On 4 June 2007 Stephen Rapp, the Prosecutor of the Special Court for Sierra Leone, described the civil war in that country as among ‘the ugliest scenes of viciousness in recent memory’:
Human beings, young and old, mutilated. Rebels chopping off arms and legs, gouging out eyes, chopping at ears. Girls and women enslaved and sexually violated. Children committing some of the most awful crimes. The exploitation of the resources of Sierra Leone used not for the benefit of its citizens but to maim and kill its citizens. The very worst that human beings are capable of doing to one another.Footnote 1
This was the opening statement in the trial of Charles Taylor, the former president of Liberia. Although Taylor had never set foot in Sierra Leone, he was accused of having supported the Revolutionary United Front (RUF), a rebel group that had committed these atrocities to control the country's diamond resources. The trial – which is the first of an African head of state before an international tribunal – has been hailed as an important step in ending the culture of impunity in which tyrants and rebel leaders believe they will never be held accountable for their crimes. As Rapp noted in an interview with the Christian Science Monitor, ‘The world has turned a page in the wake of Taylor's arrest . . . The days are gone when leaders accused of atrocities could escape into exile.’Footnote 2
An attempt to turn that page had been made four years earlier by David Crane, the first Prosecutor for the Special Court, when he unsealed the indictment of Taylor while the latter was still president of Liberia. He had just arrived in Accra for a meeting with other west African leaders in an attempt to negotiate an end to Liberia's civil war by persuading him to step down and accept asylum in Nigeria. Crane's goal was to put the participants on notice that they should not negotiate with a war criminal but, instead, extradite him to the Special Court. His hosts, however, were committed to the negotiations and were unwilling to hand over a fellow head of state. Not taking any chances, Taylor returned to Liberia as soon as he learned of the indictment, triggering the collapse of the peace talks.Footnote 3 While Taylor's departure was secured two months later, an opportunity to end the violence earlier was effectively vetoed by the Prosecutor. Roughly 1,000 people died in stepped-up political violence between the breakdown of the talks in Accra and Taylor's eventual resignation.Footnote 4
These two episodes in the Taylor case illustrate one of the central conflicts confronting the International Criminal Court (ICC), namely the tension between the legal goal of enforcing the rule of law to end impunity and the political requirements of negotiating an end to armed conflicts. Should the Prosecutor hold back from criminal proceedings if he is persuaded that prosecution could interfere with negotiated transitions? Article 53 of the ICC's founding Rome Statute allows the Prosecutor to exercise his discretion not to investigate or prosecute if it would not serve ‘the interests of justice’. Should that phrase be construed sufficiently broadly to include the interests of peace?
To many of the Court's strongest supporters among non-governmental organizations (NGOs) and the international legal community, the answer to that question is no, although some qualify that ‘no’ more than others. Two of the more systematic expositions of this position are policy papers drafted by Human Rights Watch and Amnesty International.Footnote 5 Both were issued in response to indications from the Office of the Prosecutor (OTP) that it might use Article 53 to delay criminal proceedings against the leaders of the Lord's Resistance Army (LRA) for atrocities committed in northern Uganda if they interfered with peace negotiations designed to end a two-decade civil war. Even though the arrest warrants were eventually issued, both papers objected that this statutory construction of prosecutorial discretion was contrary to the ICC's duty as a legal institution to prosecute those most responsible for the gravest international crimes – a view that the OTP has now officially endorsed.Footnote 6
The arguments presented by the two NGOs are part of a broader legalist tradition that seeks to insulate international law from international politics in order to build ‘ethical legal processes through the creation of institutions capable of promoting due process and the rule of law’.Footnote 7 When applied to the ICC, this means that narrowing the Prosecutor's discretion to preclude policy considerations will better enable the Court to act on a duty to prosecute the worst international crimes. Proponents of this view, including the two NGO papers, also make a policy argument for acting on this duty, challenging the claims of ‘pragmatists’Footnote 8 who warn against the destabilizing consequences of a rigid legalism. Long-term stability, they claim, is more likely to come from an uncompromising approach to criminal justice, in terms of both deterring gross human rights abuses and consolidating transitions to peace and democracy.
This article challenges the empirical premises that underlie the policy argument against factoring peace processes into prosecutorial discretion. Legalists assume that law can and should be separated from politics, and that in doing so, it can transform politics. This reverses the actual relationship between politics and law evident in the history of international war crimes tribunals and the cases currently under review by the ICC. Those episodes demonstrate that political factors – most notably the power of the perpetrators relative to the forces arrayed against them and the political strategies of the latter to address the conflict – determine when a criminal law approach is effective and whether it contributes to peace. Hence the Prosecutor should construe his discretion broadly in order to assess the political context in which international criminal law has to operate.
1. The Rome Statute: can the Prosecutor exercise discretion in the interests of peace?
The establishment of the ICC is the most ambitious step the international community has taken in introducing criminal accountability into the culture of international relations. Its founding Rome Statute was negotiated in July 1998, building on the precedents of the UN-created International Criminal Tribunals for the former Yugoslavia (ICTY) and for Rwanda (ICTR) by creating a permanent court that would have its own legal personality independent of the United Nations. The Court currently has jurisdiction over genocide, crimes against humanity, and war crimes committed after the treaty came into force on 1 July 2002. It can exercise universal jurisdiction if the Security Council refers a case. Otherwise, ICC jurisdiction requires some nexus to state consent, either through ratification of the Rome Statute, which subjects the nationals and territory of states parties to ICC oversight, or the voluntary consent of non-parties to the ICC's jurisdiction on an ad hoc basis. One of the ICC's most distinctive features is the principle of complementarity. Unlike the UN-created tribunals for the former Yugoslavia and Rwanda, primary jurisdiction resides with national courts and the Prosecutor must defer to them if they are doing their job. The ICC can only step in when national systems of justice are unwilling or unable to investigate or prosecute.
Does the ICC Prosecutor have a duty to prosecute crimes of sufficient gravity within the jurisdiction of the Court if national criminal justice systems do not? This is an important question, because states have often chosen alternatives to prosecution in a transition from dictatorship or armed conflict, particularly when those accused of criminal violence retain significant power and negotiation is the most viable strategy of political change. As a result, the leaders of abusive regimes and rebel movements are often granted formal or de facto amnesties or allowed to accept exile abroad in order to advance the bargaining process. These instruments are often accompanied by non-penal forms of justice, such as truth commissions, reparations, or lustration.Footnote 9 The best-known example is South Africa's Truth and Reconciliation Commission (TRC), under which amnesty was conditioned on the public confession of political crimes committed by all sides.Footnote 10 Amnesties also played a crucial role in several UN-brokered peace agreements, such as Mozambique, El Salvador, and Haiti.Footnote 11
Even without formal amnesties, conflict resolution often involves subordinating trials to expedient bargaining when the leaders whom a prosecutor might want to put in the dock are still in power and their co-operation is necessary to end violent conflicts.Footnote 12 This was the US-led NATO strategy in engaging the Serbian president Slobodan Milošević to negotiate the Dayton Accords which ended the war in Bosnia and Herzegovina.Footnote 13 It was also the premise behind inviting Charles Taylor to the peace talks in Accra.
Should the Prosecutor hold back from criminal proceedings if the parties believe that alternatives to criminal justice are necessary for a transition from repressive rule or armed conflict? Should he make an independent judgement of their political legitimacy or assess whether they are necessary for peace processes or democratization?
