home       biography       publications       speeches       organisations       images       @contact

Peace v. Justice

Guest Lecture* by Prof the Hon Gareth Evans, Honorary Professorial Fellow at The University of Melbourne and President Emeritus of the International Crisis Group, Melbourne Law School, 27 July 2010

Yesterday saw the verdict of the Khmer Rouge tribunal – the Extraordinary Chambers in the Courts of Cambodia – finding guilty of war crimes and crimes against humanity and sentencing to a long prison term, the chief of the Tuol Sleng prison, known as Comrade Duch, who was directly responsible for the torture and execution of over 12,000 prisoners under his charge in the late 1970s. It is a welcome demonstration of how the international movement against impunity for those who commit atrocity crimes or wage wars of aggression continues to gain traction.

Four separate institutional developments have occurred in recent years in international criminal law which have markedly improved the prospect of the trial and punishment of those whose behaviour richly deserves it. For all the continuing practical difficulties in a great many instances of ensuring the apprehension, timely trial and appropriate punishment of indictees or those who should be indictees, there is a strong sense of forward movement: international human rights law is on a roll.

First, there has been the establishment (following the example of the International Military Tribunal set up Nuremberg in 1945) of specialist tribunals to deal with war crimes committed in specific conflicts— in particular for the former Yugoslavia and Rwanda.

Second, there has been the hugely important establishment by treaty – the Rome Statute of 1998, of the International Criminal Court (after surviving a baptism of fire from the Bush administration) — breaking with the tradition of ad hoc war crimes tribunals by setting up a permanent court to hear cases of genocide, crimes against humanity, and war crimes, with no time limitation on its ability to prosecute. The ICC has jurisdiction where a crime is alleged to have taken place on the territory of a state party to the statute, where the accused is a national of a state party, where a country has specifically accepted the ICC’s jurisdiction, and where a case has been referred to the ICC by the UN Security Council or by a state party.

Third, there has been the development of a number of specialist national courts with international assistance, like the Special Court for Sierra Leone which is currently trying Charles Taylor and the Cambodian tribunal, whose next task will be to try the four most senior Khmer Rouge cadres still alive (including Khieu Samphan, one of my key interlocutors when I was negotiating the Cambodia peace process in the late 1980s). Other examples of such courts are the Crimes Panels of the District Courts of Dili and the “Regulation 64” Panels in the Courts of Kosovo; their common characteristic is that, unlike the first generation international criminal tribunals (the Nuremberg and Tokyo Tribunals) and the second (the International Criminal Tribunal for the former Yugoslavia, International Criminal Tribunal for Rwanda, and now the International Criminal Court), they comprise mixed international and local judges and supporting personnel, and apply a compound of both national and international substantive and procedural law.

Fourth, there have been some significant applications in recent years of “universal jurisdiction”: this is available for certain crimes committed under the Geneva Conventions and Protocols, the Convention against Torture, and, under customary international law, for genocide and crimes against humanity, and means that any state party (if it has legislated to give its courts this jurisdiction) can bring to trial any person accused of such crimes, irrespective of any connection of the accused or the crime with the state in question. The prosecution and conviction in 2001 in a Belgian court of Rwandan nuns charged with complicity in the Rwandan genocide was an important demonstration of this option.

It is probably reasonable to now assume that the future of international criminal justice lies with the ICC—and to a lesser extent with the spread of universal jurisdiction in national courts—rather than with new ad hoc tribunals, but there is still a great deal to be done to make the ICC itself effectively operational. Actually exercising the prosecution option in practice before the ICC, or any of the tribunals mentioned, is a lot easier said than done: there are a myriad of procedural and substantive difficulties to be overcome in satisfying the various jurisdictional thresholds and in getting the necessary evidence about the right indictees before the appropriate body. Not the least of the problems is in securing the arrest and physical transfer to the courts of the indictee in question, given that none of the international courts have any kind of marshals service of their own and rely entirely on the cooperation of relevant states for this purpose.

