Introduction
There has been an upsurge in armed conflicts around the world since the end of the Cold War. Although this is a worldwide phenomenon, Africa has arguably had the most significant share of these conflicts. Gross human rights violations are often among the main causes of these conflicts, and more of these abuses and even war crimes occur during the armed conflict itself. It is widely believed that punishing the perpetrators of these crimes constitutes one of the essential measures of effective conflict prevention, resolution and post-conflict peace building. However, when no single party to the conflict achieves decisive victory and the conflict needs to be brought to an end through negotiations, it is often the case that the warlords and other combatants responsible for these crimes refuse to agree to any ceasefire except on the condition of an amnesty. This seems to have led to a situation where, since the mid-1970s, more than 15 states on four continents have passed amnesty laws exonerating past regimes or warlords from criminal procedures, in order to allow for a peace agreement or supposedly reconcile a post-conflict nation (Burke-White 2001, p. 42, Huyse 2008, pp.1–23).
Since at least the 1990s, however, the tendency of the globalisation of justice and the increasing arch of international human rights law, vigorously defended by a number of global human rights defence organisations, have led to a growing campaign for post-conflict ‘justice’. This campaign has sought at least to question state sovereignty in granting these amnesties, and it firmly opposes any peace agreement that does not allow for criminal justice proceedings, often with the slogan ‘no justice, no peace’ (Penman 2007, p. 8). The campaign has further resulted in a redefinition or attempted redefinition of customary international law with regard to amnesty provision in peace agreements (Cassese 2004, p. 1133). This is also true for international criminal law. The establishment in 2002 of the International Criminal Court (ICC) as a ‘permanent institution [that has] the power to exercise its jurisdiction over persons for the most serious crimes of international concern’ (Article 1 of the Rome Statute) may be seen as a culminating point of these campaigns.
By October 2008, the ICC had opened investigations into four situations that the Court believed met the criteria set in its Rules of Procedure and Evidence. Interestingly, all of these cases relate to African countries (i.e. the Democratic Republic of Congo and Uganda, both opened in 2004, and the situations in the Central African Republic and the Sudanese region of Darfur, both opened in 2005). In addition, all the 12 warrants of arrest issued by the Court's Pre-Trial Chambers by this date concerned African elements, including five in Uganda.1 On 14 July 2008, the Chief Prosecutor of the ICC indicted the Sudanese leader, President Omar al-Bashir, charging him with genocide, war crimes and crimes against humanity, and requested a Pre-Trial Chamber to issue an arrest warrant to this effect. Issued on 4 March 2009 without the charge of genocide, this is the thirteenth (African) case before the Court. (For a discussion of this specific case, see Souaré 2008b, 2009.)
This situation creates a dilemma for both researchers and practitioners in the field of conflict management and peace studies. For example, how to achieve peace and ensure justice for the victims, making sure that neither of these two objectives compromises the other? Should the mediators or the political actors of a country allow amnesty provisions in favour of peace and stability in the country concerned regardless of crimes committed, and on what justification? Is it a guarantee for peace, or should they insist on rendering justice even if this means the continuing death of thousands of innocent people? How will they morally justify this? What alternatives are there to these two options? It is with this thorny issue that this article deals, focusing on the specific situation of Uganda. It is in the effort to find this alternative solution that this article's added value lies, as this approach is rarely found in the myriad of studies written on this topic.
The article is divided into five sections. The first section makes a general diagnosis of the situation and tries to explain the factors that make it problematic, providing a concise definition of the main concepts used in the article. The second section provides a brief historical background about the armed conflict in northern Uganda and where the dilemma lies in this situation, focusing on the Amnesty Act enacted by the Ugandan Parliament in 2000 in a bid to entice armed opposition elements, including LRA leaders, to come out of the bush and sign a peace agreement with the government. This section would not be complete without also looking at the real or perceived impact of the ICC's indictment of these same leaders on the peace process in the country. The third and fourth sections deal with the arguments of those favouring one or other of the options. In the concluding and final section, an attempt will be made to move the debate from its dichotomic format (amnesty vs. criminal justice) and suggest some alternative measures with a view to reaching a middle ground that allows for peace settlement and at the same time addresses the grievances of the victims of crimes committed during the conflict.
This ‘third way’ alternative solution is based partly on the acknowledgement that ‘justice’ cannot be reduced to only one form, namely criminal or punitive justice, and that there are other forms of justice to which one could resort to address violations that occurred during the armed conflict and thus combat impunity (Henham 2007, pp. 449–468). The argument is that we need to know which of their many violated rights is more important to LRA victims in Uganda and what use can we make of ‘local’ justice systems in addition to addressing the violated rights that they consider more important.2
Diagnosis of the Problem
In contrast to interstate wars, in which the opposing sides will presumably eventually retreat to their own territories, in civil wars the members of the warring factions must live side by side after the killing stops (Licklider 1995, p. 681). Indeed, in their definition of civil war, Elbadawi and Sambanis (2000, pp. 244–269) insist that the parties to a civil war must be ‘concerned with the prospect of living together in the same political unit after the end of the war’.