The text of the Rome Statute creates a presumption against making such determinations. Its preamble lays out the ICC's mission, which is to ensure prosecution of ‘the most serious crimes of concern to the international community’ in order to ‘put an end to impunity for the perpetrators of these crimes and thus to contribute to the prevention of such crimes’. Its only reference to peace is that ‘such grave crimes threaten the peace, security and well-being of the world’. The implication is that peace is more likely to come from a consistent policy of prosecution than from deferring prosecution to political negotiations. As Darryl Robinson notes, ‘the very purpose of the ICC was to ensure the investigation and punishment of serious international crimes, and to prompt states to overcome the considerations of expedience and realpolitik that had so often led them to trade away justice in the past’.Footnote 14
A few commentators have suggested that the Rome Statute's complementarity provisions could be used to defer to an amnesty or alternative justice mechanisms. Under Article 17(1)(a), a case is inadmissible unless a state with jurisdiction ‘is unwilling or unable genuinely to carry out the investigation or prosecution’. In theory, a genuine investigation need not be a criminal one, allowing the Prosecutor to defer to arrangements like the TRC.Footnote 15 Some states, such as South Africa, advocated this position in Rome, and the United States circulated a ‘non-paper’ using the TRC as a model for allowing the prosecutor to defer to an amnesty if it was democratically ratified and necessary for peace and reconciliation.Footnote 16 This proposal, however, generated strong opposition from NGOs and many of the strongest state supporters of the ICC because of the role amnesties had played in perpetuating impunity for the former military dictatorship in Chile and elsewhere.Footnote 17 Agreement on compromise language was impossible and, as a result, amnesty is nowhere mentioned in the Rome Statute.
Nonetheless, the complementarity language agreed to at Rome is more consistent with a duty to prosecute. Article 17(2) defines ‘unwillingness’ as (a) a national process ‘made for the purpose of shielding the person concerned from criminal responsibility’; and (b) an ‘unjustified delay . . . inconsistent with an intent to bring the person concerned to justice’. Even though the TRC's amnesty was conditioned on truth-telling and was ratified by South Africa's first multiracial majoritarian parliament, there was no intention to bring to trial those perpetrators who publicly confessed political crimes. Hence a strict reading of Article 17 would not distinguish the TRC from less politically legitimate amnesties, such as the one that the Pinochet dictatorship granted itself prior to the return of constitutional rule in Chile.Footnote 18
The only provisions that can be plausibly construed as leaving open the possibility for making such distinctions are the rules governing prosecutorial discretion in Article 53. Article 53(1)(c) allows the Prosecutor to decline to investigate crimes that satisfy the Rome Statute's jurisdictional and admissibility criteria if, ‘[t]aking into account the gravity of the crime and the interests of victims, there are nonetheless substantial reasons to believe that an investigation would not serve the interests of justice’. Under Article 53(2)(c), the Prosecutor can decline to move from investigation to prosecution if it ‘is not in the interests of justice, taking into account all the circumstances, including the gravity of the crime, the interests of the victims and the age or infirmity of the alleged perpetrator, and his or her role in the alleged crime’. The phrase ‘interests of justice’ is not defined in the Rome Statute. Some commentators interpreted it as a form of ‘creative ambiguity’ which could encompass alternative justice mechanisms such as the TRC.Footnote 19 Others argued that it was intended to grant the Prosecutor broad political discretion to ‘arbitrate between the imperatives of justice and the imperatives of peace’.Footnote 20 While the Prosecutor is required to justify his decision to the Pre-Trial Chamber, which would review and possibly reverse it, it theoretically provides the Prosecutor with a means of holding back from criminal proceedings or considering alternative justice mechanisms when demanding prosecution might prolong an armed conflict or dissuade a tyrant from stepping down.
The OTP confronted this kind of situation in early 2005, just as it was ready to ask the Pre-Trial Chamber to issue arrest warrants for Joseph Kony and the leadership of the LRA, which has abducted more than 20,000 children as foot soldiers and sex slaves, and is responsible for large-scale atrocities, primarily against the Acholi population of Northern Uganda. The Court accepted the case on 28 June 2004, following a referral from the Ugandan president, Yoweri Museveni, on 16 December 2003. That decision was publicly criticized by Betty Bigombe, a former Ugandan minister and World Bank consultant, who was attempting to engage the LRA in peace talks. Bigombe and several Acholi community organizations argued that indictments would dissuade the LRA leadership from participating in negotiations. A better strategy, they argued, was to offer amnesty in exchange for demobilization and to employ traditional non-punitive reconciliation rituals to reintegrate the LRA into the community.Footnote 21 Between March and May the ICC Prosecutor, Luis Moreno-Ocampo, met representatives of these organizations and initiated several missions to Northern Uganda to interview victims and civil society groups. The Prosecutor suggested that the indictments could be delayed to give the peace process a chance. In doing so, his office cited Article 53, indicating that it might suspend its investigation if ‘it is in the interests of justice to proceed with a peace settlement’, although it also made clear that this was not the same thing as immunity or a blanket amnesty.Footnote 22 Nonetheless, arrest warrants for Kony and four of his commanders were applied for in May, issued in July, and unsealed in October.Footnote 23
2. The duty to prosecute and the case for narrow prosecutorial discretion
The suggestion that Article 53 could be used to defer to peace negotiations provoked a sharp challenge from many of the human rights lawyers and NGOs that have been most supportive of the Court. Human Rights Watch and Amnesty International each drafted policy papers that argued that the Prosecutor's duties under the Rome Statute required a narrow construction of the ‘interests of justice’ test. Citing the Vienna Convention on the Law of Treaties, Human Rights Watch argued that the phrase can only be understood in the light of the ‘object and purpose’ of the Rome Statute, which is to end impunity by holding perpetrators criminally accountable, not to assist peace negotiations.Footnote 24 Hence ‘the prosecutor may not fail to initiate an investigation or decide not to go from investigation to trial because of developments at the national level such as truth commissions, national amnesties, or the implementation of traditional reconciliation methods, or because of concerns regarding an ongoing peace process’.Footnote 25
Should the prosecutor decide otherwise, he would be making political judgements about peace negotiations or the legitimacy of alternative reconciliation methods, which are inappropriate for a judicial institution whose mission and expertise are in international criminal law. Amnesty argues that the consultations necessary to make such determinations would violate the Prosecutor's legal duties, citing Article 42(1) on the independence of the Prosecutor, who ‘shall not seek or act on instructions from any external source’.Footnote 26 Both reports also warned that adopting an expansive view of discretion would set a dangerous precedent, subjecting the Court to manipulation by warring factions and interested parties. This, in turn, would ‘undermine the perception and reality of the prosecutor as independent and beyond political influence’.Footnote 27
Deferring prosecution to political negotiations would be incompatible with the Court's legal obligations not only under the Rome Statute, but also more generally under international law. Most of the crimes subject to the ICC – genocide, crimes against humanity, grave breaches of the Geneva Conventions, torture – involve an international duty to prosecute as the result of treaty or custom that pre-dates the Rome Statute.Footnote 28 Human Rights Watch notes that this has resulted in a growing trend in international law to reject amnesties for such crimes, citing as evidence the reservation from the UN Special Representative to the 1999 Lomé Peace Accords, that there would be no international recognition of the blanket amnesty granted to all parties in the civil war in Sierra Leone if they involved serious international crimes – a marked contrast to the United Nations’ approach to peace negotiations earlier in the decade.Footnote 29 In such cases a decision to hold back from prosecution would place the Court outside the very legal developments that gave rise to the drafting of the Rome Statute in the first place.