Although such cooperation was recently forthcoming from the Belgian authorities in their important arrest of former DRC vice president Jean-Pierre Bemba Gombo in Brussels in May 2008, following his indictment by the ICC for war crimes and crimes against humanity allegedly committed in the Central African Republic in 2002–03, it was for a decade famously not forthcoming at all in the cases of ICTY indictees Mladic and Karadzic, and to this day Mladic remains at large. And for all the excitement accompanying the initial indictment of Sudan’s President Bashir in 2008, and now the recent decision of the Court to extend the indictment to the crime of genocide (which it had originally resisted), there is not much ground for optimism that he will actually be brought before the court any time soon.


There is a larger policy problem, however, which has dominated all these other considerations when it comes to weighing the utility of attempting criminal prosecutions against those reasonably believed to be guilty of any of the large range of atrocity crimes now within the jurisdiction of the ICC: the peace versus justice problem. Should the demands of justice—to bring an end once and for all to the almost universal impunity that has prevailed in relation to these crimes in the past, and to create an effective deterrent to their commission in the future—ever yield, in the case of a clash between them, to the demands of peace, namely to bring an end to some conflict that has wreaked untold destruction and misery until then and which may continue to do so if a peace agreement cannot be reached?
Despite the position which tends to be taken reflexively by human rights lawyers that peace without justice is no peace at all – that nothing is ever really lost and much gained by ensuring that there is no impunity – my own experience is unquestionably that these demands do in fact clash from time to time, and that hard choices have to be made. Above all, the problem arises when there is an ongoing conflict, and a peace negotiation is attempting to reach agreement between parties capable of perpetuating it – it is not nearly such an issue when one side or another has been clearly defeated, or has been for all practical purposes defeated and is trying to negotiate the terms of a surrender (although a problem can also arise with the competing demands of justice and accountability on the one hand against those of reconciliation on the other, to which I will return). Ongoing conflicts pose a real world policy dilemma, which I experienced in acute form on a number of occasions as President of the International Crisis Group – a human rights lawyer, but one very much in the business of conflict resolution.

Situations just do arise where the need to advance a peace process can work against the impunity principle: as much as it may shock the conscience to contemplate not pursuing prosecutions when major perpetrators of atrocity crimes are involved, this can be helpful in certain circumstances in ending conflict, and in saving as a result a great many more lives. The classic case is Nigeria’s initial grant of asylum to Liberia’s murderous Charles Taylor in 2003, not at all unreasonable given the prospect then looming of thousands more deaths in the final battle for Monrovia. The understandable joy among human rights advocates in the region and around the world when Nigeria subsequently succumbed to international pressure and handed him over, through Liberia, to be tried in the Sierra Leone Special Court should be tempered by the appreciation that this happening, in the absence of any clear evidence that Taylor had breached his asylum conditions, sent a very unhelpful message to some other serial human rights violators. Zimbabwe’s Robert Mugabe, in particular, is obsessed with the Taylor case, as evidence of what might be his own fate if he succumbs to pressure to accept some kind of agreed graceful exit from office (of the kind that has long been on offer from the opposition party) .

But cases do continue to arise where judgements are very finely balanced. Two cases which caused me a good deal of heartache at Crisis Group were the ICC indictments issued in 2005 against Joseph Kony and other leaders of the Lord’s Revolutionary Army (LRA) for atrocity crimes committed in Northern Uganda, and against senior government figures in Sudan for such crimes in Darfur. Strong views were expressed in a number of quarters that these were simply misconceived and a hindrance to the achievement of peace, and that what was required was formal amnesty, or effective immunity in practice, of the kind that has been associated with peace negotiations involving parties of some strength for time immemorial. It was not a matter of being able to fudge the issue, as some would like, and say “negotiate peace now and address the impunity issue later”: potential indictees are far too conscious of their vulnerability these days to settle for that uncertainty.