The term ‘post-conflict justice’ or ‘transitional justice’ is understood here as the interrelated processes of prosecution and accountability, truth telling, reparations and institutional reform that contribute to the long-term restoration of social relations between the former belligerents and their victims, where the war experience becomes a regrettable episode of common history (Meltzer 2004). It is also ‘formal and informal procedures implemented by a group or institution of accepted legitimacy around the time of a transition out of an oppressive or violent social order, for rendering justice to perpetrators and their collaborators, as well as to their victims’ (Kaminski et al. 2006, p. 295). The term ‘justice’ is employed here for criminal or punitive justice.
‘Reconciliation’ is often associated with ‘forgiveness’, which is defined as a negation or abandonment of vengeance (David and Choi 2006, pp. 339–367). It also means the ‘overcoming of vindictive passions, the passions of anger, resentment, and even hatred that are often occasioned when one has been deeply wronged by another’ (Murphy 2003, p. 16). It is considered a means to peace and national reconciliation. The term ‘amnesty’ is also used in the same sense, which is how it will be used in this article. From these definitions, a line may be drawn between the two concepts whereby post-conflict justice in this sense is weighted more heavily in the past, as its main locus is resolving the burden of past violations. As with reconciliation, this is turned towards making normal life possible in the future, which makes it forward-looking or future-oriented (Mani 2005, p. 514).
Peace agreements can be defined as arrangements entered into by warring parties to regulate or resolve the basic incompatibility between them (Wallensteen and Sollenberg 1997, p. 342). Because they are ‘negotiated arrangements’ between the warring parties, peace agreements usually involve undertakings for power-sharing, at least temporarily, between the parties to the conflict, because otherwise some parties will be unwilling to accept (Licklider 1995, p. 684). Thus, in the interest of reaching a settlement, alleged perpetrators of human rights abuses and even of crimes against humanity have been included in the negotiations. They have even been included in the new governments after transforming themselves into political parties, and their crimes are generally amnestied (Derrick 2002, Lambourne 2004, p. 5).
What this means is that achieving satisfactory post-conflict justice and dealing with past injustices and human rights abuses may be easier in situations where the armed conflict is ended by the outright victory of one party over the others. Armed conflicts may be ended in three main different ways: (1) by a peace agreement; (2) by the outright military victory of one party over the other or others; or (3) in other ways such as de facto ceasefires. Of relevance to this article is war termination by way of a ‘peace agreement’, owing to the fact that none of the parties achieved decisive victory over the others.
When people responsible for gross human rights abuses during the conflict are walking free or even boasting about their actions in front of their victims or relatives of their victims, this does not augur well for the peace reached. What may undermine a peace agreement still further is the sight of such people in higher public positions, as if their atrocities have been rewarded by such appointments instead of being punished. Their victims may feel very disappointed and cheated by the new arrangements. The resulting frustration may lead to some trying to take the law into their own hands, thereby rendering an already precarious peace situation more fragile.
What follows from this is that the dilemma resides in the general recognition of a number of facts. First, that impunity for past crimes must not be allowed. Second, that there is a need to overcome or transform the enmities developed during the armed conflict and build bridges among ordinary people. Third, the bitter fact that conflicts ended by way of negotiation must accommodate all parties, as no one is in a position to win outright and others will not negotiate just to see themselves incarcerated. This is what makes the task of achieving successful and satisfactory post-conflict justice without undermining the peace process one of the most daunting tasks in post- or civil war nations, and this is the case in Uganda, as the next section will show.
The Conflict in Uganda
There have been many rebellions in Uganda since the 1980s and indeed before (Branch 2005, Allen 2006, pp. 25–52, Finnström 2006). However, the one that is the subject of this article is that of the reclusive Joseph Kony's Lord's Resistance Army (LRA) in the North of the country. Most authors trace the beginning of the chain of rebellions in Uganda, of which the LRA has proved the most durable, back to 1986 and the coming to power of Yoweri Museveni and his National Resistance Army (NRA), formed in 1981 as a rebel group against the government of Milton Obote. It should be noted that from the date it regained its independence from Britain in 1962 until the forceful coming to power of Museveni, a southerner, Uganda was ruled by people from the North (Apuuli 2005, p. 37). Both in this period and in the colonial era, northern Ugandans, especially the Acholi people, constituted the main pool of recruitment to the armed forces (Doom and Vlassenroot 1999, pp. 7–9). Almost all the rebellions of post-1986 Uganda emerged from the North, especially amongst the Acholi people, who have also been their main victims (Van Acker 2004, pp. 335–357, Allen 2006, p. 30, Finnström 2006, pp. 200–220, Baines 2007, p. 100).
There are various explanations as to the root causes of these rebellions and that of Kony in particular (Finnström 2006, Allen 2006, pp. 25–52, Branch 2005, 2007). Finnström advances two explanatory variables. The first is that people from northern Uganda initiated the rebellions in an attempt to regain state power after losing to Museveni's forces from the South, something they felt unable to accept. The other explanation put forward by the author is perceived economic marginalisation of the North, ‘a history of being only peripherally included in the economic structures and processes of the country’ (Finnström 2006, pp. 202–203) (see also Omara-Otunnu 1995, pp. 223–236). Although this factor has been cited by many other authors, it is a claim that seems paradoxical for the simple reason that it was the northerners that had ruled the country since its regaining of independence, except for the parenthesis of the north-western Idi Amin's rule between 1971 and 1979.