Finally, such an approach would be an abrogation of a duty to the victims, for whom the Court was created. Human Rights Watch notes that Article 53(2)(c) mentions the ‘interests of victims’ in the same phrase as the ‘interests of justice’, and that the latter should only be understood in terms of the former's need for criminal justice.Footnote 30 This would preclude amnesties, which are incompatible not only with international law but also with the rights of victims. As Ben Chigara put it in his moral critique of amnesties,
They treat victims as if they did not have predetermined rights at the moment of abuse . . . what matters is whether there is sufficient threat of disruption of the incoming government's agenda by those that committed the alleged crimes against humanity.Footnote 31
The Court would also be defaulting on its duty to victims if it merely suspended criminal proceedings. As the Amnesty letter put it,
The suspension would create a sense of helplessness as the one court of last resort when no state was able or willing to investigate the most horrendous crimes informed them that it had suspended indefinitely the investigation that had given them hope of justice, truth, and full reparations.Footnote 32
The only approach consistent with a duty to victims is to act in line with the old adage, ‘justice delayed is justice denied’, and ignore the political consequences of that choice.Footnote 33
Both papers concede that if the international community is genuinely concerned about the impact of prosecution on peace negotiations, the only legitimate place to make that call is the Security Council, as prescribed by the Rome Statute. Under Article 16 the Security Council has the authority to suspend any investigation or prosecution for renewable 12-month periods through passage of a resolution under Chapter VII of the UN Charter. Both NGOs opposed this provision at Rome, fearing that it would serve as a conduit for politicization, and their policy papers continue to express reservations about its use. Nonetheless, Amnesty argues that the decision to subordinate prosecution to peace must be made by ‘a political power that the drafters intended to be exercised only by a political body’.Footnote 34 The logic, as one scholar put it, is that ‘political institutions should do politics and policy; judicial institutions should do justice’.Footnote 35
Moreover, pursuing justice untainted by politics is not just a matter of legal duty; it also has superior policy consequences in terms of making and consolidating peace. First, legalists challenge what Amnesty calls the ‘false premise that international justice was incompatible with political negotiations to end armed conflicts’.Footnote 36 Despite warnings from the UN Secretary-General and from international mediators, the ICTY's indictments of Bosnian Serb President Radovan Karadžić and General Ratko Mladić in July 1995 did not derail the successful conclusion of the Dayton Peace Accords. Nor did the indictment of Slobodan Milošević in the middle of the Kosovo war prevent a settlement that allowed the refugees to return to their homes.Footnote 37
Second, as the Human Rights Watch paper observes, justice can ‘have a tremendous value in contributing to peace and stability’ by stigmatizing disruptive actors, such as Milošević or Taylor, thereby ‘undermin[ing] the political weight of those individuals and marginaliz[ing] them from political life in their former countries’.Footnote 38 This is likely to happen even if those indicted are not apprehended, as was the case with Karadžić and Mladić. Richard Goldstone, the ICTY Prosecutor who obtained those indictments, subsequently wrote that they made a positive contribution to the peace process, by making it ‘legally and politically possible for the international community to insist on excluding Karadžić from the Dayton peace talks’, without which the Bosnian government would not have participated. They also isolated the two most virulent ethnic extremists from post-war politics, thereby minimizing their disruptive influence on the implementation of Dayton.Footnote 39
Third, criminal justice can contribute to peace by individualizing guilt in criminal leaders rather than allowing the victims to collectivize it in entire groups. This, in turn, is necessary to break the cycle of violence and revenge that keeps many ethnic conflicts going, particularly if it assigns guilt to criminal behaviour on all sides. This is why Antonio Cassese, the first president of the ICTY, referred to the tribunal as an ‘instrument of reconciliation’ rather than a ‘means of revenge’.Footnote 40
Finally, anti-impunity advocates see criminal accountability as necessary to establish the rule of law and deter a return to political violence in post-conflict societies. ‘Impunity for atrocities committed in the past’, by contrast, ‘sends the message that such crimes may be tolerated in the future’.Footnote 41 Some proponents of the duty to prosecute contend that the decision to negotiate with Milošević at Dayton, despite his complicity in ethnic cleansing campaigns in Croatia and Bosnia and Herzegovina, emboldened him to believe that he could return to criminal violence in Kosovo three years later without any legal repercussions.Footnote 42 The Lomé amnesty, which included a power-sharing agreement that made RUF leader Foday Sankoh vice-president and minister of mines, produced a similar outcome in Sierra Leone. In less than a year the RUF violated the agreement by attacking UN peacekeepers and taking them hostage, thereby demonstrating the tenuousness of peace without justice.Footnote 43
3. The consequentialist case for broad prosecutorial discretion
The consequentialist case for a duty to prosecute is most persuasive in a relatively stable post-conflict environment in which alleged war criminals are no longer in power and can be apprehended without a serious risk of violent backlash. It is on shakier ground during ongoing conflicts or fragile peace processes in which those accused of war crimes still command substantial power. Some of the advocacy by NGOs and legal scholars seems to imply that in such situations pursuing justice over impunity is a matter of making the right legal choice regardless of the political choices used to address existing power realities. Leila Sadat suggests such an approach when she uses the Balkan and Sierra Leone cases to argue against subordinating justice to realpolitik in peace negotiations:
Examples such as Milošević and Sankoh suggest that granting impunity, rather than definitively settling a conflict, simply encourages a resurgence of the criminal behavior. If one of the most important purposes of the criminal law is to remove dangerous individuals from society, it suggests that Slobodan Milošević and Foday Sankoh should have been indicted and tried years ago.Footnote 44
Note the implication that the key to ‘definitively settling a conflict’ is a more aggressive judicial strategy, an assumption made explicit in the Amnesty letter, which argues that such an approach will ‘likely lead to the internal and international isolation, marginalization and eventual removal from power of those indicted’.Footnote 45 However, the alternative to engaging criminal leaders is the deployment not of law, but of countervailing power, probably involving the use or threat of force. This is because the real source of impunity in places like Bosnia and Herzegovina or Sierra Leone is the perpetrators’ belief that the military balance of forces enables them to impose their will without meaningful resistance. That can only be challenged through the military power of internal forces or, failing that, through some form of international coercion or intervention. International criminal law can build on the capability and willingness of political actors to wield such instruments, but it cannot help settle a conflict without them.
In Bosnia and Herzegovina, for example, the Security Council authorized the creation of the ICTY in 1993 at time when NATO and the United Nations were unwilling to deploy anything more than a neutral peacekeeping force with no mandate to stop the violence. Even when that mandate was augmented to enforce ‘safe areas’, there was a reluctance to implement it because of the risks for peacekeepers it would impose.Footnote 46 In such a political vacuum it is unlikely that a more aggressive judicial strategy could have had a meaningful impact on ethnic cleansing. In fact, the worst single atrocity of the Bosnian war – the Srebrenica massacre – was perpetrated by forces under the command of General Mladić two years after the creation of the ICTY and three months after Goldstone had asked the Bosnian courts to defer to his investigation of Mladić for crimes against humanity.Footnote 47 The use of criminal violence with impunity was effectively challenged only after NATO was willing to use force – both directly through Operation Deliberate Force and indirectly through assisting Croatian and Bosnian military offensives – in order to convince Belgrade to rein in its Bosnian Serb allies.Footnote 48 This, in turn, was a prerequisite for the ability of the ICTY to prosecute anyone of significance and make a contribution to the peace process by removing criminal spoilers from the political scene.
