Halting Genocide: Intervention and Legitimacy
Address by Gareth Evans, President of International Crisis Group and Member of the High Level Panel on Threats, Challenges and Change, to Geneva Centre for Security Policy Forum on Global Peace and Security: Challenges and Responses, Geneva, 18 May 2005
‘Never again’ we said after the Holocaust. And after the Cambodian genocide in the 1970s. And then again after the Rwanda genocide in 1994. And then, just a year later, after the Srbrenica massacre in Bosnia. And now we’re asking ourselves yet again, in the face of more mass killing and dying in Darfur, whether we really are capable, as an international community, of stopping nation-states murdering and killing by neglect their own people – with more than 200,000 dead already, on conservative estimates, over the last two years; the Janjaweed militia still either out of control, or being controlled all too destructively by Khartoum; over 2 million displaced; and disease and malnutrition likely to kill thousands more each month. How many more times will we look back wondering, with varying degrees of incomprehension, horror, anger and shame, how we could have let it all happen?
For all the things the international community has been getting right since the end of the Cold War gave us a second chance to build under the UN Charter a genuine global collective security system (and there are a lot of largely unheralded things that we have been doing much better in the areas of peace negotiations, peacekeeping and peacebuilding), there are some things we go on getting terribly wrong. And foremost among those failings is our apparent congenital inability to react quickly and effectively as an international community when cases come along, as they have now again with Darfur, and are bound to come again, sooner or later somewhere or another in the world, of conscience-shocking mass violence occurring within the borders of a sovereign state.
We are still a long way away from reaching consensus about what kind of international intervention, in particular coercive military intervention, is justified, and when, in these situations. There continues to be enormous disagreement as to whether there is any right of intervention at all, how and when any intervention should be undertaken, and under whose authority. And while the disagreement continues, people keep dying.
Why so Intractable?
There are several reasons why this problem has proved so hard to resolve. The first is simply that there are no clear existing international law rules. The UN Charter is at best neutral on the subject, and at worst negative. The founders of the UN in 1945 were preoccupied, understandably enough, with countering aggressive war by states against each other. They expressly outlawed the threat or use of military force (Art 2.4), except for self-defence strictly defined (Art 51) and as authorised by the new Security Council "to maintain or restore international peace and security" (Art 42); then reinforced all this by enshrining as the basic norms of international relations the principle of the equality of sovereign states (Article 2.1) and their non-intervention in each other's domestic affairs (Article 2.7).
At the same time the UN founders were conscious of the catastrophic human rights violations of the preceding years – above all the Nazi genocide – and this did generate a new momentum for the better protection by international law of individual human rights. At least a toehold of recognition was gained for human rights in the terms of the Charter itself, and standards were spelled out more comprehensively in the Universal Declaration of Human Rights, adopted in 1948 and the subsequently negotiated Conventions on Civil and Political Rights, and on Economic, Social and Cultural Rights adopted in 1966, not to mention the very specifically focused Genocide Convention of 1948.
Unfortunately the limits on state sovereignty implied by human rights standards were not set out in the Charter nearly as clearly as the principle of sovereignty itself. And so the tension at the heart of the Charter – between state security and human security, between state rights and human rights – remained essentially unresolved. The criminality involved in the government of a state attacking and invading another was amply recognised, and mechanisms to deal with it amply provided.
However, in relation to the criminality of a government perpetrating or acquiescing in large-scale killings of its own people, those who wrote the Charter gave no guidance at all. The Security Council can, if it chooses, characterise these essentially internal situations as amounting to a "threat to international peace security", and as such use its authority under the Charter to authorise enforcement measures, but there is an obvious element of artificiality and discomfort involved in doing this which makes the achievement of consensus extremely difficult.
The second reason for the issue of humanitarian intervention being so difficult to resolve is the emotional attachment to state sovereignty by so many countries. All states hate the idea they might be the target of a coercive external intervention, and the memory is strong in the developing world as to how such intervention has been used and misused in the past. Africans remember all too well what French missions civilisatrices meant for them. Latin American states are particularly sensitive given the history of their region: they understandably think of the Monroe Doctrine as not so much designed to keep the Europeans off their backs as the Americans on them. Asian states, with their own history (although in many cases for other less elevated reasons), retain an almost visceral discomfort with the whole idea of intervening in what goes on inside other countries.