But at the same time, the arguments for not giving amnesty were stronger in both these cases than some of us in the conflict-resolution business were initially inclined to acknowledge. In the Northern Uganda case, the ICC indictments do seem to have clearly concentrated the minds of at least the lesser LRA commanders, to the extent that there has been a dramatic improvement in the overall security situation since their issue, and do seem to have given Kony himself some sense that the net is closing in. But that said, as of mid 2010, Kony and his associated seem no closer to being arrested, and the conflict no closer to being resolved.

In the case of Sudan, was always a strong argument that moves to prosecute senior regime figures, including President Bashir himself – were needed to pressure the country’s leadership into recalculating the costs of further defying the international community, and that any softer line, trading away justice for the hope of peace, would simply go on being ignored. Although the more optimistic hopes about the utility of the prosecutions in this respect have not yet borne fruit, the fears expressed by others that prosecution would only drive Bashir into a corner and lead to a dramatic re-escalation of the Darfur conflict, or impact strongly adversely on the still very fragile north-south peace process, have not been realized.

There are two important principles that must govern any decision to preference peace over justice.
The first is that justice is the default position, and that it is only in the most exceptional cases, where the evidence really is clear that very major peace benefits are involved, should serious consideration be given to discontinuing investigations under way or granting formal amnesties.

Justice serves too many public policy goals to ever be lightly traded away: retribution (helping channel revenge through institutional rather than freelance channels), incapacitation (physically removing from the scene potential post-conflict spoilers), rehabilitation (giving some hope to offenders that they will have a post-justice future), truth telling (focusing on reality, stripping away myths, and minimizing the prospect of repetition), delegitimization (again, exposing and discrediting), institutionalization of human rights norms, and – of huge importance to those of us in the conflict prevention and resolution business – deterrence (the power of example preventing future misbehavior by others).

All these benefits—and most of all deterrence—are significant over the long term. It is only when the shorter-term costs of prolonging an ongoing conflict clearly outweigh these benefits that non-prosecution of clearly prosecutable cases should be contemplated. The obvious downside risk of these situations is that the more the ICC’s work is perceived as “negotiable,” the more its role as a deterrent of atrocity crimes is undermined: the cases really do have to be very exceptional.

The second principle is that if decisions to give primacy to peace over justice do have to be made in certain hard cases, those decisions are best made not by the ICC or its prosecutor but by those with appropriate political responsibility: in the case of this court, the Security Council has that power, if it chooses to use Article 16 of the Rome Statute enabling it to suspend prosecutions for renewable periods of twelve months. The prosecutor’s job is to prosecute, and he should get on with it, with bulldog intensity. His task is to end impunity for the worst atrocity crimes: of course, given that resources are usually scarce and potential targets many, he must retain a discretion, as to how many potential indictees to pursue at any given time, and as to which crimes to give priority focus. And Article 53 gives the prosecutor a certain formal discretion not to pursue matters if the “interests of justice” so require. But the interests of justice do not necessarily coincide with the interests of peace. Giving the prosecutor the primary role in resolving the peace v. justice dilemma, making the determination as to when and how to weigh the demands of conflict resolution, puts him in an impossible situation. So he has to get on with justice.

If the judgment has to be made, on occasion, that the interests of peace should override those of justice, as I have suggested it will, then that should be not for the ICC but the Security Council to decide—however difficult that will no doubt prove to be in practice—and the pressure and weight of expectations should be taken off the prosecutor’s shoulders in this respect. There is a reluctance—particularly by those who come at these issues from a human rights rather than conflict resolution perspective—to give so central a role to Article 16, and this general balancing role is probably more than was originally contemplated for it. But the international community has to recognize that because there are competing principles of more or less equally compelling moral force involved here, there has to be some mechanism for accommodating them, and that Article 16 seems to be the best available option.


A variation of the peace v. justice problem in the case of ongoing conflicts is the issue of accountability v. reconciliation in the case of conflicts that have reached some stage of resolution. There is the same potential for legal purists to take the position that justice must be pursued come what may, and the same need in practice to make careful case-by-case judgements based on the characteristics of each particular situation.