In Sierra Leone it is clear that engaging the RUF was a mistake, as was releasing Sankoh from prison to lead the RUF delegation in Lomé after he had been prosecuted in Freetown and sentenced to death for treason. However, keeping Sankoh in prison and issuing indictments of rebel leaders still at large would not have ended the RUF's reign of terror since it had not been defeated on the battlefield. More importantly, Nigeria, which led the West African peacekeeping force that had kept the RUF at bay, was no longer willing to keep its troops in Sierra Leone and the United Nations was only willing to replace them with a neutral peacekeeping force.Footnote 49 Therefore the alternative to peace with amnesty was not peace with justice, but the continuation of the civil war – without foreign military assistance – against a rebel group that was still being armed by Charles Taylor's Liberia. Impunity was a symptom of these political realities that no justice and accountability mechanism could have erased. And as in Bosnia and Herzegovina, it only ended with military intervention, in this case from the United Kingdom, in response to a plea from UN Secretary-General Kofi Annan after the RUF had returned to criminal violence and taken 500 UN peacekeepers hostage. This strengthened what had been a neutral (and up to that point, ineffectual) peacekeeping operation into an enforcement mission that eventually defeated the RUF.Footnote 50
Moreover, the feasibility of prosecution and its contribution to peaceful transitions are dependent upon the political or military strategies designed to bring a conflict to an end. If the perpetrators’ forces have been defeated or weakened to a point where negotiations are unnecessary, criminal leaders could be put on trial without a serious risk of violent backlash. This was the historical experience after the Second World War with the Nuremberg and Tokyo War Crimes Tribunals, which were made possible by the unconditional surrender of Nazi Germany and imperial Japan.Footnote 51 In Rwanda, the defeat of the genocidal Hutu regime by the Rwandan Patriotic Front meant that the architects of the genocide could face national and international prosecution without destabilizing the post-conflict transition. Similarly, the RUF's surrender to a more robust international force made possible the creation of the Special Court for Sierra Leone and the prosecution of RUF leaders, who were no longer in a position to disrupt a peace settlement. Prosecutions under these circumstances entail the risk of partisan justice in which the losers are prosecuted and the victors are immune.Footnote 52 Nonetheless, defeating criminal actors, or weakening them to the point where their co-operation is not needed to end a conflict, is a prerequisite for discounting the potential impact of criminal justice on peace.
If, by contrast, the perpetrators are not defeated and retain significant power, conflict resolution will have to be based on a bargaining paradigm that involves some compromises with criminal justice. This most clearly applies when the United Nations or other mediators adopt an impartial Chapter VI approach to conflict resolution. As explained by Haile Menkerios, the former head of the African division of the UN Department of Peacekeeping Affairs,
So many leaders of these conflicts have committed abuses, crimes, and the killing and the suffering of innocent civilians continues as long as the conflict continues. So you have two choices. You can consider them criminals, bring them to face justice, and get them out of the way, or invite them to negotiate for a peaceful settlement. When there is neither the internal capacity nor the external will to do the first, there is no other choice but to invite them to negotiate, often for power sharing in a transitional arrangement to stop the wars.Footnote 53
These arrangements are often problematical in terms of human rights, since those implicated in war crimes are allowed to enter the political arena. If, however, the international community is unwilling or unable to take enforcement actions against them, then there is no choice but to engage them in peace negotiations, and it is difficult simultaneously to criminalize them and to try to solicit their voluntary co-operation.
That conflict would still exist, although it would be mitigated, if the strategy of conflict resolution moved beyond neutral mediation to coercive diplomacy, since its purpose is to use pressure to change a perpetrator's behaviour. In Bosnia and Herzegovina, for example, NATO's use of force was not designed to defeat the Serbs the way international forces had defeated the RUF. Rather, it was to convince Milošević that he was overextended, that time was not on his side, and that it was in his interest to end the war. A central feature of the negotiating strategy of US envoy Richard Holbrooke was to bypass the Bosnian Serb leadership – whom he characterized as ‘useless interlocutors’ who had reneged on every commitment made to international mediators – and rely on Milošević to deliver them.Footnote 54 Goldstone's indictment of Karadžić and Mladić assisted this strategy. Indicting Milošević, by contrast, would likely have had the same impact on Dayton that unsealing the arrest warrant for Charles Taylor had on the collapse of the Accra talks.
The reason why indicting Milošević was impractical in 1995 was because his co-operation was necessary both to negotiate and to maintain Dayton, given the political–military constraints under which NATO and the United Nations were operating. This changed with the Kosovo war in 1999. When Belgrade did not withdraw after the first few weeks of bombing, the United States and NATO concluded that Milošević was not only no longer the key to the peace process, he was now the main source of instability in the region.Footnote 55 Moreover, unlike Dayton, NATO's war aims did not require Belgrade's continuing co-operation to stabilize the post-conflict environment. Given NATO's political strategy, the ICTY's indictment of Milošević – which was encouraged by US and UK officials, who released previously classified information to the ICTY ProsecutorFootnote 56 – did not interfere with NATO's strategies of war termination and post-conflict nation-building.
4. Peace-versus-justice dilemmas facing the International Criminal Court
The lesson that should be drawn from the cases described above is that the feasibility of prosecution and its impact on peace are shaped and constrained by the political strategies designed to end a conflict. This is relevant to the first three situations under investigation by the ICC, since each is taking place in the context of ongoing political violence, creating tensions between international criminal justice and conflict resolution. As a result, the Prosecutor will need to find some means of injecting political prudence into his discretion in order to be effective without foreclosing negotiated solutions or disrupting fragile peace processes.
4.1. Northern Uganda
The LRA case has presented the ICC with its most overt controversy regarding the impact of prosecution on peace. Its strongest critics have been Acholi civil society leaders, who opposed its involvement from the start because they feared it would drive the LRA from peace talks.Footnote 57 When negotiations resumed in the south Sudanese capital of Juba in July 2006, they asked the Prosecutor to withdraw the arrest warrants, which were seen as an obstacle to their completion – a view that persisted even after the Juba process collapsed in April 2008 when Kony refused to sign the peace accords.Footnote 58 Their alternative was to extend the 2000 Amnesty Act to the indicted LRA leaders if they agreed to end the war. Instead of prosecution, they advocated a traditional reconciliation ritual, the mato oput, in which a wrongdoer drinks a potion made from a bitter root, confesses to his victims, and makes amends.Footnote 59
The reason why most Acholi leaders have prioritized peace over prosecution is because the war has been the principal source of humanitarian problems in the region. When the arrest warrants were unsealed in October 2005, 1.7 million people, or roughly 90 per cent of the population in the three most war-ravaged districts, had been relocated to internally displaced persons (IDP) camps, where the Uganda People's Defence Force (UPDF) have provided inadequate protection from the LRA. In addition, local and international NGOs have documented military abuses against civilians, including torture, rape, arbitrary detention, and extortion.Footnote 60 The health and sanitary conditions in the camps have also had devastating social consequences. According to a study conducted by the Ugandan Ministry of Health in collaboration with UN agencies, roughly 1,000 people were dying each week from preventable disease and malnutrition.Footnote 61
Despite these concerns, Human Rights Watch and Amnesty International have consistently opposed trading justice for peace. They were joined by several international lawyers during the Juba negotiations when Museveni and other government officials suggested that they might offer an amnesty to the indicted LRA leaders in exchange for a genuine commitment to end the war.Footnote 62 As for the conditions in the IDP camps, both NGOs have called on the OTP to investigate Ugandan military and civilian leaders for the forcible relocation of civilians – a crime under the Rome Statute – and for the abuses in the camps.Footnote 63 In other words, human rights are best promoted not by compromising justice in the interests of peace, but by applying criminal law even-handedly.