The principle of non-intervention has proved wonderfully attractive to the legions of new states who have joined the UN in the decades since 1945 – increasing its numbers from 51 to the present 191 as the tide of decolonisation and post-Cold War state fragmentation proceeded. Sovereignty hard won is sovereignty not lightly conceded, and it is not surprising that for many of the new states an extreme interpretation of Article 2, seen as immunising them from almost any external scrutiny at all, has become something of an article of faith. But there are regional differences within the developing world: sovereignty sentiment continues to run much more strongly in Asia at one end of the spectrum than it does in sub-Saharan Africa at the other, where the despair in recent years has been as much about the lack of intervention as its happening.
There is a third reason why the issue of "humanitarian intervention" – or, as most of us, not least here in Geneva, would prefer to call it, "intervention for human protection purposes" – has, after a period when hopes rose that a new consensus might be emerging around the concept of the "responsibility to protect" (which I will come to in a moment) become difficult to resolve all over again. Suspicions have been reinforced once more by the use of this rationale – i.e. action to depose a tyrant, to protect the country's people – to justify the war in Iraq: not from the outset, but as other rationales fell away. As weapons of mass destruction failed to turn up, and the evidence of Saddam Hussein's links with terrorists failed to get any stronger, Tony Blair and other defenders of the war were forced back to supporting it on straightforward so-called humanitarian intervention grounds: Saddam's gross mistreatment of his own people, which reached genocidal levels in his use of chemical weapons against Kurds in the late 80s and his massacre of southern Shiites in the early 90s.
There are a number of problems with this rationale. It certainly did not appear to be the real motive for intervention, or anything like it, at the time. It sits uneasily with the West's blind-eyed acquiescence in Saddam's genocidal behaviour at the times in question. It suggests that the rationale for humanitarian intervention, or intervention for human protection purposes, is as much punishment as it is to avert here and now threats which are actually occurring or imminently about to occur. It suggests that any regime guilty of extreme human rights violations in the past, however muted its behaviour since, can be militarily attacked by others at a time of their choosing. And it doesn't acknowledge that even if a prima facie can be made for going to war on this ground, the case for actually doing so depends on multiple other criteria being satisfied, not least that the results of military action will not be worse than taking no action.
A Way Forward: ICISS and the Responsibility to Protect
So what are the principles that should govern our responses if and when these kind of situations come along? The most substantial effort so far to identify them, and build an international consensus around them, has been the work of the Canadian government-sponsored International Commission on Intervention and State Sovereignty (ICISS), which I had the privilege of co-chairing, with Mohamed Sahnoun, and which presented its report to the UN Secretary General at the end of 2001.
The Commission made, I think it is fair to say, four main contributions to the international policy debate.
The first, and perhaps ultimately the politically most useful, was to invent a new way of talking about the whole issue of humanitarian intervention. We sought to turn the whole weary debate about the "right to intervene" on its head, and to recharacterise it not as an argument about any "right" at all, but rather about a "responsibility" – one to protect people at grave risk – with the relevant perspective being not that of the prospective interveners but, more appropriately, those needing support. This new language has been helpful already in taking some of the heat and emotion out of the policy debate, requiring the actors to change their lines, and think afresh about what are the real issues are. Our hope - and so far our experience - is that entrenched opponents will find new ground on which to more constructively engage, just as proved to be the case between developers and environmentalists after the Brundtland Commission introduced the concept of "sustainable development".
The second contribution of the Commission, perhaps most conceptually significant, was to come up with a new way of talking about sovereignty: we argued that its essence should now be seen not as control but as responsibility. The UN Charter's explicit language emphasises the respect owed to state sovereignty in its traditional Westphalian sense, but actual state practice has evolved in the nearly 60 years since the Charter was signed: the new focus on human rights and, more recently, on human security, emphasises the limits of sovereignty.
We spelt out the implications of that change by arguing that sovereignty implies responsibilities as well as rights: to be sovereign means both to be responsible to one’s own citizens and to the wider international community through the UN. The starting point is that any state has the primary responsibility to protect the individuals within it. But that's not the finishing point: where the state fails in that responsibility, a secondary responsibility to protect falls on the international community acting through the UN.