In any event, what interests us here is that Joseph Kony formed the LRA after the defeat by the NRA of Alice Lakwena's Holy Spirit Movement (HSM II), another northern rebellion, in 1987. Kony proclaims himself a messianic prophet, declaring that he aims to overthrow the government in Kampala and rule Uganda according to the Ten Commandments (Allen 1991, Doom and Vlassenroot 1999). Notwithstanding this manipulation of religion and going against the commandment ‘you shall not murder’, the rebel movement has been accused of some of the most egregious crimes committed against innocent civilians, including war crimes, crimes against humanity and torture. By the end of 2004, an estimated 1.6 million people were living in internally displaced persons' (IDPs') camps, having fled the group's savage attacks (Apuuli 2005, p. 41). The group is also notorious for the systematic abduction of children, many of whom are then forcefully conscripted in its ranks and/or used as human shields, porters, labourers, sex slaves, etc. (Human Rights Watch (HRW) 1997, p. 4, Doom and Vlassenroot 1999, p. 25).
Others see this account as one-sided. Adam Branch (2007, p. 181), for example, maintains that ‘the Ugandan government's counterinsurgency has also been brutal toward Acholi, as the NRA and its successor, the Uganda People's Defence Force (UPDF), have focused their use of force on destroying suspected rebel support among civilians’. He attributes the formation of IDP camps to the Ugandan Government's ‘policy of forced displacement’ and he qualifies them as ‘internment camps’, contrary to their qualification by the Ugandan Government as ‘protected villages’ (see also Otunnu 2006). Although he tends to give more credibility to the ‘gravity’ counter-argument of the ICC when this is used to argue that indicting only elements of the LRA is biased (Branch 2007, p. 181, Dowden 2007, Happold 2007), Tim Allen (2006, pp. 97–102, 2007) seems to acknowledge this fact.
Leaving aside this debate and notwithstanding the alleged crimes, owing to a number of factors, including in particular the failure of government forces to defeat the rebellion militarily, the Ugandan Government introduced an amnesty Bill in parliament in 1999, which was passed as an Act in 2000 (Hovil and Lomo 2005, Apuuli 2005, Allen 2006, pp. 74–78, Branch 2007, p. 184). The Bill offered pardons to all Ugandans engaged or engaging in acts of rebellion against the government since 26 January 1986 who voluntarily renounced rebellion. The Amnesty Act is the product of a great deal of activism from human rights and civil society groups as well as strong mobilisation of political, traditional and religious leaders in Acholiland (Hovil and Lomo 2005, Allen 2006, p. 72, Baines 2007, p. 101, Branch 2007, p. 184). The Act was aimed at all Ugandans without discrimination, which if not explicitly aimed at them clearly included the LRA members (Apuuli 2005, p. 44, Hovil and Lomo 2005). Indeed, within a year scores of rebel elements took advantage of the Act, renounced violence and were reintegrated into the society in different parts of the country (Refugee Law Project (RLP) 2005, p. 4, Apuuli 2005, p. 45). By mid-2004, over 5000 adult former LRA fighters reportedly had surrendered and applied for amnesty under the Act (Allen 2006, p. 75)3.
In December 2003, strangely, the government of Uganda asked the ICC to investigate crimes committed by the LRA in the country since the start of its military operations some 20 years earlier. The Chief Prosecutor of the ICC soon established that there was enough evidence to begin an investigation. Eventually, on 27 September 2005, the Court issued arrest warrants for the five most senior leaders of the LRA, including its Chairman Joseph Kony (ICC 2005, Apuuli 2006).4 By this, ‘there appears to be a contradiction between the Amnesty Act, seen by many as an alternative to punishment, and the investigations and subsequent punishment by the ICC’ (RLP 2005, p. 4). Branch (2007, p. 184) goes further to argue that ‘the warrants eviscerate the Ugandan Amnesty Act’ and thus ‘fly in the face of the popular demand for general amnesty, rendering the Act inapplicable to the very people to whom it most needs to be applied for peace to arrive’.
As a result, although peace talks began between the government and the rebel group in mid-2006 under the auspices of the government of south Sudan, the indictments of the ICC appear to be a complicating matter. For while they have been important in bringing the LRA to the negotiating table (International Crisis Group (ICG) 2006, Allen 2006, pp. 116–117, 2007), they seem to be ‘the main stumbling block to peace’ (Atkinson and Finnström 2006, Baines 2007, p. 102, Branch 2007, p. 183, Happold 2007).
In February 2008, the Ugandan Government and the rebel movement signed the sixth in a series of documents expected to eventually lead to a final peace agreement to end the conflict. They fixed dates in March, April and May 2008 for this signing that never happened. Speaking on Ugandan radio in October 2008, the Acholi Paramount Chief, David Acana II, claimed that Joseph Kony called him the previous month and apologised ‘for not meeting the Acholi cultural leaders, for switching off his phone and for his failure to sign the final peace deal’. Kony is reported to have said this, notwithstanding the fact that the chief facilitator of the talks, south Sudan's Vice-President Riek Machar, is said to have decided to ‘close the chapter’ on the painstaking Juba peace talks. But even with this apparent change of heart by Kony, the main problem seems still to be the LRA's demand that the ICC withdraw its arrest warrants (Olupot 2008, Sserwanga 2008). This is not to deny the fact that there are other issues, such as the amount of money that LRA leaders are calling to be paid to them, but these cannot be matched to the role of ICC warrants. The mobilisation of funds, if money were the main problem and the Ugandan Government could not provide it, is far easier than getting the ICC to disengage.
