The best way to make this point may be identify clear examples, case studies, of where the international community has been engaged in exactly this kind of preventive support but without labeling it as an R2P exercise. Take Burundi, for example, which could so easily have followed Rwanda’s path in 1994, and which remains extremely fragile to this day, but where multiple international players have been working hard and long to ensure that it didn’t — by efforts such as Nelson Mandela’s initial political mediation; the deployment, essentially preventively, of peacekeeping troops particularly by South Africa; the detailed analysis and recommendations for a decade now by my own International Crisis Group as to what is necessary to achieve sustainable peace; and the attention being given to it Burundi as one of the first two cases taken up by the Peacebuilding Commission.

If the first big conceptual requirement is that R2P not be seen too narrowly, the other is that it not be seen as having too wide a coverage. It’s linguistically, and perhaps emotionally, tempting to say, as some have, that there is a responsibility to protect people from HIV/AIDS, or the proliferation of small arms or nuclear weapons, or those in fragile ecosystems from the impact of global warming. But to put everything that could be broadly defined as a human security problem into the R2P basket is to denude it of content and utility in those situations, the mass atrocity situations, where we need it most as a rallying cry.

These are not the only ways, and possibly not even the most dangerous ways, in which R2P can be given too wide a coverage. It is also tempting to use the concept too widely by not distinguishing R2P situations from those of conflict generally or human rights violations generally, a distinction I insisted upon at the beginning of my remarks. It is tempting because issues of civilian protection (from loss of life, from injury, from economic loss, from assaults on human dignity) are always involved in any deadly conflict, whatever its cause and whatever its scale, and in any significant human rights violation. But dangerous, again, because if everything is an R2P situation then at the end of the day nothing will be — in the sense, again, that we will just not have that necessary global reflex response of which I spoke at the outset, ensuring that when the next extreme, conscience-shocking situation comes to stare us in the face, as one surely will, the issue will not be whether the international community should act, but when and how it should act.

There are around the world at any given time some 70 different situations which can reasonably be described as ones of actual or potential deadly conflict — I know that because the International Crisis Group documents them each month in our CrisisWatch bulletin — and there are at any given time at least as many countries, not all the same ones, where significant human rights violations are occurring, as documented by Human Rights Watch and Freedom House surveys among others. But the truth of the matter is that there are perhaps no more than ten or a dozen countries at any time which can reasonably be claimed to be R2P situations, i.e. where genocide, or ethnic cleansing, or other crimes against humanity, or war crimes were either actually occurring or could foreseeably occur at some time in the future — immediate, medium term or long term — unless appropriate preventive measures are taken.

It is a hazardous business to identify those situations, because there will always be differences of view — strongly and honestly held — about how to characterise particular cases. Is what is happening now in Darfur properly described as a continuation of the atrocity crimes which unquestionably occurred in 2003 — 2004, or has it, while still bearing all the scars of those days and still requiring immense international effort to resolve, morphed into a kind of conflict much less distinguishable than many others? Is the situation in Kosovo, if the final status issue is not resolved peacefully in the next few months, capable of deteriorating back into the large scale killing and ethnic cleansing that prompted military intervention, unhappily without Security Council approval in 1999? Is Sri Lanka best understood as a particularly bloody and intractable separatist conflict, or does it — given its ethnic dimension, and the kinds of atrocities which have occurred in the past — count as the kind of R2P situation in the making which demands preventive action, by the Sri Lankan government with the support of the wider international community, to ensure that further deterioration does not occur?

Is the current situation in Burma/Myanmar best responded to as a human rights and democracy problem, requiring whatever mix of pressure and persuasion that will best work, or as an R2P situation in the making? Is that the right characterisation, too, of the situation in Zimbabwe, or his Robert Mugabe’s immiseration of his own approaching the level of what we could reasonably begin to call atrocity crimes? Is the recent re-emergence of ethnicity-manipulating hate speech in Cote d’Ivoire, which we last saw in 2004, an alarming reminder of what we saw with Radio Milles Collines before the Rwandan genocide, or just a product of irresponsible political leadership which demands censure on human rights grounds but no particular wider anxiety?