The third contribution of the Commission was to make it clear that the "responsibility to protect" was about much more than intervention, and in particular military intervention. It extends to a whole continuum of obligations:
the responsibility to prevent: to address both the root causes and direct causes of internal conflict and other man-made crises putting populations at risk;
the responsibility to react: to respond to situations of compelling human need with appropriate measures, which may include coercive measures like sanctions and international prosecution, and in extreme cases military intervention; and
the responsibility to rebuild: to provide, particularly after a military intervention, full assistance with recovery, reconstruction and reconciliation, addressing the causes of the harm the intervention was designed to halt or avert.
Of these three dimensions to the responsibility to protect, the Commission made very clear its view that prevention was the single most important. But that said, the question of military action remains, for better or worse, the most prominent and controversial one in the debate. Whatever else it encompasses, the responsibility to protect implies above all else a responsibility to react - where necessary coercively, and in extreme cases with military coercion - to situations of compelling need for human protection. So the fourth contribution of the Commission was to come up with guidelines – going both to legality and legitimacy - for when military action is appropriate.
Five Criteria of Legitimacy
The effectiveness of the global collective security system, as with any other legal order, depends ultimately not only on the legality of decisions, but the common perception of their legitimacy - their being made on solid evidentiary grounds, for the right reasons, morally as well as legally. The Commission identified five criteria of legitimacy which we argued should be applied by the Security Council -- and be used by the world at large -- to test the validity of any case made for a coercive humanitarian intervention.
Adopting and applying these five criteria of legitimacy may not guarantee that the objectively best outcome will always prevail, but it would maximise the possibility of achieving Security Council consensus around when it is appropriate or not to go to war; maximise international support for whatever the Security Council decides; and minimise the possibility of individual member states bypassing or ignoring the Security Council.
(1) Just Cause: is the harm threatened sufficiently clear and serious to justify going to war? We set the bar for military intervention deliberately high, and tight, excluding many kinds of unconscionable behaviour (e.g. imprisonment and torture of political opponents, overthrow of a democratically elected government) that would certainly other forms of coercive response, (e.g. targeted sanctions):
There must be serious and irreparable harm occurring to human beings, or imminently likely to occur, of the following kind:
A. large scale loss of life, actual or apprehended, with genocidal intent or not, which is the product either of deliberate state action, or state neglect or inability to act, or a failed state situation; or
B. large scale ‘ethnic cleansing’, actual or apprehended, whether carried out by killing, forced expulsion, acts of terror or rape.
As I have already suggested, in the context of the invasion of Iraq in 2003, this would certainly have been satisfied a decade or more ago (when the West was indifferent or worse to the plight of the Kurds and southern Shiites), but much less obviously so in recent years. In the case of Darfur, there is not much doubt that this test is satisfied, on all available evidence, here and now.
Genocide? It is important to note that the ICISS report deliberately downplays the "g" word, making it clear that coercive action can be triggered in the face of certain horrific behaviour, whether that is occurring "with genocidal intent or not". Our report was written before the Darfur events erupted, and before the ongoing furious controversy as to whether the crimes there committed should be described as "genocide -- as US Secretaries of State Colin Powell and now Condoleezza Rice have now done (not that it seems to have led either of them to do anything more than they otherwise would have done to "prevent and punish" that behaviour, as the Convention obliges states to do) – or rather just "crimes against humanity" as the UN’s own Commission on Darfur has concluded they probably are. But if it was being written now I suspect the ICISS report would have been even more explicit.
My own strong view, very much reinforced by the Darfur controversy, is that focusing on the "g" word is a distraction -- and an often terribly counterproductive one -- from the real issue, which is the need to act to protect people when atrocity crimes of any kind are being committed, or about to be committed, and to hold the perpetrators to effective account. The problem with the Genocide Convention is that its language is very precise, and susceptible as a result to endless legal argument: it requires, above all, that certain defined acts be "committed with intent to destroy, in whole or in part, a national, ethnical, racial or religious group", and this requisite intent, as the UN Darfur Commission made clear, can in the real world be very difficult to establish, even on a prima facie basis. Any accusation of genocide tends to invite the response, exactly as has happened with Khartoum, that whatever killing and ethnic cleansing and rape may have occurred, it wasn't genocidal in character: when lawyers find themselves lacking sufficient evidence to prove the contrary, a propaganda victory can be handed on a plate to those whose behaviour, on any view, remains ugly, indefensible and deserving of strong international response.