The ICC, for its part, still insists on the validity of its indictments. The closest it has come to softening its position was when it asked the Ugandan Government to furnish it with information on the competence of a proposed war crimes court that Kampala had agreed with the rebels to set up in a section of the High Court of Uganda (Nyakairu 2008). What this points at is that criminal proceedings are still a paramount process in this case from the perspective of the ICC and those that favour its approach, and this is where the dilemma lies for Uganda (Southwick 2005, Atkinson and Finnström 2008).
Post-conflict Justice: No Justice, No Lasting Peace
As noted above, the Amnesty Act of 2000 and the proposed amnesty provision in the yet-to-be-signed peace accord between the government of Uganda and the LRA rebel movement is greeted with opposition from some quarters, both within Uganda itself and outside. Prominent amongst the opponents are the ICC itself and some international human rights organisations. The ICC seems to have pressed the Ugandan Government not to allow the top leadership of the LRA to benefit from the Act. In a press release confirming the referral of the situation concerning the LRA by the Ugandan Government, the ICC notes that:
President Museveni has indicated to the Prosecutor his intention to amend this Amnesty [Act] so as to exclude the leadership of the LRA, ensuring that those bearing the greatest responsibility for crimes against humanity committed in northern Uganda are brought to justice. (ICC 2004)
For example, in reaction to the Ugandan Government's announcement of its intention to offer amnesty to the LRA leaders if they signed a final agreement, Amnesty International issued a public statement. In it the organisation expressed its dismay at this news, noting that the government of Uganda did not have any authority to offer ‘protection’ to any person charged by the ICC and that even the latter would have ‘no power to revoke arrest warrants on grounds that it might assist political negotiations to end the armed conflict’ (AI 4 August 2006, emphasis added). Others have insisted on the criminal prosecution of the rebellion's top leadership which has been responsible and directed the crimes, but seem ready to accept a pardon for the children who were abducted and conscripted in the LRA movement (Apuuli 2005, p. 49, 2006).
There are three main arguments for the proponents of this approach. First, they argue that amnesty laws are incompatible with international law. Second, that contrary to the claim of the proponents of amnesty provisions, who justify their argument on the basis of moral considerations for future victims, amnesty laws are not after all morally justifiable. Third, that amnesty laws do not offer any guarantee that they will lead to durable peace in the country or prevent others from occurring. There will now be a look at these arguments in some detail.
With regard to the first argument, it is advanced here that amnesty provisions constitute an acceptance of impunity and may thus be regarded as condoning the crimes committed by those amnestied. Chigara (2002, p. 2) and Ssenyonjo (2007, p. 376) argue that amnesty laws that expunge criminal liability promote a culture of impunity and continuous violence that all instruments of international law reject categorically.
One legal instrument often cited to substantiate this argument is the Rome Statute of the ICC that stipulates in its preamble that the State Parties to the Statute are ‘determined to put an end to impunity’ for the perpetrators of serious crimes of concern to the international community. Nonetheless, as will be seen below, some proponents of amnesty laws are of the view that the Rome Statute also allows for amnesty provisions when peace is at stake. They point to paragraphs (1)(c) and (2)(c) of Article 53 of the Statute as allowing the Prosecutor of the ICC to suspend or abandon an investigation or prosecution where there are substantial reasons to believe that the investigation or prosecution ‘would not serve the interests of justice’. They also point to Article 16 of the Statute that gives a similar prerogative to the UN Security Council who, in a resolution adopted under Chapter VII of the UN Charter, can request the Court to suspend an investigation or prosecution for a renewable period of 12 months, based on considerations of international peace and security.
In their response to this apparent questioning of one of the main arguments of their thesis, many legalists tend to interpret the term ‘interests of justice’ in a way that does not allow the proponents of amnesty laws to use it as a legal basis for their arguments. For example, Stahn (2005) and HRW (June 2005, p. 2) argued in 2005 that since the term is not clearly defined in the Statute, the Office of the Prosecutor (OTP) of the ICC ‘should adopt a strict construction of the phrase “interests of justice” in order to adhere to the context of the Rome Statute, its object and purpose, and to the requirements of international law’ (in HRW's words). For them, these things are clearly defined in the preamble of the Statute, which is to combat impunity.