In post-conflict societies trying to rebuild themselves after mass atrocity crimes have been committed (often, as in the case of Rwanda, by very large numbers of people, not just isolated psychopaths), both national leaders and internationals trying to advise and assist them face hugely difficult dilemmas in deciding how, on the one hand, to ensure accountability for those crimes but, on the other, somehow to achieve social reconciliation. In one society investigations of mass murder and prosecutions of those responsible for it in national, international, or hybrid courts may well be thought to contribute to a sense of political catharsis that clears the air, relaxes tensions, and opens the door to restorative justice. But in another it may be seen as increasing instability and deepening hostility among adversarial groups: Mozambique and Namibia are among many examples where reconciliation was pursued to the exclusion of accountability.

Between the polar extremes of emphasizing retributive justice through trial and punishment, and hoping that time itself will be sufficient to heal the wounds and memories of past atrocities, there lie several other options, a number containing elements of both approaches.

One is simply postponement, delaying the day of reckoning—as essentially Chile and Argentina did, and to some extent also Cambodia, until the society was felt to be strong and cohesive enough to face its past and itself impose accountability for those crimes without fear of destabilizing consequences.

A second is lustration, the barring, by administrative rather than judicial process, of a whole class of individuals from public employment, political participation, and the enjoyment of other civil rights—as with denazification in Germany and demilitarization in Japan (for both of which there was an administrative appeal apparatus), and debaathification in Iraq (where there was no such process). This can work where the class in question is clearly defined and comprehensively discredited, with its members all evidently sharing some moral responsibility, but it tends to be counterproductive, generating resentment and resistance, where, as in Iraq, these conditions are not met. A more selective version of this approach may be described as vetting, as for example in Bosnia where, after the Dayton Peace Accords, many members of military, paramilitary, or police forces responsible for serious violations of basic rights of ethnic or minority groups were dismissed from service; or in the Czech Republic, Lithuania, and post-unification Germany, where administrative purges temporarily removed those affiliated with past abuses from certain positions in the public sector, with a particular emphasis on those who were alleged to have collaborated with the former secret police.

The third option is a truth commission, premised in its pure form on the notion of fixing responsibility without attendant punishment, in effect amnesty in return for full disclosure (thus addressing key accountability objectives such as truth telling and delegitimization of state-sponsored violence). In the familiar South African Truth and Reconciliation Commission, there was a heavy additional emphasis on forgiveness as a key to reconciliation and less in practice on the truth for amnesty trade-off . In fact almost no very senior figures from the apartheid regime appeared nor were any prosecuted for their failure to do so, evidently because such prosecutions could have derailed the fragile transition.

Truth commissions and court processes can also be seen as complementing each other rather than being somehow in competition. In Argentina the significant amounts of information produced by the truth commission established in 1983 were then utilized by the authorities in prosecuting members of the military junta that had ruled the country. In Sierra Leone both a truth commission and a special court were established, and they were held simultaneously for a time. Certainly trials as well as truth commissions may be pictured as vehicles of emotional expression and cathartic transformation: it has been argued that trials are beneficial because (among other things) they can express the community’s abhorrence of the atrocities committed and because they “can placate a victim’s desire for vengeance.”
A fourth option is to provide reparations to victims of human rights violations, which might be either substantive (compensatory or restitutive) or symbolic. The restitutive example most often cited, and frequently touted as a success, is of the Canadians and Americans of Japanese descent who were interned during the Second World War; a good example of the symbolic approach is closer to home – the apology offered to the indigenous peoples of Australia on behalf of all Australians by Prime Minister Kevin Rudd in 2008.