I am very attracted, as I believe we all ought to be, with the proposal of David Scheffer, former US ambassador at large for war crimes issues and now at George Washington University, who argues (e.g. in his Financial Times article of 2 February 2005) that in order to avoid these unedifying and counterproductive semantic arguments, which constantly distract attention from the need for effective action, we should all just use the generic expression "atrocity crimes" and leave it to the prosecutors and judges in the international courts, or courts exercising international jurisdiction, to work out what is genocide, a war crime, a crime against humanity or the like in the particular circumstances of each case. Some situations, like Rwanda in 1994, might be so obviously genocide that an exception to this general approach could be made, but most of the time we would be much better off, in terms of focusing attention on the basic moral guilt involved, and the need for action, just talking about the horror of atrocity crimes.
(2) Right Intention: is the primary purpose of the proposed military action to halt or avert the threat in question, whatever other motives may be in play?
In the case of Iraq, it is probably fair to say that this test was satisfied in the case of the UK, where Tony Blair has long been passionate about the monstrous behaviour of Saddam towards Iraq's own people. In the case of the U.S., the jury would I think have a harder time. In the case of the third Coalition member, Australia, perhaps there was some other motive than following the leader (and earning, as local commentator Phillip Adams unkindly said, some more "Frequent Fighter Points") -- but I've been out of the country over five years now and I'm afraid it has escaped me…
In the case of Darfur, if the international did decide to coercively intervene, through the African Union or anyone else, I don't think there is much doubt this test would be satisfied.
(3) Last Resort: has every non-military option for the prevention or peaceful resolution of the crisis been explored, with reasonable grounds for believing lesser measures will not succeed?
For Iraq, the answer to this depends essentially on which rationale for intervention you regard as most important. If it was the issue of possession of weapons of mass destruction, manifestly this criterion was not satisfied. But if you accept Saddam's tyranny over his own people as sufficient justification, then perhaps it was.
In Darfur, this criterion must be very close to being satisfied (although whether that means military force is actually deliverable is another question). Non-military international pressure has been intensifying on Sudan -- not least with the welcome recent decisions of the Security Council exposing its leaders to targeted sanctions and International Criminal Court prosecution -- and it may be that it will as a result call a full halt to the killing, mount and allow a relief effort massive enough to prevent any more dying, and allow in a big enough force to fully monitor the situation and act forcibly to protect civilians as circumstances require. But it hasn't done enough of these things so far, and time to demonstrate its bona fides is fast running out.
(4) Proportional Means: is the scale, duration and intensity of the planned military action the minimum necessary to secure the defined human protection objective?
In the case of Iraq, a massive military operation was mounted, leading to some 3,500 civilian deaths and 10,000 military deaths at the time, quite apart from the thousands who have died since; it is a matter for judgement whether one regards those casualties as, for the Iraqis supposedly being protected, an appropriate trade for the end of Saddam Hussein's capacity to persecute.
In Darfur, it remains difficult -- for reasons of practical deliverability as much as anything else, given that those needing protection are scattered over a region the size of France -- to contemplate a full-blown coercive invasion. But it has to be recognised that a much larger military force than the 2000 or so troops so far deployed by the African Union -- and the International Crisis Group argues for one at least five times larger -- is clearly necessary; that this needs, to make it viable, much stronger international support than has so far been offered (though the Canadians have recently moved to help fill the gap); and that it also needs a stronger, more effectively applied, civilian protection mandate than has so far been accepted. There could be no argument that a response of this kind, still premised on some kind of consent from Khartoum, is disproportionate to the need.
Even if thing deteriorated to the point that Sudan withdrew its consent for any external military presence, and the international community were to decide that it would intervene anyway, being prepared to confront a hostile national army as well as scattered militia, in a way that clearly required a much larger military operation, it would still be difficult to argument that this response was disproportionate to the need as we know it.
(5) Reasonable Prospects: is there a reasonable chance of the military action being successful in meeting the threat in question, with the consequences of action not likely to be worse than the consequences of inaction?