For his part, Moreno-Ocampo claimed jurisdiction to evaluate allegations against all parties in Northern Uganda, including the government and the military. He justified the current focus on the LRA rather than the UPDF because his investigations found that ‘the crimes committed by the LRA are of dramatically higher gravity’.Footnote 64 As to Museveni's statements that he would protect Kony from the ICC, Moreno-Ocampo held that the arrest warrants cannot be withdrawn and that Uganda, as a party to the Rome Statute, has a duty to execute them. While acknowledging Uganda's right to adopt alternative justice mechanisms, he insisted on prosecuting those who bore the greatest responsibility and refused to use his discretion to suspend the arrest warrants.Footnote 65 Nonetheless, the ICC chose to maintain a low profile to minimize interference with the peace process from the initiation of the Juba talks in July 2006 to their breakdown in April 2008.Footnote 66
Despite the controversy, the ICC's involvement coincided with advances in the peace process. The most important achievement was the negotiation of a ceasefire in August 2006, which involved the relocation of LRA bases from southern Sudan to the Democratic Republic of the Congo (DRC). This dramatically improved the security situation, facilitating humanitarian access and allowing more than half the IDPs to return to their homes.Footnote 67 The negotiations in Juba were characterized by the International Crisis Group as most promising in the history of the conflict and eventually led to a final peace deal in March 2008.Footnote 68 The agreement included a controversial justice and accountability mechanism that would apply traditional means of reconciliation, except for those responsible for the most serious crimes, who would be tried in a newly created special division of the High Court of Uganda.Footnote 69 This latter provision was designed to circumvent the ICC arrest warrants by claiming that the Ugandan courts were willing and able to prosecute, though it did not explicitly identify the indicted LRA leaders as candidates for this new judicial mechanism.Footnote 70 The prospects for such an agreement, however, may have been a mirage, since Kony refused to sign, pointing to the ambiguity of the war crimes provisions as among his central reasons.Footnote 71
During the negotiations the ICC and its NGO supporters made a persuasive case that the referral contributed positively to the peace process.Footnote 72 First, it concentrated the attention and resources of the United Nations, the African Union (AU), and Western donors on what had been a neglected humanitarian crisis. Second, the criminalization of the LRA isolated it from external assistance, particularly from Sudan, which had provided it with sanctuary and arms in a proxy war with Uganda in response to Kampala's support for the Sudan People's Liberation Army (SPLA) in its rebellion against Khartoum.Footnote 73 Finally, the reduction of outside support, combined with the threat of prosecution, increased the pressure on the LRA to negotiate its own safety.Footnote 74
Therein, however, lies the conundrum for any possible resumption of the peace process. While the arrest warrants might serve as a prod to negotiations, insisting on their execution is likely to prevent their completion. It is implausible that Kony would sign a peace agreement that would leave him vulnerable to trial either in Kampala or The Hague – unless his forces have been so weakened that the likely alternative is being killed or captured by the UPDF. Moreover, some analysts attributed the problems at Juba to the unrepresentative nature of the LRA delegation, given the unwillingness of the indicted LRA leaders to attend for fear of being arrested.Footnote 75 Any prospective negotiations will consequently require at least some temporary compromises with criminal justice, such as guarantees of safe passage to attend peace talks or an arrangement to provide Kony and his commanders with asylum in a state that has not ratified the Rome Statute.
Given the nature and scale of the LRA's crimes, most proponents of international criminal justice would object to such arrangements as condoning impunity and weakening the ICC's deterrent impact elsewhere.Footnote 76 Others argue that such compromises are likely to be futile, since the LRA remains a spoiler and is using the negotiations as breathing space to regroup and return to violence, particularly if the Sudanese peace agreement collapses and its former patron in Khartoum once again sees it as a useful proxy against the SPLA.Footnote 77 Kony's refusal to sign the peace agreement and reports of the LRA's return to the forced conscription of abducted children lend support to this view.Footnote 78
If one accepts these arguments, the logical consequence is to dismiss the possibility of meaningful negotiations and prepare for a military solution. Many supporters of prosecution seem to elide this dilemma by using the language of domestic law enforcement, calling for the execution of arrest warrants.Footnote 79 However, apprehending the leaders of a well-armed rebel movement that has not been defeated requires something more than a police operation, as is demonstrated by the inability of the UPDF for two decades to capture Kony. This also explains why the Prosecutor's agreements with neighbouring states and the United Nations to bring Kony to justice have not yet led to his arrest. The Sudanese government, the LRA's former (and possibly current) backer, has allowed the political authorities in Juba to mediate a negotiated solution, which it currently prefers to a military confrontation.Footnote 80 In the Ituri region of the DRC there is a UN peacekeeping force (MONUC) operating under a Chapter VII mandate to enforce a peace agreement against warring factions, not far from where the LRA had sought refuge. While MONUC agreed to co-operate with the ICC, arresting LRA leaders is a significant military operation (a skirmish with the LRA in January 2006 left eight Guatemalan peacekeepers dead), which would divert resources from its mandate to protect civilians and disarm militias in Ituri.Footnote 81
Another option is allowing the UPDF to enter the DRC. Some analysts believe that Museveni's ulterior motive for going to the ICC was to legitimize such an intervention by having the LRA branded as international criminals.Footnote 82 Yet during the Congolese civil war this was the part of the DRC that Uganda had occupied, both directly and through ethnic militias, in order to control its resources. As a result, attempts by Uganda to persuade the Security Council to grant it authority for such an intervention have been rebuffed.Footnote 83 These concerns might be assuaged if the UPDF acted in concert with MONUC and the Congolese army. There were indications that this might happen following a September 2007 meeting between Museveni and the DRC president Joseph Kabila threatening joint military action if Kony does not evacuate his base in north-east Congo, and the military option may once again be on the table following Kony's refusal to sign the Juba accords.Footnote 84
Military solutions, however, have not worked in the past, and will involve risks not only to the interveners, but also to victim communities in Northern Uganda. It could have the unintended consequence of intensifying violence in the north – repeating the experience of Operation Iron Fist in 2002 when the UPDF attacked LRA bases in southern Sudan – and reversing improvements to the humanitarian situation since the ceasefire. It could also jeopardize the lives of abducted children whose families want them returned – an outcome more likely to be achieved through negotiation than through force.Footnote 85 Prosecution, on the other hand, requires a commitment by international and regional actors to assume those risks because the threat posed by the LRA is proportionately greater. Absent that commitment – either because of the lack of will or because the risks to peace and other values are too high – the alternatives are either to live with and contain the LRA threat or to try again to end it diplomatically. If there ever is a return to the latter option, the ICC's role will likely be as a source of leverage in negotiations, whose successful completion will probably require compromises with criminal justice.