Human Rights Watch goes further in its Policy Paper, inviting the OTP to publish new regulations narrowing the reading of the phrase ‘interests of justice’ to prevent it from being manipulated for political purposes. This notwithstanding, the organisation acknowledges that ‘the only means by which the Rome Statute explicitly permits concerns about peace process to “trump” prosecutorial efforts is through a deferral by the UN Security Council’, as outlined in Article 16 of the Statute. This means that the organisation does not reject this Article, albeit it ‘has concerns about how Article 16 may be applied’, namely the danger of political interference in the judicial process (HRW 2005, pp. 5, 7–8).5
Indeed, in September 2007 the OTP published its Policy Paper on the Interests of Justice in nine pages.6 In this document, the Prosecutor of the ICC states, among other things, that the criteria for the exercise of Article 53 of the Rome Statute ‘will naturally be guided by the objects and purposes of the Statute – namely the prevention of serious crimes of concern to the international community through ending impunity’. He also noted that ‘there is a difference between the concepts of the interests of justice and the interests of peace and that the latter falls within the mandate of institutions [such as the UN Security Council] other than the Office of the Prosecutor’ (p. 1 of the document). These arguments are almost identical to the ones advanced in HRW's Policy Paper. But what seems to follow from these arguments is that the debate about the principle of peace concerns justifying the suspension or halt to ICC's investigations or proceedings is rather a ‘technical’ one, as the question is about who has the authority to ask for the suspension or end of investigation or prosecution, and not about whether the decision can be made.
As noted above, the second argument of the legalists is that amnesty laws are not morally justifiable. Chigara (2002, p. 5), Apuuli (2005, p. 47) and Ssenyonjo (2007, p. 376) argue that national amnesty laws for their part disregard the rights of the victims, as they treat them as if they did not have pre-determined rights at the moment of abuse, and if they did, as if those rights had not been breached at all. For this, they argue that the supposed moral justification that the tenants of amnesty laws claim to have is invalid, for the victims have a right to justice and nothing can justify a state or a mediator denying them this right.
Orentlicher (1991, pp. 2548–2549) suggests that ‘by generally requiring prosecutions, international law helps assure that governments do not forego trials simply because it seems politically expedient to do so’. Based on the assumption that prosecutorial ‘trials secure pre-eminent rights and values’, she insists that governments should ‘assume reasonable risks associated with prosecutions, including a risk of military discontent’. Cherif Bassiouni (1996, pp. 9–28) echoes the same sentiments, arguing that the obligation to render justice to the victims should never be sacrificed for political considerations.
The third argument of the proponents of this approach relates to their concern about how to dissuade parties to the conflict to commit similar crimes in the future with the belief that the amnesty provision covers them from any prosecution. In other words, they argue that amnesty provisions do not offer any guarantee for durable peace, but merely postpone the eruption of discontent, regurgitating old conflicts into the future, and that only prosecution can provide this guarantee (Akhavan 2001, pp. 7–31). The outbreak of the Liberian civil war in 1999 after a peace agreement and even a successful presidential election in 1997 is often cited as an example to illustrate this argument (Penman 2007, p. 11). In the case of Uganda, Ssenyonjo (2007, p. 381) argues that only by enforcing ICC arrest warrants against the indicted leaders of the LRA would the international community send them and generally other warlords around the world a powerful signal that will dissuade them from committing such grave crimes in the future.
However, it should be noted that most of the arguments in this regard are based on situations where there was a clear victor, which is different to the case of Uganda. In particular, the study by Akhavan is focused on such situations in the former Yugoslavia and Rwanda. In fact, it was the failure of the Ugandan Government to defeat the LRA militarily that led to the situation that is the subject of this article. This has led some to argue that the whole story of referring LRA to the ICC was a political strategy by the Ugandan Government to achieve international criminalisation of the group and garner foreign support for its military operations against the rebels (Branch 2007, pp. 182–183, Happold 2007).
The Case for Amnesty as a Last Resort
Like the opponents of amnesty provisions, those in favour of them are many, and so are their arguments. Most of the proponents of this approach claim that their arguments are based on the specific circumstances surrounding the particular case in question, and not a general and blanket argument for any conflict situation. In total, they have three main arguments or counter-arguments. First, they argue that the pursuit of criminal justice should be up to a limit where it will not hamper the peace process or endanger national reconciliation in the country concerned (David and Choi 2006, Kaminski et al. 2006). In a situation where credible evidence shows that prosecuting a suspected war criminal will most probably hamper the peace talks or result in a renewal prolongation of fighting (if an agreement has already been signed), caution must be taken in how to deal with such a situation (Snyder and Vinjamuri 2003, Mandeloff 2004, Williams 2005, Moy 2006, Mallinder 2007).
In fact, Fletcher and Weinstein (2002, p. 585) maintain that there is a ‘paucity of empirical evidence to substantiate claims about how well criminal trials achieve the goals ascribed to them’. In the specific case of Uganda, Branch (2007, pp. 179–198) calls into question the ability of the ICC to achieve either peace or justice.
How do they then respond to the argument of the legalists that this means the denial of justice to the victims of the people concerned? They respond that insisting on a strict application of punitive justice by pursuing the prosecution may lead to a renewal of fighting or, in the Ugandan case at hand, stand in the way of resolving the conflict. Thus, they suggest that opting for measures to avoid a recurrence of armed conflict or to stop it makes more sense because it prevents the creation of future victims (Williams 2005, p. 295, Moy 2006, p. 272, Slaughter and Burke-White 2006, p. 347). Thus, the amnesty clauses are seen here as a ‘necessary’ measure to attain peace and halt further crimes.
Happold (2007) brilliantly, if somewhat cynically, captures these sentiments in the last paragraph of his commentary:
The Court [ICC], it appears, is on the horns of a dilemma. If it is too assertive, the Court risks being rendered ineffective. If it is too deferential, it risks damage to its credibility. Paradoxically, the best result for the Court might be the failure of the peace negotiations.