In the case of Iraq, this criterion is one of the hardest for defenders of the war to meet, even without the wisdom of hindsight which has made the war less defensible on this ground with every passing month. It should have been foreseen that the country was sadly divided, that the ‘liberators’ would not be universally welcomed, and that the overthrow of an Arab government was far more likely to generate terrorist recruits than to deter them
The "balance of consequences" test is, and should be, a very important constraint. Apart from anything else, it effectively rules out military action against any one of the five permanent members of the Security Council -- e.g. against Russia over Chechnya, or against China over some imaginable course of events in Xinjiang -- even if all other conditions for intervention were to be met.
The same is true for other major powers -- which is why Indonesia's permission was required for the East Timor intervention: that permission was reluctantly given, and under much international pressure, but it was given, and the case was, accordingly, not one of coercive "humanitarian intervention" at all. This is why a military intervention to respond to the situations in Indonesia’s Aceh or Papua provinces, however much it deteriorated, can be effectively ruled out. Coercive external intervention would simply trigger a much larger conflict, multiplying the human catastrophe involved.
All this raises the familiar question of double standards, to which the only answer can be this: the reality that interventions may not be able to be mounted in every case where there is justification for doing so, is no reason for them not to be mounted in any case.
In the case of Darfur, the country’s population (some 40 million) and armed forces size (over 100,000) would not in themselves be knockout reasons against coercive intervention, but the consequences of mounting a full scale invasion, not only for Sudan itself but for its surrounding not very stable region, would have to be very carefully evaluated. Obviously the East Timor option – getting Khartoum’s consent, however reluctant, to a large external civilian protection force – continues to be the more attractive solution, and the one that has prevailed in the UN so far.
The Legality Test
What if, taking into account all these five criteria of legitimacy, a very clear case can be made for coercive intervention, but the Security Council – under the UN Charter the only source of authority for the use of military force, except in cases of legitimate self-defence - simply won’t vote to authorise it? This was essentially the issue that had to be confronted with Kosovo in 1999, when all the elements of a horrific new ethnic cleansing operation were falling into place but Russia made clear it would veto any military intervention.
In these cases a very real dilemma arises as to which of two evils is the worse: the damage to international order if the Security Council is bypassed, or in the damage to that order if human beings are slaughtered while the Security Council stands by.
The ICISS Commission’s response to this dilemma was not to try and establish some alternative basis for the legality of interventions in these situations - we saw our role as not to find alternatives to the Security Council as a source of authority, but to make it work better. We opted instead for a clear political message: if an individual state or ad hoc coalition steps in, fully observes and respects all the necessary criteria of legitimacy, intervenes successfully, and is seen to have done so by world public opinion, then this is likely to have enduringly serious consequences for the stature and credibility of the UN itself. That is pretty much what happened with the U.S. and NATO intervention in Kosovo, and the UN cannot afford to drop the ball too many times on that scale.
There is another side to this credibility argument which arose in the context of the invasion of Iraq in 2003, where world opinion was manifestly not in favour of the intervention: here the French and others were - and still are - able to claim, with some effectiveness, that the credibility of the UN would have been put more at risk if the Security Council had gone along for the ride rather than resisting, as it did, the US pressure.
With the publication of the ICISS report in December 2001, even overshadowed as it was by 9/11 and the massive international preoccupation with terrorism, the concept of the ‘responsibility to protect’ began to gain international recognition. It was embraced enthusiastically by Secretary-General Kofi Annan, who acknowledged, very graciously, that it had rather more potential to bridge the sovereignty v. intervention divide than his own earlier attempt to find consensus around the idea of ‘state sovereignty’ needing to be balanced by a recognition of the competing claims of ‘individual sovereignty’. It began to be embraced, importantly, in the doctrine of the newly emerging African Union. And it won support among many academic commentators and international lawyers who accepted, to a greater or lesser extent, the Commission’s own rather heroic assessment of the ‘responsibility to protect’ as already an emerging international norm which might in due course become accepted as customary international law.
International law being the rather odd beast that it is – capable of evolving through practice and commentary as well as through formal treaty instruments - these embraces and acknowledgments are to some extent self-fulfilling. But that is only the case if the momentum is maintained: if the responsibility to protect concept is to really catch hold, and become the primary frame of reference within which catastrophic human rights violations are assessed and responded to in the future, its acceptance has to become considerably more visible and universal.