4.2. The Democratic Republic of the Congo
Unlike Uganda, the ‘peace versus justice’ dilemma in the Democratic Republic of the Congo (DRC) is latent rather than active. That is because the peace process that ended the Congolese civil war – which killed 3.3 million people from its outbreak in 1998 until its formal resolution in Sun City, South Africa, in December 2002 – subordinated prosecution to power-sharing. That war involved more than twenty armed factions and the intervention of nine neighbouring states, all of which were directly or indirectly involved in deliberate attacks on civilians. The Sun City agreement secured the withdrawal of foreign forces and then established a transitional government of the five main armed groups in anticipation of elections.Footnote 86 The strategy of pacifying other militias was non-retributive. It involved co-opting their leaders with government positions or military commissions in exchange for allowing UN peacekeepers to assist in a process called brassage, in which ethnic militias were demobilized, retrained, and either integrated into the national army or allowed to return to civilian life with amnesty.Footnote 87
On 16 July 2003 Moreno-Ocampo indicated that he planned to open the ICC's first investigations in Ituri, a province in the north-eastern DRC, which he referred to as ‘the most urgent situation’ in the Court's jurisdiction.Footnote 88 Ituri has been the scene of brutal violence against civilians perpetrated by militias representing Lendu agriculturalists and Hema pastoralists, which, in turn, was fuelled by Rwandan and Ugandan intervention. Although the peace process was able to secure the withdrawal of Rwandan and Ugandan troops from Ituri, the militias themselves were not part of the agreement and the fighting continued.Footnote 89 What triggered the Prosecutor's attention was a massacre of civilians in Bunia, the regional capital, after a battle between rival militias to control the city. At the encouragement of the Prosecutor, the transitional government referred the case to the Court in April 2004 and the OTP began its investigation in June.Footnote 90
The decision to start in Ituri was a prudent one in terms of the peace process. First, since the militias were not part of the power-sharing accord in Kinshasa, there was little risk of destabilizing the transitional government.Footnote 91 Second, the United Nations had deployed a peacekeeping mission (MONUC) in Ituri that, in co-operation with a French-led EU force, had undertaken an enforcement mission to assist the fledgling Congolese army in protecting civilians and disarming the militias.Footnote 92 Since MONUC was authorized to take military action against those the ICC was likely to investigate, the Court had a potential enforcement arm in the region that does not exist in Uganda (or elsewhere in the DRC). In fact, the first two warlords extradited to stand trial in The Hague – Thomas Lubanga Dyillo and Germain Katanga – agreed to lay down their arms as a result of MONUC's more robust mandate, and then were arrested in March 2005 after the murder of nine Bangladeshi peacekeepers.Footnote 93
The Prosecutor will need to tread carefully, however, as he moves beyond Ituri. That is because virtually all the politicians and militia leaders who have been co-opted into the government are complicit in substantial human rights abuses, including support for warlords in Ituri.Footnote 94 This subordination of justice to expedience was necessitated by the fact that none of the parties won a military victory and many retained forces under their control who could return to political violence should they be threatened.Footnote 95 That choice has entailed substantial costs in terms of good governance and human rights. As James Traub notes, ‘the men who had provoked Congo's unspeakable civil war would be rewarded by sharing the spoils of statehood’.Footnote 96 Moreover, the Congolese army, comprising former combatants, has often been ineffective in campaigns against insurgents in conflict areas in the eastern Congo, and as abusive towards civilians as were the militias from which they were recruited.Footnote 97
The ICC is unlikely to be the principal instrument to remedy the defects that flow from this arrangement, particularly with respect to powerful actors, since it could not rely on enforcement by the Congolese army, or from MONUC, whose Chapter VII mandate has so far been limited to Ituri. What is needed is co-ordinated donor pressure for greater transparency, and security-sector reform, involving a stronger system for vetting, training, and disciplining new recruits.Footnote 98 Ending impunity will also require strengthening the local justice system, including military courts, such as the one in Bunia, which has a mixed record in prosecuting crimes committed by the security forces.Footnote 99 The ICC may be able to make a contribution to this process by monitoring these courts with the prospect of asserting jurisdiction if they fail to do their jobs.Footnote 100 Nonetheless, the Court will have to consult with those involved in the peace process, since war-related hunger and disease have taken another two million lives since the Sun City agreement.Footnote 101 The ICC will therefore have to craft a prosecutorial strategy that does not destabilize the transitional social compact in ways that exacerbate political violence and its impact on civilians.
4.3. Darfur
The Darfur case was referred to the ICC on 31 March 2005 – approximately two years after the outbreak of a civil war that involved an ethnic cleansing campaign against the region's African population by government forces and government-supported Arab militias known as janjaweed, which has left several hundred thousand people dead and more than two million displaced. The referral required authorization by the UN Security Council because, unlike Uganda and the DRC, Sudan is not a party to the Rome Statute. In fact, it was made despite the opposition of the Sudanese government, which has blocked any meaningful co-operation with the ICC. The Prosecutor nonetheless moved forward. On 27 February 2007 he identified a former interior minister and a militia leader as his first cases and on 2 May the Pre-Trial Chamber issued warrants for their arrest.Footnote 102 After more than a year of non-compliance – including the appointment of one of the indictees as minister of humanitarian affairs – the Prosecutor reported to the Security Council that the atrocities were part of a ‘criminal plan’ involving ‘the entire Sudanese state apparatus’.Footnote 103 On 14 July 2008 he applied for an arrest warrant for Sudan's president, Omar Hassan al-Bashir, on charges of genocide.Footnote 104
Anti-impunity advocates contend that by indicting the sitting head of state, the ICC can have an impact on ending criminal violence in Darfur.Footnote 105 That will only happen if the judicial process serves as a catalyst to a change in the political strategies of conflict resolution used by powerful states and intergovernmental organizations.
It has been the mismatch between legal and political strategies that explains the limited impact of the ICC up to this point. The ICC referral, like the creation of the ICTY in the early phases of the Bosnian war, created a criminal justice process unaccompanied by enforcement actions against behaviour deemed to be criminal. Although the Security Council had passed several resolutions under Chapter VII calling for Khartoum to disarm the janjaweed and end the violence, no meaningful sanctions were imposed for non-compliance or explicitly linked to future compliance. The only international presence on the ground was an underfunded and understaffed AU force that had been deployed a year earlier to monitor a non-existent ceasefire. When the Security Council authorized a more robust civilian protection force in August 2006, it never moved beyond what amounted to a pacific settlement approach in which deployment would depend on Sudanese consent. No penalties were imposed or threatened when that consent was declined, or when it subsequently accepted deployment of a hybrid UN–AU Mission in Darfur (UNAMID), but imposed conditions that amounted to obstruction.Footnote 106
Given the unwillingness of the Security Council to authorize enforcement, the principal instrument of conflict resolution has been impartial mediation through the good offices of the United Nations and the AU. This is incompatible with the ICC referral because one cannot simultaneously subject the government to criminal scrutiny and non-coercively seek its co-operation. It is also incongruent with conditions on the ground. For such an approach to succeed in ending a civil war, all parties must be in a ‘mutually hurting stalemate’Footnote 107 in which they recognize that they cannot achieve their objectives militarily and that the continuation of the war will make them all worse off. The ruling party in Khartoum, however, believes that maintaining its power is better served by the continuation of the war than by a negotiated settlement, which the International Crisis Group pinpoints as one of the central reasons for the failure of the peace process.Footnote 108 In fact, the most significant outcome of the mediation effort, the Darfur Peace Agreement, signed on 5 May 2006 by Khartoum and one faction of the two main rebel groups, resulted in increased fighting and the deterioration of the humanitarian situation as the faction that signed joined government forces in escalating attacks on rebel-controlled areas.Footnote 109 As in Bosnia and Herzegovina, a pacific settlement approach to large-scale ethnic cleansing is incompatible with both criminal justice and conflict resolution.