It is similarly observed that the Prosecutor of the ICC makes a similar point in his second report to the Security Council pursuant to the latter's Resolution 1593 (2005) on Sudan.8 In this report, he noted that in view of establishing whether a prosecution in Sudan is not in the interests of justice, based on the provisions of Article 53 (2)(c) of the Rome Statute, he would consider ‘the various national and international efforts to achieve peace and security, as well as the views of witnesses and victims of the crimes’ (see p. 6 of the report).9
Evidently, the probability of this approach leading to conflict settlement is high. It could be comfortably argued that it was the amnesty or protection provision in the Nigeria-led ECOWAS deal with Charles Taylor in August 2003 that helped in the resolution of Liberia's fratricidal civil war (Souaré 2008a). In addition, amnesty for South Africa's outgoing apartheid regime eased the country's transition to democracy in the early 1990s (Penman 2007, p. 9, Hay 1999).
Far from closing the debate, this argument opens the door to at least two salient questions. First, what is it that the proponents of amnesty laws suggest should happen in the case of Uganda to ensure peace without compromising the credibility of the ICC as an institution set up to combat impunity, given that it has already been involved in this particular case through its indictments and warrants of arrest, as noted above? The other question relates to the preventive role of amnesty provisions for future crimes, as this seems to be a central concern of all those that care about the rights of victims of armed conflicts. What guarantee do amnesty laws offer that warlords may not exploit them to commit similar crimes, assured that the blanket of the amnesty law covers them?
With regard to this question, International Crisis Group argues in its Briefing Paper that:
if a deal has to be done to bring peace to northern Uganda, the ‘least worst’ option might be asylum for the indicted commanders in a country not party to the Rome Statute, conditioned with their full compliance with the peace agreement. (ICC, 13 September 2006).
With regard to the question of the preventive role or lack of it in amnesty provisions, Snyder and Vinjamuri (2003, p. 20) call into doubt the deterrent effect of criminal trials, and argue that if accompanied with effective political backing and strong institutions to enforce their terms, ‘amnesties, whether formal or de facto, can help to pave the way for peace’. Indeed, some authors, who are not necessarily advocates of amnesty provisions, have raised concerns at the prospect of criminal justice or trials that are coming to be seen as if they were the only acceptable means of addressing violations of human rights that occur during armed conflicts (Villa-Vilencio 2000, p. 205, Robinson 2003, p. 483, Lovat 2006, p. 2).
Is There an Alternative? Towards a ‘Third Way’
The debate presented above centres around two main, and equally genuine, preoccupations: (1) ending armed conflicts in order to prevent further killings, human rights violations and destruction of property; and (2) combating impunity to prevent the creation of a cycle of violence in the country, because either some victims, finding themselves in a position of power, might decide to take the law into their own hands, or some perpetrators, knowing that they will not be punished, might be encouraged to do the same in the future. The question is therefore both legal and political – some would say moral.
With regard to the legal dimension of the question, it is clear that both camps presented above have some legal arguments for their positions. With regard to the legalists, it is worth noting that even aside from the Rome Statute of the ICC, no instrument of international law and no national constitution condones the grave crimes that are the subject of this debate, as Chigara (2002, p. 5) correctly notes (see also Happold 2007). However, the arguments of proponents of amnesty provisions are not void of legal basis in international law. It has been seen above how unconvincing are the attempts of some legalists to reinterpret or redefine the Rome Statute in order to deny any legal foundation for political arrangements, as opposed to criminal trials, when peace is at stake.
Even if peace processes were excluded from the provisions of Article 53 of the Rome Statute, the fact that no one disputes the presence of Article 16 means an acceptance of the principle of political arrangements, which are nothing more than some sort of amnesty provision in peace agreements. As Lovat (2006, p. 1) correctly argues, it would be preferable for the exercise of the discretion contained in these clauses of the Rome Statute to remain within the ICC than to render the UN Security Council the sole body capable of deciding whether or not to proceed with an investigation and/or prosecution. In fact, this makes the process even more subject to political manoeuvre, given the nature of the Security Council.
It is true that Article 16 of the Rome Statute only allows the Security Council to call for a suspension of 12 months. This provision has led Stahn (2005, p. 699), Gavron (2002, p. 109) and many others to claim that this Article cannot serve as an instrument to ensure permanent respect for an amnesty law. Their argument is that amnesty laws are usually designed to exempt a group or a class of persons ‘permanently’ from criminal responsibility, while Article 16 is only a temporary device, designed to delay proceedings for a limited period of time. But this argument is questionable, for Article 16 provides for a renewal of the 12-month period without specifying a limit of the number of these renewals. This logically means that it can regularly be renewed to a point where resorting to criminal proceedings against suspected criminals will not make any sense, if this were to happen after, say, two or three decades. These people may by then be dead or have converted themselves into well-respected national and/or international personalities.
A case in point (with regard to unlimited renewals) is the United Nations Disengagement Observer Force (UNDOF), established in 1974 following the agreed disengagement of the Israeli and Syrian forces on the Golan Heights. In October 2008, this force was still stationed in the Middle East with its mandate being renewed every six months. The same is true for the United Nations Interim Force in Lebanon (UNIFIL), created by the Security Council in 1978 to confirm Israeli withdrawal from that country.