A lot of effort has been going into maintaining that momentum. Two big milestones have now been passed in that respect, and a third lies ahead. First, the High Level Panel on Threats, Challenges and Change, whose report A More Secure World: Our Shared Responsibility was submitted to the Secretary-General in December 2004, squarely adopted the whole concept in these words (para 203):
We endorse the emerging norm that there is a collective international responsibility to protect, exercisable by the Security Council authorizing military intervention as a last resort, in the event of genocide or other large-scale killing, ethnic cleansing or serious violations of international humanitarian law which sovereign Governments have proved powerless or unwilling to prevent.
The High Level Panel also effectively endorsed the criteria of legitimacy on which the ICISS Commission had insisted be a basis for any resort to military action: the only difference was that the Panel recommended that these criteria be applied by the Security Council not only when considering the use of military force in internal ‘humanitarian intervention’ situations, but indeed to all decisions to use military force in any context whatever.
The second big milestone that has been passed is the embrace of these recommendations by the Secretary-General himself in his own proposals for reform going forward to the September Millennium Summit, in the context of this UN 60th Anniversary, published in March 2005 as In Larger Freedom: Towards Development, Security and Human Rights for All. After repeating, in effect, the language of the High Level Panel, he went on to say (para 135):
While I am well aware of the sensitivities involved in this issue, I strongly agree with this approach. I believe we must embrace the responsibility to protect, and, when necessary, we must act on it.
There are indeed sensitivities still involved in this issue, and they are on full display as the third milestone approaches – the adoption of the Secretary-General’s recommendations by the heads of state and government assembled at the Millennium Summit itself. The concept of the responsibility to protect and all it entails is, like the rest of the package that will go forward, being vigorously discussed within a declaration-drafting process that is being conducted at two levels – on the one hand, the working General Assembly level, with its familiar cast of New York diplomats (what might be described as ‘down in the weeds’); and on the other, at the level of political figures and senior officials in capitals and those in the UN Secretariat liaising with them ( perhaps describable as ‘up in the tower’). The picture that has so far emerged about whether ‘responsibility to protect’ language will ultimately be endorsed is one that, in this instance seems to reflect the different altitudes involved: there is considerable pessimism still down in the weeds, but a little more optimism being expressed up in the tower, where there is some hope that strong commitment from African leaders, and some emerging support from Latin America, will end up neutralising and overcoming the strident hostility that is being expressed elsewhere within G-77 ranks.
It has become apparent that one of the factors inhibiting greater consensus on this issue is the perception that the ‘responsibility to protect’ is only about military action, and indeed is effectively a synonym for military action. It can and should be made very clear (as it was in the original ICISS report, but less so in the later High Level Panel and Secretary-General’s reports, which addressed the issue in a primarily military context) that the responsibility to protect involves a continuum of responsibilities, starting – crucially – with the responsibility to prevent, only extending to the responsibility to react if prevention fails (and even then with military coercion being the least preferred, and always last resort, form of reaction), and being accompanied as well by a responsibility to rebuild those societies shattered by internal catastrophe. I would strongly urge that these different dimensions of the concept – and in particular the primacy given to prevention – be incorporated in the draft language for which the leaders’ approval is sought.
The issue at the moment hangs delicately in the balance. The challenge for those who see the responsibility to protect concept as creating the conditions for a far more effective response to conscience-shocking situations than the international community has managed in the past is to pull out all stops to ensure that the momentum of its evolution doesn’t grind to a halt in September. If we can achieve acceptance of the approach first mapped by the ICISS Commission and subsequently embraced by both the High Level Panel and the Secretary General, I believe we really will be well on the way to building an international legal order that, despite all the challenges which continue to buffet it, will give us some grounds for optimism that we are
not forever condemned to repeat the mistakes of the past,
not only the mistake of going to war when we should not be, but what can be the even bigger mistake of not going to war – to protect our fellow human beings from catastrophe – when we should be.
When it comes to genocide and all the other atrocity crimes by which we have been horrified (not only in the more distant past but very much over the last fifteen years under the nose of a reborn UN Security Council with no more Cold War excuses to explain its impotence), I for one never again want to be standing before an audience such as this saying "Never Again".