For the Bashir indictment to be something more than naming and shaming, it will have to provoke the Security Council into moving from pacific to some form of coercive conflict resolution. This could involve the use of sanctions targeted at Sudan's leadership or its oil revenues, followed possibly by a credible threat of force – actions that so far have been blocked by China's threatened veto, given its economic and strategic relationship with Sudan.Footnote 110 The publicity surrounding the indictments – augmented by civil society groups – could increase the reputational costs to China in playing this role, dissuading it from vetoing sanctions resolutions or encouraging it to put pressure on Sudan to comply with Security Council mandates.Footnote 111 It could also cause Western governments to press the matter more forcefully in the Security Council or even to take action outside it. In fact, some ICC supporters have suggested that the Bashir indictment itself could be used as a source of leverage to get Sudan to end the conflict, or at least allow greater humanitarian access and a more robust peacekeeping presence.Footnote 112
If the indictment does indeed serve either as a prod to enforcement or as a source of leverage, success will almost certainly require compromises with criminal justice. That is because the purpose of pressure is to change regime behaviour – that is, to increase the costs and risks of the war for Khartoum to a point where ending it serves its interests. A principled approach to criminal justice, however, amounts to a demand for regime change, since the Prosecutor and major UN and NGO studies have attributed responsibility not only to President Bashir, but also to Sudan's most influential political and military officials.Footnote 113 Moreover, as with Dayton, the successful deployment of an effective peacekeeping force will probably require the continuing co-operation of the Sudanese government to negotiate and maintain a ceasefire and to disarm the militias, which the UN peacekeepers would monitor rather than enforce. If this is ever a realistic prospect, it will require some short- to medium-term compromises on criminal justice, such as the use of Article 16 by the Security Council to suspend criminal proceedings on renewable 12-month periods, restrictions on UN peacekeepers in enforcing arrest warrants, or the Prosecutor exercising his discretion and lowering his profile if genuine progress is being made. A more aggressive criminal law approach would only be feasible if international intervention resulted in the removal of the regime – either directly or by triggering a leadership change – or if the civilian protection mandate does not require the active co-operation of the Sudanese government, as was the case with the Milošević regime after the Kosovo war.
4.4. Implications for prosecutorial discretion
Since there is a potential ‘peace versus justice’ dilemma in each of the cases under his purview, the ICC Prosecutor will have to exercise political prudence in deciding when and how to proceed. The NGO studies are correct that the OTP is not ideally suited to play this role, since it is a legal institution whose mandate and expertise are in the law, not in diplomacy. Preferably, these calls should be made by the Security Council, a political body responsible for maintaining peace and security. If, for example, the Ugandan government believes that amnesty is needed to end the war with the LRA or mediators view the ICC as an obstacle to negotiations in Darfur, the proper venue for making such an appeal is the Security Council, not the Prosecutor's office. As a former ICC Deputy Prosecutor, Serge Brammertz, put it, ‘The priority of the Rome Statute is to prosecute[;] it's not here for political stability.’Footnote 114
The problem with establishing such a categorical division of labour is that it is not clear that the Security Council will play its designated role when there is a genuine conflict. While the Security Council is a political institution, its resolutions often reflect a least-common-denominator compromise of national interests rather than a coherent political strategy. Article 16 is also a blunt instrument, suspending the entire criminal process for renewable 12-month periods, and is only likely to be invoked after a peace deal has been finalized and at the request of the parties or of international mediators. Given the fluidity of negotiations during ongoing conflicts, most of the cases in which an aggressive prosecutorial strategy might have destabilizing consequences will fall below the Security Council's radar screen.
For example, in the run-up to the 2006 Congolese elections, the transitional government offered amnesties and military commissions to three Ituri-based warlords and MONUC assisted in the demobilization and reintegration of their militias. Human Rights Watch criticized this policy for rewarding impunity, although a MONUC spokesman supported it for reducing the level of violence in Ituri.Footnote 115 On 6 February 2008, three months after the implementation of the agreement, the ICC unsealed the indictment of one of the warlords – Mathieu Ngodjolo Chui, for the massacre of two hundred people in the Hema village of Bogoro in February 2003 – who was immediately arrested and extradited to The Hague.Footnote 116 Human Rights Watch subsequently wrote that this should embolden the Prosecutor to move against other militia leaders and the officials who supported them in Kinshasa.Footnote 117 The ICC Prosecutor is not bound by a national amnesty and the Security Council has not invoked Article 16. Does that mean that he should proceed without consulting those involved in the peace process and making an independent evaluation as to the likely impact of criminal justice on political violence on the ground?
The answer is that the Prosecutor should not, and probably will not, for both ethical and practical reasons. First, acting on a duty to prosecute blind to its consequences can increase the risk of violence to victim communities with the recurrence of the very kinds of atrocity that led to the involvement of the ICC in the first place. Second, ignoring such considerations would likely undercut the Court's co-operation with the political actors on whom it depends to be successful. If the announcement of investigations or arrest warrants is linked to the breakdown of a ceasefire or a return to violence, it could alienate the ICC from war-torn communities, as has occurred in Northern Uganda. It could also jeopardize its relationship with the Congolese government or MONUC by increasing the political violence with which they would have to contend or in foreclosing negotiated solutions in those regions in the eastern DRC that are still at war. Finally, it could alienate the Court from donor states involved in post-conflict stabilization and reform and which are among the strongest supporters of the ICC. Given the fact that the ICC lacks an independent source of financing, as well as its own police force to provide security for investigations or apprehend those indicted, it is likely to hold back from potentially destabilizing prosecutions, at least until conditions change.Footnote 118
The Prosecutor will consequently have to act as both lawyer and diplomat in exercising his discretion, evaluating not only the gravity of the crime and the admissibility of the case, but also the likely impact of an investigation or prosecution on prolonging a conflict or undermining a political transition. In doing so, he should consult with a wide variety of stakeholders likely to be affected by criminal proceedings, as well as the United Nations, regional organizations, and governments and NGOs involved in mediation efforts. This does not make the Prosecutor a political actor who takes instructions from others. While he should maintain his independence as a legal officer, he also needs to acknowledge the interdependence of politics and law, the boundaries the former sets for the latter, and the risks of trespassing across them. Hence one OTP analyst wrote that, to be effective, a prosecutor should ‘pursue a process of consultation and a sufficient degree of international consensus-building’ to ‘operate in co-ordination with, rather than in opposition to, international efforts to address conflicts’.Footnote 119 If he instead emulates David Crane's indictment of Charles Taylor through the aggressive pursuit of cases that could undermine peace processes, he is likely to dry up the capital he has with political actors, thereby weakening the Court as an institution.Footnote 120
The final question is whether Article 53 is the best means of reconciling legal duty with political constraint, and there is a case to be made that it is not. The OTP's interpretation of the ‘interests of justice’ test has moved closer to the legalist position since the LRA controversy in 2005. A September 2007 policy paper stated that it should not be ‘conceived of so broadly as to embrace all issues related to peace and security’ and that ‘the broader matter of peace and security is not the responsibility of the Prosecutor’.Footnote 121 In fact, it implicitly rejects the need to adapt prosecution to peace processes when it asserts that any ‘political or security initiative’ must conform to ‘the new legal framework and this framework necessarily impacts on conflict management efforts’.Footnote 122 In other words, the Court should follow its legal duty to identify those responsible for criminal violence without political considerations and this should constrain political actors from offering amnesty or exile as instruments of peacemaking.
Nonetheless, the document suggests more flexibility than the NGO papers when it acknowledges that Article 53's reference to ‘the interests of victims’ could include interests other than prosecution, such as their security and protection.Footnote 123 This, in turn, requires an ‘ongoing risk assessment’ and a dialogue with victims and their representatives, which could influence the timing and profile of investigations and indictments. Such an approach, broadly defined, could be the rough equivalent of factoring peace processes into prosecutorial discretion since their destabilization would almost inevitably lead to further victimization. Nonetheless, Article 53 is unlikely to be the vehicle for exercising such discretion. It could limit the OTP's flexibility, since any deferral would have to be submitted to the Pre-Trial Chamber, which could overturn the decision, and a decision to resume criminal proceedings would require the submission of new evidence. Moreover, the publicity surrounding the submission could subject it to political controversy and compromise its appearance of impartiality.