However, even if the 12-month period were not renewable, what would be the use of suspending criminal proceedings ‘to allow for a peace process to proceed’ if you exclude from that process political arrangements, including the offer of an amnesty? What would be the use of that arrangement if it would not be respected on a permanent basis, as Stahn and Gavron suggest? Would this not be a flagrant violation of a core principle of international law, namely the principle of pacta sunt servanda enunciated in Article 26 of the 1969 Vienna Convention on the Law of Treaties? This Article stipulates that ‘every treaty in force is binding upon the parties to it and must be performed by them in good faith’.
Following these remarks, it is obvious that amnesty provisions cannot be discarded based purely on legal argument. With regard to customary international law, Macaluso (2001, p. 359) argued that ‘prior to World War I, it was fairly common to insert into a post-war treaty an amnesty provision that would pardon the worst of the offenders’. In fact, it could be argued that this was the case until the campaigns of the globalisation of justice started in the 1990s, and even continued well into the decade. Of course, this is only true when none of the parties to the conflict achieves decisive victory, as seems to be the case in Uganda.
Kaminski et al. (2006, p. 297), for example, wonder whether the 1989 roundtable talks in Poland would have worked if the two main negotiators for the communist regime had had reason to worry that once they stepped down they would be prosecuted for human rights abuses committed during the martial law.
Similarly, in the United Kingdom, the April 1998 Good Friday Agreement signed between the British and Irish Governments and the parties to the 30-year conflict in Northern Ireland had strong human rights provisions in it. However, although they were not formally described as an amnesty, the arrangements regarding the early release of paramilitary prisoners, which were a key aspect of the agreement, in practice operated in this way. Prisoners who had been convicted of scheduled offences (offences under UK anti-terrorism and emergency legislation) were released, subject to a review process outside the normal realms of sentence review (Aoláin 2005). The only category of prisoners that did not benefit from this measure was those ‘affiliated to organisations which have not established or are not maintaining a complete and unequivocal ceasefire’.
There is also another problem with the Rome Statute that even makes insisting on the ICC process in the specific case of Uganda quite problematic. Given that the conflict in Northern Uganda reaches back to 1986 and much of the most atrocious violence took place prior to 2002, ICC's temporal jurisdiction, otherwise known as ratione temporis, makes the Court a highly inappropriate vehicle for rendering justice to the majority of victims of these crimes (Branch 2007, p. 186).
All this notwithstanding, it must be acknowledged that international law, both customary and humanitarian, is undergoing significant changes with regard to crimes committed during armed conflicts, as noted in the introduction of this article. But while these are noble efforts in principle, because they aim to combat impunity, there could be justification in wondering whether the absolutist approach supported by some advocates of these changes is realistic, given that the reality of many armed conflicts may militate against it. If one can rightly reproach the proponents of amnesty provisions for the lack of preventive guarantees in their position, it will be hard to find the example of an ‘armed conflict ended by negotiations’, which did not recur ‘solely because of criminal trials’ for those suspected of committing crimes during that conflict.
It is true that armed conflicts that end through negotiated settlements are more likely to break out than those ended by the victory of one party over the others. However, this is simply because in the latter scenario the loser's organisation is destroyed, making it very difficult to resume the war, while in the first scenario that is the subject of this article, segments of the post-conflict structure usually retain the capacity for resorting to war (Licklider 1995, pp. 681–690).
What follows from this discussion is the necessity to take the debate from its current form and consider alternatives, based on the recognition that the two preoccupations identified above are not mutually exclusive. To do this, it is necessary to acknowledge that ‘justice’ cannot be reduced to only one form, namely criminal or punitive justice. It must be recognised that there are other forms of justice to which one could resort to address violations that occurred during the armed conflict and, by that, combat impunity (Henham 2007, pp. 449–468). In other words, alongside civil and political rights, there are economic, social and cultural rights. After all, justice is a process that seeks to repair violated ‘rights’; yet, it is often all these rights that are violated during armed conflicts.
Once this is recognised, one must establish how to prioritise these rights in addressing those that were violated. Here, without falling into excessive relativism, heed must be paid to ‘perceptions’ of the people concerned about the concept of justice. It must also be established to which of their different violated rights they would themselves give priority in specific circumstances such as in Uganda.
This shows the importance of surveys such as the one conducted by researchers from the Human Rights Center (HRC) of the University of California, Berkeley, and the International Center for Transitional Justice (ICTJ) in northern Uganda in March and May 2005. The survey was carried out in the areas most affected by LRA violence, with a view to understanding the position and perceptions of the people most affected regarding the different issues, particularly the amnesty offer. Of the 2585 respondents surveyed, 40 per cent had been abducted by the LRA, 31 per cent had had a child abducted, and 58 per cent had witnessed a child being abducted. Almost half (45 per cent) had witnessed a member of their family being killed. Approximately one-quarter (23 per cent) of the respondents had been physically mutilated, maimed or injured by the LRA at some point during the conflict (Pham et al. 2005, p. 21). According to the report, respondents were asked to identify their immediate concerns and, if they listed more than one, to rank them. ‘Food and peace’ were the top two main concerns mentioned (34 per cent and 31 per cent respectively). Less than 1 per cent mentioned criminal justice as their most immediate concern (p. 25).