Of all the current situations that might feature on an R2P watchlist, the one that probably concerns me most of all is that in Iraq, because there is every reason to fear that — particularly in the context of a precipitate withdrawal of foreign forces from the centre and north of the country (where the argument for a significant continuing presence is much stronger than in the south) — that the present situation, horrible as it is, will rapidly deteriorate into a massive further outbreak of communal and sectarian violence and ethnic cleansing beyond the capacity of the Iraqi government to control, and from which it would be unconscionable for the wider world to stand aloof. If anyone thinks that so much ethnic cleansing has already occurred the conditions for a genocide, if they ever existed, are no longer there, they just haven’t been concentrating.

I say this as conscious as anyone possibly could be of the bitter irony involved. I was wholly opposed to the coalition invasion of Iraq in 2003, and argued strenuously at the time that it was not justified on R2P grounds any more than it was on WMD or responding to terrorism grounds, however much Tony Blair and some others still to this day want to argue to the contrary. And I am deeply aware how much that conflation of Iraq 2003 with R2P has damaged the R2P cause among those who argue that whatever its noble intent, if R2P is accepted in principle it is likely to be misused in practice.

The truth of the matter is that while Iraq in 2003 would have been on anyone’s R2P country watchlist, this was not a case, even at the threshold level when one is weighing the gravity of the threat involved, for a coercive military intervention. Although there were clearly major human rights violations continuing to occur (which justified international concern and response, e.g. by way of censure and sanctions), and although mass atrocity crimes had clearly occurred in the past (against the Kurds in the late 1980s and the southern Shiites in the early 90s) such crimes were neither actually occurring nor apprehended when the coalition invaded the country in early 2003.

My point in all of this is simply that if the responsibility to protect principle is going to become genuinely embedded in international consciousness, and to become genuinely operationally effective, then we have a lot more work to do. There’s not only the institutional preparedness agenda, and the political preparedness agenda, about which others will speak; inextricably linked with these there’s the conceptual preparedness agenda - getting our, and everyone else’s, heads clear about what we mean by R2P situations, actually occurring and in the making.

Taking forward all these agendas, but above all the agenda of unfinished conceptual business, is going to be the task of the new Global Centre for the Responsibility to Protect, to which the Secretary General has graciously given his imprimatur this morning, and which will be launched early in the new year. You’ll hear a lot more about this Centre as its organisation evolves, but in short, it’s the brainchild of a group of global NGOs (ourselves, Human Rights Watch, Oxfam International, WFM and Refugees Inernational), supported generously so far by a number of like-minded governments (Netherlands, Canada, Australia and Rwanda, with more in the pipeline) and by a major pledge from at least one highly committed individual, Scott Lawlor. The Centre will be based at the City University of New York’s Ralph Bunche Institute led by Professor Tom Weiss, and will have from the outset a strong North-South dimension, in its professional staff, its International Advisory Board which I will be co-chairing with Mohamed Sahnoun, and in the globally represented group of Associated Centres that will be working closely with it. And it will be working very closely indeed in support of Francis Deng and Ed Luck, in their genocide prevention and R2P advisory roles which we will be discussing in the first panel session.

It is very easy — in this world we inhabit full of cynicism, double-standards, the crude assertion of national interest, high-level realpolitik and low-level manouvring for political advantage — to believe that ideas don’t matter very much. But I believe as passionately now as I have ever done in my long career, starting and finishing in the NGO world but with a lot of time in politics and government in between, that they matter enormously, for good and for ill. And I don’t think that there are many ideas that have the potential to matter more for good than that of the responsibility to protect.

I hope that in this conference, and others like it, and in the ongoing work of all those in this room today, and others like us, we really can create a situation, once and for all, where it is simply no longer possible to claim with any credibility that what happens within state borders, however grotesque and morally indefensible, is nobody else’s business — that we can create a world in which we never again have to say ‘never again’.