A fifth option for dealing with post-conflict accountability and reconciliation issues is to rely on traditional justice mechanisms—informal tribal, local, grassroots, and/or village-level justice systems that owe nothing to European-derived state-level justice. In some contexts traditional justice may be a viable complement, or alternative, to conventional state-administered justice systems, given that establishing conventional forms of governance in post-conflict settings, or in countries where states are fragile, is complex and time consuming.
Rwanda provides an example of efforts made to reinvent a form of traditional justice. In an attempt to address in a manageable way the massive number of crimes committed during the 1994 genocide, the government reintroduced in 2002 a form of traditional justice known as “gacaca”, whereby a panel of lay judges coordinates a process in which survivors and arrested suspects confront each other without lawyers, and the community gives testimony, drawing on traditional, rural conflict resolution mechanisms normally used to settle minor village disputes. But applying this process to crimes of the scale involved here—where those found guilty of murder can receive sentences of up to twenty-five years (although in practice most have been for much shorter terms or commuted) —poses some real problems, in that the accused are denied legal representation, many judges are illiterate and have little legal knowledge, and the system’s reliance on voluntary confessions encourages self-incrimination, and the whole process remains quite controversial.

A British project to restore Sierra Leone’s rural paramount chiefs in 2000, in the aim of stabilizing the post-conflict situation there and encouraging displaced people to resettle in their villages, pointed up another potential hazard of attempts to revive traditional justice. In strengthening the customary authority of paramount chiefs—considered a major source of 1990s conflict—the conditions for renewed conflict may have been inadvertently recreated. As the International Crisis Group has pointed out in Liberia and Sierra Leone, the at-first- sight attractive reliance on the traditional may have missed an opportunity to encourage citizens to reconsider the modes of domination and governance that had been among the initial causes of conflict.
These considerations are by no means applicable just to Africa, or only in a post-conflict context. In a report on Pakistan’s judiciary system, for example, Crisis Group argued that traditional tribal justice systems, while speedier and less expensive than formal courts, were more open to abuse and subjective, coercive forces, and unlikely to deliver justice. Women fared especially poorly under traditional justice mechanisms that condoned or prescribed punishments such as gang rape and honour killing. Since informal dispute resolution mechanisms, involving tribal or village-level councils, were unlikely to deliver justice, conventional courts (in this case district and civil courts in the subordinate judiciary) would need to remain a critical part of dispute resolution strategies.
The key consideration for those wrestling with these options is that it must ultimately be a matter for the people of the distressed country itself to decide which, or what combination, they choose. Certainly there must be a strong presumption in favor of conducting whatever process is decided upon on local soil. As David Crane has argued in the case of the Special Tribunal for Sierra Leone:
The tribunal is set up for the victims, their families, towns, and districts. At the end of the day they are the ones who will have to live with the result and to try to put their destroyed lives back together in some small semblance of order. A tribunal that loses that focus will drift into history under a very dark cloud indeed. That is why in the future tribunals need to be where the horrors took place, in the middle of the crime scene. The people need to see their tribunal in action.


It is impossible to offer a definitive, one-solution, answer to the dilemmas I have been describing. There just often are inherent tensions, and agonizing choices involved, between the demands of justice and those of peace, and the demands of accountability and those of reconciliation, just as there famously are between justice and equality, or freedom and equality and other competing pairs of moral principles.

The only way through the morass, I believe, is to treat justice, accountability and no-impunity as the default positions – but to recognize that they are just that, not absolute, and that in the real world cases will arise where the arguments for making exceptions really are compelling. Human rights lawyers should not feel that they are letting the side down if they accept that. Those of us in the conflict prevention and resolution business, and those of us in the business of hunting down and punishing the guilty, all ultimately want the same thing: an end to violent conflict and the horror and misery of war and mass atrocity crimes, and to ensure the dignity and common humanity of our fellow human beings. Getting there will require open minds on both sides of the debate and, above all, a willingness to listen carefully to what those most hurt by the conflicts and crimes in question are trying to tell us. This is one area, among many in international affairs, where outsiders – whatever their qualifications and experience – should never be over-confident that they know best.


*The text of this lecture is drawn substantially from Gareth Evans, The Responsibility to Protect: Ending Mass Atrocity Crimes Once and For All (Brookings Institution Press, 2008), Chs 4, 5 and 7.