Interestingly, the survey found that in Soroti, one of the villages most hit by LRA violence, 80 per cent responded affirmatively when asked whether those who committed human rights violations should receive amnesty. Sixty-nine per cent of respondents in Gulu, 57 per cent in Kitgum and 53 per cent in Lira supported amnesty for past offenders. Most importantly, respondents were also asked, ‘If the only road to peace was amnesty, would you accept that?’ The overall support for amnesty increased slightly to 71 per cent from 65 per cent (Pham et al. 2005, pp. 28–29).
It would seem from this survey that personal security and the restoration of economic rights are more important to the victims of the LRA than political rights or criminal trials for those that wronged them. Allen (2006, pp. 78–88, 102–109) and Southwick (2005) cite many examples that prove the strong support that the Amnesty Act has among people in Acholiland and the fears that they have about the impact of the ICC's involvement in Uganda, although Allen (2006, pp. 114–117) tries rather unconvincingly to prove that this support is waning. The argument that ICC indictments for LRA leaders played an important role in bringing the LRA to the negotiating table has been acknowledged above. However, to argue that by simply succeeding in bringing the LRA to the negotiating table means that these fears are not founded (Allen 2006, pp. 115–117, 2007) is to assign a totally new mission to the Court – that of inducing the revival of stalled peace processes, rather than investigating alleged crimes, trying the suspects and punishing those found guilty, as can be seen in Figure 1.
Therefore, it could be argued that in order to solve the conflict in northern Uganda, the government in Kampala should be allowed to pass its amnesty law for LRA leaders, and the ICC should revoke its indictment and warrants of arrest against these leaders. Both the government of Uganda and those members of the international community that are ready to fund the very expensive criminal proceedings of these indicted leaders should help fund projects aimed at addressing the victims' economic rights that they consider more important.
The peace agreement should also make it clear that any violation of its terms will mean the nullification of the amnesty provisions for those responsible for that breach, in conformity with Article 60 of the Vienna Convention on the Law of Treaties. Finally, use must be made of local justice mechanisms in the country, such as Mato oput and Nnyono tong gweno in view of addressing other violations that might not be covered by solely attending to issues of economic rights. (For a detailed description and historical background of these local justice mechanisms in Uganda, see Baines, 2007, pp. 91–114, Latigo 2008, pp. 85–119.) Efforts should also be made to address the shortcomings of these local justice mechanisms so that they are compatible with the magnitude of crimes committed. Indeed, the Prosecutor of the ICC appeared to acknowledge the need for such an approach in his keynote address to a conference on ‘International Criminal Justice in the 21st Century’ hosted by the American University in Washington in September 2005 (see Moreno-Ocampo 2005).
Conclusion
This article looked at the dilemma of how to achieve peace and ensure justice for the victims of armed conflicts, making sure that neither of these two objectives undermines the other. The conflict in northern Uganda and its stalled peace process was used as a case study. It noted that the debate about this issue is dominated by a dichotomic (amnesty vs. criminal justice) approach with two main camps. On the one hand, there are those that are opposed to any kind of political arrangements that might expunge suspected war criminals from any responsibility. On the other hand, it found there are those that justify such arrangements as a last recourse. It noted that both camps have strong legal and moral arguments and have genuine preoccupations.
For this reason, the article called for a ‘third way’. This alternative solution should be based partly on the acknowledgement that ‘justice’ cannot be reduced to only one form of it, namely criminal or punitive justice, and that there are other forms of justice to which one could resort to address violations that occurred during the armed conflict. With that in mind, it may be possible to combat impunity without necessarily compromising peace or hampering its attainment, for example by knowing which of their many violated rights are more important to LRA victims in Uganda and what use can be made of local justice systems in addition to addressing the violated rights that they consider more important.
The article used the ICC's own statute and a survey conducted by researchers from the HRC (University of California, Berkeley), and the ICTJ in 2005. Based on this, it argued that in order to solve the conflict in northern Uganda, the ICC should revoke its indictment and warrants of arrest against LRA leaders, since that is the main problem in the peace process. The peace agreement might include some sort of amnesty for these leaders, but make it clear that any violation of its terms would mean the nullification of the amnesty provisions for those responsible for that breach. This will make it different from peace agreements such as the Lomé Accord that included an amnesty provision for warlord Foday Sankoh in Sierra Leone, but was repeatedly violated by him and his combatants, eventually leading to a renewal of conflict. Use must then be made of local justice mechanisms to address the criminal charges against the indicted LRA leaders. Finally, both the government of Uganda and those members of the international community that are ready to fund the very expensive criminal proceedings of the indicted leaders should help to fund projects aimed at addressing the victims' economic rights that they consider most important.
In the final analysis, and as Robinson (2003, p. 484) correctly observes, while the ICC and all those concerned with combating impunity for crimes committed during armed conflicts should remain firm on this, one should keep an open mind to creative and good faith alternatives to criminal prosecution. There is no inherent contradiction between the objectives of the ICC and these good faith alternatives, such as bona fide truth commissions.