If the Prosecutor were to hold back from politically sensitive proceedings, he would be more likely to rely on his inherent discretion. Under the Rome Statute, there are no guidelines for when the Prosecutor must initiate formal investigations or applications for arrest warrants. In deciding when and how to proceed, he will confront more crimes than he has resources to prosecute. Consequently, those resources are more likely to be deployed to cases that are pragmatically feasible than to those that might disrupt peace processes and risk the relationships the Court has with the states and international institutions upon which it depends for its long-term success. In a situation like the DRC, for example, the Prosecutor need not invoke Article 53 to hold back from a potentially destabilizing case. Rather, he could delay criminal proceedings and avoid public statements, at most discreetly collecting information until the political situation changed. The principal cost of such an approach is its lack of transparency. This could create suspicion that the Court is protecting powerful actors, a view widely held in the DRC.Footnote 124 Nonetheless, if the flexibility and reputational costs of invoking Article 53 are too high, this is the only means of holding back from actions that could trigger renewed conflagrations while at the same time retaining the flexibility to pursue cases after the peace process solidifies.
In fact, Human Rights Watch makes a similar proposal when it concedes that the Prosecutor has limited discretion in terms of the timing of a potentially destabilizing prosecution, but should not publicly announce what he is doing because it could compromise the integrity of the Court.Footnote 125 Nonetheless, Human Rights Watch views this as an exceptional device that should be used rarely and only for a limited time so as not to amount to a de facto amnesty. The analysis above, however, suggests that this will be a far more regular feature of prosecutorial discretion, often for indeterminate periods of time, given the kinds of ongoing conflict with which the Court is involved.
5. Conclusion
The central philosophical question underlying this legal controversy is whether justice in the aftermath of war and atrocity requires the application of criminal justice to its sponsors. In a book that takes politicians, diplomats, and even the ICTY Prosecutor to task for subordinating justice to accommodation during the Bosnian war, Paul Williams and Michael Scharf answer that question affirmatively and support that claim with a passage from Michael Walzer's classic work on just war theory:
[T]he assignment of responsibility is the critical test of the argument for justice . . . If there are recognizable war crimes there must be recognizable criminals . . . The theory of justice should point us to the men and women from whom we can rightly demand an accounting, and it should shape and control the judgments we make of the excuses they offer (or are offered on their behalf) . . . There can be no justice in war if there are not, ultimately, responsible men and women.Footnote 126
This, however, is an incomplete reading of Walzer's treatment of the subject. Following the excerpted quote, he explains that he is discussing moral responsibility rather than guilt or innocence as determined by judges, and he goes on to argue that considerations of proportionality mean that there are ‘often prudential reasons for not calling judges’.Footnote 127 That is because the apprehension of criminal leaders requires something different from domestic law enforcement and that difference is more potentially threatening to the rights and lives of innocent civilians. Walzer illustrates this point through a quote from US Secretary of State Dean Acheson defending the US decision to cross the 38th parallel during the Korean War as necessary to ‘round up the people who were putting on the aggression’. This, however, required military escalation rather than police work, and it inflicted harm ‘far beyond the people who are rounded up’.Footnote 128 Walzer is prepared to accept those costs on proportionality grounds in order to remove from power and try the leaders of a radically evil regime, such as Hitler's Germany, but not for lesser despotisms, such as imperial Japan or communist North Korea. Whether or not one agrees with this position, the argument forces us to consider the humanitarian costs that would have to be accepted to bring to justice leaders who have not yet been defeated.
Yet many of the arguments marshalled by NGOs and international lawyers ignore these trade-offs by using the language of domestic law enforcement as if what is needed is the simple execution of arrest warrants. To illustrate, the former ICTY Prosecutor Richard Goldstone attributed Milošević's belief in his own impunity to ‘the failure of the international community to prosecute Pol Pot, Idi Amin, Saddam Hussein and Mohammed Aidid’.Footnote 129 But bringing any of these malefactors to justice would have involved something more than executing arrest warrants. Take Goldstone's last two examples. With Saddam Hussein, it would have required a decision in 1991 to march to Baghdad, with all of the attendant costs and risks that have been made evident by the second Iraq war. With Aidid, one could argue that the Security Council issued its first arrest warrant when it passed Resolution 837 (1993), which called for the bringing to justice of those who ambushed and murdered 24 Pakistani peacekeepers serving in the UN mission in Somalia. The attempt to apprehend Aidid led to an escalation of conflict between US and international forces and Aidid's clan, culminating in the tragic battle of Mogadishu, which precipitated the withdrawal of those forces.Footnote 130
Similarly, executing an arrest warrant against Kony, a yet-to-be-defeated rebel leader, is a military operation, not a police action, and involves a much broader set of political and moral calculations than does law enforcement. As Walzer notes, ‘war affects more people than domestic crime and punishment, and it is the rights of these people that force us to limit its purpose’.Footnote 131 If military escalation is necessary to apprehend the criminal actors and one cannot justify the ensuing loss of innocent life on proportionality grounds, then the perpetrators cannot be ‘removed from the (moral) world of bargaining and accommodation’.Footnote 132
Allowing criminal actors to enter this world rather than the world of crime and punishment is likely to be challenged by many within the human rights community as compromising accountability. Bassiouni put the issue starkly in a way that informs much of the anti-impunity movement:
The human rights arena is defined by a constant tension between the attraction of realpolitik and the demand for accountability. Realpolitik involves the pursuit of political settlements unencumbered by moral and ethical limitations. As such, this approach often runs directly counter to the interests of justice, particularly as understood from the perspective of victims of gross violations of human rights.Footnote 133
The goal of institutions like the ICC is to insulate international criminal law from politics as much as possible, so as to make justice less vulnerable to the compromises imposed on it by diplomacy and statecraft: ‘Compromise is the art of politics, not of justice.’Footnote 134
This formulation creates a false dichotomy between realpolitik and justice. As Gary Jonathan Bass notes, ‘legal justice is one good among many . . . not a duty that trumps all others’.Footnote 135 Ethical decision-making requires that it be weighed and balanced against other values, such as stability and peace, particularly since war creates the licence for the worst abuses of human rights. Factoring power realities into this equation is not necessarily surrendering to realpolitik; it is indispensable in determining what kinds of justice and accountability mechanisms are possible and assessing their consequences for other values. Pursuing justice in ways that are blind to power realities will either be futile exercises in high-mindedness or counterproductive to political settlements that are necessary to end violent conflicts. International law cannot be isolated from the political context in which it has to operate.
This latter point was the central thesis of Judith Shklar's Legalism, which challenged those legal purists who saw the Nuremberg trials simply as political justice. Shklar acknowledges that Nuremberg was a political trial, but not every form of politics is the same. The Soviet show trials of the 1930s were designed to consolidate totalitarian control; Nuremberg, by contrast, had a positive impact on the post-war transformation of the West German legal culture by exposing what had gone so frightfully wrong in the 1930s. Shklar concludes from this, ‘The question, in short, is not “Is law political” but “What sort of politics can law maintain and reflect?” . . . It is the quality of the politics pursued in them that distinguishes one political trial from another.’Footnote 136 This article extrapolates Shklar's arguments about political trials to the exercise of political discretion not to prosecute. Not every choice to set limits on international criminal justice is the same. Rather than denounce all such decisions as politicization, amoral realpolitik, or disreputable expedience, they should instead be assessed on the judgement and moral character of the politics that sets those limits.
